NatAmi 68070 design draft

[bloodline]

HenryCase wrote:

bloodline wrote:
NATAMI is anything but elegant... the Amiga was an elegant solution to the computing problems of the mid 80's... Take any generic mainstream PC/Mac/whatever and it will be far more elegant a solution to the modern computing environment than an weird Amiga like kludge...

You and I have different interpretations of the word 'elegant' it seems. I am looking from the perspective of less bottlenecks in the system, which the PC/Mac/whatever have. What's the point of all that CPU power if you can't use it to its full potential? Why use the CPU for everything when co-processors can do a better job? As I said PC architecture is moving in the co-processor direction (and has been for a while now) but it's not quite where I'd like it to be yet.

There are more "Co-Processors" in even a 10 year old PC than there ever was in the Amiga...

I would love for you to list these "bottlenecks", that PC's have... I can think of plenty present in the AGA chipset, off the top of my head...

bloodline wrote:
A5000 is a bit of a meaningless idea, But I do understand that it would have been nice to see what could have been... but as the great Dave Haynie has pointed out, The future of Amiga development would have moved to off the shelf parts... AAA was OK, in 93... but nothing but a joke by 97...

AAA != SuperAGA, but I see where you're going by mentioning it.

Personally I have no issue with the PARISC strategy that was planned for the new Amiga CPU, as at least the 68k functions could have been built in to the new CPU core.

That was NYX, that was not Amiga compatible, it was a totally new architecture... I was meant to replace the Amiga, though Chris Ludwig did mention that AmigaOS would have been ported to it, the main OS would have been WinNT... Had commodore started on NYX earlier and got it going, then it might have push Commodore ahead again like the Amiga did al those years before...

 However, if Commodore were planning on going with all off the shelf parts then I'm glad they fell on their arse when they did because they clearly didn't see the unique benefits of the Amiga architecture.

Moving to off the shelf parts would have been a lazy attempt to renovate the platform to make up for the years they wasted by not investing highly in R&D.

Yes, that's right... and it was the only realistic solution after they failed to capitalise on the Amiga technology....We should have been phasing the AGA chipset out by 1990... perhaps with something like AAA filling the gap until something like NYX could be brought to market by 1992...

Using your same Dave Haynie AAA reference 'it would have been revolutionary if released in 1990'. Commodore had 5 years to get from OCS to AAA, which is plenty of time IF they properly invested in R&D.

Yes, that's sad. But that's what happened and the rest of the world moved on!

[bloodline]

Plaz wrote:
@bloodline

like just how suited the Coldfire actually is to running 68k code

You're correct, it isn't. That's why you put in the JIT.

No, I mean running a sequence of coldfire instructions that perform the same function as the missing 68k instruction... it's not a simple 1 to 1 mapping... The 68k Addressing modes can require a huge amount of additional logic and arithmetic code to emulate and we haven't even covered Flag calculation... The 68k is horrible to Emