Amiga.org
Amiga computer related discussion => Amiga Hardware Issues and discussion => Topic started by: Xenepp on July 25, 2007, 02:54:38 PM
-
To save bogging down the announcment thread, I've created this one to get an idea of what the majority of people want from the PCB.
I'm not sure about how many will be made yet, I guess if you post here that your interested it'll give me an idea.
I need to translate the PCB to a format my guy in china can work with so while I'm at it, I may as well configure it they way everyone wants.
There's been talk of making the ports (joystick, power etc) modular, i.e. on ribbon cables so they can be mounted however you like in a custom case, how does everyone feel about that?
I'm open to other suggestions like shape of the PCB for mounting in HDD cases (if possable). Just throw around some ideas and I'll amend this first post to collect them in a list or sorts.
A VERY rough idea of cost at this point is about £40 for an assembled PCB and £20 for the kit (PCB plus parts to solder yourself). I'd imagine assembled PCBs would be however be pretty limited ;)
Joel
-
I want an assembled PCB. :-)
However, I don't have strong opinions about port layout, PCB shape etc.
-
I'm interested, to have one PCB :P
-
I am interested too...
A smallest possible case, completely assembled with ports soldered on to the PCB. Is in my opinion the best solution.
(similar to to the Efika)
-
I want one two ;-)
I would like the modular design if i may.
Regards
-
I'll take one board, unpopulated is fine.
The original design is fine, but of course if you make it into something standard like mini-itx, then that'd make finding a case easier.
-
So assembeled as in a fully functioning minimig for £40 ?
sounds brill, count me in
-
A Minimig in an original Amiga 1000 Casing would be cool but then it comes to my mind,that there aren't enough of them for a production run so only a professional Chinese Copy of this would work:)
Also a MacMini like design would be fun.
-
You'll never make an assembled one for £40 not without a huge production run.
I have my hat here... and anyone who knows me will know I am capable of eating it.
-
Just so we're all on the same track, I can only produce ONE version of the PCB. If I were to make two, the cost would double, so that's why I wanted to see what everyone thought first, then go with that.
On a related topic, I'm considering designing a case for the Minimig in CAD and have another guy I know fabricate it so I can then mold it myself and cast it in resin. This way it'd fit perfectly and look however we like ;)
What does everyone think about that idea?
Joel
-
@alexh
Go buy some tomato sauce,mustard and mayonaise for your hat meal:)
It's perfectly possible to produce it at such a low cost because China is cheap and the Chinese workers are nearly perfect assemblers (for the little money they get).
-
Because I'd be into experimentation, I'd be looking for a board with full 24-bit RGB out (so I can have a go at an AGA implementation), an IDE port, 68020+ processor, and possibly a CPU bus. Basically, the A1200 equivalent of the Minimig (minus clockports and other similar stuff. That would be a fair amount of work though.
For those who just want a working Minimig, Dennis' current design should be fine.
Hans
-
I'd like just a working board, no case or anything else... Just a fully populated and working minimig board...
-
If that would be the price, I could take one assembled one.
And one idea for those interested in a bit more custom design:
How about an option for board without any connectors soldered? (and maybe also supplied without them) They are the easiest parts to solder anyway.
That way those wanting to have them on external board wouldn't have the job of removing them from the board first. (or having useless second set of connectors)
Also, any idea for "board only, I buy my components locally" -price?
-
assembled, A1000 type case
-
I'm in for two boards
-
I think those prices are excellent!
I promise to buy one of the kits for £20 as soon as they come out. I enjoy soldering and don't do it much these days so look forward to building the kit.
As far as the ports go, I think its easiest just to have them on the PCB. People can easily buy/make little extension cables for their own custom cases.
-Chris
-
I'ld like to know more about the seller and preferably some reference from more long time users that you're a real and solid person before I order anything.
40 UKP (80$ ?) is a great price for an assembled unit, however I don't see any possibility that you can pull it off at that price. For how much do you think you'll get a spartan3 chip ? Add to that the 68k, PIC and ram. I think you'll need to double your prices for both the assembled and kit.
-
I'll be able to get a much better idea about prices once I can show the design to my guy, but I can tell you the estimated £20 would not include parts, other than posts and simple bits, it would just be the PCB. This is another thing that will influence cost of the preassembled PCBs since I've not looked into cost of components (nor avaliability).
You'll never make an assembled one for £40 not without a huge production run.
I have my hat here... and anyone who knows me will know I am capable of eating it.
This is why I was hesitant to give a fixed price. The PCBs would however be assembled by me, which is why those would be pretty limited.
Also, depending on the hat, might I suggest burger relish? Really goes well with tweed. :lol:
Joel
-
It's the upfront cost which is the thing putting me off, to make say 200 boards, populate them, quality test them, ship them.
You are looking at £10,000 up front :-(
That is also assuming that the Minimig sources that have been released are actually functional (Which I will be trying to prove this weekend.)
-
The XC3S400-4PQG208C alone is about $22 in 100+ quantities.
-
£10,000?? Where are you getting your boards done?!
My guy does my boards as small runs, so costs nothing like that. I estimate £20 for a bare unpopulated PCB based on previous runs I've done.
I think it may be a good idea to move the ports to the back of the PCB with the MMC slot on the side or front. This way, people can decide if they want to have built on ports or they can just wire them to the PCB and put them where they want in their case.
So what does everyone think about the size of the PCB? If I moved the ports to the back I might have to make the board wider than it is tall to properly route the traces so I suppose it really depents on port placment prefrence.
Joel
-
Hows about making it easy to fit into a Mini-ITX case? I'm not saying easy as in what you'd need to do to mount an Efika in a Mini-ITX as I don't want to be drilling new holes and crap like that, but something with screw holes in the correct places on the Minimig PCB, connectors at the back, add standard power supply connector if needed, etc. to minimize our own personal case fabrication requirements. I'd also put the MMC slot on the back panel for easy access instead of tearing apart the case to swap if you ever need/want to.
-
I definitely want one, or possibly more. I would be prepared to pay up-front to aid costs for the production run, provided I was guaranteed a refund were there not enough interest to warrant a production run.
--
moto
-
I am definitely interested in a populated board!
Look into making it Mini-ITX size, 17x17 cm. There are lots of cases available for that motherboard size. Making your own case should not be priority at this point I think, more important is getting the motherboard out!
Also putting pin-headers for joystick and other connectors is a great idea! Keep us informed. :-)
-
I'm interested in this concept too.
Rude!
-
I'd be intereseted in a couple of boards.
As for layout, ports on the back would be great and things like the sd slot on ribbon would be good to, as long as the board width is less than 5.5" I would be happy, then it could be built into a external disc case like Mavmiga and myself did with his Efika. Here :- http://bbrv.blogspot.com/2007/07/macmigaefika.html
Gaz
http://www.lincsamiga.org.uk
-
Sign me up for one, maybe two assembled boards. Ribbon cables or not, I'm in.
-
It looks like there is enough interest to do it, but I'll not be taking any payments until I have the PCBs in hand. It's far less complicated that way.
Once I have the design done, the price is set and everyone's happy I'll start making a list of those who wants one. Obviously, the more that want them, the cheaper it will end up being.
It's a good idea I think to design it to fit standard Mini-ITX cases. This way, people have a lot of options with how they'll case it up and I can make an Amiga type case to Mini-ITX specs, giving me a firm guideline. Nano-ITX is too small to route traces for all those connectors.
Thing only drawback is that it'll get bigger. The beauty of Minimig is that it's so small but then it doesn't lend itself well to casing with ports everywhere.
So what does everyone think about that?
Joel
-
I'd be in for an assembled one for 40 quid.
-
You can add me to the list of those interested.
-
You can count me in as well. I like the idea of the ports on ribbons but whichever design you choose i would be still interested.
Paul.
-
Hi.
Sign me up for one. Assaembled please. :)
BuzzBrain
-
cv643d wrote:
Look into making it Mini-ITX size, 17x17 cm. There are lots of cases available for that motherboard size.
Also putting pin-headers for joystick and other connectors is a great idea! Keep us informed. :-)
I take full credit for mayking the headers available!! :-)
But why make the board BIGGER (12cmx12cm) to make it mini-ITX. Why not reshape it to 8.8cm X 16.5cm and then the board would fit in any 3.5" external HD case. They all come with PSUs and are cheap too. If the connectors are headers its just a case of squeezing the sockets into the backpanel!
Anyway these are re-posts of my earlier ideas from comments on the News ite,m
-
Xenepp wrote:
£10,000?? Where are you getting your boards done?!
My guy does my boards as small runs, so costs nothing like that. I estimate £20 for a bare unpopulated PCB based on previous runs I've done.
200 * 20 = £4000 and that is before you've paid tooling costs, started buying parts, assembling, testing etc.
To make 200 boards will cost about £10,000 upfront.
Make less than 200 and I think the price per unit will go up beyond most peoples price point.
-
Sign me up.
How small are the solder points on the PCB likely to be? If it's reasonable I'd take one unassembled.
-
Too small for all but the most skilled hand solderer. The pitch on the FPGA is very small and there are a lot of pins.
-
I'm in for 2 put together.
-Tig
-
edit, just read the open source news....
cool, I would like an assembled one please
-
I'll be up for one fully assembled.
-
I'll take 2 fully assembled please!!!
-
I would be willing to pay for 2 assembled MiniMigs :D !!
-
I'd be up for a fully assembled one.
-
For those that want a custom case you can model one with Rhino, Solidworks or most any solid modeler and send an .stl file to a rapid prototyping service.
3d art to part is an online firm that provies low cost printouts. Action figures and collectible type statues run $65-$100, So a small custom case is possible.
The final output is rough and must be sanded down for a fine finish. You can then paint the casework to your liking or send it out for molding. For limited runs low cost silicon molds can run off a $100 or so units. Resin catsting prices vary...Steel molds are required for more numbers but those molds cost tens of thousands of dolllars.
But this is all possible...
A far cry from just 5 years ago...
-
Count me in too, I'd like a board that is at least partially assembled (SMD components at a minimum). I like the idea of making the board miniITX sized, but it's not that important to me because I'll probably make a custom case for it.
If you sell unassembled kits, it's best to include all the parts because buying parts in small quantities is expensive. If you buy them in large quantities it will save money for all the buyers.
-
Ditto, small is beautiful, who needs a big case? The smallest case possible please.
-
Yeah I`m interested. Price dependant of course.
-
Ok, I'm starting work on this tomorrow morning so from here we need votes. The choice is between:
1. Mini-ITX standard to fit those cases (17cmX17cm)
2. Original Minimig design (12cmX12cm)
Whichever has the most votes by the time I check this thread in the morning is what I'll go with.
Personally, I vote for the original design. I know it's non-standard but it's nice and compact. Cost will also be lower due to less PCB area.
Voting starts from here!
Joel
-
I'm interested too, but it depends on offer and price.
If the 2xRAM, FPGA, CPU is pre-assembled most of the pain is avoided.
As for size, I think a 3,5" size is the best due it fits in a drive slot both in pc-boxes and external chassi for hdd.
The only reason to go for 12x12 is that you don't need to modify the design.
I think however the design could use an onboard regulator for the +5VDC & capacitor to avoid to issues with the power supply. A 5V supply won't provide 5V after a sufficient length of wire.
As for buildin or use on-pcb-connectors. Why not replace the connectors with IDC connectors that is wired to a breakout pcb via flatcables in any manner desired ..?
-
Based on the comments, I'd definately take an assembled one then, and the mini-ITX form would be best.
-
Put me down for 1, mini-itx standard, fully assembled.
-
Ugh, I think its better to follow a standard. Just look at Efika, do we really want Minimigs built into scanners, external 3.5 enclosures etc... it does not look professional I think.
Actually Mini-ITX is 17x17 cm and Nano-ITX is 12x12 cm. But Nano-ITX cases are more expensive than Mini-ITX ones I think.
-
I'm kind of a fan of the ribbon cables. It would make custom cases easier to make. I would expect that we will see lots of custom cases for the mini-mig. With how small the board is, I would think that it would be trivial to make an adapter plate for anyone wanting to mount the mini-mig in a mini-atx case. Personally, I would like to put the mini-mig into an old gutted laptop.
-
the Amiga was for people with imaginations, a system that the user could configure on their own to represent them. "Think Different" rings a bell?
mini-itx.
-
I personaly think the original 12x12 size would be the better formfactor. I would like to put it in a Amiga500 or 600 Case.
-
I would definitely be interested in a preassembled board. Not too bothered about how the ports connect (soldered on or ribbon cables.)
-
I have been lurking here ever since I found out about the Minimig progect and what can I say? Hurrah and about time!
You can put me down for 1 semi-assembled unit (all smd stuff done with the rest loose). Or if its all or nothing then fully assembled with the ports left out please.
I cast my vote for the original size board with pin headers for the ports. Im sure I can find a junk digibox at work to fit it into so it can sit on my desk with everything else.
As for the price, if you can keep the price for a assembled unit below the cost of the NES PowerPak (http://www.retrousb.com/index.php?productID=133)(another recenly long awaited retor FPGA project) then I will be a happy man!
-
12x12, anything bar that HUGE 17x17cm monstrosity suggested!
-
I would take one *S*
-
May I give some suggestions?
If you do a new board design please change the following from the orginal design:
1) Use a single chip ram like this (http://www.cypress.com/portal/server.pt?space=CommunityPage&control=SetCommunity&CommunityID=209&PageID=259&fid=37&rpn=CY62167DV30&ref=sch) one, available at digikey (http://www.digikey.com/scripts/DkSearch/dksus.dll?Detail?name=428-1860-ND) for example.
2) Include a +5V regulator on board OR use an ATX-style powerplug if you go Mini-ITX. The +5V regulator should be capable of providing about 1A to safely power all peripherals one might attach.
3) Hardwire the patch needed to get the current board to run. Alternatively, you can also swap pin81 and pin 79 on the FPGA. This way you still have four user-IO's left. You do need to change the .UCF file though and recompile the core.
4) The MMC card interface has a resistor based clock gate circuit around R50,R51. This should be replaced with a proper (single) gate "OR" chip. The margins on this signal are pretty tight on the current board. Also, R49 should be 0 ohm ideally to avoid problems when upgrading the PIC to a newer PIC18LF2620 or something similair. Margins are tight on that signal too atm.
And you could do the following optional improvements:
5) You could consider using a single 20MHz oscillator and figure out how to program the DCM to generator the proper clocks, this would spare a crystal.
Have fun!
PS Offcourse I would like one too! :-)
-
I'd like an original design fully assembled PCB.
(I'm also fine with the mini-ITX size if that's the resulting choice)
-
I think about ordering one but then it would be fun to build one myself.
BTW: What I would prefer over this both options would be an original board created by Dennis himself if this is somehow possible:)
-
Belial6 wrote:
I'm kind of a fan of the ribbon cables. It would make custom cases easier to make. I would expect that we will see lots of custom cases for the mini-mig. With how small the board is, I would think that it would be trivial to make an adapter plate for anyone wanting to mount the mini-mig in a mini-atx case. Personally, I would like to put the mini-mig into an old gutted laptop.
Maybe there is a standard for those 9pin RS232 serial connector brackets often found on pcs? ..would be really neat to be able to just plug in :-)
As for adapter any plate with a 11x11 cm square hole should allow mounting in MiniATX, problem solved. It's not problem making that hole fit a 3.5" (10,2 x 14,7 cm) either.
Is there any standard formfactor of laptop interiors?
Maybe you could just rewire the digital RGB before the resistor A/D and vsync/hsync to the builtin TFT, and let some D/A port on the MCU control the brightness and similar ..?
-
I'd like a fully assembled one.
I also prefer using standards. So I vote for Mini-ITX.
-
/me bows to Dennis, the custom chip master :-)
-
I'm interested in two to four unpopulated boards.
It would also be nice if you could sell components, as the parts are much cheaper if bought in bulk.
Having headers on the board instead of the connectors is a good idea, but I don't like the idea of the mini-ITX form factor. Headers would not take up much more space than the original connectors, probably even less, so it should still be possible to make it 12x12. 17x17 is just over twice as large, or twice as large as it needs to be in other words. Also, I suspect the boards might get much more expensive if they were made for the mini-ITX form factor, seeing as how you usually pay for the area when you order circuit boards. Having it fit in cabinets made for mini-ITX would be practical, but think of the drawbacks, much more expensive boards, and harder to mount them in smaller cases. At 12x12 they could probably still be mounted in mini-ITX cases without too much trouble, and they will also fit in nano-ITX cases.
Anders M.
-
Two assembled ones for me :-)
I really dont see the point of using mini-ITX form factor at this point - people who insist can mount it on a mini-ITX sized card board or something (sheesh!)
The smaller the better is my like, and as someone else here suggested, it might be interesting to build it into for example an original A600 casing. Uhm.. is there a way to connect original A1200/A600 keyboards? :-)
-
I am interested in one and it doesn't matter if its prepopulated board or not.
-
Two for me, the smaller the better is my preference too.
I hope people don't gut working A600's and A1200's to put a Minimig inside???
Will the proposed PCB run incorporate the changes/upgrades that Dennis has suggested? I hope so.
Thanks for starting this thread.
-
Whatever the Minimig is (no I didn't read the whole threads but yes I have a vague idea), I am interested if
1. It can do at least what a basic A500 does (no doubt about that obviously) ;
2. There's no soldering to do ;
3. You can supply it with power and install it in a standard PC case.
If the PCB is plug-and-play - already installed in a case - it is even better.
Edit : Of course please don't sacrifice any working Amiga - or I'm not interested anymore !
Good luck !
-
I want an assembled PCB.
Me too!
TrevorDick
-
12 x 12 would be ideal, all it really needs is about 6 USB ports and a video port or some kind, composite would be nice, then yo could plug it into an iPod like device (Cowon, Archos) hookup a projected keyboard and you could carry the whole thing in an winter jacket pocket!
-
As much as I love custom hardware, I think the standard mini itx layout with Dennis' suggested design impovements would be best in the long run. That way you can use comercial cases or build your own design with only one run.
Mini ITX would be my choice.
Plaz
-
I thought about this for a bit, and have decided that I like the 12x12 board size better. I'd prefer to buy a board with some improvments made, the improvements suggested by Dennis, as well as pin headers for the serial port and joystick ports. The video, power, kb, and mouse ports should probably be on board at the back (short wires = less noise).
Would it be possible for the final schematic to be posted if and when the big order is placed? If some of us wants more boards in the future we can have them made by Futurlec (http://www.futurlec.com/PCBService.shtml) or some similar company.
EDIT: forgot to mention, that having a SO-DIMM socket (or similar) in place of the onboard ram would be awesome. I know I've got a 32MB SO-DIMM around here somewhere.
-
I´ll take one fully assembled.
-
Considering:
* We pay per square area.
* Larger boards means it will fit in less boxes.
* Dennis essentially contributed an pcb errata that needs to be fixed.
* 3,5" drive space is just slightly larger than the 12x12cm.
Some wants to build it into specific machines and others want it standalone. IDC 2.54mm spaced pin headers, with flat cable connectors solve that.
And while at it, let those pin headers be compatible with pc IDC <-> DB9.
IDC DB9 http://pinouts.ru/SerialPorts/Serial9_pinout.shtml
CDrom<->Soundcard http://pinouts.ru/Multimedia/analog_creative_pinout.shtml
Why not modify minimig v1 with the errata and fit it into a 3,5" sized pcb (15.2 x 10.2 cm) ..?
-
I will buy one, I would prefer to assemble it myself but I don't think I have the required skill level.
Tell you what, I will have one of each, one to use and one to screw up with my soldering iron.... :idea:
Keep The Faith!
-
Hi there!
I'm very interested in purchasing a Minimig so please count me in!
Have you considered making it small enough to fit in a 5.25" drive bay? This way the Minimig could easily be integrated in a standard PC or Amiga One case and could drain power from the existing PSU. I'm thinking of something along the lines of the Access clone.
Also maybe the expansion ports could be facing outwards? I'm not entirely sure how that would work but i could draw a quick 3d model of the idea and post it here if other people find this idea appealing.
Just my £0.02 :-)
-
a couple of boards here too, pleeze! ;)
-
Count me in for a fully assembled Minimig.
When it comes to 'fully assembled': a friend of mine is the proud owner of one of the One chip MSX-machines. Nice case, sturdy, compact. You can see it overhere (http://www.bazix.nl/onechipmsx.html).
Maybe something to think about and make a comparable casing?
-
@Dennis
What software did you use to Route and produce the Gerber files ?
Maybe update the Gerber files on your website with the Board fix (If not too difficult), but yeah, at least if we are going to do a production run for A.Org, DEFINITELY include the fix into the Gerber files before submission to the Board manufacturer.
I am VERY interested, I just want to sit back for a while and see what mods are done, and costs etc. (Me being in Australia and all).
-
the One chip MSX-machines [snip] Maybe something to think about and make a comparable casing?
At €229 euro's I think it is a TAD expensive ;-)
-
You can count me in for at least one assembled and one kit.
-
dbrads wrote:
I will buy one, I would prefer to assemble it myself but I don't think I have the required skill level.
Tell you what, I will have one of each, one to use and one to screw up with my soldering iron.... :idea:
Keep The Faith!
I'll do the same if it turns out cheap enough. if not then one fully assembled.
The smaller the better for me. You can always make it bigger by sticking it to some kind of mounting plate for bigger cases, but you can't make it any smaller!
-
Hi,
At 40 quid a pop, if this really materializes, I'd take 5 - 10 fully assembled units. The pin header->DB9 idea seems pretty solid to me.
If anyone can refresh my memory: does Denis' design include standard VGA out?
-(e)
-
sweetlilmr wrote:
Hi,
At 40 quid a pop, if this really materializes, I'd take 5 - 10 fully assembled units. The pin header->DB9 idea seems pretty solid to me.
If anyone can refresh my memory: does Denis' design include standard VGA out?
-(e)
15/31 Khz (Jumperable Scandoubler) @ 50 Hz.
So you will need a VGA display that supports 50hz (That counts out most LCD panels). And explains why i have great difficulty getting LCD displays to work properly on my 4000's Arxon Scandoubler.
-
With an A4000 you can just select NTSC and have a 60Hz output (and with WHDLoad you can force it in games too). That will work with most LCD monitors.
MiniMig cannot (yet) switch to a 60Hz screen mode.
-
I'm another guaranteed customer for one of these. I'd definitely prefer pre-assembled and 12x12 (as small as possible is best), but can handle the kit form and larger format if that's what ends up being available.
I'd also be very happy to buy a custom case, if one is made. Otherwise I'll make my own.
I'm not sure if these have a serial port on them, but if they do then I'd HAVE to get two, just for 4 player Lotus II! :-D
Edit: Checked Dennis's Minimig site, it does appear to have a serial port YAY!
-
I want one too! Fully assembled, if not too expensive.
-
Oeh! I want one too! Actually, i want 2. One assembled and one unassembled.
Here's an idea: How about you put a RAM slot on the board (SD-RAM or so), so we can have extra FAST or CHIP ram? :idea:
That would be VERY nice :)
Don't know if it's possible though...
Edit: would it be possible to "upload" new "firmware" to the FPGA? For example, an AGA chipset? Maybe through the serial port?
-
Where do I send the money? :)
-
Edit: would it be possible to "upload" new "firmware" to the FPGA? Maybe through the serial port?
Erm, you put the FPGA image on the MMC card. I thought everyone knew that?
For example, an AGA chipset?
You wont be able to update to AGA though. Not with the original MiniMig PCB.
o No 32-bit processor (No real point in AGA without an 020+).
o 16-bit RAM (need 32-bit for AGA).
o 12-bit R2R ladder / Video DAC (need 24-bit for AGA)
You might be able to work around the RAM thing if you ran the RAM interface at 2x speed. (Dunno if that is possible, probably have to run everything at 2x speed and divide down inside FPGA).
I guess you could ignore the lack of colours on the screen. (I've not looked that close at the PCB design, perhaps there are more than 12-bits, OCS/ECS only needs 12-bits but you never know.)
But there is no getting round the CPU bit.
-
I'd be interested in buying one!
-
12x12 cm = Nano-ITX, so people can buy ready-made Nano-ITX cases, and probably fit them in mini-ITX cases as well.
This size will also fit in a 5,25" drive bay or 5,25" external drive case.
My wishes:
- IDC pin headers for the ports.
- Fix the errors Dennis mentioned.
Would there be any way of making the VGA port output composite sync? If so, that enables the usage of a RGB SCART cable for TV hookup.
Count me in (assembled), as long as i can stick it on a TV.
-Paul
-
alexh wrote:
Erm, you put the FPGA image on the MMC card. I thought everyone knew that?
I didn't...
alexh wrote:
You wont be able to update to AGA though. Not with the original MiniMig PCB.
o No 32-bit processor (No real point in AGA without an 020+).
o 16-bit RAM (need 32-bit for AGA).
o 12-bit R2R ladder / Video DAC (need 24-bit for AGA)
You might be able to work around the RAM thing if you ran the RAM interface at 2x speed. (Dunno if that is possible, probably have to run everything at 2x speed and divide down inside FPGA).
I guess you could ignore the lack of colours on the screen. (I've not looked that close at the PCB design, perhaps there are more than 12-bits, OCS/ECS only needs 12-bits but you never know.)
But there is no getting round the CPU bit.
Hmm... Ahh well, can't have everything in the world, can you? ;)
-
Hmm I'll settle for whatever board size wins. Don't really care about the size. 12x12 is fine, but if not, then something else.. Also I don't care how the connectors will be, so I'll go with whatever majority decides.
Or well the VGA connector should probably be mounted directly on the PCB, as the image quality suffers quite quickly if it's tethered on a ribbon connector.
-
Hi !
I would like to have a fully assembled one, depending on the price...
-
I will take one in any format i can get my hands on. Cheers
-
I am interested in two to four unpopulated boards.
:-)
-
I want one please, assembled, and preferably in a nice case! I dont want to solder anything.
One question: will the picture output be via SCART/RGB or VGA?
Thanks
-
Holy Tobasco! This is the most exciting Amiga news in years.
Count me in. Fully assembled, 12 x 12 would be better (so long as someone comes up with a custom case that we can buy as an extra). Otherwise mini ITX might be better.
I think it's fair to say that you'll get an initial run of 100 of these things, no problem.
As a suggestion, don't under-price them. Figure out how long it takes you to assemble and do basic tests on a unit, and then price accordingly. You'll need to be motivated to keep going!
-
As long as there is no soldering for me to do, I'll go with whatever.
-
Yep, the 12 x 12cm format sounds perfect. That way it can fit in any size case you want.
The option of the external connectors (VGA, Joystick, etc) on 'stand-off' ribbons cables is also a good idea, since then it would make it easier to fit nicely into larger cases.
But you want to keep it simple (like the EFIKA).
Is there any way we can have an online poll to make it easier to see what people would prefer (rather than wading through these replies)?
Also ready-assembled is much better. I don't see the point of a bare PCB (for the majority of potential customers...)
Maybe you could get the bloke in China to do the whole lot (PCB, assembly, etc.)? I'd say the current projected price is more than a bargain. If assembly of a complete product in China bumps the price up to say $US 100 I'm sure you'd still have LOADS of buyers.
Cheers,
Mike.
-
*raises hand* Me too! Me too! :-D
I'd love to get my hands on an assembled board.
Now for my two cents :-)
1) I'd like to see a mini-itx version.
While I understand that there are case modders out there that would love to get there hands on a smaller board (smaller board = bigger possibilities) I think for the rest of us out there it would just be a lot easier if we stick with standards.
2) Please incorporate Dennis' improvements if you can. If you can't no big deal :-)
I'm sure you're going to be hearing all sorts of idea's for improving on the design (in fact I guarantee it :roll:) and while I am all for the idea of making the board AGA compatible in the future I think for now it's best to keep things simple and maybe just incorporate the fixes Dennis spoke of (if possible).
3) Custom case would be cool but focus on one thing at a time.
I for one have been dieing to get a new A1000 case molded (one that doesn't yellow with age :P) but alas I think it's better that you focus on this project for now.
Though in the future I hope you'll consider it (or maybe even size reduced A1000 case) :-)
Side Note:
Now am I mistaken or does the minimig already supported VGA out :-? This is defiantly a must have for me.
-
"swap pin81 and pin 79 on the FPGA. This way you still have four user-IO's left. You do need to change the .UCF file though and recompile the core."
This means non-patched and patched boards won't be compatible. Please let's at a minimum have this patch incorperated. Or we might end with two fpga binaries.. (for no good reason).
The other fixes doesn't seem to affect the plain software aspect of the board.
-
count me in for 2 compleet boards.
-
Well it looks obvious to me that the majority are looking for the original board size of 12x12, which makes sense. The whole point of Minimig afterall is it's a titchy Amiga!
Is everyone in agreement about the pin headers? It would be the best option for case fitting plus allow people to decide what ports they really need.
Dennis, I'll make the neccessary changes then as you have suggested. Thanks for letting me know.
I'll get started now then and let you guys know when I'm done.
Joel
-
Just to clear a couple of things up:
Would there be any way of making the VGA port output composite sync? If so, that enables the usage of a RGB SCART cable for TV hookup.
Composite or S-video would need a separate video encoder. There's no room on-board for this. You can probably modify an old Amiga RF modulator for RF output though, they ran through RGB, which afterall, is all VGA is. You can probably use an RGB SCART lead with most European TVs too as the Minimig outputs 50hz. All you need to do is tie the horizontal and vertical syncs together to get composite sync for normal RGB (doesn't mean you can use composite video this way).
Also ready-assembled is much better. I don't see the point of a bare PCB (for the majority of potential customers...)
I beg to differ. Most guys who'd want to have a go at this project don't have the means to fabricate the PCB. It's really the hardest part to get past, the rest is trial and error.
-
Ya can count me in
-
Xenepp wrote:
Just to clear a couple of things up:
You can probably use an RGB SCART lead with most European TVs too as the Minimig outputs 50hz. All you need to do is tie the horizontal and vertical syncs together to get composite sync for normal RGB
It's composite SYNC I want, for RGB.
'Tie together': as in simply join the two pins together, or should I stick something in between?
-
As far as I remember, the sync from VGA is rated higher than standard RGB so might need to be resisted slightly, but generally, you should be fine just combining the two.
Joel
-
Put me down for one too, Assembled!
(/me looks right at the wall of SMD rework machines and smiles)
-
PaSha wrote:
It's composite SYNC I want, for RGB.
'Tie together': as in simply join the two pins together, or should I stick something in between?
You have a chuffin FPGA, just change the verilog!
'OR' the H&V signals internally, wire out the new Csync instead of one or the other, voila ;-)
Even better, wire up the functionality to the 15KHz/31KHz jumper.
31KHz = Separate H & V
15KHz = CSync
Ask nicely and I'll provide you with a modified Amber.v
Oh, BTW: Those wanting to make mods to their MiniMig CORE will need to get hold of this:
http://www.xilinx.com/ise/logic_design_prod/webpack.htm
-
I would also suggest adding a few extra dollars per unit so that Dennis could be sent one or two free units.
-
I will take one fully assembled, any size is fine. Maybe you can set up a paypal account for taking funds. I am willing to prepay to help out on the upfront cost. I am pretty sure others would be willing to put money upfront also (for proof refer to those coupon deals; at least this one is real.) I also feel that a cost between US$100 to $125 would be very reasonable for this. Distribute a portion of any extra funds to the related parties to this project.
-
I suggest that European orders are shipped directly from an EU country, so to save on shipping + import taxes!!
-
I'll purchase one if Dennis is getting a cut of the profit.
-
Just checked out some PCB pages to see the cost of manufacturing the board.
I don't know too many details about the minimig board (didn't bother to count SMD pads, for example), so I used some numbers that seemed reasonable. I just wanted to get a rough quote anyways.
I checked with http://www.4pcb.com
I specified a 5'' by 5'' board, double sided, 0.062'' FR4, 2 layers, lead free, 1oz copper, 0.008'' min trace width, min. 0.02'' holes, 500 SMD pads on top, 0 on the bottom, 0.01'' SMD pitch, green solder mask on both sides and silkscreen on the top.
Qty Same-day 1-day 2-day 3-day 4-day 1-week 2-week 4-week
50 $49.10 $39.32 $22.21 $18.29 $16.83 $15.85 $13.89 $9.98
150 $33.43 $26.76 $15.08 $12.41 $11.41 $10.74 $9.41 $6.74
250 $30.30 $24.24 $13.66 $11.23 $10.33 $9.72 $8.51 $6.09
The NRE is $149.
Surprisingly cheap!
I was going to check with Gold Phoenix too, but their page is down ATM.
Anders M.
-
For everyone wanting to keep 12x12cm size, if that is chosen as the thing to do, at least go with Nano-ITX which I believe is that size. I think there are some Nano-ITX to Mini-ITX coversion kits to fit standard Nano boards into standard Mini cases, and there are probably Nano-ITX cases by now. Whatever is done, at least pick some standard to do it in.
-
I would buy at least two fully assembled\populated... Put my name on the list if your taking advance orders or just getting a head count...
trip6
-
Check with Olimex: http://www.olimex.com/pcb/index.html
Less startup costs at least.
-
Xenepp wrote:
Just to clear a couple of things up:
Would there be any way of making the VGA port output composite sync? If so, that enables the usage of a RGB SCART cable for TV hookup.
Composite or S-video would need a separate video encoder. There's no room on-board for this. You can probably modify an old Amiga RF modulator for RF output though, they ran through RGB, which afterall, is all VGA is. You can probably use an RGB SCART lead with most European TVs too as the Minimig outputs 50hz. All you need to do is tie the horizontal and vertical syncs together to get composite sync for normal RGB (doesn't mean you can use composite video this way).
Also ready-assembled is much better. I don't see the point of a bare PCB (for the majority of potential customers...)
I beg to differ. Most guys who'd want to have a go at this project don't have the means to fabricate the PCB. It's really the hardest part to get past, the rest is trial and error.
IIRC, Dennis had jumpers on the Minimig to switch from 15kHz to 30~31kHz for the RGB output to be able to use old Amiga monitors, or modern VGA monitors. In fact I had written him about it and have the reply somewhere on another computer at my home (on the laptop today).
One question: There should also be a jumper to switch from 50hz to 60hz for the North American and Japanese markets interested in this project. Please don't make it 50hz only.
And lastly, GO JOEL!!! This is absolutely the most exciting Amiga news in years and I think it is great that it is Open Source and was developed by one single person that was not a greedy company out to screw the Amiga community (again).
What is the current count of people interested in boards? Many of us want more than one board, what is the total count of boards wanted?
-
The olimex boards wouldn't be good for this, as it wouldn't be possible to get more than 4 minimigs on the biggest board (2 if the 12x12cm form factor is chosen, as the width of the olimex board is 20cm), and it costs 132$.
This is 33$ (or 66$) per board, as opposed to as low as 6$ per board from 4pcb. The NRE of 150$ is negligible when several hundred boards are ordered.
It's a very good offer for prototypes and single projects, but for something like this, when hundreds of boards are needed, the bigger board houses have the best prices to offer.
Anders M.
-
it'll be funny if this outsells A1, Pegasos and Efika boards.. lol
-
Xenepp wrote:
I'll get started now then and let you guys know when I'm done.
Keep us posted all the way until they're shipped.
Oh my word. This is so cool I almost wet myself.
-
I´m in on two assembled boards, for up to 150 pounds - it´s just a fantastic project!
-
amigadave wrote:
One question: There should also be a jumper to switch from 50hz to 60hz for the North American and Japanese markets interested in this project. Please don't make it 50hz only.
AFAIK the design of the MiniMig internals and the V1.0 PCB means that you are stuck with a 50Hz PAL Amiga (for the time being at least).
Because of the way MiniMig's clocking currently works, I dont think you can even change to 60Hz in software like you can a real PAL Amiga. This was discussed on the main MiniMig thread at least once.
You may be able to change the main Crystal from a 4.433619MHz PAL one to a 3.579545MHz NTSC one but you will almost certainly have to play with the DCM to generate the (2x) 7.15909MHz etc.
Although how a 4.433619MHz clock was used to generate a (1.6x) 7.09379 MHz one, I dont quite know.
That alone might not be enough, you *may* need to change some constants (horbeam and verbeam comparisons) in agnus.v for NTSC timing.
I may well be wrong, but you'd have to ask Dennis to be sure.
BTW: Dennis We do appreciate your work of art mate ;-)
-
So, will we know a price and estimated shipping date any soon? :D
-
Does Dennis get any money for this?
-
If Dennis wanted to make any money from this he would have. He could have made and sold boards... THEN at a later date release the JUST the binary FPGA file.
There was an opportunity for him to make money from this and he chose not to AND provide the source.
He is a great man for doing that. A better man than myself I fear.
I for one would like to contribute to him, if I bought a MiniMig capable PCB.
-
I'd like ONE assembled board ,in a standard size,and would prefer if work with either American standard Amiga monitors or standard vga monitors.Am assuming delivery/availability of 8 weeks or more(but in time for Christmas?) from now AND no huge tax on boards sent to the U.S.
It is surely possible for anyone who wants to thanks Dennis to even now send him cards and letters with or without money.Presumably it isn't yet against the law to give cash presents.
-
I for one would like to contribute to him, if I bought a MiniMig capable PCB.
Agreed. I too would be willing to pay something extra, to say "thank you" to Dennis. Whoever makes this, please allow for an optional donation! I think the community should reward someone for delivering...
Oh and count me in for a couple...
-
I would like two assembled boards :)
I agree dennis should get at least some money
even if it is only $5 or $10 a board
-
i'll pay an extra $20 USD if thats acceptable for my board if it goes directly to Dennis. He delivered when even a "company" like Amiga Inc. couldnt. He even beat out that Clone-A machine.
-
I'd be interested in a populated board, or at least a board with all the SMT stuff populated.
I would also like to see all of the Major IC's socketed
wildstar1063
-
Yeah, Xenepp smack a couple of bux on to the cost of each board and put it towards the growing truckload of alcohol deliveries for Dennis (Assuming he likes Alcohol).
Put me down for one fully populated board (DEFINATELY with the Hardware fix added, preferably directly to the Production run, and not just a jumper wire). I'd like to populate it myself, and solder myself, My SMC skills are shocking, whereas my through Board soldering is okay, and BULK orders of boards and FPGA should reduce costs.
-
Just checked out some PCB pages to see the cost of manufacturing the board.
He may want to have a look at MYRO (http://www.myropcb.com/) as they do the full turn key solution (they get the parts, make the PCB and assemble and the prices they quote are to good to be true.
I was going to check with Gold Phoenix too, but their page is down ATM.
Your gonna need to e-mail them at sales at goldphoenixpcb dot biz
-
TheMagicM wrote:
i'll pay an extra $20 USD if thats acceptable for my board if it goes directly to Dennis. He delivered when even a "company" like Amiga Inc. couldnt. He even beat out that Clone-A machine.
This even beat the One Chip MSX (http://www.bazix.nl/onechipmsx.html) computer that someone mentioned here a while back.
-
I would also like to apply for one unit. Best is ready to run board or at least all smd parts soldered.. To my knowledge, having the smd part soldered at the pcb factory should also be the least amount of costs.
I prefer all components on a single pcb incl. port adapters etc. Mini-ITX form factor eases the housing in mini-pc cases..
Peter
-
I think there should be a proper website so people who are interested fill up a form or something :)
Any idea about what the price could be?
How will games run? ADF/DMS disks running from a memory card via floppy drive emulation?
-
Another thought if a new motherboard is made. If there's enough available pins on the FPGA (perhaps pick a different FPGA package to add more pins), put on some additional things that may not be in the current MiniMig design but which could be added with time to the Verilog design. An IDE port, ethernet, and perhaps PCI slot as examples. There's a ton of things on opencores which might be combined into MiniMig. Maybe a larger motherboard will be made someday which could have Zorro slots and video slots as well in addition to PCI and maybe PCI-Express with a PCI-?PCI-Express bridge chip), sortof like early Boxer ideas were.
-
I'm in for a fully assembled board.
-
I have created a Minimig wiki:
http://www.opencircuits.com/Minimig_Project
My hope is that it will help people to accumulate knowledge in this matter such that one doesn't have to read long.. threads to find information.
Especially now that there are several.
No login required, works almost like wikipedia.
-
The problem with the "bigger FPGA" idea is that the 208-pin package is the biggest one with pins from Xilinx IIRC. If you want to go higher, you need to go BGA. And BGA soldering equipment is expensive, you need at least a four layer board, inspection of the soldering requires even more expensive equipment, and for cheaper BGA soldering machines (that still cost several thousand dollars), the failure rate is high.
It would, in other words, get much more expensive.
It's an idea for the future, for those who want to experiment, but for the majority, who just want a very compact and practical OCS A500 to play with, staying with the 208 pin FPGA is the best solution
-
Here's a thought... How using the original design, with headers instead of the ports... Then having different boards to mount it to via screws and ribbons. The different boards would then have ATX and ITX layouts or whatever else. The mainboard would be the same, it's the board it's attached to that varies.
-
This is great news and I would really want one. So count me in for at least one fully assembled Minimig maybe two depending on the price :) .
-
Switching from the Spartan 3 XC3S400 to a Spartan 3E XC3S500E would give you about 17 extra I/O pins while still keeping the 208 pin package. It would also give you a good bit of space for bug fixes and/or future enhancements. The current Minimig design uses up about 82% of the XC3S400, but only 65% of an XC3S500E.
Personally, I'd like to see a real floppy controller, IDE controller, a Kickstart ROM socket and possibly a Zorro-II slot added to the Minimig.
There is also an open source Atari ST clone. It would be nice if one board could be used for both the Minimig and the ST.
-
I'm glad to see that there is a lot of support for Dennis getting some much needed reward and i'm glad to see that there is some progress towards getting a product out. This grassroots stuff really makes it a lot more fun. Count me in for a board, and for whoever is making this possible, please allow us to pay over, or donate a portion so that we can get Dennis fat as a cow on pizza and thoroughly drunk.
-
Baby steps folks. Nevermind the PCI and IDE and all that crap. If future expansion is in our minds, we should focus on Zorro-II and possible RAM/CPU expansions first. Zorro-II being the first priority, as there are many A2K expansions out there (like SCSI interfaces) that would be well within the realm of MiniMigs current specs.
-
Please please please make it to some recognised standard size and mounting hole placement.
Mini-itx or nano-itx I don't mind but smaller is better if possible.
-
Baby steps folks. Nevermind the PCI and IDE and all that crap. If future expansion is in our minds, we should focus on Zorro-II and possible RAM/CPU expansions first. Zorro-II being the first priority, as there are many A2K expansions out there (like SCSI interfaces) that would be well within the realm of MiniMigs current specs.
IDE would be cheap and easy to add. Really not much point in using a Zorro-II SCSI card.
Zorro-II would be easy to add too, since it's pretty much just the 68k bus with a few extra signals added.
I wouldn't bother with PCI though, too much hassle and too little gain. Especially since the Minimig is currently 68000 only.
16 bit sound and RTG compatible video cards could be integrated within the FPGA itself, no need to use an external device for these.
-
ATA (IDE), ZorroII, SCSI all wants a large number of I/O pins. There a currently 4 remaining. A few more possible by reusing boot configuration pins. Getting larger fpga means BGA and that is a path you don't want to go.
But fpgas are capable of really fast communications, esp the standard spartan3. So a possibility is to use the leftover I/O to connect another fpga to hookup these things. Essentially a "south bridge" running on a pseudo serial bus.
Not minding the cost of an extra fpga, special copper trace requirements, support circuitry, power, and connectors.
-
freqmax wrote:
ATA (IDE), ZorroII, SCSI all wants a large number of I/O pins. There a currently 4 remaining. A few more possible by reusing boot configuration pins. Getting larger fpga means BGA and that is a path you don't want to go.
But fpgas are capable of really fast communications, esp the standard spartan3. So a possibility is to use the leftover I/O to connect another fpga to hookup these things. Essentially a "south bridge" running on a pseudo serial bus.
Not minding the cost of an extra fpga, special copper trace requirements, support circuitry, power, and connectors.
Alternately you could connect the Zorro bus bridge directly to the CPU bus, leaving the FPGA to do the rest of the chipset. Using a serial link to another FPGA (or even a CPLD) would be fine for slower peripherals such as IDE.
Hans
-
It would save costs possibly to connect directly. However in the eyes of an FPGA, Amiga electronics is like snails.
-
freqmax wrote:
ATA (IDE), ZorroII, SCSI all wants a large number of I/O pins. There a currently 4 remaining. A few more possible by reusing boot configuration pins. Getting larger fpga means BGA and that is a path you don't want to go.
I was looking at the schematics earlier toady. It might be possible to pick up 5 extra fpga pins by multiplexing the joystick ports. This would require adding another chip to the board though.
Personally, I don't see the need for PCI, Zorro, or SCSI. It would fun to add USB and MP3 to the SPI ports, though. Hook up a Vinculum Vmusic board and you've hot both.
-
The thing with Minimig is access to software primarly, and hardware secondarly.
PCI, SCSI can be had on any PC, Or networked to the amiga (another issue.. ;-))
Zorro could be useful for expansion cards that you simple can't run on any other hardware.
Then there's the situation with software that requires certain hardware.
I think USB might become neccessary due that keyboards & mouse might become only available in that format. But ps/2 is a simple solution that will work for a long time. And I think the effort is better spent on other parts.
Maybe USB memories/ethernet could make it worthwhile anyway?
-
Alternately you could connect the Zorro bus bridge directly to the CPU bus, leaving the FPGA to do the rest of the chipset. Using a serial link to another FPGA (or even a CPLD) would be fine for slower peripherals such as IDE.
The Zorro bus IS the CPU bus.
Same deal with the IDE, for the most part.
-
Ok, so how much extra circuitry is needed?
Any idea how long flatcable Zorro manages?
-
Ok, Girls now I´m hot... :madashell: I want my Minimig as fast as possible !
@Xenepp:
When do you think you´ll be able to order our minimig-babies and what will be the price for one minimig ?
-
Sounds good to me!!!
Xenepp wrote:
It looks like there is enough interest to do it, but I'll not be taking any payments until I have the PCBs in hand. It's far less complicated that way.
Once I have the design done, the price is set and everyone's happy I'll start making a list of those who wants one. Obviously, the more that want them, the cheaper it will end up being.
It's a good idea I think to design it to fit standard Mini-ITX cases. This way, people have a lot of options with how they'll case it up and I can make an Amiga type case to Mini-ITX specs, giving me a firm guideline. Nano-ITX is too small to route traces for all those connectors.
Thing only drawback is that it'll get bigger. The beauty of Minimig is that it's so small but then it doesn't lend itself well to casing with ports everywhere.
So what does everyone think about that?
Joel
-
I'd be interested in an ATX board, as long as there is a floppy and IDE controller on it.
Well, any form factor that will fit in an ATX case would do.
-
alexh wrote:
If Dennis wanted to make any money from this he would have. He could have made and sold boards... THEN at a later date release the JUST the binary FPGA file.
There was an opportunity for him to make money from this and he chose not to AND provide the source.
He is a great man for doing that. A better man than myself I fear.
I for one would like to contribute to him, if I bought a MiniMig capable PCB.
Well deserved money is never to be ashamed of earning it, at least not a reason to be any worst (or do you think only bad people deserve to make money :-P), nor anything stopping Dennis or anyone else for that matter to make money with it. In the end Dennis might make money, direct or indirectly from it, or at the very least I hope to be so as it he rightly deserves it!
On a side note I think both bounties #23 (http://thenostromo.com/teamaros2/?number=23) and #24 (http://thenostromo.com/teamaros2/?number=24) would fit perfectly towards MiniMig goals, presenting an Open Source solution completely free from (sc)Amiga IP.
-
mongo wrote:
IDE would be cheap and easy to add. Really not much point in using a Zorro-II SCSI card.
Given a machine as Minimig what would you use an IDE drive for? At least when against a 2Gb USB pen, where you could fit almost any games and change it on the fly and cheaply.
It would add space also...
-
Given a machine as Minimig what would you use an IDE drive for? At least when against a 2Gb USB pen, where you could fit almost any games and change it on the fly and cheaply.
USB requires a USB controller and a USB stack, both of which are more complex to implement than IDE.
Besides, Compact Flash cards work quite nicely on IDE. No need for a hard drive if you don't want one.
-
No pain, no game... ;-)
-
I'd say: Produce it as it is now. Forget about USB / IDE for the moment. The compact-flash implementation in place right now should be enough to 'act' as a harddrive. The cards have sufficient capacity and cost next to nothing (said that, the bytes/euro is way smaller than with IDE harddrive but hey, who needs 200GB of storage on a OCS/ECS Amiga?). If there really is a need for IDE/USB solutions will surface soon enough.
And I too like the idea to raise the PCB price a few euros and donate those euros at Dennis' bank-account. He deserves it.
-
If I can stick one in a Shuttle case (mini-ITX) then I'll have one, fully assembled - thanks! What a great project!
-
A600/A1200 IDE would be really easy to add here, it is after all mostly an address decoder and the drives hang off the bus.. Perhaps one surplus IO pin for the drive select.. I can't remember what else you need.
Then just load an A500/600/2000 ROM.
If it can be incorporated to the new board, please do it.
Probably wouldn't even be a cross-incompatible fork of the core.
-
I have looked at the files. And there are some things I miss:
* Bill Of Material list (BOM) ?
* Component sourcing (where to buy) ?
* 68AW512M datasheet (need package mechanical details) ?
If the original pcb is to be used, this data is needed or a new board will have to be created.
As for production run. I think it's easiest to stick with the current one, and make the patch suggested by dennis regarding pins swap on fpga. Such that it actually get's done.
In the meantime we could tinkering with options for the next revision. And the PIC18 should be programmed with board revision such that fpga configuration for several boards can exist in the same flashcard.
-
I'd be perfectly happy with the released board design, one unpopulated and one populated, no alterations required.
If you do go making alteration then I guess I'll have to see but I wouldn't object to IDE, a faster 68k or more ram :-D
andy
-
68AW512M ,an ST 512Kx16 SRAM in a 44 pin TSSOPII package, is no longer produced. We will have to substitue a Cypress (CY62157EV30LL-45ZSXI), ISSI (IS62WV51216BLL-55TI), or Renesas (HM62V16514LTTI-5) SRAM.
:-)
-
I agree on the 'ignore the other stuff for now, like usb, ide, floppy'.. that can wait. change the mobo design and build it as is, it'll get the job done.
-
Put me down for one as a DIY-kit, just the PCB with the SMD stuff in place would be cool.
-
I'm up for one assembled, case or no case. Price is irrelevant.
Cheers!
-
Do these other srams have the same mechanical package (esp pads). And same pinout?, if not we can't use the layout as is.
Might be a good idea as a minimum to provide circuits aswell as pcb board ie "kit". As to avoid not getting the parts that fit..
As for costs, I know from a fab visit that things that makes cost run away is multilayer (>2L), through hole assembly, components on both sides.
So letting end users solder connectors by themselfes might be a real cost saver (and possible time-to-delivery).
Some selftest configuration flashdisc would be really useful aswell.
-
All of these SRAM parts below (55nsec or faster) share a common pinout and TSSOPII footprint:
68AW512M ,an ST 512Kx16 SRAM in a 44 pin TSSOPII package
Cypress (CY62157EV30LL-45ZSXI) 512Kx16 SRAM in a 44 pin TSSOPII package
ISSI (IS62WV51216BLL-55TI) 512Kx16 SRAM in a 44 pin TSSOPII package
Renesas (HM62V16514LTTI-5) 512Kx16 SRAM in a 44 pin TSSOPII package
What could be different and may cause an issue between these SRAM parts is ground/VCC bounce since the PCB is only 2layer and lends itself to a higher inductive volatge drop Ldi/dt on the power and ground connections to the ICs. We can only tell this by testing some of these SRAM parts in an actual 2layer MiniMig PCB.
:-)
-
Looking at the dennis layout. And consider it's an 2 Layer (it is?). There seems to be no groundplane under the SRAMs, maybe adding such would benefit stability.
Hopefully we will not get an Minimig-o-Freeze ;)
-
interested.
clockport?
-
I believe there was some talk about using a 1Mx16 RAM earlier in this thread, as that would be much cheaper.
-
What could be different and may cause an issue between these SRAM parts is ground/VCC bounce since the PCB is only 2layer and lends itself to a higher inductive volatge drop Ldi/dt on the power and ground connections to the ICs. We can only tell this by testing some of these SRAM parts in an actual 2layer MiniMig PCB.
That can be a problem. On the current board I have had none of such problems however. A friend of mine used to pause Lemmings the whole night and the board was still running every morning. The ground path from RAM to FPGA is pretty wide so inductance is probably not too bad. A single ram chip would be even better though, as no signal routing is necessary on the ground plane.
Dennis
-
Should be better to construct board as was made by commodore? with custom Chips, etc?
and then a Zorro II/III and PCI's ports to add additional and compatible Amiga hardware. :-o
-
Dennis,
For a 2layer PCB, you did a nice job on the layout and helped to keep the blank pcb very affordable.
The concern with the other brands of SRAM (Cypress (CY62157EV30LL-45ZSXI), ISSI (IS62WV51216BLL-55TI),
Renesas (HM62V16514LTTI-5) on a 2layer PCB is the differing levels of Ldi/dt drop for those vs the ST SRAMS. The different brands of SRAM will all draw different amounts of instantaneous current when the SRAM is being accessed (read or write cycle). If you have any pcbs still not fully populated and could validate one of these other brands of SRAM, it would be helpfull.
:-)
-
@Dennis
Will digikey 428-1860-ND do?
Datasheet:
http://download.cypress.com.edgesuite.net/design_resources/datasheets/contents/cy62167dv30_8.pdf
I assumed it had to be SRAM, not SRAM-async.
Maybe it's possible to use the 2'nd layer to add groundplane?
Is the 74HC4060 wired in parallel or serial mode?
Has meaning when looking for crystals (par/ser).
-
The CY62167DV30 1Megx16 SRAM is in a TSSOP 48 pin package, and the 512Kx16 SRAMS are in a TSSOPII 44 pin package. The Cypress 1Mx16 and 512Kx16 SRAMs have different packages and differnet pinouts.
If the MiniMIG schematic and PCB layout were updated for the CY62167DV30 1Megx16 SRAM (TSSOP 48 pin package), the CY62167DV30 1Megx16 Cypress SRAM could possibly be used instead of two of the 512Kx16 SRAMS.
:-)
-
Sorry, Pixie.. just curious if your "it's" in the AROS slogan is a typo or not... "you know it has roots", you know it is roots", or maybe "you know its roots"? :-)
As for minimig, some general purpose connector port/slot for future add-ons would be nice. Personally I would very much like to build it into an old dead A600's case, and would then need some way to hook on the original keyboard.
-
For a 2layer PCB, you did a nice job on the layout and helped to keep the blank pcb very affordable.
For small quantities, a 2 layer board is much cheaper, but for large quantities (100+) a 4 layer board isn't that much more expensive.
-
it's as in 'AROS roots'... I was told that it would be it's as opposed to its by an US speaker, but I had that very same doubt as you.
-
"It's" and "Its" are both valid possesives in English AFAIK.
-
So, can we now also type "he's" instead of "his" and "she's" instead of "her"? :-D
-
As I said in a previous post:
Google FTW! (http://www.apostrophe.fsnet.co.uk/)
-
I would take atleast 1
For extras...
- Clockport
- A600 RAM expansion, if possible.
Btw, does anyone knows if it is possible to add an A600 accelerator to the minimig?
-
count me in, 1-2 boards, assembled or not 8-)
-
odin wrote:
"It's" and "Its" are both valid possesives in English AFAIK.
No, they are NOT!!!
-
Long time lurker here...
TheMagicM seems to have it right - the project is open source so those handy extras: IDE, clockport, fastram etc may well come in time.
The most important thing however (at the mo) is to get the design Dennis has produced modified (with the suggested fixes) to fit a standard mini/nano-itx case. This way everyone can easily and conveniently enjoy the open source nature of this little masterpiece.
Oh and my congratulations to Dennis btw for finally completing this after 3 years work. You are a saint to this community after a lot of broken promises and general cr*p they have received!
Oh and in regards to the sub-argument, "It's" can not be used for possession - I was assuming it was supposed to be some kind of pun (in that members sig).
-
jkonstan wrote:
What could be different and may cause an issue between these SRAM parts is ground/VCC bounce since the PCB is only 2layer and lends itself to a higher inductive volatge drop Ldi/dt on the power and ground connections to the ICs. We can only tell this by testing some of these SRAM parts in an actual 2layer MiniMig PCB.
Sticking a decoupling capacitor right next to the SRAM chip should solve this problem. If the Most datasheets have some kind of recommendation for decoupling/circuit-layout that you can follow.
Hans
-
The decoupling cap next to the IC will help VCC dip/bounce; however, it will not help with ground bounce. Ground bounce is fixed by use of a low L inductance ground path connection for the IC/SRAMs (i.e. a ground plane).
:-)
-
I have noticed that digikey.com have started to stock the XC3S400-4PQG208C at 25.45 USD for one off:
http://www.digikey.com/scripts/DkSearch/dksus.dll?Detail?name=122-1519-ND
They didn't before. So it was a real pain in the ass to get those chips at reasonable ease and price.
-
The joys of sourcing parts. Here today and gone tomorrow, or worse, company lies about having them on hand and missing deadlines.
-
Here are some interesting links for those who want to make their own minimig PCB:
Make PCB's with a laser printer (http://www.makezine.com/blog/archive/2006/02/how_to_make_pcbs_with_a_laser.html)
Easy PCB Fabrication (http://www.fullnet.com/~tomg/gooteepc.htm)
Express fabrication of pcbs from gerber files (https://www.freedfm.com/!freedfmstep1.asp)
:-D
-
We have a small run of original (Minimig ver. 1.0)bare PCBs.
The only change is the silk screen just to identify the source for quality control.
I stress that these are suitable only for the serious electronics dudes, experimenters and people interested in 1st wave prototypes. They will involve smd work and so are not for the faint hearted!
We are expecting them in next week. As soon as they arrive we will post a link to a paypal shopping cart.
Pricing will be;
£11 - Shipped U.K.
£13 - Shipped Europe (Airmail)
£15 - Shipped World (Airmail)
We will also try to source the original ST memory and make those available as well.
Any profit will be donated to the TeamAros Kickstart II (http://thenostromo.com/teamaros2/?number=24) bounty! :-)
-
CD32Freak wrote:
Here is an interesting link for those who want to make their own minimig PCB: Make PCB's with a laser printer (http://www.makezine.com/blog/archive/2006/02/how_to_make_pcbs_with_a_laser.html) :-D
Been there, done that. Heat and pressure often smudge the traces on the qfp tracks, thin tracks often get lifted off the board when removing the wet paper, etc. Applying heat and pressure uniformly is difficult, the larger the board, the higher the incidence of over pressure and heat on some spots, under pressure and heat on others. Some traces smudge, others don't stick. Getting the top and bottom layer aligned right will make you go postal.
Yes, the minimig is pretty simple, but I wouldn't bother trying to ghetto fab the board. It would be cheaper and less stressful to get it done right.
-
Nusim wrote:
We have a small run of original (Minimig ver. 1.0)bare PCBs.
The only change is the silk screen just to identify the source for quality control.
I stress that these are suitable only for the serious electronics dudes, experimenters and people interested in 1st wave prototypes. They will involve smd work and so are not for the faint hearted!
We are expecting them in next week. As soon as they arrive we will post a link to a paypal shopping cart.
Pricing will be;
£11 - Shipped U.K.
£13 - Shipped Europe (Airmail)
£15 - Shipped USA (Airmail)
We will also try to source the original ST memory and make those available as well.
Any profit will be donated to the TeamAros Kickstart II (http://thenostromo.com/teamaros2/?number=24) bounty! :-)
That is good news, but I wouldn't buy the PCB without the parts kit as well. Buying the PCB and then sourcing the parts from 3 or more vendors would make the cost prohibitive in shipping alone...
-
it's as in 'AROS roots'... I was told that it would be it's as opposed to its by an US speaker, but I had that very same doubt as you.
To clear this up.....
it's = the contracted form of "it is"
its = possesive pronoun, belonging to
This seems contradictory even to native english language folks because MOST of the time the 's is tagged on the end to show possesion. Not in the case of the word "it".
So in Pixie's case the correct usasge would be "its roots".
(Notice the contradiction in the previous sentence?)
Now back to minimig.... I've update my preference.
Mini or microATX is fine as long as we can stick to a standard.
Plaz
-
Methanoid wrote:
odin wrote:
"It's" and "Its" are both valid possesives in English AFAIK.
No, they are NOT!!!
Indeed you are right, I stand corrected =).
-
Nusim,
The MiniMIG pcb 1.0 run sounds good to me.
I am interested in a MiniMIG blank pcb 1.0, and I am one of those serious electronics experimenters.
Please let us know when these become available.
:-)
-
jkonstan wrote:
The decoupling cap next to the IC will help VCC dip/bounce; however, it will not help with ground bounce. Ground bounce is fixed by use of a low L inductance ground path connection for the IC/SRAMs (i.e. a ground plane).
:-)
Actually, it helps with both VCC dip/bounce and ground bounce as it provides a low-inductance path for transient currents on both sides. If you draw out the circuit schematic for a decoupling-capacitor connected to an IC, you'll see that it affects both VCC and ground.
A ground plane is still a good idea. You should be able to create a reasonable ground plane on a 2-layer board if you keep most signal lines on one side and keep any signal traces on the lower layer short.
Hans
-
Been there, done that. Heat and pressure often smudge the traces on the qfp tracks, thin tracks often get lifted off the board when removing the wet paper, etc. Applying heat and pressure uniformly is difficult, the larger the board, the higher the incidence of over pressure and heat on some spots, under pressure and heat on others. Some traces smudge, others don't stick. Getting the top and bottom layer aligned right will make you go postal.
What about photosensitive bare boards and printing to transparencies for exposure?
But yes, I'd agree that having it made by a real manufacturer would give best results.
-
The toner-transfer method: http://max8888.orcon.net.nz/pcbs.htm
Might be useful for throughhole, small boards.
But this is many components, highdensity, doublesided board. I think that requires proper methods.
-
Definitely interested in a working minimig board.
I think it would be good to not rush into this - the sources have just been released, so let's see if we can save some pins on the FPGA for future use (I liked the suggestion to use the larger capacity FPGA that still had pins but had more usable pins available, but I'm sure there are good reasons not to use it). On the other hand we don't want to wait forever and future enhancements can be turned into a new product revision (the second run of boards).
Is the reason for the SRAM so that the design doesn't need a memory controller for SDRAM on the FPGA (which would require a huge bunch of pins)? It would have been nice to have 8MB, but not at the price of the SRAMs.
I'm all in favour of the smaller board, tweaked to have a nano-iTX compatible layout myself.
Could be cool to use the product to try and implement other computer systems too - the Atari ST comes to mind immediately (although it has a slightly higher CPU clock that could prove an issue with the current design) as well as a host of 8-bit computers (with their CPUs implemented on the FPGA). So being able to select which 'system' to load in the beginning would be awesome, if that is possible from the SD card...
-
hmm, could probably use PSRAMs... they are supposed to be as easy to interface as SRAM, but offer the capacity and price of SDRAM.
i personally like the idea of headers instead of ports, and would prefer a pre-assembled board (my soldering skills need more work, SMD is still out of reach for me).
A larger capacity FPGA would be nice, but not necessary.
the actual form factor, i don't care about, as long as it's still small.
I'd like at least an 020, but thats not really important for a first run.
All in all i definitely want to get one.
-
I'd take one, mini-itx for me.
-
I don't think there are any good reasons why an XC3S500E wouldn't work, the board layout would just have to be changed a bit. It has more available I/O, more capacity and it's cheaper.
Personally, I don't think we should change too much of the original design before the first board run is ordered (or else, it'll never be finished), but changing the FPGA is no bad idea at all. In addition to the things already mentioned, it's newer too(AFAIK), and therefore less likely to get obsoleted soon.
Having to redesign the board and testing the new design could mean that we would have to wait a bit longer for the final board, but getting a board with a higher capacity FPGA and more I/O at a lower price is attractive.
PSRAM seems perfect for this, the only problem is, it's only available in BGA. At least it was the last time I checked. Too bad, as it's damn cheap and available in relatively large sizes. Some sort of dynamic RAM would probably be a good idea for the future, provided we could find a solution that didn't tie up too many FPGA pins, but implementing something like this in the core before the first board run would take a lot of time, so I think we should stick with one 1Mx16 SRAM chip for now.
Anders M.
-
Hi everyone,
Thank you for proposing a group order of a pcb I don't know if I could personally afford the AUD $60 for a board though the exchange rate is a killer.
Would it be possible to use a fpga like this one they reckon it costs like US$1.50 I suppose that's in bulk though. I think it has 16000 gates is that too small?
http://www.infectus.biz/developer-features.php
regards
John
-
Thanks... there's nothing worst then having a bad formed sig... like walking on naked
-
Unit01 wrote:
Hi everyone,
Thank you for proposing a group order of a pcb I don't know if I could personally afford the AUD $60 for a board though the exchange rate is a killer.
Would it be possible to use a fpga like this one they reckon it costs like US$1.50 I suppose that's in bulk though. I think it has 16000 gates is that too small?
http://www.infectus.biz/developer-features.php
regards
John
You figure it out, the SPARTAN-3 FPGA 400K 208-PQFP that Dennis used, contains 400,000 gates, chances are, 16,000 gates is a TAD small.
-
No this is a complete reproduction of a 500 in a 12 x 12 cm board. Custom graphics chips are emulated, it eliminates the need for old electronics which is becoming scarcer and more expensive. There is no need for Zorro slots, anything you would put in them would be better done through USB.
-
Anyone have a list with the exact video modes currently supported by Minimig v1.0 ..?
-
Thanks Whiteb for pointing that out.
I was just interested in that one as you can buy it already soldered onto a pcb with a flashrom and usb uC chip.
I guess it really needed to be a ProAsic3 A3P400 or better. :roll:
http://www.actel.com/products/pa3/
never mind just have to wait till until someone wants to do a pcb run here :-)
-
Seems their software is Libero IDE v8.0 and is supposed to work with Redhat 4.x
-
I am definitely interested in a assembled one.
-
This is my 1st post. :-P
Now I have to confess that I've never owned an Amiga before (I own Z-80, 6502 and Intel-based PC's) but after reading every thread about the Minimig, I made two conclusions...
1.) Dennis is an absolute legend! From a fellow designer, a good job on the PCB artwork too!
2.) I am interested in owning a Minimig board too :crazy:
Cheers,
Valentin
-
Has anyone calculated what all the parts to produce a completed Minimig should add up to, minus the PCB itself?
I am tempted to try the Laser printed PCB trick and buy all the parts to put on it. If I am unable to produce the PCB myself successfully, I would still have all the parts needed to put on a PCB produced by one of the members here that eventually produces a run of PCBs.
I think I saw somewhere in one of these threads that Dennis has also provided the Gerber file for his v1.0 Minimig PCB layout.
I also wonder what the total count of boards wanted is from this and all other threads. It must be well over 100 by now.
-
I was going to do the laser printer transfer myself, until I printed the layout and realized how many vias there are, and how small they are. They all need to be drilled, and connected on both sides of the board. I'll have to see if I can find a small enough drill bit first.
The other problem is parts, some can only be purchased in large quantities. The SRAM appears to be obsolete, which is an even bigger problem.
-
basman74 wrote:
This is my 1st post. :-P
Now I have to confess that I've never owned an Amiga before (I own Z-80, 6502 and Intel-based PC's) but after reading every thread about the Minimig, I made two conclusions...
1.) Dennis is an absolute legend! From a fellow designer, a good job on the PCB artwork too!
2.) I am interested in owning a Minimig board too :crazy:
Cheers,
Valentin
So.. are you saying that the Minimig could well end up being your first Amiga? :-)
-
**double post**
-
amigadave wrote:
Has anyone calculated what all the parts to produce a completed Minimig should add up to, minus the PCB itself?
For 100 off, My guess would be like this (prices are in USD) most prices were obtained from Digikey:
Per unit cost - Minimig Rev. 1.0
FPGA: $19.25
PIC: $5.15
PCB: $7.00
2 x SRAM: $14.50
68k CPU: $11.00
misc IC's: $1.80
MMC slot: $2.20
Connectors: $5.50
Headers: $0.50
Passive parts: $4.00
--------------------
Bare PCB: $7.00
--------------------
parts delivery: $3.00
--------------------
Subtotal: $80.90
--------------------
Assembly: $25.00(hand assembled?)
--------------------
Unit cost: $105.90
Deduct $32.00 from the above to give you your answer for parts cost minus PCB and assembly.
Cheers,
Valentin
PS: Using the 'Toner transfer' method to make the PCB would prove challenging, given the presence of the PQFP
-
basman74 wrote:
This is my 1st post. :-P
Now I have to confess that I've never owned an Amiga before (I own Z-80, 6502 and Intel-based PC's) but after reading every thread about the Minimig, I made two conclusions...
1.) Dennis is an absolute legend! From a fellow designer, a good job on the PCB artwork too!
2.) I am interested in owning a Minimig board too :crazy:
Cheers,
Valentin
Welcome to the Amiga.org, the more the merrier. Hope you enjoy your stay. :pint:
BTW, I like your avatar, cute puppy.
-
So.. are you saying that the Minimig could well end up being your first Amiga? :-)
Yes. 8-) However, this may only be true if I don't buy (ideally) a '1200 first...
@ ottomobiehl: Thanks!
Cheers,
Valentin
-
Ok, let's get off the USB track of thinking here folks. For all intents and purposes MiniMig is an A500 clone. USB will not work worth a flip in this current code / CPU state, nor in the near future.
Yes, the eventual possibilities are endless, but until one of you geniuses out there figures out how to code the complete AGA chipset, FORGET USB! Besides, the only USB stack I know of for Amigas is a little hard to license right now.
From a development standpoint, it'd be far easier for expansion purposes to focus on creating an A2000 clone in ATX form. I personally think that's the direction we should be heading. Zorro II expansion, with a push towards ECS chipset compatibility, fitting in a nice ATX (not ITX) case.
Once we have Zorro II slots, there are numerous ways of applying ATAPI interfaces to the system already available from your friendly Amiga retailer.
Hang PCI for a while too. Let's simply focus on making this thing compatible with and comparable to the real thing before we go fumbling around with things that far outside of the scope of the project.
Dennis gave us A500 level, next step is A2000 level.
Meanwhile, with regard to the CURRENT design, I think the board could (and should) easily be formed to fit the ITX formfactors, just to give it something better than two sheets of Lexan for a case... And to appease all of the ITX fanatics that seem to be littering this site.
Personally, I love my big burly manly computer cases that take hand trucks to move around. But that's just me. I'd cram the thing into an old SGI Origin 2000 server if I could! :lol:
-
Basman, make that 11$ for the RAM: http://www.digikey.com/scripts/DkSearch/dksus.dll?Detail?name=428-1860-ND
-
Dr_Righteous >> Actually that is true, but, once, if, the minimig progresses to the state of the A2000, it would need a little more than what the original A2000 was or had.
I do think, how ever, that a few PCI ports could do though.
-
Basman, make that 11$ for the RAM: http://www.digikey.com/scripts/DkSearch/dksus.dll?Detail?name=428-1860-ND
Mikkel,
Yes, if someone were to modify the PCB artwork (there's a patch on that board that needs rectifying if i'm not mistaken), then your suggestion would be cheaper, of course :-)
While we're considering modification of the PCB, let's also look at:
http://www.digikey.com/scripts/DkSearch/dksus.dll?Detail?name=TC51WHM616AXBN70-ND
These PSRAM's would provide four times the memory density of the current setup for less than an SRAM equivalent and are 70ns rated, sufficiently fast enough for the Minimig. Not so easy to install though :-(
Cheers,
Valentin
-
Dr_Righteous wrote:
Yes, the eventual possibilities are endless, but until one of you geniuses out there figures out how to code the complete AGA chipset, FORGET USB! Besides, the only USB stack I know of for Amigas is a little hard to license right now.
You could just use the AROS USB Stack... In fact AROS is a better fit for minimig than AmigaOS anyway... long term.
Hang PCI for a while too. Let's simply focus on making this thing compatible with and comparable to the real thing before we go fumbling around with things that far outside of the scope of the project.
Since there are numerous opensource PCI controlers for FPGA's coupled with the fact that PCI devices are cheap, plentiful and diverse... PCI would be far more sensible than ZII. Again AROS would be able to provide PCI drivers.
-
@Dr_Righteous:
Ok, let's get off the USB track of thinking here folks. For all intents and purposes MiniMig is an A500 clone. USB will not work worth a flip in this current code / CPU state, nor in the near future.
Didn't knew my A500 had a MMC flash card! :-P
-
@bloodline:
Since there are numerous opensource PCI controlers for FPGA's coupled with the fact that PCI devices are cheap, plentiful and diverse... PCI would be far more sensible than ZII. Again AROS would be able to provide PCI drivers.
I seriously think this is out of MiniMIG's scope. What value would have a 68000 machine expandable? Expandable to do what with that raw processing power?
MiniMIG will allow to run Amiga games and probably faster in a little case, now you start to add space, and 'functionality' based on a 68000 processor and you're stretching to far, IMO...
BTW, I hope people interested in MiniMIG also help him untie from current and future IP {bleep}-slapping through AROS kickstart replacement, this way it could be presented to the rest of the world and not just to a selected few.
-
@Mikkel
Basman, make that 11$ for the RAM:
Yes, that's 22$ for both parts.
I found a source for the M68AW512ML70ND6 SMD TSSOP package STM Brand. However it would be a significant investment to buy these in at a reasonable pricepoint.
We might try the Cypress parts first to see if they are o.k.
Of course the advantage of supplying the ST parts means that everyone will, initially, be working from the same 'baseline' as Dennis so it's likely we will get these parts.
-
No Zorro slots, it's a stupid move in the extreme, you are tied to pricey old hardware for a new machine, just put PCI slots.
PCI = simple, available, cheap.
Zorro = Old, limited suppliers, expensive.
-
YES!!!!
USB and PCI, finally reasonable priced hardware! Use the AROS USB stack and the AROS PCI controller, it's all there, Amiga reborn for the 21st Century!
-
I have rebuilt most of the schematic in gEDA
http://rapidshare.com/files/46334959/minimig070801.tar.bz2.html
http://geda.seul.org/
If someone feels to build on it.
Btw, By useing a 1Mx16 ram one can free RAM_SEL1 by using RAM_SEL0 as additional address pin. And one sram instead of two saves a lot of wiring.
@Dennis:
Why does the sram use /UB and /LB when data is 16-bit anyway?
It works as a "enable" for each 8-bit part from what I can see.
The 28MHz fat agnus clock is generated by DCM from the PAL clock?
-
The other problem is parts, some can only be purchased in large quantities. The SRAM appears to be obsolete, which is an even bigger problem.
:idea: Rochester Electronics http://www.rocelec.com/
They used to buy up new old stock of obsolete parts. Not sure with Rohs, how their business is? :-?
-
count me in, 1-2 boards, assembled or not
Is a list around to sign up? I'll go for 2 boards assembled.
-
Thanks for the schematic.
The UB & LB strobes on the SRAM are due to data bus addressing needed for the 68K cpu. 68K CPU has UDS* strobe for upper byte lane D15-D8 access, and 68K CPU has LDS* strobe for lower byte lane D7-D0 access. When 68K does a read, D15-D0 can be accessed as a word with 68K reading what it requires off of its Databus; however, a write access on a 68K CPU requires the byte lane be qualified. When UDS* active on a write cycle, UB* on SRAM must be active. When LDS* active on a write cycle, LB* on SRAM must be active.
D15 ........D8,,,,,,,,,D7........D0
Byte 0= Even,,,,,,,,Byte 1= Odd => (Word 0)
/UDS ,,,,,,,,,,,,,,,,,,,/LDS
etc ....
:-)
-
What a mess :-)
Any idea how to make the fpga translate it into plain 16bit access? Or even 32bit.. as any sdram is likely to be organised that way.
I think the catch here is partial write where one could need to first read 16-bit (or32), modify, and then write back. Meaning the ram would have the double access time in practice.
Regarding the schematic. The most needed is the .inc files for the parts. Particulary the sd-card socket seems to be a pain (an image -> .inc program would be nice). At least the supplied would do for a hacker setup by having a ll signal solder pads done. Maybe should use a digikey.com one instead?
A clarification of the various component packages used would be useful.
The LM1117MP-adj for example is likely to be SOT23 asfair, because that's the only pkg size digikey have in stock in one-off quantities.
-
Is there an official place to sign up?
Thanks.
-
basman74 wrote:
amigadave wrote:
Has anyone calculated what all the parts to produce a completed Minimig should add up to, minus the PCB itself?
For 100 off, My guess would be like this (prices are in USD) most prices were obtained from Digikey:
Per unit cost - Minimig Rev. 1.0
FPGA: $19.25
PIC: $5.15
PCB: $7.00
2 x SRAM: $14.50
68k CPU: $11.00
misc IC's: $1.80
MMC slot: $2.20
Connectors: $5.50
Headers: $0.50
Passive parts: $4.00
--------------------
Bare PCB: $7.00
--------------------
parts delivery: $3.00
--------------------
Subtotal: $80.90
--------------------
Assembly: $25.00(hand assembled?)
--------------------
Unit cost: $105.90
Deduct $32.00 from the above to give you your answer for parts cost minus PCB and assembly.
Cheers,
Valentin
PS: Using the 'Toner transfer' method to make the PCB would prove challenging, given the presence of the PQFP
Thanks Valentin,
I have reconsidered and will wait until someone (maybe you) produces fully assembled boards to purchase and complete the rest of the project to have my own Minimig.
Unlike some others here, I only want it to run games on in the smallest form factor possible. If it could be fit into the base of a joystick that would be ideal.
I don't see the point of making it with PCI, USB, etc. extra connectors and 68030 or faster capabilities. The majority of Amiga games will run on the 68000 w/1mb RAM.
I would like someone to take Dennis' work and expand on it to make a better Amiga in the future, but that is a completely different project that will take considerably more work. When it has progressed to a point which exceeds my current 68060/50mHz & 604e PPC/233mHz Amiga I will buy that project too, but I think that is a year or three off into the future. Someone, or some group of geniuses here please prove me wrong.
For now all I would want is Dennis' original concept, with his suggested fixes, produced in the smallest package to fit into the base of a joystick, or if not possible, perhaps into an old 1010 external floppy drive case.
Lots of interest here, let's get this done before Christmas so we can all enjoy the fruits of our labor putting a Minimig together.
:-D
-
@amigadave
I think you hit the nail on the head. If the first run waits for improvements, we will be another two years out. Our best bet is to do the simple modifications that Dennis has already suggested, and get it out as is.
Slap a MiniMig 1k sticker on it, and THEN start talking about all of the cool additions that could be made for future models. The MiniMig 1/2k could be for those of us that want a small form factor, and are not concerned about upgrades, while the MiniMig 2k could be the larger form factor with expansion ports.
What is needed now is working MiniMig 1ks that people can start playing with, and can find the limitations, so the second run can be better.
Maybe the real question shouldn't be, who would be interested in a MiniMig, but
Who would be interested in spending $150 now on a MiniMig 1k, knowing that there will be a better model available a year from now?
I definitly am. I have no problem spending a $150 today, so that production gets started. After all, no matter how much talking is done before the first run, as soon as the units are in our hands, people will start finding things that could have been done better. That is what future models are for.
-
I may be interested on a fully assembled (or at least with the tiny SMD soldered) depending on price.
-
About the SRAM: No, it isn't 22$ totally, no we don't need to look for expensive, obsolete RAM chips of the same type that Dennis originally used. The chips I linked to earlier are 1Mx16, so we only need one. This will also simplify the board layout. This should not be any problematic, as this chip was originally suggested by Dennis a bit earlier in this thread.
-
After reading this thread, I think it's best to just keep the first "batch" of Minimig's simple and plain, so no PCI, no USB etc. The Minimig is already complex enough in it's current form and if we get the hang with the rev1.0, we can move to rev2.x. That board can be mini-itx with larger fpga, usb, sdram, compactflash-harddisk... but I would stay away from PCI slots and ram slots. SDRAM is so cheap nowadays you can just solder 32mbytes onto the board for a couple of dollars. Saves the hassle of making the board compatible with any brand of ram sticks out there. Also, who needs PCI if you can code your own graphics card into the FPGA?? Just a simple framebuffer would suffice to create an 800*600 16bit color desktop. Switching the monitor from rtg to ocs? Just a couple of lines of verilog. Coupled with a 20Mhz 68000 or even a 68020 it would be very nice and still simple to build, maybe even on 2 layers.
So, my proposition is that the minor revisions be all core/firmware compatible (so rev1.1 would have the bugs fixed but still be 100% compatible with 1.0) and 2.0 will have different fpga, different ram etc. If we keep the verilog code modulair, we can build cores from the same source tree for both rev1.x and rev2.x boards. If you look at the code now, only minimig1.v and minimig1.ucf are board specific, all other sources could be re-used for a rev2.0 board.
Dennis
-
I agree SDram is cheap enough that they are a bargain path for memory upgrade. Provided that we can figure out a sdram "driver" in verilog. Loot those evil x86 pc's :-D
As for compability. It would save some headache if boards can use the same fpga configuration binary. And in the case they can't the MCU (pic18) should select the appropiate one based on board revision.
A cheap upgrade path is to have unpopulated areas where one could add chips for sdram, composite output etc..
-
Dennis,
Thanks for giving your input here. There are a lot of dreams, and desires on what the MiniMig should and could be. People have been waiting to have an Amiga compatible for a long time, and you have made it possible.
Because you have single handedly raised the dead, you by default have become the Amiga's second coming. This means your word concerning all things MiniMig are going to be taken largely without question. This also means that you have the power to do what no one else in the community can do, which is to end arguments by just stating the way it should be. We will all understand if this is not a responsibility you want to take on, but the community will grow better if you do.
So, please, if you are willing, step in and state your opinion to help those that build on your work stay on the right track.
-
http://rapidshare.com/files/46443453/minimig.sch.html
Almost complete.
-
I agree with Dennis to keep the 1st batch/run of MiniMIGs compatible so that the base set of Verilog HDL and 68K boot code runs on it so enhancements to these can be added on a stable hardware platform. Some kind of very limited expansion could be via the four pins of SPARE I/O FPGA expansion on MiniMIG.
For MiniMIG Gen2 (A1200 like), a 68020 (or maybe even a Freescale 683xx CPU32) would be probably be the way to proceed. With the dynamic bus sizing on the 68020, we could choose a 32bit or 16bit databus back to the FPGA where some similarity to 68000 Bus exists in 16bit 68020 bus mode. A 68020, SDRAM, IDE, compact FLASH, Amiga Clock port, and maybe USB would be good for Gen2. The issue that we are going to have is runnig out of pins on a QFP Xilinx FPGA. Gen 2 MiniMIG could really use a bigger Xilinx FPGA in a BGA package; however, this would push us into a 4 or 6 layer PCB and also require BGA rework equipment to assemble. SDRAM controller is available in Xilinx MIG/Core generator and also on www.OpenCores.com. I have in the past written Compact Flash and PIO IDE controllers in Verilog as well as designed 68020 embedded hardware. I am willing to help on a MiniMIG Gen2.
:-)
-
I hope that most here will agree with the common sense that Dennis and others are suggesting for the first run of v1.1 Minimigs. Then v2.0 can be discussed and designed over the next two weeks :-P (couldn't resist the Amiga joke of the decade) and another production run with the new design and compatible verilog code to make it work can be tested by a few brave souls.
I am good friends with the owner of a plastic injection molding business that I can talk to about doing a limited run of cases for a Minimig joystick, if the PCB can be designed to a reasonable size for such an implementation. Dennis' original size would work for such and only be a little over sized compared to most other Amiga joysticks. I like the 12cm by 12cm form factor with the ports as Dennis had on the v1.0 design. I could easily design a replacement bottom half for a joystick I already have in mind to accept the Minimig and complete the casing with the joystick on top.
-
Belial6 wrote:
@amigadave
Who would be interested in spending $150 now on a MiniMig 1k, knowing that there will be a better model available a year from now?
I definitly am. I have no problem spending a $150 today, so that production gets started.
If someone is willing to do a 'pilot run' of 100 units (and 100 people commited to purchase :-P ), then I believe this is what it will take to get the process started. This would allow for further 'tweaking' of the design (if necessary) or hacking on 'improvements' etc.
Also, I for one would be happy to put my hand up to organise such a pilot run . Unfortunately, I am committed to my current (work) projects for another week or so, so I would not be able to do anything until then.
Probably at some point before then, we might receive an update from the original poster of this thread. ;-)
Cheers,
Valentin
-
I've been wondering, how feasable would an old fashioned largely-throughhole PCB be? That'd make it possible for pretty much anybody to produce it in the garage (and not to mention MilSpec!). Are there sockets available for the SMD FPGA, mem chips etc?
Hmm../me runs off to download a PCB CAD prog.
-
Are there sockets available for the SMD FPGA, mem chips etc?
Yes but they are very expensive, specially ones that go through hole.
You cna get cheap ones that use the same surface mount pads for around £20, the clam sockets with through hole pins cost a bomb though, simply because they are only designed to be used on prototypes.
You can get QFP to PGA adaptors though, but even so.. surface mount is the best way tbh.
Well unless you could break the code down to use a few 86pin PLCC package FPGA's.. call one Agnus, another one.. I dunno.. dennis ;-)
Thats probably the best bet for real DIY'ers
-
Throughhole would increase cost significantly and introduce more EMI.
-
Throughhole would increase cost significantly
well thp assembly is actually cheaper, but the more chips would bump up the cost... that said it would be a real DIY project then, build a jtag programmer on the board and have some fun...
just like the good old Sinclair Spectrum
and introduce more EMI.
Hands up who care :-D
the A500 was all thp, as was the A2000, 1000 etc.
-
I like Dennis idea of keeping FPGA firmwares compatible and well structured.
BTW, about form factor I think MiniMig is nearer to nano-itx (12x12cm) than others like mini-itx.
For minimig v2 there are cheap 68LC060/75Mhz that could be a very interesting cpu as it has good compatibility, great speed, MMU (it lacks FPU but if the 060 was socketed people could put a full 060 if they prefer it).
I think Minimig v2 could be a substitute of an A1200 Blizzard 060. It should be able to interface with the 040/060 directly. There's no need ATM to add A1200 cpu connectors because if you change FPGA code to work with 040/060 correctly you would save a lot of logic and get a much faster system. An option could be added to "slow down" memory accesses to a similar speed of a Blizz060/50, but I guess that since the 040/060 would be supported by the fpga a lot of goodies would work like burst modes and access to ram would be much faster than on real Amiga 060 boards.
If it has RTG and other features it would be great, but 64MB would be mandatory (latest amiga scene demos are starting to require more than 32MB)
About the RTG software AROS CGX system could be modified to run on AmigaOS 68k and patch standard libraries like CGX/P96 do.
An AHI driver could be developed and sound output could be mixed with paula output without the need of cables :-)
Apart from network and USB I don't see much need of PCI stuff. I think that the basic computer functionality should be provided as standard.
My wishlist:
1.- Minimig v1 should be released more or less as is ASAP. Otherwise nothing will be done.
2.- Minimig v1.1 may have different form factor and slight changes but the core should be compatible as Dennis has said. Includes a dumb IDE port (similar to the A500IDE from aminet) and more ram that can be configured as chip/fast ram. Up to 2MB of of Chip ram (I guess that a plain 68000 wouldn't accept 8MB of chip). The rest could be configured as fast ram or slow/ranger ram.
3.- Minimig v2 with AGA and 060 socket. 64MB chip soldered on board. LC/EC 060s are not much more expensive than other motorola cpus. ¿ISA slot for old NE2000 cards? ¿faster PIO modes or even DMA? ¿keep an extra 68000 for 100% comaptibility with A500 games?
4.- Minimig v3, retargetable edition. Like v2 but with RTG, RTA, Network.
5.- Minimig v3.1 USB support.
Of course it's just a dream as I guess Minimig will reach my dream Minimig v1.1 as maximum, but it will be ok anyway. Dreaming is cheap :-)
-
I want one!
A minimig for me
:-D
-
I want two minimigs PCB complete !!
So count me in !!! :-D
-
Thanks for the initiative. I'd like one, where do I get in line?
-
basman74 wrote:
If someone is willing to do a 'pilot run' of 100 units (and 100 people commited to purchase :-P ), then I believe this is what it will take to get the process started. This would allow for further 'tweaking' of the design (if necessary) or hacking on 'improvements' etc.
I'll be one of those 100 people, if it goes ahead :-)
Matthew
-
Crumb wrote:
Of course it's just a dream as I guess Minimig will reach my dream Minimig v1.1 as maximum, but it will be ok anyway. Dreaming is cheap :-)
Dreaming of things like this can also be healthy, in certain doses.
I think your list is fairly good and realistic. Though I'm not sure how long AGA will take to impliment, considering how long it took OCS. There is more support though now it's out, and I do remember reading somewhere that AGA wasn't that big a leap over ECS.
Do we even need AGA for this thing I wonder - is Dennis even considering it?
-
Fixer wrote:
Do we even need AGA for this thing I wonder - is Dennis even considering it?
No, Dennis has already stated before that he is not interested in doing AGA, but would hope that others would build on what he has started.
-
I just did a rough count thrugh the thread. There are ~130 boards requested (some people want more than 1 or 2) with the majority of people wanting fully assembled boards. ~10 people have asked for miniITX.
-
Thanks Hojo_Norem,
You saved me from making that count. I have been wondering how many there were and was going to start a spreadsheet with all the member names that wanted boards and a count of who wanted more than one Minimig and another column that showed who wanted completed boards as opposed to those who want bare PCBs and a bag of component parts.
130 boards should be enough to make it worth someone's time and trouble and get a better price on some of the parts. Who is going to step up and be the first to take charge of getting the first run into production?
I am ready to buy at least one, if not two boards. Preferably the original size of 12cm square.
Ready...... Set........ Go Minimig! The replacement A500 in the palm of your hand.
-
I'm interested in one too! Fully assembled please.
-
Xenepp, why no feedback?
-
Just looking at the jpeg of Dennis' v1.0 Minimig again and thinking what a beautiful job he has done with the design. Maybe I am easier to impress than some of you, but I think Dennis' design is nothing short of a work of Amiga computer ART! (Love the Galaxy Quest quote too)
I would really like to have a duplicate of HIS EXACT BOARD layout, including the same silkscreen text. The only changes I want are the fixes he has suggested and I would prefer NTSC video timing, or the option to switch from PAL to NTSC and back via a jumper.
Can anyone make this request a reality?
Dennis, if you are reading this and have access to one more of the original 4 boards you produced, I would gladly buy it. (I know this is not a real possibility)
-
i'm interested too... for an assembled kit
-
http://rapidshare.com/files/46669345/minimig.sch.html
Minimig schematic complete, minus the joystick port.
I used 1Mx16 ram instead of 2x 512kx16.
Headphone back-emf?
* The +1.25V and +2.5V can be connected to +3.3V to save power. Due P_heat=U_drop*I.
* Howcome the many connections to the fpga isn't in numerical order?
* What is the value of C37 100uF/6.3V ?
* Why is R42 present?, it's just sitting between +1,25V and GND.
* Howcome (54,55,56) M1,M0,M2 is connected to VccAUX, doesn't it need 3.3V?
* The 'SPI_DOUT' from the fpga and SPI_DOUT from the sdcard (via 1k) can drive eachothers output, maybe this can be resolved with an or gate or such instead?
* Maybe the keyboard and mouse can share connection?, I've seen that on laptops asfair (saves i/o).
* There's no over/under potential protection for keyboard or mouse?, like video output have. Howcome?
* Maybe the rs232 output of mcu and fpga can be shared without a jumper by setting an AND-gate between. Because the default level for rs232 is "1". And thus any transmission will produce "0", and if the other maintain the rs232 "1" resting state. IT should work.
-
freqmax wrote:
http://rapidshare.com/files/46669345/minimig.sch.html
Minimig schematic complete, minus the joystick port.
I used 1Mx16 ram instead of 2x 512kx16.
Headphone back-emf?
* The +1.25V and +2.5V can be connected to +3.3V to save power. Due P_heat=U_drop*I.
* Howcome the many connections to the fpa isn't in numerical order?
* What is the value of C37 100uF/6.3V ?
* Why is R42 present?, it's just sitting between +1,25V and GND.
* Howcome (54,55,56) M1,M0,M2 is connected to VccAUX, doesn't it need 3.3V?
* The 'SPI_DOUT' from the fpga and SPI_DOUT from the sdcard (via 1k) can drive eachothers output, maybe this can be resolved with an or gate or such instead?
* Maybe the keyboard and mouse can share connection?, I've seen that on laptops asfair (saves i/o).
* There's no over/under potential protection for keyboard or mouse?, like video output have. Howcome?
* Maybe the rs232 output of mcu and fpga can be shared without a jumper by setting an AND-gate between. Because the default level for rs232 is "1". And thus any transmission will produce "0", and if the other maintain the rs232 "1" resting state. IT should work.
Posting a image link to a site that requires us to register is a waste for those (like me) that don't want to bother registering at that site. Can you upload the image here?
P.S. Dennis, I hope you don't mind me using a reduced version of your jpeg for my Avatar for a while.
-
I need one too....
Preferably assembled, or at least with the SMD components mounted (soldering SMD just isn't my thing).
ATX form factor would be nice, but the original design is ok too.
-
Posting a image link to a site that requires us to register is a waste for those (like me) that don't want to bother registering at that site. Can you upload the image here?
@amigadave
You don't have to register to download the file, just click the free button down below :-D
-
Well done to convert the schematic.
I'll try to answer all of your questions/observations.
* The +1.25V and +2.5V can be connected to +3.3V to save power. Due P_heat=U_drop*I.
The LM1117 has a dropout voltage of 1.3V, so the +2.5 must be connected to +5V to be able to regulate properly.
Besides, power dissipation will be the same; it doesn't matter if the heat is dissipated in +3.3V regulator or in the +2.5/+1.25V regulators.
* Howcome the many connections to the fpa isn't in numerical order?
The connections are made that way to ease the PCB layout.
* What is the value of C37 100uF/6.3V ?
Yes.
* Why is R42 present?, it's just sitting between +1,25V and GND.
R42 is to provide a minimum load for the LM1117. Just to be sure.
* Howcome (54,55,56) M1,M0,M2 is connected to VccAUX, doesn't it need 3.3V?
No, they need +2.5. All config pins on the spartan do. If you want to connect them to +3.3V, proper current-limiting should be applied.
* The 'SPI_DOUT' from the fpga and SPI_DOUT from the sdcard (via 1k) can drive eachothers output, maybe this can be resolved with an or gate or such instead?
A gate would be better. During normal operation, the SPI out of the MMC card is tristated by disabling the MMC, the same goes at the FPGA side (it's a tristate port). This way, they can share the SPI bus.
* Maybe the keyboard and mouse can share connection?, I've seen that on laptops asfair (saves i/o).
That is possible, but you still need 4 I/Os. It does save a PS/2 connector though.
* There's no over/under potential protection for keyboard or mouse?, like video output have. Howcome?
It is a minimal design. The spartan has built-in ESD protection that this design relies on. However, monitors and TV's are notorious for blowing up video ports so extra protection has been added there!
* Maybe the rs232 output of mcu and fpga can be shared without a jumper by setting an AND-gate between. Because the default level for rs232 is "1". And thus any transmission will produce "0", and if the other maintain the rs232 "1" resting state. IT should work.
Yes, no problem. However, all debug output of the PIC has to be disabled as not to interfer with the FPGA serial output.
Dennis
-
...and now we are mentioned on reddit.com (http://programming.reddit.com/info/2btgo/comments).
-
CD32Freak wrote:
Posting a image link to a site that requires us to register is a waste for those (like me) that don't want to bother registering at that site. Can you upload the image here?
@amigadave
You don't have to register to download the file, just click the free button down below :-D
My Bad :-(
-
I am ready to buy at least one, if not two boards. Preferably the original size of 12cm square.
Then sir, you might as well get what you want (http://aros-exec.org/modules/newbb/viewtopic.php?post_id=18028#forumpost18028)! ;-)
At the same time contributing for the openness of it by directly contributing to the Team AROS Kickstart II bounty.
-
amigadave wrote:
Posting a image link to a site that requires us to register is a waste for those (like me) that don't want to bother registering at that site. Can you upload the image here?
Any tip on a better site than rapdishare.com is appriciated. I find them a bit messy. But just took what I knew at the spot when I was tired.
-
The assembled PCB would be fine for me!
My original Amiga 500 Plus broke down on me some time ago...
Anyone knows if I can recover the Kickstart?
-
take a look on the amigakit.net about the kickstart roms...
-
Here is a crazy thought.. Would it be possible to add more than one CPU on future revisions?
What about an DSP, like on the Delphina cards?
-
I'm also interested in a minimig in any shape. Fully assembled though if the price is right.
BR
Jörgen
-
Would it be possible to add more than one CPU on future revisions?
What about an DSP, like on the Delphina cards?
The neat thing with FPGAs is that you can do just about anything you want, so long as there is enough room in them for your logic design and enough pins to interact with the outside world. You'd need to add some control logic to the FPGA allowing it the capability to arbitrate and talk to the DSP properly, and assign some FPGA pins to wire up to the DSP on the PCB. Perhaps some of that could be shared with the 68000 bus, perhaps not. It sounds like there's cocnern that the largest pincount QFP style package which is easier to work with does not have enough pins for an IDE port and other things in addition to Dennis's design features. Larger pincount versions of the FPGA exist, as do larger logic capacity FPGAs. Unfortunately the larger pincounts come in BGA, which is much harder to work with, and it's expensive to send out for assembly by a proffessional company. And larger logic capacity FPGAs are just freakin expensive chips. You of course could look for some balance in there, but you'd at least need more pins and thus BGA version to add much more control signals or address/data bussing, even if the additional logic design fits into the same FPGA.
-
I would like 2 PCB with parts in kit, unsoldered.
-
I would like 2 pre-assembled and tested.
-
freqmax wrote:
Any tip on a better site than rapdishare.com is appriciated. I find them a bit messy. But just took what I knew at the spot when I was tired.
imagebay.com works for me.
-
I'd like to have at least one assembled and maybe a kit.
-
Put me down for two Minimig v1.0 please.
One fully assembled and one only with SMB.
I'd like it in a mini-ITX or nano-ITX *compatible* form. So if the current 12x12 cm design could just be fitted with wholes in the correct places I would be happy.
-
Count me in for a kit. Ports on ribboncable sounds good :)
-
I'd like to have at least one assembled one. No soldering for me, please...
-
I would like one Minimig, assembled please.
And as small as possible, doesn't matter if it fits a standard tower as I would do a custom one anyway.
What sreenmodes are supported?
-
Troels_E wrote:
I would like one Minimig, assembled please.
And as small as possible, doesn't matter if it fits a standard tower as I would do a custom one anyway.
What sreenmodes are supported?
Dennis' design has a jumper to switch from 15kHz to 31kHz RGB output so it will work on both old Amiga monitors like the 1080 and 1084, or will work on newer VGA monitors. I think the timing is PAL only at this point, but hope that will change to allow NTSC output from some boards, or switchable back and forth between PAL and NTSC. The software mimics an OCS A500, so display resolutions should be in the same range as an original PAL A500 (since I am in an NTSC display country I can't remember the exact numbers for PAL displays at the moment).
It is great that Dennis built in a scan doubler so modern monitors can be used. Some members are asking for composite output so TVs can also be used.
-
After thinking about it for a bit, I'd take a Minimig kit. Sign me as interested.
The NTSC timings would be a huge benefit to me, though having it be switchable would be even better.
-
Is it possible to generate both pal and ntsc timings using the same crystal? By using the clock management circuits on the spartan fpga.
I'm interested in at least one PCB.
-
It could be possible with a 27 MHz XTAL.
See the page:
http://www.opencircuits.com/Minimig_NTSC
Here is the computation:
27,000,000 / 429 x 455 = 28,636,363 (NTSC clock, 0 ppm)
27,000,000 / 432 x 454 = 28,375,000 (PAL clock, 6 ppm)
The DCM from the Spartan 3 accepts only coefficients from 1 to 32 :-(.
The Cyclone III accepts coefficients from 1 to 512.
The NTSC clock generation can be simplified to:
27,000,000 / 33 * 35 -> would fit into 2 cascaded DCMs.
We are less lucky with the PAL clock generation:
27,000,000 / 216 * 227 -> 227 is a prime number !
-
amigadave wrote:
It is great that Dennis built in a scan doubler so modern monitors can be used. Some members are asking for composite output so TVs can also be used.
Talking about that (and I don't mean to sound ungrateful) I would have thought Composite/S-Video would have been a necessity on this thing, considering how console-esque it is.
Anyone know how easy this would be to integrate.. or even standard Amiga-RGB for us Scart users?
-
FrenchShark wrote:
It could be possible with a 27 MHz XTAL.
See the page:
http://www.opencircuits.com/Minimig_NTSC
Here is the computation:
27,000,000 / 429 x 455 = 28,636,363 (NTSC clock, 0 ppm)
27,000,000 / 432 x 454 = 28,375,000 (PAL clock, 6 ppm)
The DCM from the Spartan 3 accepts only coefficients from 1 to 32 :-(.
The Cyclone III accepts coefficients from 1 to 512.
The NTSC clock generation can be simplified to:
27,000,000 / 33 * 35 -> would fit into 2 cascaded DCMs.
We are less lucky with the PAL clock generation:
27,000,000 / 216 * 227 -> 227 is a prime number !
Could you explain your equations a little bit more?
27/4.433619 = 6.0898331 ie divide by 6 ..?
-
Talking about that (and I don't mean to sound ungrateful) I would have thought Composite/S-Video would have been a necessity on this thing, considering how console-esque it is.
An Analog Devices AD723 could be used to generate both Composite and S-Video output from the MiniMig's RGB.
Probably add about $6-7 to the cost of things.
-
Adding composite/s-video output would take up precious board space. You need several components in addition to the encoder chip, including some large electrolytic capacitors, and possibly one or two clock sources.
-
narmi wrote:
Adding composite/s-video output would take up precious board space. You need several components in addition to the encoder chip, including some large electrolytic capacitors, and possibly one or two clock sources.
The AD723 does not require output capacitors and the clock can be supplied by the FPGA.
-
Anyone know what the max size ROM a minimig can use?
Dammy
-
It currently allocates 512k out of 2M as kickstart memory. With writes disabled. Any change to this can be done by changing fpga configuration.
-
freqmax wrote:
Could you explain your equations a little bit more?
27/4.433619 = 6.0898331 ie divide by 6 ..?
28,636,363 Hz and 28,375,160 Hz are the base frequencies of an NTSC and PAL Amiga respectively.
As discussed before, they can be generated with a single 27 MHz crystal.
Why did Commodore used these frequencies ? Because :
- the PAL color clock is 5/32 of 28.375160 MHz (i.e. 4.433619 MHz)
- the NTSC color clock is 1/8 of 28.636363 MHz (i.e. 3.579545 MHz)
-> That way, you have got a video compatible computer.
Dennis used a 4.433619 MHz crystal to generate the 28.375160 MHz and all the sub-multiples.
I guess he did that because the 28.375160 MHz crystal was not available.
-
Oh, add me to the list of the interested. Form factor doesn't matter. Original would suffice. I'm hoping by the time they're ready I'll be out of my financial woes.
-
I want minimig with PAR: so I can play 4 player adapter games.
And I need a hard drive so I can use Deluxe Paint 5.
And I need it to work like my REAL Amiga 500 which REALLY lets me switch back and forth from a NTSC to PAL display in REAL time anytime I want. A REAL Amber chip happily deinterlaces a 50hz or 60hz signal so I don't get what the problem is. A REAL Fat Agnus chip instantly switches to PAL or NTSC by setting the correct bit. If minimig was REALLY hardware compatible then it would work already.
-
@JosephC,
Minimig's creator never claimed that it was "hardware compatible" with all A500 chips, just that he was trying to make it able to run "most" OCS Amiga games. I don't think he ever intended it to be a complete replacement for Classic Amiga computers. It was just a fun challenging project to see if it could be done. See Jens "A-Clone" project, which is more like what you are asking for, a complete hardware replacement of the custom chips on an A500. Jens demonstrated it at AmiWest last year where he had completed the replacement of 2 or 3 custom chips with 100% accurate function by plugging his PCBs into the chip sockets. Once he has completed the designs, I believe his plan is to put the replacement designs into silicon and we will finally have a source for replacement custom chips for the Amiga and possibly a way forward to some enhancements to the original Amiga designs.
I am writing all of this just from my questionable memory without going to Jens website to confirm, so take it with a grain of salt and go check out what Jens is doing for yourself. Minimig is a completely different project with a different purpose.
My 2 cents,
-
Very interested in an assembled board.
-
i'd be interested in two or three populated or unpopulated, and add my vote for modular ports.
-
I think the unfortune fact is that there are no free pins for a parallel port on the fpga.
-
@amigadave
See Jens "A-Clone" project, which is more like what you are asking for, a complete hardware replacement of the custom chips on an A500. Jens demonstrated it at AmiWest last year where he had completed the replacement of 2 or 3 custom chips with 100% accurate function by plugging his PCBs into the chip sockets. Once he has completed the designs, I believe his plan is to put the replacement designs into silicon and we will finally have a source for replacement custom chips for the Amiga and possibly a way forward to some enhancements to the original Amiga designs.
I am writing all of this just from my questionable memory without going to Jens website to confirm, so take it with a grain of salt and go check out what Jens is doing for yourself. Minimig is a completely different project with a different purpose.
My 2 cents,
Correct me if I'm wrong but I don't think it was Jens intention to create replacement custom chips for the Amiga from the Clone-A project :/
(though admittedly my memory may be a bit off about this)
From what I remember reading in the article from Total Amiga a few months back, it sounds to me like the aim of that project was more along the lines of a Amiga in a joystick and/or a mini-itx board with scan doubling/flicker fixing built in.
Either way it goes I personally plan to support both the Minimig project and Clone-A. Though I am really[/] excited at the prospect of what Jens might create in a Clone-A mini-itx format :-D
-
techie wrote:
Correct me if I'm wrong but I don't think it was Jens intention to create replacement custom chips for the Amiga from the Clone-A project :/
(though admittedly my memory may be a bit off about this)
From what I remember reading in the article from Total Amiga a few months back, it sounds to me like the aim of that project was more along the lines of a Amiga in a joystick and/or a mini-itx board with scan doubling/flicker fixing built in.
You're correct it was never the ultimate goal to replace the custom chips, but it WAS one of the stages.
I remember at one point cycle-accurate custom chip replacements were being showcased in an original A500.
This is how the move forward to recreating and improving the original Amiga chipset is being done!
Both the Minimig and Clone-A projects are interesting and serve different purposes. However the fact of the matter is we have only one completed so-far and that is Dennis' Minimig.
I'm sure Jens will come through though - one of the few good guys left.
-
I'll buy a completely fabricated one, no matter the form factor...
-
so whats the status? did we ever get a price each and a delivery date?
-
Little update here, Dennis is updating the PCB afterall to V1.1, we decided it was best if he did it since it'd take too long for me to recreate it in Eagle and don't have a gerber capable editor.
Patience is key here as he is very busy with real world commitments. Once I have the new PCB though, I'll send it to my guy in China and from that point it will be two weeks before the boards arrive with me to test.
I'll keep you guys posted.
-Joel
-
Id be very interested in 1 assembled board, depending on final pricing, please.
-
I'll buy a few...assembled and in kit form.. :-)...
Im pretty handy @ soldering chips on boards..
Firewire ports?? hehe
Rich
ny
-
I used Eagle at one time, very cool and easy to use program.
Please, can you put all the ports on one side of the board?
It would make it easier to stick in some ready made computer case.
-
Xenepp wrote:
Little update here, Dennis is updating the PCB afterall to V1.1, we decided it was best if he did it since it'd take too long for me to recreate it in Eagle and don't have a gerber capable editor.
Patience is key here as he is very busy with real world commitments. Once I have the new PCB though, I'll send it to my guy in China and from that point it will be two weeks before the boards arrive with me to test.
I'll keep you guys posted.
-Joel
Great news Joel, I don't know how I missed your last message earlier.
Can you reply with a brief summary of what changes are being made for the v1.1 Minimig PCB? Also, can you answer the question which form factor is Dennis using for v1.1, is it still 12cm by 12cm? Will your assembled boards have the ports Dennis used, or pin headers, or will we have a choice of either? Do you have a price for the following:
1) Bare PCBs
2) Bare PCB w/kit of all parts needed to complete a Minimig
3) Assembled Minimig boards with all parts installed
Thanks for taking this on and getting these boards manufactured.
-
I have summarised issues that may need attention:
http://www.opencircuits.com/Minimig_Board_v1.0_documentation
-
i am very interested in an assembled board.
and i am sure some of my friends are too.
-
I'm only interested in a bare PCB, so I can put it in a frame :-D
-
Now that is a cool idea :)
-
I'd like one as well, fully populated.
-
I would like to have the cpu and other chips in sockets, so that they can be switched if they fail for some reason.
Is it possible to run kickstart 3.1 rom and OS3.9 on the minimig? Is it possible to change the CPU to a 030 or 040?
BR
Jörgen
-
Page 106 of 442:
http://www.freescale.com/files/32bit/doc/ref_manual/MC68040UM.pdf?fpsp=1
A quick count gives 43 ctrl + 32 adr + 32 data = 107 i/o pins.
So there's at least 27 new pins to connect. And the current setup only have 4 free. Thus a serious reorganisation with a southbridge solution or a enterpoint/uk PGA socketed solution would be needed.
So it proberbly can be done. But not as a plugin solution to the current setup.
As for kickstart/WB I think dennis have to tell ;)
Btw, found another assembly house: http://www.creationtech.com/printed_board_assembly.shtml
-
jonssonj wrote:
Is it possible to run kickstart 3.1 rom and OS3.9 on the minimig? Is it possible to change the CPU to a 030 or 040?
I would imagine that Kickstart 3.1 would work, as long as you had the kickstart image. I doubt OS3.9 would work though, doesn't it need at least an 020? Plus AFAIK there's no hard drive support on Minimig yet.
Anyone know if there are plans to add hard disk support? I know there might not be enough pins for an IDE interface (my preferred option, with the low price of Compact Flash cards and adapters) but UAE hardfile support would be much appreciated :-)
Another thought - if hard disk support was added, could the 68000 be replaced by a 68010? I understand that WHDLoad works better with 68010+.
Matthew
-
UAE hardfile support is plain verilog hacking :-)
Though the MCU would need some additional code to support write operations.
As for replacing the cpu, it's mainly a question about weather the number available fpga i/o's are enough.
-
freqmax wrote:
As for replacing the cpu, it's mainly a question about weather the number available fpga i/o's are enough.
I believe you'll also have to deal with the fact that other 68k family CPUs use 5v I/O while the FPGA uses 3.3V.
As to IDE support... has anyone considered SATA? It needs fewer pins.
BTW, I'm interested in an assembled board, if the ballpark cost mentioned earlier can be met. At my age, I can barely see those 0.5mm pins, let alone solder them. ;-)
-
JimS wrote:
I believe you'll also have to deal with the fact that other 68k family CPUs use 5v I/O while the FPGA uses 3.3V.
Does this mean that 68010 isn't an option? I know things get more complicated with later 68k CPUs but IIRC the 68010 could be used as a drop-in replacement for the 68000 in A500s. If it is possible to use a 68010, would there be any disadvantage? I'm wondering if there would be any value in having the CPU socketed so users could choose between the two CPUs.
As to IDE support... has anyone considered SATA? It needs fewer pins.
I had wondered about SATA, but I thought it might be too complex at this stage.
What I would like to see is USB for keyboard, mouse and joystick, and SATA for storage (although HDF support would be enough for me.) The reason being that PS/2 and IDE are becoming obsolete, it makes more sense to me to make it compatible with the most up-to-date technology.
Matthew
-
The datasheet for MC68040 I looked at says Vcc 0V-7V. So I guess 3.3V operation is ok. S-ATA is nice but requires either an transceiver/PHY => more pins. Or an FPGA with S-ATA port => Virtex-4 FX => Significantly more cost & use of bga packages.
-
freqmax wrote:
The datasheet for MC68040 I looked at says Vcc 0V-7V. So I guess 3.3V operation is ok. S-ATA is nice but requires either an transceiver/PHY => more pins. Or an FPGA with S-ATA port => Virtex-4 FX => Significantly more cost & use of bga packages.
Ahh.... it's always something. ;-) I'd *guess* that if you ran a 68040 at 3.3V, you'd need to roll down the clock speed to get reliable operation. But that's just a guess on my part. I seem to recall Dennis saying he picked this particular CPU because it was 3.3V and compatible with the FPGA. Be great if faster CPUs were an option.... either with an actual CPU, or another FPGA running a 68K core.
The folks at Schmartboard have announced BGA versions of their prototype boards. Last time I looked it wasn't shipping yet...
For the USB interface, how about the Vinculum chip? It has it's own CPU that runs a USB host stack. Interface is via either a UART, SPI or FIFO. I believe there's a spare SPI on the Minimig...
-
But would all of those pins necessarily need to be connected to the FPGA. Or could the FPGA be made to have the data lines, minimum address lines, minimum control lines. Since the additional address lines would be for extra memory therefore not need to go to the FPGA and the control lines would not need to go there also. Maybe set the MiniMig up the it has a header which connects to a CPU/Memory/Additional (IDE, ZORRO, etc). Just use the MiniMig core as the custom chipset replacement (sound, graphics, I/O). The MiniMig should still have the onboard memory but be used as ChipRAM and Kickstart. All other FastRAM would be with the CPU. Just a thought. Wouldn't be much different than some other Amiga upgrades (jumper wires everywhere)
-
JimS wrote:
For the USB interface, how about the Vinculum chip? It has it's own CPU that runs a USB host stack. Interface is via either a UART, SPI or FIFO. I believe there's a spare SPI on the Minimig...
That chip looks quite useful. From my scan thru of the data sheet, it could allow access to USB mass storage devices, USB mice and keyboards, and perhaps more. All with only 4 pins to interface it (in SPI mode).
The one thing that gets me though, is it claims USB 2.0, but is only capable of 12Mb/s. That makes it USB 1.1 not 2.0, not that it matters much since the CPU can't handle USB 2.0.
-
ivier:
I think it's possible to get usb by simple using two pins on the fpga, albeit maybe only usb1.1 at 12Mbps. For usb2.0 at 480 MBps it may need a PHY chip.
But any cpu to shuffle data will likely slow things down a lot.
Seeing a lot of suggestions for Minimig. I think actually USB makes most sense. As one can get keyboard, mouse, joystick, mass storage, rs232, soundcard, ethernet (netboot!), etc.. from ONE usb port via an hub.
As for USB2 it's a real FAKE, your equipment can be USB2 complaint but unable to perform 480 Mbps operation!
JimS:
Actually to enable higher speed, voltage is lowered. This is related to flank times and P=U²/R ie the energy dumped during switching.
-
freqmax wrote:
The datasheet for MC68040 I looked at says Vcc 0V-7V. So I guess 3.3V operation is ok.
There is a specific 3.3V version of the 68040 it is called 68040V.
If you want to save some pins on your FPGA, you can design the board like on a real Amiga : the Chip-RAM bus and the Fast-RAM bus are separated by buffers. Agnus takes care of the Output Enable of these buffers.
-
A USB Host Controller usually connects back to a host processor on a high bandwidth bus (parallel CPU bus, PCI, clock port, etc ..) in order to be able have high bandwidth transfers. The Vinculum parallel FIFO interface mode with 8 bit bi-directional data bus and simple 4 wire handshake for data I/O and command monitor interface would work well interfaced back to the Xilinx FPGA or to the 68K bus. The issue with USB host on the present Minimig PCB would be the speed of a serial SPI connection from 68K/Xilinx_FPGA to a Vinculum USB HOST controller that would most likely limit the bandwidth/transfer rate. Data sheet has Vinculum SPI port max SPI Clock = 10Mhz => 10Mbits/sec Max SPI transfer rate; thus, a Vinculum SPI connection in any application will have a limit on the SPI tranfer rate of 10Mbits/sec or maybe lower due to possible signal integrity issues on a 2layer PCB.
Also, USB 2.0 compliant device does not have to support USB 2.0 High speed (480Mbits/sec) transfer mode. A USB 2.0 compliant device such as Vinculum only have to support "Basic-Speed" of 1.5Mbits/sec (Low speed) & 12Mbits/sec (Full Speed) while ignoring the high speed handshake from a USB 2.0 High speed (480Mbits/sec) device so that the High Speed device (480Mbits/sec) will kick back down to the old 12Mbits/sec (Full Speed) speed mode.
:-)
-
JimS wrote:
As to IDE support... has anyone considered SATA? It needs fewer pins.
The FPGA in the MiniMig currently has 4 unused I/O pins. That's more than enough to add IDE.
-
Count me in for original (1.0 or 1.1, wathever) 12x12 kit (not assembled, only the board and parts). :cheers:
-
http://www.fujiyama.cn/Company.asp
Found the above factory in Shenzen China.
They sell some curious products.
Most intersting is a nifty gadget... Get this, a classic Nintendo 16,8bit,gameboy emulator,AVI,WMV,ebook,picture viewer & camera features, 1 gig ram, sd card slot, FM radio, usb2 with a TVT monitor, and TV output.
It's intersting to note they also offer industrial services with payment terms clearly listed.
Seems if larger numbers with a complete system are ever considered an outfit like this can do the job. From my expereince factories like this have cheap labor and technical expertise to pull off some of the things the community is looking to do like usb, faster CPU, more ports.
-
Has anyone tabulated the list of interested people? Are we going to get some kind of email if this run actually happens?
-
Well, the authors of this thread could set up an "official" poll, about how many you would like to get, and then sign with an e-mail addy.
-
I'm surprised that after more than 2 weeks since Dennis released the Minimig sources, no one has even produced a bare Minimig board. So what's your excuse? Amiga fatigue? :lol:
-
How is software compability for running with a 68020, 68040 or 68060 instead of 68000 ..? esp for games.
If there's any minor incompabilities is it realistic to let the fpga hide those from the software. As the FPGA sits in between the cpu and ram ..?
CD32Freak:
I think people want some hardware misses patched, like pin81/19 swap, or-gate for the sd/mmc interface etc..
Then there's the issue of finding an pcb maker that will assemble the boards for a reasonable price. And that you can rely on.
-
I have actually had two bare MiniMIG 1.0 PCBs produced for myself based on Ver1.0 gerber files/layout, and I am gathering up the parts in order to assemble them by hand.
:-)
-
Any particular parts that are hard to source ..?
Any tip on FPGA sources except digikey?
-
freqmax wrote:
How is software compability for running with a 68020, 68040 or 68060 instead of 68000 ..? esp for games.
If there's any minor incompabilities is it realistic to let the fpga hide those from the software. As the FPGA sits in between the cpu and ram ..?
It would probably be easier to implement some sort of hard drive/file support and run the games through WHDLoad.
-
My idea was to use a "better" cpu for other purposes than to run WHDLoad.
WHDLoad is just a bonus :-)
-
@jkonstan
Very good, that's more like it :-D
-
DigiKey is the source for a fair portion of my parts for my MiniMIG. I may also buy some parts from Mouser Electronics & Jameco. Digikey & Ebay are options for the Xilinx Spartan3 208QFPs FPGA.
:-)
-
Is it possible to use bluetooth with Minimig? It would be nice to have a keyboard and joystick without the wires...
-
Is it possible to use bluetooth with Minimig? It would be nice to have a keyboard and joystick without the wires...
Oh please... Will this ever stop?
"Bluetooth with Minimig?"
"68060@80Mhz with Minimig?"
"4 PCI-Express slots?"
"Integrated popcorn machine?"
Shouldn't we first bother about the _basics_, i.e. getting a small production run organized somehow?
-
So when is this happening?
My money is gathering dust :(
-
jkonstan wrote:
I have actually had two bare MiniMIG 1.0 PCBs produced for myself based on Ver1.0 gerber files/layout, and I am gathering up the parts in order to assemble them by hand.
:-)
What??? You did a production run of ONLY 2 Minimig PCBs with well over 100 members here waiting and asking for someone to produce a board, either assembled or bare, for them?
Congratulations, but I must say I am disappointed that you did not ask for the money to run more PCBs for members here. I will gladly pay anyone that is willing to produce a Minimig for me, as I do not feel confident in taking on such a project myself, without help from those more knowledgeable and skilled than myself. I am good with a soldering iron, but would prefer to obtain an assembled board that is an exact copy of Dennis' original v1.0 design with any v1.1 fixes that can be integrated.
So, if there is anyone else that is preparing to produce another run of Minimig boards? I am ready to buy one or two.
-
Dave, organising group buys on boards is not for everyone and that sort of thing can and will lead to headaches. Can't blame the guy for getting the boards for himself rather than worrying about people not paying/changing their minds/boards getting lost in the post/more complaints/complaints when people put theirs together and it doesn't work. etc.
Having said that, if anyone knew someone in or near London who could make and assembler a board or two for some ££ I would love to hear about it!
-
Amigadave,
I fabbed these two blank PCBs so that I could try to help with the development/enhancement of MiniMIG. The two Ver 1.0 blank boards were fairly pricey (over $25/each without electrical test) as compared to some of the other better efforts that were already on going to produce a bunch of cheaper blank PCBs. An AROS related individual/group was going to produce a batch of the original Ver 1.0 PCBs for sale, and the original starter of this thread had the mass production covered for Ver 1.1 of this updated PCB artwork. I thought that would have everyone pretty much covered without someone else mucking up the waters. I really do try to stay focused on the technical side of things.
:-)
-
Hi,
To anyone making a PCB or updating the design I can add a few suggestions.
Disconnect the HSWAP_EN (pin 206) from 0V.
This stops the FPGA driving all I/O during configuration and should negate the need to disconnect SPI_DOUT from pin 81.
If you need any pins pulled to defined states, you can edit the UCF file and recompile.
If you are buying the XC3S400-4PQ208C FPGA, it is easier to obtainm the RoHS compliant replacement, the XC3S400-4PQG208C!
Also the earlier rev BFQ silicon has a number of bugs that could possibly affect the design, in particular relating to the DCM (Digital Clock Manager). Newer parts are much better. Take a look at the Errata for the Spartan 3.
Adding a 6 pin header for the PIC would allow the cheap PicKIt 2 programmer to be used to program the PIC18LF252, providing you add series resistors to RB6/7.
www.pcbtrain.com (http://www.pcbtrain.com) 2 layer, 12cm x 12 cm, plated through hole PCB, with solder resist and silksreen on one side, quantity of 50, £9.20 each including VAT.
Hope this helps.
So where can I order my Minimig kit from? :-)
Ian
-
I think what is required to satisfy most people. Is a board with various fixes (or-gate to sd/fpga/mcu, fpga spi_dout pinswap etc..). And this board should have at least all large smd components assembled.
I have compiled a list of fixes that should be considered:
http://www.opencircuits.com/Minimig_Board_v1.0_documentation
-
humppa wrote:
Is it possible to use bluetooth with Minimig? It would be nice to have a keyboard and joystick without the wires...
Oh please... Will this ever stop?
"Bluetooth with Minimig?"
"68060@80Mhz with Minimig?"
"4 PCI-Express slots?"
"Integrated popcorn machine?"
Shouldn't we first bother about the _basics_, i.e. getting a small production run organized somehow?
total agree with you.
the more important to the project its the basics', then came the extras.
-
I want a espresso machine (Firewire!) on mine!
-
rkauer wrote:
I want a espresso machine (Firewire!) on mine!
Ok, serious now. Please, let's concentrate in deliver the plain, original board (or the 1.1 version, whatever).
-
I have bluetooth keyboards on all my computers, I hate cords, they serve no useful purpose unless you want to strangle a cat.
-
jkonstan wrote:
Amigadave,
I fabbed these two blank PCBs so that I could try to help with the development/enhancement of MiniMIG. The two Ver 1.0 blank boards were fairly pricey (over $25/each without electrical test) as compared to some of the other better efforts that were already on going to produce a bunch of cheaper blank PCBs. An AROS related individual/group was going to produce a batch of the original Ver 1.0 PCBs for sale, and the original starter of this thread had the mass production covered for Ver 1.1 of this updated PCB artwork. I thought that would have everyone pretty much covered without someone else mucking up the waters. I really do try to stay focused on the technical side of things.
:-)
I didn't mean it to sound like I was jumping down your throat.......just disappointed that a larger production run of Minimig boards has not been completed yet.
I meant it when I wrote congratulations. I am happy that you have two boards to work on and I am also glad that you appear to be working on improvements to the original design and/or code.
Hopefully it won't be too long before some assembled boards are available for general distribution (sale).
-
persia wrote:
I have bluetooth keyboards on all my computers, I hate cords, they serve no useful purpose unless you want to strangle a cat.
Wireless does not always == bluetooth and also a wireless dongle won't strangle any cats ;) ;)
-
Xenepp wrote:
Little update here, Dennis is updating the PCB afterall to V1.1, we decided it was best if he did it since it'd take too long for me to recreate it in Eagle and don't have a gerber capable editor.
Patience is key here as he is very busy with real world commitments. Once I have the new PCB though, I'll send it to my guy in China and from that point it will be two weeks before the boards arrive with me to test.
I'll keep you guys posted.
-Joel
Just wanted to see where we are today? Has Dennis done the fixes yet? By no means is the meant to be a whine. I'm sure will all be happy to accept that it's done when it's done. Just wanted to see if the design stage was complete or not.
-
I want one!
Is this the right place to notify? Am I counted now?? :)
I'll have one almost no matter the price ;-)
Just a question, will it be possible to firmware upgrade it later with aga support and things like that? Or is it fixed in stone in terms of what it can emmulate? Thinking that it has a 68000 cpu, and not an emmulated cpu or a 020.
Not a biggy, since we hopefully one day also can by Clone-A... :)
-
An novel idea for someone enterprising would be to get a batch of the pcbs fabricated, buy and equal amount of chips necessary and sell the kits on ebay...
That is if there is not going to be a mass production run as was originally planned and discussed in this thread, at least it would make these somewhat available...
Just a thought...
-
@arnljot,
No AGA and no 68020 or faster until a revised board design is completed. Could be two weeks, or two years before that is done, if ever.
@trip6,
I doubt that eBay is the right place to market this kind of project. Right here on A.org and perhaps also on AmigaWorld.com will get you 90%+ of all potential buyers for such a kit.
Now to anyone that is nearing the completion of a production run of the original v1.0, or the revised v1.1 Minimig PCBs, I would like to be put on a list of buyers (get in line to make sure I get one of the first produced) with or without the chips assembled, but preferred with chips on board.
Thanks in advance,
David
-
Though, to my question.
Is there an offical que, and if so - is this it?
-
Thats my point I thought that a "official list" was being compiled from this discussion thread. I myself asked to be put on the list for two fully assembled boards... Now that idea has seems to have gone to pot... Or has it? The person that started the discussion thread please respond.
-
It appears that more than one individual is producing boards and to my knowledge, no official list is being compiled yet by the persons responsible for producing those efforts.
I guess when they decide to let us know more, they will do so.
Until then, unless we (you or I) intend to try to produce any boards ourselves, we will just have to wait until someone with boards ready to sell lets the rest of us know how we can buy them.
I am hoping that I can get on a list so I don't miss the opportunity to get a Minimig, but if I can't get on a list, I will just have to keep watching and try to contact any persons that finally do produce something.
Patience........it's an Amiga kinda thing.
-
To save bogging down the announcment thread, I've created this one to get an idea of what the majority of people want from the PCB.
I'm not sure about how many will be made yet, I guess if you post here that your interested it'll give me an idea.
I need to translate the PCB to a format my guy in china can work with so while I'm at it, I may as well configure it they way everyone wants.
There's been talk of making the ports (joystick, power etc) modular, i.e. on ribbon cables so they can be mounted however you like in a custom case, how does everyone feel about that?
I'm open to other suggestions like shape of the PCB for mounting in HDD cases (if possable). Just throw around some ideas and I'll amend this first post to collect them in a list or sorts.
A VERY rough idea of cost at this point is about £40 for an assembled PCB and £20 for the kit (PCB plus parts to solder yourself). I'd imagine assembled PCBs would be however be pretty limited ;)
Buddy..... how far along is the project now? Are you still in the redesign stage for the port on ribbon cable idea? A progress report would be cool.
Me and my friend both want one. I will take it in ANY form. ANY formfactor.
EDIT: I'm only interested in a system that only requires a case.. I dont want to solder anything. Will many of these boards be assembled already?
-
There's been talk of making the ports (joystick, power etc) modular, i.e. on ribbon cables so they can be mounted however you like in a custom case, how does everyone feel about that?
-----------------------------------------------
I missed that part.. that would be a very very good idea :)
-
Users have talked about putting a MiniMig in a real Commodore Amiga case. Is it possible to utilise its keyboard as well somehow? I'm thinking the easiest method would be building an adapter for the PS/2 keyboard port, and plugging the Commodore Amiga keyboard into that. That way, no changes to the MiniMig design need to be made.
The ports connected to the ribbon cables could match up to the port holes in the case, and the unused ports could be blocked off with a thin sheet of metal (just for neatness)..
Then the other issue is the disk drive. It sounds like too much trouble to add a disk drive controller to the MiniMig design. And I think many people will like being able to stick their entire games collection on a single memory card anyway.
I wonder if MiniMig cases could be made, based closely on real Commodore Amiga cases..
-
I'd have updated the thread sooner if there was something to update. ;)
I've not had word from Dennis yet but I'm sure he's just busy. I'll not pester him as he is busy at the moment so we just need to play that old waiting game folks. You know it's worth it :)
I can't see it being that much longer anyway but if there are folk who just can't wait maybe I'll do a small run of the v1.0 PCBs.
There is no "list" yet as nothing's set in stone but if there are folk who want the raw v1.0 PCB (no components) post here and if there's enough I'll get some of those manufactured. I'll post later today/tomorrow as to exactly how many need to be interested.
-Joel
-
PS2 ports are a waste, just have USB ports then you can use them for anything.
How about a case that looks like a scaled down Amiga case so it fits a 13x13 board?
-
@jahc
I wonder if MiniMig cases could be made, based closely on real Commodore Amiga cases..
I don't think this would make a lot of sense. The Minimig is great because of the form factor. It's a great device to carry around or to use in you living room hooked up to your television. I doubt many would be interested to have a case with the size of a real Amiga case. Much too bulky...
Besides... why is it that people love talking about cases when it is still unsure how the product itself is manufactured? The last time this happened was with the pathetic Troika case design contest. So now we have 457 different case designs but not a single working Amy-mainboard. Great.
-
Well, I want one too...
I don't see much discussion of compatibility. Has any software been tested on it yet?
I would rather sacrifice compatibility for SPEED not size.
Someday I hope someone builds a faster minimig with a PCI video slot and a pcmcia slot. Boy I would be in heaven.
-
Ok, I've decided I will be doing a run of 8 V1.0 Minimigs to get something running and answer any questions you folks have about the hardware.
The PCBs will be un-populated (you need to supply components) but I'll do the needed patch on the PCB as instructed by Dennis.
It'll be about 2 weeks from now (once I have my funds gathered to pay for manufacture) and each PCB will be £20 + £2 postage in the UK and £5 worldwide.
I'm not accepting any payments until I've got the PCBs in hand and have tested one of them, however the first 8 to post here confirming they want one will get them. Consider it a pre-order.
As promised before, I'll be doing runs of the V1.1 PCBs once Dennis is done so if you want to wait for that, DON'T post that you want a V1.0 PCB (unless you want both)!
Joel
-
Count me in. UK based ASIC Engineer with SMT rework station ready ;-)
-
I'll take one and follow this up with a Pmail. If you can't or won't supply the rest of the components, can you provide an easy link to a list of ALL other components needed to complete the v1.0 Minimig? It would just save the rest of us the time of sifting through all the threads and messages to find it, or searching Dennis' site for the correct location of such a list (if Dennis has provided a comprehensive list on his site I have not seen it).
Thanks, and I hope I am one of the first 8 to reply. Can't you produce more in the run, as I am sure there are more than 8 that want to take you up on your offer?
-
Sign me up, I'll take one. I think that comes up to about $35US?? Solderstations and components are standing by. Now if I could just add a few more extra days to the week....
Plaz
-
£25 = $50 given recent exchange rates.
I'll be waiting for the V1.1 board myself, but I hope the V1.0 buyers get a lot of fun out of it.
-
arnljot wrote:
I'll have one almost no matter the price ;-)
Me too - within reason. What's the likely timetable for these??
-
I'll take a V1.0.
For those wondering, it is usually about $2US for each £. So, expect the price to be in the $50 range. I figure that if the V1.1 comes out before I get the V1.0 populated, the V1.0 will become a framed nostalgia art piece.
-
when you say we need to supply the components, what components would that be?
-
TheMagicM wrote:
when you say we need to supply the components, what components would that be?
It's just a bare pcb with no parts on it. You'll need to buy the fpga, cpu and other supporting components your self and assemble the device yourself.
-
aah ok.. I prefer a fully assembled board (I'd rather not solder myself). I'll wait for 1.1 if its fully assembled.
-
Hattig wrote:
£25 = $50 given recent exchange rates.
I'll be waiting for the V1.1 board myself, but I hope the V1.0 buyers get a lot of fun out of it.
Typically I use www.oanda.com to check exchange rates. Nice site. And of course I made a mistake and did a Euro<>Dollar conversion instead of Pound<>Dollar. Duh.
I plan to have much fun, oh and I hope it works too when I get done. :lol:
Plaz
-
TheMagicM wrote:
when you say we need to supply the components, what components would that be?
FPGA, resistors, caps, IO connectors.... any thing that is not a trace or circuit board :-P
Do we have a parts list yet or do we need to generate one from the scematics? www.digikey.com has one of the best sites/search engines for this type project and the prices aren't bad either. Works well for North America. Not sure if that's the case for Europe, Oz and other places.
Plaz
-
So whats the best PCB software to do the 1.0 patch with, I don't want to wait another year for the patch to be updated in the Gerber files.
What software did Dennis use to create the board with ?
Anyone in Australia know of a place to get boards done ?
-
Count me in for a v1.1 fully assembled. If you need any cash up front then let me know.
-
I dont know about you lot, but I am going to phone up sales of Motorola, Xilinx, Microchip etc. and try and blag some free samples.
-
Any idea when we can expect delivery on the assembled units? I know Dennis is making changes to the PCB but possibly before christmas, after christmas? And any further update on the cost of an assembled unit?
Cheers
-
TheMagicM wrote:
when you say we need to supply the components, what components would that be?
Regarding needed parts, have a look at
http://home.hetnet.nl/~weeren001/minimig.html
and
http://www.opencircuits.com/Minimig_Project
-
tonyyeb wrote:
Any idea when we can expect delivery on the assembled units? I know Dennis is making changes to the PCB but possibly before christmas, after christmas? And any further update on the cost of an assembled unit?
It looks like v1.1 is on hold for now until Dennis has done the updates. It could be soon but if not we can only assume he's busy.
The original poster is doing a run of v1.0 PCBs though.
-
Fixer wrote:
It looks like v1.1 is on hold for now until Dennis has done the updates. It could be soon but if not we can only assume he's busy.
The original poster is doing a run of v1.0 PCBs though.
Yeah but i want a complete one and preferably a 1.1 - dunno why - just newer seems better! lol
-
tonyyeb wrote:
Fixer wrote:
It looks like v1.1 is on hold for now until Dennis has done the updates. It could be soon but if not we can only assume he's busy.
The original poster is doing a run of v1.0 PCBs though.
Yeah but i want a complete one and preferably a 1.1 - dunno why - just newer seems better! lol
Not always! But I want both anyway. The original v1.0 is a work of original Amiga art and genius. The v1.1 boards will hopefully be even better, but I really like Dennis' PCB layout. It is so clean and compact.
It is really amazing to me that Dennis has accomplished, single handed, what no other company has taken the time, effort and initiative to complete! I know it was not easy for him to finish, but looking backward at what he has done it makes me wonder why no one had done it sooner.
My hat is off to Dennis. He will be remembered (at least by me) as the first to complete such a project and crack the code of the Amiga's custom chips. A remarkable feat that we all should be thankful for. Dennis you are a part of Amiga history now and forever, enjoy your 15 minutes of fame. :-D
-
I would buy a v1.0 but like I said, I dont want to do any soldering.
-
Another update on these.
Dennis emailed me today and yes, he's been busy but we should see V1.1 next week so if that's the case, I may as well just start taking pre-orders for that instead of V1.0 as I can't place the order until next week anyway.
If you are seriously wanting a Minimig PM me and I'll add you to the list of pre-orders. Don't worry too much if you don't get a look in this time as I'm sure I'll do another run at some point.
As I said before there are two flavors of order, bare PCB or pre-assembled, but the pre-assembled boards are going to be limited to 5 per run. There is no limit to bare PCBs obviously.
Once I get the new files from Dennis I'll give you guys a week to get your orders in then it's off to China! :D
It's getting close guys!
-Joel
-
As it goes, this thread's gotten pretty long, should I start a new thread for pre-orders so no-one misses out?
Admin?
-Joel
-
PM sent
-
Xenepp wrote:
As it goes, this thread's gotten pretty long, should I start a new thread for pre-orders so no-one misses out?
Start a new thread, take PM's, leave your email or home phone number - I don't care as long as I'm on that list :)
I'm still in for two pre-assembled boards. I have a few friends who are interested too, so it might even be five. I'll let you know the final number once we figure out how to pre-order properly.
-
PM Sent.
Preferably a pre-assembled, but if they are all taken, I will take a bare board.
-
All pre-order discussion should now be directed here:
http://www.amiga.org/forums/showthread.php?t=30997
Full details are outlined there and those who have PM'd me (already!) should check there to make sure they are still in and let me know.
-Joel
-
What will the v1.1 board improvements be ..?
* Or gate between fpga/mmc spi?
* Spi_dout fpga patch?
* Pin headers?
* DCM generated color clock?
* One or two SRAM chips?
* Video D/A resistor ladder linearity fixed?
Btw, is it possible to make the mc68000 behave like a mc68020 or such via fpga tweaks? (and make aga or such feasable)
-
Improvments are mostly fixes and part reduction from what Dennis tells me.
Swapping the 68k for an 020 is possable with changes to the schematic and the FPGA core. Both require lots of time and patience, but I would expect someone will have a go at making upgrades at some point.
-Joel
-
I will take a pre-assembled V1.1!!! I'll PM you NOW!!
:)
-
Xenepp:
I think you misunderstood me. My idea is weather it's possible to "fake" a mc68020 with a mc68000 using the fpga that sits like a spider in the wire network. Ie using only hdl (verilog) tweaks.
-
Sorry freqmax, I'm not following you :-?
To "fake" the 020 is folly. As soon as any software that needed it was run it'd crash, as to the software, the hardware would be "failing".
The only way 020 programs would run would be to impliment the hardware and reconfigure the design to match.
Unless you mean to just impliment the 020 exclusive features the the FPGA then have the 68k handle the rest, that could be possable I guess, but if it was I'd expect it to run real slow :-)
-Joel
-
Xenepp:
Well the idea was to just implement the 020 exclusive features in the FPGA. And let the present 68k handle the rest.
(makes me wonder how hard it's just to swap the current 68k to something else, pin compatible?)
Dennis:
Are crystals X1 and X2 operating in parallel or serial resonance mode?
For pcb v1.1 makers:
I have sorted the issues I have discovered and found mentioned on amiga.org here:
http://www.opencircuits.com/Minimig_Board_v1.0_documentation
There's a non obvious issue with the video D/A. The original resistor values cause a dent in the desired vs actual output value. The math and results are here:
http://www.opencircuits.com/Minimig_Video_d/a_resistor_ladder
My patch also make better use of the full signal swing available.
In short the resistor ladder should use 4220 ohm, 2100 ohm, 1050 ohm, 523 ohm (1% tolerance). In order to avoid this.
I also found that the output D/A also needs a translation table to match input data with correct output amplitude due how ohms law works. Anyone found this in the verilog sources? Amber & Agnus is likely candidates..
-
freqmax wrote:
makes me wonder how hard it's just to swap the current 68k to something else, pin compatible?
Only thing you could have possibly swapped it for would have been a 68010 but they dont make them anymore, and I doubt they ever made SE variants. 020+ wouldnt be pin compatible. With a few wires and a lot of patience I am sure you could wire on any CPU mounted A500/A600 accelerator.
freqmax wrote:
In short the resistor ladder should use 4220 ohm, 2100 ohm, 1050 ohm, 523 ohm (1% tolerance).
Q. Can you get SMT resistors of such wild values?
-
The thing is, you would need to alter the FPGA program even if the 020 was pin compatable since the architecture is still slightly diffrent and the rest of the hardware on board would need to know this, or be upgraded to cope with it.
A good example is the Sega Megadrive. I remember a few years ago, a few people were tinkering with it trying to squeeze more speed out of the 68k. They upgraded it with a faster chip and all kinds of wacky things happened in some games, most wouldn't even boot. That was still using a 68k.
-Joel
-
Apologies for this slightly OT post... Anyone in London able to help me assemble one of these things? I can solder but just barely. Wish I put more time into trying to improve things in that department! Would be good to have a big group Minimig buildathon :)
-
In the other thread it says preassembled will be limited to two per person. I think this should be 1. Then at least more forum members will get a chance at owning one. If i could solder well enough i would build it myself but as i cant i would like an assembled one.
-
alexh wrote:
Q. Can you get SMT resistors of such wild values?
Value - Digikey ID
523 - RR12P523DCT-ND
1.05K - RG20P1.05KBCT-ND
2.1K - RR12P2.1KDCT-ND
4.22K - RR12P4.22KBCT-ND
Package 0805, 1/10W, Thin film, 10 pcs/order asfair.
Total cost ~2 USD.
Actually when I did the math the thing I was looking for were the available resistor values. The most important is to check that Vpp-peak is lower than 0.7V at all times. So the values may be wild for humans. :-D But they fit digikey ordering and the equations.
As for improvements, how about wire the mcu spare i/o to another pin header? (just like it's done on the fpga)
http://www.opencircuits.com/Minimig_MCU_user_I/O
-
AROS could use some devs to bring AROS 68K out of unmaintained.
Dammy
-
Q. Can you get SMT resistors of such wild values?
The values listed are the "preferred value" for 1% resistors as defined by an EIA standard.
http://www.pc-control.co.uk/resistor-eia.htm
-
It would be nice if someone with fpga and pic coding skills added a "gayle" + hardfile support. It could work with the current design (even if hardfile would be a little slow) but could be useful with OS3.x :-)
-
Crumb wrote:
It would be nice if someone with fpga and pic coding skills added a "gayle" + hardfile support. It could work with the current design (even if hardfile would be a little slow) but could be useful with OS3.x :-)
As soon as MiniMig hardware arrives this will be one of the first changes. Should be viable.
Simultaneous Floppy Emulation and Hard File access might not be viable though.
We'll see.
-
I didn't want to put this on the pre-order thread, as it states it is just to track the number of people that want a Minimig PCB.
Joel, can you tell us why only 5 assembled PCBs are available from this first production run and an unlimited number of bare boards? Is it because you only have enough parts to complete 5 assembled boards and don't want to put up the money in advance to purchase more parts? Or, is it because you are assembling the boards yourself and don't want to commit to doing the work on more than 5 boards at a time?
Is there any way we can increase the number of assembled boards?
Is there anyone else here on A.org that has the abilities to expertly complete the assembly of Minimig PCBs and components to make assembled Minimigs, so those of us that are able to receive bare Minimig PCBs can send them to you for assembly and get the completed boards back in a week or two?
I would like two completed Minimig computers, so I am trying to find any and all ways to obtain my goal. Please send me a Pmail if you can help with the assembly of a completed Minimig.
I have a feeling there may be many others watching the Minimig threads that have similar needs for completing their projects.
-
alexh:
Why would not simultaneous floppy and harddisc (file) access work?
Btw, Write operation is not even implemented atm :roll:
amigadave:
I know some local companys that can produce a minimig from drawings. But they cost a lot more than china. Otoh they are reliable and will deliver. And I can personally inspect the companies.
One prototype one-off board would proberbly go for around 220 EUR. (way less for many boards in a batch)
Assembly is approx 428 EUR start cost, and 7.5 EUR per board.
Considering this I prefer the China route ;-)
-
Xenepp,
Does "pre-assembled" means "fully soldered and working" or will I need to do any soldering? I can put parts together but not soldering :) I melt everything.
-
amigadave wrote:
Is there any way we can increase the number of assembled boards?
Is there anyone else here on A.org that has the abilities to expertly complete the assembly of Minimig PCBs and components to make assembled Minimigs, so those of us that are able to receive bare Minimig PCBs can send them to you for assembly and get the completed boards back in a week or two?
What's involved? Is it just a matter of soldering the parts onto the board? I took surface mount soldering training classes years ago. I used to work in a computer factory applying engineering change orders under a microscope.
But I don't have any of that equipment now. I'd be willing to try and put one together as a hobbyist and see if if it works...If so, perhaps I can help out?
Fester
-
freqmax wrote:
alexh:
Why would not simultaneous floppy and harddisc (file) access work?
There is only one interface to the MMC through the PIC and they would have to share.
-
alexh:
One could implement a queue or semaphore that sends out one request to the MCU at a time. Floppy & Harddisc doesn't have to access the MMC simultaneously in the physical sense.
-
A lot of people are asking for 020+ CPU and AGA.
I just had a crazy idea : for the Minimig v1.2, let's put a real 64 pin DIL 68000 in a socket.
Any A500 accelerator board that plugs into the 68000 socket can be used !!
For the 3.3V <-> 5V translation, we can use Pericom chips like the PI5C3384.
It might be interresting to put also 512KB or 1024KB of flash directly hooked up on the 68000 bus with the Flash's chip select being managed by the FPGA. That will free up the full 2MB of SRAM for Chip RAM use only.
The AGA chip RAM is 4x faster than the ECS/OCS chip RAM (32-bit access with double CAS). The ECS/OCS chip RAM runs at 3.5 MHz or 280ns access time. If you have a 16-bit SRAM with an access time of 70ns or less, you can be as fast as the AGA chip RAM.
-
Xenepp wrote:
As I said before there are two flavors of order, bare PCB or pre-assembled, but the pre-assembled boards are going to be limited to 5 per run. There is no limit to bare PCBs obviously.
-Joel
Is there any chance you might sell kits? i.e. board + all parts? I don't mind doing the assembly myself, but sourcing parts for just one board is not easy or cheap.
-
freqmax wrote:
One could implement a queue or semaphore that sends out one request to the MCU at a time. Floppy & Harddisc doesn't have to access the MMC simultaneously in the physical sense.
Of course.
Primarily it is dependant on: Data Rate.
Something that you cannot change (without altering the clocking, or the data bus width)
If you cannot get the data from the MMC fast enough to interleave floppy disk & hard disk accesses (not necessarily 1:1) and keep compatibility then you wont be able to do it. Maybe I am worrying over nothing...
Secondly it is dependant on how much room is left in the PIC and how much room / block RAM is left in the FPGA.
-
Make it fit inside a coffee cup ;P
I dont have a spare 68000, however i do have a spare 060.
So i'll be twisting my brain to see if i can make it work..
I can get full 68060@75's sorta cheap btw. I'll have 7 next month, if anyone is interested pm me.
-Mike
http://amigaweb.net
-
Joel, can you tell us why only 5 assembled PCBs are available from this first production run and an unlimited number of bare boards? Is it because you only have enough parts to complete 5 assembled boards and don't want to put up the money in advance to purchase more parts? Or, is it because you are assembling the boards yourself and don't want to commit to doing the work on more than 5 boards at a time?
Is there any way we can increase the number of assembled boards?
Well I raised the limit to 10 but yes, it's due to the time I can invest into assembling these myself. The more there are the longer it would take and the less enthusiastic I'd be to do another run ;-)
If I could invest more time and do more I would, believe me. I'll be doing more runs though once this one's over so I hope everyone will get a Minimig that wants one in time. :)
-Joel
-
I checked the MCU firmware:
http://www.opencircuits.com/Minimig_Board_v1.0_mcu_firmware
And these changes seems to be needed in order to allow an harddisc emulation:
Define "struct adfTYPE dh0;"
Update User() to handle selection 'dh0' aswell. Current implementation decides you ALWAYS want 'df0' :-)
Make HandleFpgaCmd() to take more parameters and make them variable size per request.
Modify ReadTrack() such that it doesn't presume 11 sectors per track.
Make a SectorToFpga() function that don't stuff sector header formatting into the transfered sector.
The rest is to make the FPGA to present the SPI transfered sectors to appear magicly from the Gayle ide controller.
What part would impose a data rate limit faster than is currently available? Even hardware banging floppy loaders can't get around that each track-to-track move needs 3 ms. Which should allow one to send a harddisc sector in between if so required.
-
Hardfile support would be great, but that should be able to be done without any hardware changes, right?
Now that I am on the list for a 1.1 board, I can indulge in a little future dreaming. I don't have the board yet, so only a little dreaming....
Of all the things people want for v1.2+, the one that I think would be the biggest benefit would be adding a USB port. Almost any hardware can be had to run off of a USB bus. Everything from CD drives, to hard drive, to mice, to video capture cards. It might not be the best interface for any particular piece of hardware, but it is the most... universal. With a USB port, drivers become the only problem to add new hardware to the MiniMig.
I'm not saying that in the long run, other interfaces wouldn't be great. The more the merrier after all, but given the jump start that I expect the MiniMig to give to the hobby, USB would bring the most benefit with the least effort.
The second biggest boost for our hobby would be getting AROS fully running, and compatible enough to run most Amiga software.
-
Hardware isn't the main issue with usb, thats the easy part. The difficult part is the software. You better get crackin on that usb stack. I know I wouldn't touch it with a 10 foot pole. I've done enough usb development to know better.
-
Is /FPGA_SEL2 unused ..?
I can't find any use of it in either the FPGA or the MCU firmware.
-
Belial6 wrote:
Of all the things people want for v1.2+, the one that I think would be the biggest benefit would be adding a USB port.
Probably time to end this now. There will never be a USB host for a 7MHz 68000 with 1.5Mbytes of RAM. The USB software stack is too demanding.
-
even if USB 1.0???
what is the requests to have a USB ports working nice?
-
1NOM155 wrote:
even if USB 1.0???
Yes, even USB 1.0 or 2.0 Full Speed. The software memory and processing overhead is about the same for both.
-
alexh wrote:
Probably time to end this now. There will never be a USB host for a 7MHz 68000 with 1.5Mbytes of RAM. The USB software stack is too demanding.
I agree with some notes of reservation: The FPGA has some powers of it's own, and it's two 8MB banks in the 1.1hack, all though it's 1.5Mbytes available to the miggy running.
But I think that we have to focus on two things:
One team should focus on core and firmware (adding more features from UAE and removing bugs). And one team focus on the hardware. DVI/HDMI connectors, push pop memorycard slot. Upgraded CPU? I dunno, I'm not really capable of saying what is, and what is not possible within the minimig hardware design confinements.
Dennis van Weeren states on his homepage:
Dennis wrote:
Future
What does the future hold for Minimig? I don't know. My hope is that due to the GNU public license people will debug it, expand it and generally make it better. What I would like to see first is the implementation of some form of harddisk support, ethernet support and offcourse a debugged sprite engine :-). It would also be nice if the verilog sources of Minimig would make it into a sourceforge project. I could really need some help there. And the rest? Only time will tell!
I think people with the skills should take his word for it. Structure the firmware code in an open accessible basis as an sourceforge project, and refine the hardware on openciricuts.com. Dennis should head the hardware, and someone else with UAE and lowlevel programming should head the software part.
Right now it seems to me that we (the users) are screaming at them like a pack of hungry dogs for kits, features and stuff, and to me it doesn't seem like there is one unified coordinated effort to help dennis out here to handle the horde of hungry amigians...
So this is what I think could help:
1) Split the project into two under one umbrella.
- Minimig Hardware
- Minimig firmware
2) The umbrella project should (third toplevel project)
- Handle feature request, bug reports and coordinate effort from "third parties"
- Organize bounties:
-- Firmware coding bounties
-- Hardware design bounties
-- Hardware manufacturing bounties (the minimig shop?)
This might all sound very ambitious or negative about all the good and valuable effort a lot of people are doing right now.
But I think that a lof of best practices from the open source project management world could be employed here. It would delay some of the quick wins that we might see in the near future (10 assembled boards ie). But I think that after a while when a project organization is working, that a lot more people could benefit a h*** of a lot more...
Maybe I'm sticking my neck out here now and ramble and rant... But I hope that some people with the right skills read this and agree, and organize the right people.
-
arnljot wrote:
So this is what I think could help:
1) Split the project into two under one umbrella.
- Minimig Hardware
- Minimig firmware
2) The umbrella project should (third toplevel project)
- Handle feature request, bug reports and coordinate effort from "third parties"
- Organize bounties:
-- Firmware coding bounties
-- Hardware design bounties
-- Hardware manufacturing bounties (the minimig shop?)
I really like the idea of adding hardware/firmware bounties because it assigns tasks and allows for a way to try to recover some hardware development costs. It can get a little pricey building low volume prototypes.
:-)
-
jkonstan wrote:
[I really like the idea of adding hardware/firmware bounties because it assigns tasks and allows for a way to try to recover some hardware development costs. It can get a little pricey building low volume prototypes.
While I think that reverse auctions, or tenders for building the hardware won't be lucreative for any hardware manufacturer to deal with a minimig opensource project.
I'm sure that many amigians (migians?) have connections or know people in the hardware industry. And a scheme could be set up where say the project agreed with a manufacturer to build say 100 minimigs at an agreed price.
The minimig opensource shop would then collect prepaid order (a counter on the front page!) and dispatch the money and order to the manufacturer when the target number of orders where met.
One problem here is that it would have to be "binding non refundable" orders which is in conflict with most eu/us/asian consumer laws. The reason is to handle the logistics and problems with commitments with the manufacturer and waiting time whilst eager amigians wait for their kits.
But you could say that:
- The vendor get €40 per assembled minimig board (€4000 for 100 boards)
- The project gets €60 euro from each user per board
- €20 "surplus" goes to bounties (posible refunds and paypal disputes) and posting and handling upon delivery from manufacturer to each end user.
- If 100 orders cant be met within 12 months, all order ar null and void and money is refunded (either full €60 or the €40, depends on what the user agreed when preordering, the rest €20 goes to bounties).
These are just ideas for manufacturing bounties. There are a lot of problems and challenges that needs to be sorted out, and eventualities that needs to be planned for. But I'm sure that we (the amiga community) could make something genuinly unique here now.
For the bounties I think it should work this way:
- Requests and bug reports are received by the minimig umbrella project
- The umbrella project sorts and categorises them and defines milestones, targets and plans for each item (hardware and software)
- Each target or "deliverable" if you will becomes a bounty
- Either it's solved for free by the community (preferably) or bounty donators vote and assign money to different targets, and people or parties deliver the target and claim the bounty.
These are just some ideas, and all probably need massive refining before they can become viable.
But I think that someone who knows Dennis should approach him (I dunno how often he reads here), and ask him if he's interessted in supporting some people in organising something like this.
It seems to me that the aros guys are pretty organised and also quite experienced. I'm sure one or two from that team wold mentor such a project like this if it were head up by the right people (known entities in the amiga community).
Suggested "Minimig community project" startup:
1) Identify a spear head person
2) Approach Dennis and get his blessing/involment
3) Beg for mentorship from aros guys
4) start building the minimig community
--) This means that 10 developer boards should be built
--) Developer boards are given to proven amiga developers with track record
--) Developer boards are returned to the minimig community project if no significant deliverables are made after say 6 months
--) Developer boards can in them self be bounties for the developer if a contribution is deemed worthy of the project.
5) Nominate Dennis as head hw designer with a second in command head designer
6) Nominate a software head designer with seccond in command designer
-
arnljot:
The url is: http://www.opencircuits.com/Minimig_Project
(just noticed the spelling :-) )
I think people will do whatever they feel is rewarding because of what can be achieved rather than any monetary gain.
As for USB, maybe it's possible to do some "USB Light" in verilog, invisable to the Amiga? (or a small soft cpu to do the data management).
After looking at what the FPGA ports can be configured into. It seems both USB-12Mbps and 10Base-T can be accomplished by directly connect said interfaces to 3-4 resp 4 io pins.
What I think should go into Minimig v1.1 are the issues mentioned below, to be solved and integrated:
http://www.opencircuits.com/Minimig_Board_v1.0_issues
I have added a lot of info on the minimig worthwhile for anyone interested in starting lowlevel bit poking in minimig. At present it looks like hardfile r/w support can be implemented by a simple modifications of verilog sources, and mcu firmware. Same goes for 10Base-T, as it happens user0+1 and user2+3 consist of two differential capable io's.
As for actually producing a minimig v1.1. I have component list and schematic. It's just the "pcb" program that don't seem to want to autoroute the board.
-
freqmax wrote:
arnljot:
The url is: http://www.opencircuits.com/Minimig_Project
(just noticed the spelling :-) )
I am aware of this site. The page seems more "wikipedian" rather than a work tool for an open source project. It should be more documentary of the project (like a part list!).
freqmax wrote:
I think people will do whatever they feel is rewarding because of what can be achieved rather than any monetary gain.
And thats the "magic" and power of open sourcing! But I do also believe in bounties as an added tool for fast tracking some developments and sourcing some help from other parties that normally wouldn't be available to an opensource project. Like for instance an bounty to fix the designs and make them more widely available to more people, like the file format and incorporate the 1.1 memory hack in the 1.0 design. These seem to me like trvial fixes a PCB board manufacturer should be able to do when doing a first batch run to manufacture the board.
freqmax wrote:
As for USB, maybe it's possible to do some "USB Light" in verilog, invisable to the Amiga? (or a small soft cpu to do the data management).
This was my thought too. As far more feeble devices than the minimig can handle simple USB operations (like stereops etc). This should be handled in a "WinUAE" way so that periperhals (mice joystic) is available as control devices in the amiga, and storage devices should come up as harddrives (UAE style).
freqmax wrote:
After looking at what the FPGA ports can be configured into. It seems both USB-12Mbps and 10Base-T can be accomplished by directly connect said interfaces to 3-4 resp 4 io pins.
Nice, btw. You seem like a fellow who should be recruited by dennis in an open source project as someone either a doer (if your skills match) or at least as admin as you can assess the merit of goals and their complexity?
freqmax wrote:
What I think should go into Minimig v1.1 are the issues mentioned below, to be solved and integrated:
http://www.opencircuits.com/Minimig_Board_v1.0_issues
Agree, a version 1.1 should be a cleanup and update of what exists so far.
freqmax wrote:
I have added a lot of info on the minimig worthwhile for anyone interested in starting lowlevel bit poking in minimig. At present it looks like hardfile r/w support can be implemented by a simple modifications of verilog sources, and mcu firmware. Same goes for 10Base-T, as it happens user0+1 and user2+3 consist of two differential capable io's.
As for actually producing a minimig v1.1. I have component list and schematic. It's just the "pcb" program that don't seem to want to autoroute the board.
See, now there is atleast three people capable of taking care of the minimig community: Dennis, xenepp and yourself! ;)
-
The issue with USB host on the present Minimig PCB would be the lack of CPU MIPs (the slow clock speed of the 68SEC000 CPU) and lack of spare FPGA I/O pins. A seperate CPU on a USB HOST controller such as the FTDI Vinculum USB HOST controller might work becasue it has a seperate CPU of its own running the USB stack although we would have to use a SPI interface to connect to it since we only have 4 spare FPGA I/O pins. The speed of a serial SPI connection from 68K/Xilinx_FPGA would most likely limit the bandwidth/transfer rate. Data sheet has Vinculum SPI port max SPI Clock = 10Mhz => 10Mbits/sec Max SPI transfer rate (if no signal integrity issues arise). A FTDI Vinculum USB HOST controller small daughtercard that connected to MiniMIG1 Spare I/O header J9 could be built. We may have enough 68K CPU MIPS to handle this kind of implementation.
http://www.vinculum.com/
Other soft USB option (not feasible on MiniMIG1.0):
A usb phy connected to an FPGA would require @ 7 FPGA I/O pins and an embedded Xilinx microblaze soft CPU core to run the USB Stack. The USB stack is large; thus, the microblaze CPU would need external RAM access. The RAM is tied up by the 68K and FPGA_AMIGA_Chipset. Thus, the soft USB option does not look to good for MiniMIG1.0.
http://www.fairchildsemi.com/ds/US/USB1T11A.pdf
:-)
-
@all & @jkonstan
I'm really impressed with you guys and your tech knowledge!
If only all these investigations where logged in a project and kept organised... ;)
-
My idea with usb was to exploit the User-I/O header. Connecting User0+User1 to D+ & D-. And let say User2 select pullup on one D line to indicate speed. Thus eliminating the need for any external PHY, and any resulting io consumption.
Also USB might save pins. The SD/MMC slot, Keyboard, Mouse, Joystick0, Joystick1 can be replaced.
Maybe replacing the MCU with an CPLD is a viable option?
That CPLD would then be fast enough to handle Ethernet, USB aswell as booting the system etc.. And it doesn't need a boot eeprom.
Another option with CPLD is FPGA reconfiguration via ethernet.. :-)
-
Spartan3 I/O is programable/flexible; however, I am not so sure that it is flexible enough to eliminate the need for an external USB HOST Physical layer. I would probably count on needing an external USB HOST physical layer such as the Fairchild part below.
http://www.fairchildsemi.com/ds/US/USB1T11A.pdf
Actually, replacing the 68SEC000 CPU with part of the Xilinx Spartan3 might be an option. I have the downloaded the VHDL source to an ATARI ST project that includes an almost finished 68K CPU. This would free up lots of FPGA pins.
I need to get my MiniMIG(s) going so that I can get some development going.
:-)
-
Xenepp, can we have an update on what is happening re building boards?
-
source to an ATARI ST project that includes an almost finished 68K CPU
This CPU Core is not finish. The Debuging must go on. Currently hangs up this CPU Core with Kickstartrom 2.04 at #$F81914.
Viele Grüße
TobiFlex
-
TobiFlex wrote:
source to an ATARI ST project that includes an almost finished 68K CPU
This CPU Core is not finish. The Debuging must go on. Currently hangs up this CPU Core with Kickstartrom 2.04 at #$F81914.
Viele Grüße
TobiFlex
What set of Amiga hardware are you debugging the 68K VHDL CPU core with? Do you have an estimated clock rate for the 68K core in a Spartan3 FPGA ? I was going to try the 68K VHDL CPU core when I got my MiniMIG going.
Thank you for the update on it.
:-)
-
I have adapted the minimig Core to the DE2 Board from Terasic with a ALTERA Cyclone 2C35. With an external CPU MC68HC000 runs the Minimig perfekt. I have change the RAM Timing from SRAM to SDRAM - so i can use the on Board SDRAM Chip.
And now i use the Board to Debug Wolfgangs 68K Core.
-
TobiFlex wrote:
I have adapted the minimig Core to the DE2 Board from Terasic with a ALTERA Cyclone 2C35. With an external CPU MC68HC000 runs the Minimig perfekt. I have change the RAM Timing from SRAM to SDRAM - so i can use the on Board SDRAM Chip.
And now i use the Board to Debug Wolfgangs 68K Core.
Sounds as though you are using a stock FPGA evaluation/development board possibly with some adpater PCBs that you built.
More questions because I have a few evaluation FPGA boards as well ...
Q1) Did you build a PIC adapter PCB as Dennis had for the Spartan3 Eval board or did you use the Altera DE2 SD socket and write a bunch of code in order to replace the floppy ?
Q2) The current Verilog implementation on MiniMIG essentially runs 0 wait states with the external SRAM. Have you modified the MiniMIG Verilog core to allow for wait states so that SDRAM or DDR can be adapted ?
Q3) How do you find an affordable Altera DE2 ? Those are expensive unless you get the student discount ?
:-)
-
Unless you can find or write a 68K core that can fit in the XC3S400 currently used by the MiniMig there is very little point.
To use Wolfgang's 68K core, you would need a much bigger FPGA.
Not much point replacing a $7 CPU when you need an FPGA that costs $20+ more to do it.
-
Since I have been busy working on trying to get a couple MiniMIGs put together, I have not yet compiled Wolfgang's 68K VHDL core; thus, I was not aware of its size in terms of Xilinx CLBs.
I agree completely that the 68SEC000 is pretty cheap $7 to $11; however, I was thinking more of getting around some of the present limitations of MiniMIG1:
1. The lack of more FPGA pins (soldering hobby/home project limit of 208 pin QFP on FPGA). The new extra FPGA pins could be used for IDE, Compact FLASH, parallel interface to Vinculum VNC1L (USB host controller with a CPU core and USB stack), etc ....
2. The 68K bus clock rate limit on a two layer PCB. A larger Xilinx FPGA (more CLBs) with a soft 68K core embedded in it could possibly clock the soft 68K faster than the external 68SEC000. This would have to be investigated by compiling Wolfgang's VHDL code and running Xilinx timing analyser on it.
:-)
-
jkonstan wrote:
1. The lack of more FPGA pins (soldering hobby/home project limit of 208 pin QFP on FPGA). The new extra FPGA pins could be used for IDE, Compact FLASH, parallel interface to Vinculum VNC1L (USB host controller with a CPU core and USB stack), etc ....
2. The 68K bus clock rate limit on a two layer PCB. A larger Xilinx FPGA (more CLBs) with a soft 68K core embedded in it could possibly clock the soft 68K faster than the external 68SEC000. This would have to be investigated by compiling Wolfgang's VHDL code and running Xilinx timing analyser on it.
:-)
There are enough free pins on the FPGA to do IDE as it is with a few external components and Compact Flash can also be connected via the IDE port.
Wolfgang's 68K core will run up to about 17 MHz in a Spartan 3.
The easiest was to gain more I/O pins is to change to a Spartan 3E. The design changes are minimal, you can still use the same 208 pin QFP package, and you gain more gates and more I/O pins.
-
I wish i had the foggyest what you lot are talking about. It all sounds very impressive all this tech speak! I only know software... and nothing low-level :-(
-
Q1) Did you build a PIC adapter PCB as Dennis had for the Spartan3 Eval board
No.
or did you use the Altera DE2 SD socket and write a bunch of code in order to replace the floppy ?
Yes.
Q2) The current Verilog implementation on MiniMIG essentially runs 0 wait states with the external SRAM. Have you modified the MiniMIG Verilog core to allow for wait states so that SDRAM or DDR can be adapted ?
I use SDRAM. With DDR I have no experience. The SDRAM Clock is 112MHz. The SDRAM and the minimig runs synchron.
The SDRam needs 8 Clocks to read or write one Data word. This is a half CPU Clock. That enougth time to access.
Q3) How do you find an affordable Altera DE2 ? Those are expensive unless you get the student discount ?
This is my secret ;-)
Viele Grüße
TobiFlex
-
mongo wrote:
There are enough free pins on the FPGA to do IDE as it is with a few external components and Compact Flash can also be connected via the IDE port.
I agree that Spartan3E is better way to go on an update to MiniMIG.
I have implemented several IDE interface in Verilog over the years; thus, I am pretty curious. There are 4 spare FPGA spare I/O pins left in MiniMIG1. How do you intend to support/implement an IDE interface (IDE_RESET, CS0*, CS1*, INTRQ, DMREQ, DMACK, IORDY, IORD*, IOWR*, ICS16*, DASP*, DA0-DA2, DB0-DB15) which require at least 12 FPGA I/O pins and some external CBT16245 level shifters used on 68K bus ?
Also, I was not thinking of using the compact FLASH card in IDE mode (where Pin9 is pulled to ground and only good for compact flash memory cards). My thought was to have a compact flash socket that was actually supported compact flash peripheral cards as well such as ethernet and serial port cards (i.e. Full Compact FLASH socket mode = PCMCIA with reduced number of address lines).
:-)
-
Any better fpga than the XC3S500E would require the use of a BGA socket.
As for benefits of a soft 68k core one is that supply just won't end. So it might be worthwhile to make debug and refine it. But maybe wait with using it until non-bga FPGAs become cheap and dense enough to accomodate it.
-
jkonstan wrote:
I agree that Spartan3E is better way to go on an update to MiniMIG.
I have implemented several IDE interface in Verilog over the years; thus, I am pretty curious. There are 4 spare FPGA spare I/O pins left in MiniMIG1. How do you intend to support/implement an IDE interface (IDE_RESET, CS0*, CS1*, INTRQ, DMREQ, DMACK, IORDY, IORD*, IOWR*, ICS16*, DASP*, DA0-DA2, DB0-DB15) which require at least 12 FPGA I/O pins and some external CBT16245 level shifters used on 68K bus ?
Address and Data lines connect to the 68K via level shifters, same with the Reset line. IORD*, IOWR* CS0*, and CS1* can easily be generated completely externally, though you can reduce the number of external components if you use one output line of the FPGA as a IDE Chip Select line. IORDY has to go to the FPGA for it to generate wait states for the 68K if needed, but even that might not be totally necessary unless you plan on using a really old drive. INTRQ has to go to the FPGA, unless you want to poll the drive instead of using interrupts, but I don't recommend doing that if you don't have to.
DMARQ, DMACK, ICS16*, and DASP* aren't needed.
3 FPGA I/O pins and a few external components are all that are required.
-
mongo wrote:
Wolfgang's 68K core will run up to about 17 MHz in a Spartan
How many % of the fpga logic does it use?
Which Spartan were used?, esp which speed grade -4 or -5 ?
Page 7 (of 208)
http://direct.xilinx.com/bvdocs/publications/ds099.pdf
-
TobiFlex wrote:
I have adapted the minimig Core to the DE2 Board from Terasic with a ALTERA Cyclone 2C35. With an external CPU MC68HC000 runs the Minimig perfekt.
Any idea if Freescale's 68306 (http://www.freescale.com/webapp/sps/site/prod_summary.jsp?code=MC68306&nodeId=0162468rH3YTLC61650795) would work fairly easily? ( straightaway I can see some work needed to convert IPL2:0 to the 306's separate IRQ lines).
Looks like its built in DRAM controller would make adding upto 64MB pretty easy, plus there's some programmable chip selects and IO pins for extra hackable goodness :hammer:
-
Doobrey wrote:
Any idea if Freescale's 68306 (http://www.freescale.com/webapp/sps/site/prod_summary.jsp?code=MC68306&nodeId=0162468rH3YTLC61650795) would work fairly easily? ( straightaway I can see some work needed to convert IPL2:0 to the 306's separate IRQ lines).
Looks like its built in DRAM controller would make adding upto 64MB pretty easy, plus there's some programmable chip selects and IO pins for extra hackable goodness :hammer:
MC68306 could work since it is based on the 68HC000 core.
This is a nice idea (especially with the JTAG port on the MC68306); however, the MC68306 would have issues. The MC68306 DRAM controller works only with old obsolete DRAM which is no longer produced anymore. MC68306 is a 5Volt I/O part; thus, we would have to use level shifters between 5V MC68306 <=> 3.3V FPGA. Thus, there is not as much benefit in using MC68306 as we would like.
:-)
-
mongo wrote:
Address and Data lines connect to the 68K via level shifters, same with the Reset line. IORD*, IOWR* CS0*, and CS1* can easily be generated completely externally, though you can reduce the number of external components if you use one output line of the FPGA as a IDE Chip Select line. IORDY has to go to the FPGA for it to generate wait states for the 68K if needed, but even that might not be totally necessary unless you plan on using a really old drive. INTRQ has to go to the FPGA, unless you want to poll the drive instead of using interrupts, but I don't recommend doing that if you don't have to.
DMARQ, DMACK, ICS16*, and DASP* aren't needed.
3 FPGA I/O pins and a few external components are all that are required.
Your description shows that you need more than the 4 programmable FPGA Spare I/O pins that are left on MiniMIG Ver1.0 J9 header in order to build the IDE interface that you described and that the MiniMIG PCB would need a new PCB layout. Of course if you add a CPLD/external logic and a bunch of level shifters to MiniMIG, a PIO mode IDE interface could be built.
PI/O IDE is ok; however, DMA can be nice when one can DMA directly into SRAM. Thus, I would not write off the need for DMAREQ and DMACK.
IORDY for IDE would be nice when some users trys to hook there old small IDE hardrives from their old Amigas onto a MiniMIG at some point.
:-)
-
jkonstan wrote:
MC68306 is a 5Volt I/O part; thus, we would have to use level shifters between 5V MC68306 <=> 3.3V FPGA. Thus, there is not as much benefit in using MC68306 as we would like.
Bummer :(
I couldn't see anything in the datasheet that specifically said it was a 5v part.( Or rather, I couldn't see a minimum Vcc in the DC specs..maybe I'm blind).
Ta for the reply.
-
jkonstan wrote:
Your description shows that you need more than the 4 programmable FPGA Spare I/O pins that are left on MiniMIG Ver1.0 J9 header in order to build the IDE interface that you described and that the MiniMIG PCB would need a new PCB layout. Of course if you add a CPLD/external logic and a bunch of level shifters to MiniMIG, a PIO mode IDE interface could be built.
I need 3 I/O pins and a little external logic to implement an A600/A1200 compatible IDE port, which would allow booting from an IDE hard drive with an appropriate A600 ROM.
I could do it with 2 pins, if I really wanted to.
A new PCB layout is a given if you are going to add IDE.
PI/O IDE is ok; however, DMA can be nice when one can DMA directly into SRAM. Thus, I would not write off the need for DMAREQ and DMACK.
Sure DMA is nice, but it's really not necessary. It also requires new drivers and a separate bus for the IDE, unless you want to shut off the 68K during DMA transfers.
IORDY for IDE would be nice when some users trys to hook there old small IDE hardrives from their old Amigas onto a MiniMIG at some point.
:-)
I don't have a problem with that. You still don't need more than 3 I/O pins.
-
The 68306 product summary only says "Available in 5V". Looking in the datasheets there's no mention of 3.3V. I still fail to see the benefits besides address space. Any DRAM is better controlled by the FPGA where it can be used both as Chip and Fast ram.
If 'TobiFlex' succeeded to use sdram. It would be interesting to find out more in order to replace the 2M Async SRAM with 32-64 MB DRAM or more.
One possible setup is to have two FPGAs and one XCF02S configuration memory.
FPGA0 bootloads configuration from eeprom. And loads data from SD/MMC to configurate FPGA1. FPGA1 loads data from SD/MMC to re-configurate FPGA0. Both bootloaders are deactivated. And normal operation is initiated.
This could allow elimination of the MCU and allow one extra FPGA for m68k vhdl/verilog core. Aswell as plenty of I/O pins to play with.
-
freqmax wrote:
The 68306 product summary only says "Available in 5V". Looking in the datasheets there's no mention of 3.3V. I still fail to see the benefits besides address space. Any DRAM is better controlled by the FPGA where it can be used both as Chip and Fast ram.
If 'TobiFlex' succeeded to use sdram. It would be interesting to find out more in order to replace the 2M Async SRAM with 32-64 MB DRAM or more.
One possible setup is to have two FPGAs and one XCF02S configuration memory.
FPGA0 bootloads configuration from eeprom. And loads data from SD/MMC to configurate FPGA1. FPGA1 loads data from SD/MMC to re-configurate FPGA0. Both bootloaders are deactivated. And normal operation is initiated.
This could allow elimination of the MCU and allow one extra FPGA for m68k vhdl/verilog core. Aswell as plenty of I/O pins to play with.
Use of the dual FPGA scheme could get us more free/open FPGA I/O pins and potentially replace the PIC. The soft 68K CPU core or an additional soft Microblaze CPU could control the SD card as well. In order to lower the cost of the Xilinx FPGAs and expensive Xilinx Configuration serial device, the use of the Spartan3E FPGA which is lower cost and can use lower cost SPI FLASH (ATMEL, ST, and Spansion) for configuration should be considered (see App note below).
http://www.xilinx.com/bvdocs/appnotes/xapp445.pdf
Also, maybe TobiFlex will share some of his FPGA SD card (floppy replacement)development so that we do not have to re-invent it.
:-)
-
Looking at the core files, only LVCMOS33 is used. So using XC3S500E would be feasable. Any cons/pros of using LVTTL instead?
And 622+ Mbit/s operation is possible between FPGAs.
More config eeprom choices seems to translate into easier to source and lower price.
-
Hmm, if this is the price, I'd consider it.
-
Hi Xenepp,
I'd like one bare board and one assembled please.
Andy
-
I was just on the schmartboard web site t'other day. They have a prototyping board for BGA chips up to 400 pins. They claim it can be hand-soldered.
-
Impossible, they must have a BGA rework station.
-
Another update, Dennis alerted me to V1.1 yesterday and I sent it off to China already.
Two weeks and thing will start to happen here! :-D
-Joel
-
alexh wrote:
Impossible, they must have a BGA rework station.
Well, they do say that it works better with a reflow oven. However, they do show instructions for doing it by hand.
-
The largest Xilinx Spartan-3 FPGA that has less than 400 pins (20x20 matrix) is the FG320 package:
http://direct.xilinx.com/bvdocs/publications/ds099.pdf
(FG320 XC3S1500 is also the largest FPGA with Webpack support)
The FG320 package is an 18x18 matrix with 1mm pitch:
http://www.xilinx.com/bvdocs/packages/fg320.pdf
Schmartboard 202-0026-01 is BGA 400 Pins, 1.0 mm Pitch:
http://www.schmartboard.com/index.asp?page=products_bga&id=110
Diagram of 202-0026-01:
http://www.schmartboard.com/schmartboard_pd_202-0026-01.pdf
It would be benefitial if someone can clarify this matter. I have noticed this company in the past. Debunk time? :-D
-
Xenepp wrote:
Another update, Dennis alerted me to V1.1 yesterday and I sent it off to China already.
Two weeks and thing will start to happen here! :-D
-Joel
Good news. You're the man!
Just out of interest, have you asked what it would cost them to actually produce an assembled board in various quantities?
I have all sorts of stuff in my house from when I worked in China and if they can assemble a complete set of Calloway golf clubs with travel bag for $170 then they should be able to produce an assembled minimig for a reasonable price. ;-)
-
Xenepp wrote:
Another update, Dennis alerted me to V1.1 yesterday and I sent it off to China already.
Two weeks and thing will start to happen here! :-D
-Joel
Great news! Let me know when you want some cash.
-
Has anyone got a list of parts that could be bought from say CPC or RS or (even) Maplins!? Im interested in the cost of the parts and how available they are. Im thinking i might build one to improve my soldering skills then maybe sell my prebuilt one from Joel. ;-)
-
I have one work-in-progress:
http://www.opencircuits.com/Minimig_Board_v1.0_components
-
"It would be beneficial if someone can clarify this matter. I have noticed this company in the past. Debunk time?"
I have also been seeing their ads a lot recently. They have been advertising in Nuts and Volts for a while, and I think Servo also. I have also noticed that Fry's has started carrying their products in their stores.
It doesn't seem like an outrageous claim to me. I can think of a couple of ways to do this. The first would be to make a mask that would have pins that fit between the pins. If solder does not stick to the material that the schmartboard is made from, but does stick to the solder pads, you can easily tin the board and chip. Then you just hold the schmartboard in place and heat it. Even better if the schmartboard has a highly heat conductive material with leads down to the pins.
I'm not saying that this is how they do it, but that is just one solution of the top of my head. I'm sure with some thought, several solutions for hand soldering these parts could be worked out. Particularly if the tools to do it were going to be made in large quantities.
-
Ok, I looked up their address, and found their PDF that explains how it works. Basically it is a board that you solder your part to, so that you can add it to your project. The description seems perfectly reasonable to me.
http://www.schmartboard.com/schmartboard_dc_ez_instructions.pdf
-
freqmax wrote:
I have one work-in-progress:
http://www.opencircuits.com/Minimig_Board_v1.0_components
Does anyone know where one can buy the MCU? I found it at digikey, but minimum order was 84 units. I was kind of looking for one or two.
Yes, I'm thinking about building the minimig :-)
--
Thomas
-
Belial6:
Well but HOW do you solder it. Because you won't be getting the soldering iron between the chip and the pcb.
ThomasML:
Check Mouser.com, Maplin.co.uk ..?
Try google aswell. Sometimes it works.
Which country or continent do you live in?
-
Good question. The next time I'm in a Fry's, I'll look and see if it is obvious from just looking at the board. It will probably be a few weeks, as I don't live close to one anymore.
-
ThomasML >> www.conrad.de they have it all :)
-
yssing:
My impression when I got their brochure is that they sell mostly beginners kit and misc junk.
Seems my local branch doesn't have PIC18LF252, and they certainly don't have Xilinx stuff. My impression is that they suck.
Btw, their url of 1059 bytes doesn't impress me either.
Digikey have the PIC18:
http://www.digikey.com/scripts/DkSearch/dksus.dll?Detail?name=PIC18LF252-I%2FSP-ND
-
My impression is that they suck.
They do. Reichelt (http://www.reichelt.de/) or RS-Components (http://rswww.com/) are much cheaper. RS-Components has multiple stores (http://www.rs-components.com/index.html) in various countries (UK, Germany, etc.).
-
Reichelt: http://www.reichelt.de/
I presume.
-
freqmax wrote:
Belial6:
Well but HOW do you solder it. Because you won't be getting the soldering iron between the chip and the pcb.
The way I understood it, all of the solder balls are on via holes, so you can solder it from the underside of the board.
-
Why dont they sell the PCB *WITH* components? It would be a great idea and save us loads of hassle. I would love to buy a pre-assembled Minimg, and a pcb WITH components would be my second option, but I dont see myself chasing components which I dont know even what they are for.
-
Everblue wrote:
Why dont they sell the PCB *WITH* components?
Because there is no "they"? The "they" you refer too is called Xenepp who will be buying and building them by hand :-D without charging for his time or skill judging by the prices he's asking.
I've PM'd him asking for one bare board and one assembled when he does another run of them. The bare board is so that with a friends help and the smt tutorials on sites like Make (etc) I can attempt building one myself.
Andy
-
I think a more sustainable model is by delivering a "kit" ie pcb + parts. This should also allow for lower prices, even if Xenepp took a small profit. That will allow people to concentrate on soldering. Some repair shops will do soldering work for pay aswell.
-
freqmax wrote:
I think a more sustainable model is by delivering a "kit" ie pcb + parts. This should also allow for lower prices, even if Xenepp took a small profit. That will allow people to concentrate on soldering. Some repair shops will do soldering work for pay aswell.
The problem is that the only way it would allow for lower prices is if parts are bought in larger quantities (25+) which would require a fairly large initial cash outlay for the person putting together the kit.
If anyone has an extra few thousand dollars lying around and wants to do so, they could probably make a few bucks profit, but I don't know if it's worth the effort for anyone, especially since I don't know how much demand there would be for a kit. Most people seem to want a pre-assembled one.
-
freqmax wrote:
Belial6:
Well but HOW do you solder it. Because you won't be getting the soldering iron between the chip and the pcb.
You can use an oven to do BGA-soldering at home, here's one starting point:
http://www.instructables.com/id/EBXC76M6V5EP28623C/
(there's a lot more of those, I have one pdf which tells you how to make an oven controlled by a MCU to get the spot-on temperature changes).
freqmax also wrote:
ThomasML:
Check Mouser.com, Maplin.co.uk ..?
Try google aswell. Sometimes it works.
Which country or continent do you live in?
I live in Norway, Europe. And, I didn't find the 68000-CPU on mouser either. The PIC-MCU I don't have any problems finding :-)
--
Thomas
-
The problem with soldering of bga is that you can't verify the soldering operation before applying power.
Unless you (ab)use an old radio tube with an HV supply as a x-ray source. And use ccd? as a image receiver (the principal way it's done in a professional enviroment).
So soldering is complicated, but verification is the hard stuff (unless blowing chips is acceptable).
-
Could we use something like this to solve the BGA problem?
http://www.ironwoodelectronics.com/catalog/Process/GHz_BGA_Sockets.html
-
Belial6 wrote:
Could we use something like this to solve the BGA problem?
http://www.ironwoodelectronics.com/catalog/Process/GHz_BGA_Sockets.html
Sure, but it's extremely expensive and therefore I don't see this going into the minimig project.
--
Thomas
-
Any word on where this run of MiniMigs is?
-
Here are pictures of my white Minimig rev 1.0 PCB (http://www.amiga.org/forums/showthread.php?t=31344):
(http://img410.imageshack.us/img410/7477/nakedminimigfrontis0.th.jpg) (http://img410.imageshack.us/my.php?image=nakedminimigfrontis0.jpg)
(http://img509.imageshack.us/img509/1829/nakedminimigbackhe7.th.jpg) (http://img509.imageshack.us/my.php?image=nakedminimigbackhe7.jpg)
:-D
-
Well, there's good news and bad..
First the good.. the Minimig PCBs have arrived! They look great (see here) (http://www.ghost3usters.com/minimig.jpg)
The bad news, it looks like I'll not have the time to do a run of complete PCBs. :cry:
I've just gone back to college and it's more work that I ever imagined. That plus life equals no time for much of anything. The free time I do get I just want to sleep!
I can always get the bare PCBs though and have 9 here with folk's names on them so if you want them just drop me a PM, but a big sorry to all those waiting for complete boards, it just isn't going to happen from me. :cry:
Anyway, I'm going to start emailing those who pre-ordered the bare boards and collecting payments and getting them shipped out.
-Joel
-
@Xenepp
Yes, lack of time, I know what that means :S
But I think we must all thank you for all your effort, although personnally I wanted only a populated PCB :(
-
@CD32Freak,
Pretty cool, did you only have the one board made? Did you make the v1.1 changes that Dennis suggested?
I love the Galaxy Quest quote on the board, it fits the Amiga spirit so well. :-D
-
Xenepp wrote:
Well, there's good news and bad..
First the good.. the Minimig PCBs have arrived! They look great (see here) (http://www.ghost3usters.com/minimig.jpg)
The bad news, it looks like I'll not have the time to do a run of complete PCBs. :cry:
I've just gone back to college and it's more work that I ever imagined. That plus life equals no time for much of anything. The free time I do get I just want to sleep!
I can always get the bare PCBs though and have 9 here with folk's names on them so if you want them just drop me a PM, but a big sorry to all those waiting for complete boards, it just isn't going to happen from me. :cry:
Anyway, I'm going to start emailing those who pre-ordered the bare boards and collecting payments and getting them shipped out.
-Joel
Joel, did you only have 10 bare boards manufactured? Am I one of the lucky nine?
-
Three questions then...
1) Have you been able to assemble your own MiniMig to make sure the boards work? (I hope you will have enough time to assemble at least one board for your own use.)
2) Did you already get the components for the boards you were planning?
3) If you did, can you sell the parts and board to us so that we basically have a kit?
-
CD32Freak wrote:
Here are pictures of my white Minimig rev 1.0 PCB (http://www.amiga.org/forums/showthread.php?t=31344):
:-D
OMG that is a Gorgeous board......
What company did the board, how much, how long etc ?
Website ?
-
Thats really cool that you've got some of the boards and I would still be interested in a blank board from you either in this or another batch should you do one :) Well done getting them done this far!
Andy
-
Very Cool :-) (http://img410.imageshack.us/img410/7477/nakedminimigfrontis0.th.jpg)
-
@Xenepp
I'm still happy to be on the list for a bare board. Even if things don't always turn out as expected, thanks for all the work you've done so far.
Plaz
-
@Xenepp,
The bad news, it looks like I'll not have the time to do a run of complete PCBs.
Sorry to hear that, I wanted to acquire a ready-made PCB for 'tinkering'.
Guess I will be having another look at building my own minimig board...
Valentin
[ Edit: My knowledge of PIC and 68k architectures are very limited, so this might take a while.. :-) ]
-
I'm not sure about how many will be made yet, I guess if you post here that your interested it'll give me an idea.
A little late to the party here, but sign me up for a rev1.1
Preferably with at least SMD done, even better with all soldering done .. but I'll take a bare board if that's what's going.
(form-factor not fussy, plugs vs headers not fussy, real life vs usb/firewire/ide/pci/blah/blah not fussy :crazy: .. )
-
I'm also interested in a Minimig but only in one with SMD parts allready soldered.
My personal wish list for a Minimig V2 would be:
- 68EC020 instead the 68SEC000 for better compatibility with the Amiga 1200 if AGA is implemented. ;-)
- IDE Port
- And a 65816 CPU. The CPU which should be used is selected with the FPGA core. Why an 65816? Because it would be a good CPU for a C64 FPGA core. This way the minimig would be a great device for C64 Fans too. And often Amiga Fans are C64 Fans too. ;-) Yes I know, the 6502 for the C64 coold be implemented in the FPGA too, but I think a CPU like the 65816 would even be better.
-
blachner wrote:
My personal wish list for a Minimig V2 would be:
- IDE Port
Why would you want an IDE port? I am almost certain we'll be able to emulate hard disk using the MMC card using existing MiniMig
-
alexh wrote:
Why would you want an IDE port? I am almost certain we'll be able to emulate hard disk using the MMC card using existing MiniMig
Yes you are right, this would be enough, because the size of MMC is big enoug for needed hard disk size for Amiga. Couldn't this be done on Minimig V1 with an other PIC program?
But a 68EC020 switchable with 65816 for C64 Emulation would be very nice on Minimig V2.
-
I have read the MCU firmware. And I don't see any reason as to why not it should be possible to do adf-write and harddisc-read/write operations. Only program memory could hinder it as I see it.
I don't see any reason to have a 65816. Both that and 6502 can likely be done in HDL.
Also there simple is NO free i/o for another CPU with XC3S400 and not likely XC3S500E either.
-
freqmax wrote:
I have read the MCU firmware. And I don't see any reason as to why not it should be possible to do adf-write and harddisc-read/write operations. Only program memory could hinder it as I see it.
I don't see any reason to have a 65816. Both that and 6502 can likely be done in HDL.
Also there simple is NO free i/o for another CPU with XC3S400 and not likely XC3S500E either.
Do you know if for the 65816 a free FPGA implementation is available which is fully compatible with it and reaches the same speed? For 6502, I know, there are som good free FPGA implementations available.
If a good 65816 FPGA implementation is available, than a real 65816 on Minimig might not necessary. So my whish list for Minimig V2 would be shrinked to 68EC020. ;-)
But I see an other problem, the program memory of the PIC. Because the PIC should than handle Amiga hard disk and disk image files read and write and also C64 disk image read and write the memory is likely be to small.
-
If the FPGA is upgraded from 400k to 500k gates there should be enough room for a 65816 or a 68ec020 core inside it. At the present time I do not think somebody has written any of those cores, but when it is mass produced I think somebody will write them since the source of software emulators for those (and many other) CPUs has already been made public.
-
little wrote:
At the present time I do not think somebody has written any of those cores
http://www.opencores.org/projects.cgi/web/t65/overview
Opensource 65C816 VHDL
-
@alexh
maybe you want to plug in a real hard disk from your old miggy, CDROM, DVD...
-
@crumb
Plug it into your PC, copy the files to the SD card (or the HDF on the SD card). Just like you would for WinUAE etc.
I wonder what's next to be suggested?
Bluetooth I/O and sensor bar input for a Wiimote to allow "lightgun emulation" on LCD panels? :-)
-
Nah you don't need sensor bar input... its just a couple of IR LEDs :-D
Andy
-
How much ROM does CDTV / CD32 use ..?
-
Plug it into your PC, copy the files to the SD card (or the HDF on the SD card). Just like you would for WinUAE etc.
Absolutely, Why use one computer when you can use two. Maybe you could work a Mac into the process too.
I wonder what's next to be suggested? Bluetooth I/O and sensor bar input for a Wiimote to allow "lightgun emulation" on LCD panels?
I know, what is it with these people who want to make a new Amiga that supports modern hardware standards?
________
HAWAII MEDICAL MARIJUANA DISPENSARY (http://hawaii.dispensaries.org/)
-
The blank pcbs are back on Ebay. (http://search.ebay.co.uk/_W0QQfgtpZ1QQfrppZ25QQsassZnusim1)
-
Actually one possible reason for real IDE would be to have CDTV emulation once ECS is working.
It'd also be good for later v2.x MiniMigs with a 32bit bus and AGA so that the CD32 can be properly emulated.
If it's as easy to do IDE as people have been saying then it makes sense to try it now as one of the low-hanging-fruit kind of tasks.
alexh wrote:
@crumb
Plug it into your PC, copy the files to the SD card (or the HDF on the SD card). Just like you would for WinUAE etc.
I wonder what's next to be suggested?
Bluetooth I/O and sensor bar input for a Wiimote to allow "lightgun emulation" on LCD panels? :-)
-
freqmax wrote:
How much ROM does CDTV / CD32 use ..?
CDTV 256Kbytes
CD32 1Mbyte
-
AJCopland wrote:
Actually one possible reason for real IDE would be to have CDTV emulation once ECS is working.
CDTV is OCS ;-) (Well half OCS, it has a FATTER Agnus)
The CDTV doesnt use IDE :-(
It'd also be good for later v2.x MiniMigs with a 32bit bus and AGA so that the CD32 can be properly emulated.
1) Put ISO images on SD card?
2) Solve other v2.x problems (RAM, CPU, FPGA) first?
-
AndrewBell wrote:
Why use one computer when you can use two.
It's a one off operation. Not like you'd have to keep the PC connected all the time.
AndrewBell wrote:
Maybe you could work a Mac into the process too.
Last time I looked a Mac was a PC.
AndrewBell wrote:
I know, what is it with these people who want to make a new Amiga that supports modern hardware standards?
Exactly my point. Glad we agree.
-
@alexh:
Guess it's useful to make the kickstart loader have 256kB granularity at minimum then.
AndrewBell wrote:
I know, what is it with these people who want to make a new Amiga that supports modern hardware standards?
No one hinders you from doing schematics, pcb routing, looking up datasheets, order components, and do smd soldering. Possible with BGA chip packages. And then do fault search on all components with a oscilloscope.
-
freqmax wrote:
No one hinders you from doing schematics, pcb routing, looking up datasheets, order components, and do smd soldering. Possible with BGA chip packages. And then do fault search on all components with a oscilloscope.
They'd better not.
________
Panic disorders forums (http://www.health-forums.org/panic-disorders/)
-
alexh wrote:
CDTV is OCS ;-) (Well half OCS, it has a FATTER Agnus)
The CDTV doesnt use IDE :-(
1) Put ISO images on SD card?
2) Solve other v2.x problems (RAM, CPU, FPGA) first?
I'd like to help out with some of those other problems, the SD-RAM one seemed like the biggest immediate win actually. I have no idea how to do the layout for it though.
My initial plan was to figure out what SD-RAM chip to use, someone linked to a 32Mbit (4MByte) 100/133Mhz. Then to figure out how to interface it to the FPGA, so theres some board layout to be done.
I figured that the SD-RAM would need another clock source since there are only 20MHz and 4Mhz clocks currently on the MiniMig. The example of interfacing to SD-RAM required at least 25Mhz as an input but I hadn't looked into it any further than that.
I was also hoping that Dennis or someone else would respond as to why the MiniMig would have trouble actually using SD-RAM as opposed to SRAM so there'd be a bit more information about what problems are going to be encountered.
Andy
-
freqmax wrote:
@alexh:
Guess it's useful to make the kickstart loader have 256kB granularity at minimum then.
Or do as is done at the moment, pad the file to 512kB. The CDTV uses a standard 256Kbyte A500 KS1.3.
It also has extra roms, sometimes called PROMS, sometimes called extended ROM's. These contain the CDTV menu system.
-
Colin_Camper wrote:
The blank pcbs are back on Ebay. (http://search.ebay.co.uk/_W0QQfgtpZ1QQfrppZ25QQsassZnusim1)
Thanks for that. I just purchased one. With Plaz' Digikey parts list, I may try and build myself a Minimig...
-
Xilinx have SDRAM application docs, so no big problem on the hardware side. One DCM will be occupied however. And a feedback line to the clock input of the SDRAM is needed. Maybe, maybe it can make do with a fake DCM clock from the MCLK or so.
I think the big obstacle is the pipeline architecture of the DRAM. Meaning the Minimig would request a memory access and get the answer ~3 memory clocks later (depend on Cas latency). Meaning the memory might have to be clocked at current memory speed x3.
PCB wire impedance, signal skew, rise time, reflection, decoupling, power supply ripple etc.. all come into play for this.
-
I am thinking, an added bonus (besides de extra 100k gates) the XC3S500E FPGA has over the XC3S400 in the actual design is that it has a "Low-cost, space-saving SPI serial Flash PROM". What use would that be? Take a look here (http://www.compsys1.com/workbench/On_top_of_the_Bench/MMC_Project/mmc_project.html) and scroll down to place where it says "SD card proto pcb". If such a design could be added to the minimig then no PIC controller would be required at all; that is, if the extra code to access the SD/MMC card as a floppy disk (or hard disk or CD-ROM) can be put inside the FPGA.
-
I was also thinking, if the 68k is succesfully placed inside the FPGA, that means there are enough free I/O lines to do one of the following:
A) IDE disk interface. Albeit if hfs (and maybe iso) support for the SD/MMC card is added it might be a little redundant, it does open the posibility of using one of those cheap IDE to SATA or IDE to CF adaptors.
B) Floppy drive interface. Albeit if write support for adf files is added and ipf support is written it might be redundant, but if it functions like the catweasel adapter (reading all kind of floppy disks) it might be very useful.
C) Parallel port. For people with amiga friendly printers, people that still have one of those ZIP drives or people that have one of those adapters to add two extra joysticks. It might also be used to implement a video slot.
D) SCSI port. If the A570 expansion/A2091 card is implemented in full, it would allow a real SCSI port (besides a virtual SCSI hard disk via hfs files) to connect external scanners, hard disks, removable media. etc.
-
I desperately need parallel port and IDE.
Anyone who considers himself an Amiga gamer must have a 4-player adapter that plugs into the par port. Great fun at parties!! :-D
-
little wrote:
...if the 68k is succesfully placed inside the FPGA...
The 68K is succesfully placed inside the FPGA by me! Also the SDRAM Controller and the Spihost for floppy emulation! You can try it if you have a DE2 Board from Terasic.
I hope i can finish the port for the DE1 Board next week.
Viele Grüße
TobiFlex
-
Are you going to release details about the modification to MiniMig core to use an SDRAM controller rather than SRAM?
Doesnt have to be "code", just in overview.
How does latency and burstiness affect things? Did you run the SDRAM at a multiple of the MiniMig core frequency?
Has anyone looked at PSRAM (Pseudo SRAM) as a viable alternative to minimig's SRAM? I think they are cheaper, larger, and would work just like SRAM. Perhaps a very "quick n dirty" alternative.
-
TobiFlex wrote:
The 68K is succesfully placed inside the FPGA by me! Also the SDRAM Controller and the Spihost for floppy emulation! You can try it if you have a DE2 Board from Terasic.
I hope i can finish the port for the DE1 Board next week.
Viele Grüße
TobiFlex
Cool! Please keep us posted :-)
As a side note, can anyone tell me if it would be posible to make this external parallel port floppy drives (http://www.oemliquidators.com/search_result.asp?product_id=5500728&title=EXTERNAL%20LAPTOP%20%20FLOPPY%20DRIVE%20-%205500728) read amiga formatted disks? If it would be posible then a parallel port would be the best option, specialy since it might be feasible to dinamically change it from parallel to scsi :-D
-
little wrote:
As a side note, can anyone tell me if it would be posible to make this external parallel port floppy drives (http://www.oemliquidators.com/search_result.asp?product_id=5500728&title=EXTERNAL%20LAPTOP%20%20FLOPPY%20DRIVE%20-%205500728) read amiga formatted disks?
No. It probably would not be possible.
It depends on the capabilities of the floppy to parallel port chip inside. Most are very PC 1.44Mb orientated.
No-one has _ever_ been able to make one of these drives read Amiga disks when connected to a PC and hence the reason for Catweasel controllers.
It would not be suitable for DF0, connected to a minimig via it's parallel port.
A regular PC internal floppy disk (or a DELL external laptop disk) would be perfect though.
Traditionally you cannot use PC 1.44mb floppy drives on a real Amiga because the 2x rotational speed of these drives means the data is 2x bandwidth and the Paula chip(?) cannot cope. This limitation can be removed in MiniMig.
-
alexh wrote:
Are you going to release details about the modification to MiniMig core to use an SDRAM controller rather than SRAM?
Doesnt have to be "code", just in overview.
How does latency and burstiness affect things? Did you run the SDRAM at a multiple of the MiniMig core frequency?
Hi alexh,
I don't know how exactly i can help you?
The sdram clock is 7,09Mhzx16=113MHz. I have remove the DCM from the minimig source and all clocks came from the SDRAM Controller. So I have a synchron design. Look into the attatched code. The C_7m is the clock for the minimig. If it is Low The Controller Read or Write Minimig data. If it is high the Controller read or write spihost data.
the latency is 2 and the read burst is 4.
The Spihost works with the nice old Z80 but with the power of 60MHz. For the Z80 the controller also cache 8 opcode bytes.
If i use also cache opcodes for the 68K i can make the minimig 2-4x faster i think.
--
-- Copyright (c) 2006-2007 Tobias Gubener
--
-- All rights reserved
--
-- Redistribution and use in source and synthezised forms, with or without
-- modification, are permitted provided that the following conditions are met:
--
-- Redistributions of source code must retain the above copyright notice,
-- this list of conditions and the following disclaimer.
--
-- Redistributions in synthesized form must reproduce the above copyright
-- notice, this list of conditions and the following disclaimer in the
-- documentation and/or other materials provided with the distribution.
--
-- Neither the name of the author nor the names of other contributors may
-- be used to endorse or promote products derived from this software without
-- specific prior written permission.
--
-- THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
-- AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO,
-- THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
-- PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE
-- LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
-- CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
-- SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
-- INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
-- CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
-- ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
-- POSSIBILITY OF SUCH DAMAGE.
--
library ieee;
use ieee.std_logic_1164.all;
use ieee.std_logic_unsigned.all;
entity sdram is
port
(
sdata : inout std_logic_vector(15 downto 0);
sdaddr : out std_logic_vector(12 downto 0);
dqm : out std_logic_vector(3 downto 0);
sd_cs : out std_logic_vector(3 downto 0);
ba : buffer std_logic_vector(1 downto 0);
sd_we : out std_logic;
sd_ras : out std_logic;
sd_cas : out std_logic;
sysclk : in std_logic;
reset : in std_logic;
zdatawr : in std_logic_vector(7 downto 0);
zAddr : in std_logic_vector(22 downto 0);
zwr : in std_logic;
datawr : in std_logic_vector(15 downto 0);
rAddr : in std_logic_vector(22 downto 0);
rwr : in std_logic;
dwrL : in std_logic;
dwrU : in std_logic;
zstate : in std_logic_vector(2 downto 0);
dataout : out std_logic_vector(15 downto 0);
zdataout : out std_logic_vector(7 downto 0);
c_56m : out std_logic;
zena_o : out std_logic;
c_28m : out std_logic;
c_7m : out std_logic;
reset_out : out std_logic
);
end;
architecture rtl of sdram is
signal initstate :std_logic_vector(3 downto 0);
signal slow :std_logic_vector(7 downto 0);
signal pass :std_logic_vector(1 downto 0);
signal datard :std_logic;
signal sdcom_ena :std_logic;
signal isd_ras :std_logic;
signal isd_cas :std_logic;
signal isd_we :std_logic;
signal nanostate :std_logic_vector(3 downto 0);
signal ras_pos :std_logic;
signal wr_int :std_logic;
signal rd_int :std_logic;
signal init_done :std_logic;
signal Addr :std_logic_vector(22 downto 0);
signal isdaddr :std_logic_vector(12 downto 0);
signal isd_cs :std_logic_vector(3 downto 0);
signal datain :std_logic_vector(15 downto 0);
signal casaddr :std_logic_vector(10 downto 0);
signal casba :std_logic_vector(1 downto 0);
signal wr :std_logic;
signal sdwrite :std_logic;
signal nextpix :std_logic_vector(15 downto 0);
signal zena :std_logic;
signal zcache :std_logic_vector(63 downto 0);
signal zcache_addr :std_logic_vector(22 downto 0);
signal zcache_recall :std_logic;
signal fill_zcache :std_logic;
signal match_zcache :std_logic;
signal zvalid :std_logic;
signal zequal :std_logic;
signal V_cycle :std_logic;
signal T_cycle :std_logic;
signal R_cycle :std_logic;
signal Z_cycle :std_logic;
begin
Addr <= rAddr when R_cycle='1' else zAddr;
wr <= '0' when (R_cycle='1' AND rwr='0') OR (Z_cycle='1' AND zwr='0') else '1';
c_7m <= T_cycle;
sd_ras <= isd_ras or sdcom_ena;
sd_cas <= isd_cas or sdcom_ena;
sd_we <= isd_we or sdcom_ena;
sdaddr <= isdaddr;
sd_cs <= isd_cs when sdcom_ena='0' else "1111";
zena_o <= '1' when zena='1' or match_zcache='1' else '0';
reset_out <= init_done AND reset;
process (sysclk, zAddr, zcache_addr, zcache, zstate, zequal, zvalid, match_zcache, nextpix) begin
if zaddr(22 downto 3)=zcache_addr(22 downto 3) then
zequal <='1';
else
zequal <='0';
end if;
if (zstate(2)='1' or zstate(1 downto 0)="01" or (zstate="00" and zequal='1' and zvalid='1')) then
match_zcache <='1';
else
match_zcache <='0';
end if;
case match_zcache&(zaddr(2 downto 1)-zcache_addr(2 downto 1)) is
when "100"=>
if zaddr(0)='1' then
zdataout <= zcache(55 downto 48);
else
zdataout <= zcache(63 downto 56);
end if;
when "101"=>
if zaddr(0)='1' then
zdataout <= zcache(39 downto 32);
else
zdataout <= zcache(47 downto 40);
end if;
when "110"=>
if zaddr(0)='1' then
zdataout <= zcache(23 downto 16);
else
zdataout <= zcache(31 downto 24);
end if;
when "111"=>
if zaddr(0)='1' then
zdataout <= zcache(7 downto 0);
else
zdataout <= zcache(15 downto 8);
end if;
when others=>
if zaddr(0)='1' then
zdataout <= nextpix(7 downto 0);
else
zdataout <= nextpix(15 downto 8);
end if;
end case;
end process;
process (sysclk) begin
if (sysclk'event and sysclk='1') then
if nanostate(2 downto 0)="111" then
R_cycle <= '0';
V_cycle <= '0';
T_cycle <= '1';
Z_cycle <= '0';
IF slow(2 downto 0)=5 THEN
slow <= slow+3;
ELSE
slow <= slow+1;
END IF;
case slow(3 downto 0) is
when "0001" => V_cycle <= '1'; --refresh cycle
when "0000"|"1000"|"0010"|"0100"|"1010"|"1100" => R_cycle <= '1';
T_cycle <= '0';
when others => Z_cycle <= '1';
end case;
end if;
end if;
end process;
--Datenübernahme
process (sysclk, reset) begin
if reset = '0' then
zcache_recall <= '0';
zvalid <= '0';
zena <= '0';
elsif (sysclk'event and sysclk='1') then
nextpix <= sdata;
if zequal='1' and zstate="11" then
zvalid <= '0';
end if;
case nanostate(2 downto 0) is
when "000" =>
c_56m <= '0';
if fill_zcache='1' then
zcache(47 downto 32) <= nextpix;
end if;
when "001" =>
c_28m <= '0';
c_56m <= '1';
if fill_zcache='1' then
zcache(31 downto 16) <= nextpix;
end if;
when "010" =>
zcache_recall <= match_zcache;
c_56m <= '0';
if fill_zcache='1' then
zcache(15 downto 0) <= nextpix;
end if;
fill_zcache <= '0';
when "011" =>
c_28m <= '1';
c_56m <= '1';
if Z_cycle='1' then
if zcache_recall='0' then
zena <= '1';
if zstate="00" then
zcache_addr <= zaddr;
fill_zcache <= '1';
zvalid <= '1';
end if;
end if;
end if;
when "100" =>
c_56m <= '0';
when "101" =>
c_28m <= '0';
c_56m <= '1';
when "110" =>
c_56m <= '0';
if fill_zcache='1' then
zcache(63 downto 48) <= sdata;
end if;
if R_cycle='1' then
dataout <= sdata;
end if;
when "111" =>
c_28m <= '1';
zena <= '0';
c_56m <= '1';
when others =>
end case;
end if;
end process;
process (sysclk, reset, sdwrite, datain) begin
IF sdwrite='1' THEN
sdata <= datain;
ELSE
sdata <= "ZZZZZZZZZZZZZZZZ";
END IF;
if reset = '0' then
initstate <= (others => '0');
init_done <= '0';
nanostate <= "0000";
else
if (sysclk'event and sysclk='1') then
sdcom_ena <= '1';
ras_pos <= '0';
nanostate <='0'&nanostate(2 downto 0)+1;
if ras_pos='1' then
casaddr <= addr(10 downto 0);
casba <= ba;
IF R_cycle='1' THEN
datain <= datawr;
ELSE
datain <= zdatawr&zdatawr;
END IF;
end if;
case nanostate(2 downto 0) is --LATENCY=2
when "000" => pass <= "00";
when "001" => sdcom_ena <= '0';
ras_pos <= '1';
when "010" => pass <= "01";
sdwrite <= '1';
if V_cycle='1' then
wr_int <= '1';
rd_int <= '1';
else
wr_int <= wr ;
rd_int <= '0';
end if;
when "011" =>
sdcom_ena <= '0';
when "100" =>
pass <= "11";
sdwrite <= '0';
when "101" =>
when "110" =>
when "111" =>
if initstate /= "1111" then
initstate <= initstate+1;
else
init_done <='1';
end if;
when others =>
end case;
end if;
end if;
end process;
process (initstate, pass, addr, wr, datain, init_done, casaddr, rd_int, wr_int, dwrU, dwrL, casba, Z_cycle, V_cycle, zaddr) begin
isdaddr<="XXXXXXXXXXXXX";
isd_cs <= "1111"; --NOP
isd_ras <= '1';
isd_cas <= '1';
isd_we <= '1';
ba <= Addr(22 downto 21);
if wr_int='0' then
if Z_cycle='1' THEN
dqm <= ("11"& zaddr(0)& not zaddr(0));
else
dqm <= ("11"& dwrU& dwrL);
end if;
else
dqm <= "1100";
end if;
if init_done='0' then
case initstate & pass is
when "001000" => --PRECHARGE
isdaddr(10) <= '1'; --all banks
isd_cs <="0000";
isd_ras <= '0';
isd_cas <= '1';
isd_we <= '0';
when "001100"|"010000"|"010100"|"011000"|"011100"|"100000"|"100100"|"101000"|"101100"|"110000" => --AUTOREFRESH
isd_cs <="0000";
isd_ras <= '0';
isd_cas <= '0';
isd_we <= '1';
when "110100" => --LOAD MODE REGISTER
isd_cs <="0000";
isd_ras <= '0';
isd_cas <= '0';
isd_we <= '0';
ba <= "00";
isdaddr <= "0001000100010"; --BURST=4 LATENCY=2
when others => --NOP
end case;
else
case pass is
when "00" => --ACTIVE
isdaddr <= "0"& addr(20 downto 9);
isd_ras <= '0';
isd_we <= '1';
if V_cycle='1' then
isd_cs <="0000"; --AUTOREFRESH
isd_cas <= '0';
else
isd_cas <= '1';
isd_cs <= "1110"; --ACTIVE
end if;
when "01" => --READ or Write
ba <= casba;
isdaddr <= '0'& '0' & '1' & "00" & casaddr(8 downto 1);--auto precharge
isd_cs <= "1110";
isd_ras <= '1';
isd_cas <= rd_int;
isd_we <= wr_int;
when others => --NOP
end case;
end if;
end process;
end;
-
A regular PC internal floppy disk (or a DELL external laptop disk) would be perfect though.
Mhhh, how about adding a selector header (like the one that selects 15/31 khz) to select if the minimig has an active internal floppy interface or an active external parallel port? Chances are people will not need both (or at least will not use both at the same time).
Just thinking out loud, as usual :-)
-
little wrote:
how about adding a selector header (like the one that selects 15/31 khz) to select if the minimig has an active internal floppy interface or an active external parallel port? Chances are people will not need both (or at least will not use both at the same time)
I think Parallel port and external floppy disk drive are the the same connector.
http://www.amiga-hardware.com/download_photos/a500_4_big.jpg
If you had a minimig with this connector (and an FPGA with enough I/O) All you need is a different MiniMig "image" on the SD card to switch between parallel port or real floppy disk.
-
alexh wrote:
I think Parallel port and external floppy disk drive are the the same connector.
Sadly, the amiga external floppy port is a db23 (http://pinouts.ru/Storage/ExtDiskAmiga_pinout.shtml) and not a db25 (http://pinouts.ru/ParallelPorts/parallelamiga_pinout.shtml). Besides amiga external drives (low or high density) are expensive and hard to get, while internal pc floppy drives are cheap and easy to get, that is why I though of an internal floppy interface.
If you had a minimig with this connector (and an FPGA with enough I/O) All you need is a different MiniMig "image" on the SD card to switch between parallel port or real floppy disk.
That would be cool!
-
I was looking at Actels IGLOO PLUS FPGAs, can any of these FPGAs be used for future Minimigs ?
They are very cheap - some as low as $1 :)
IGLOO SERIES (http://www.actel.com/products/iglooseries/default.aspx)
-
You might be able to use the AGL250 and above.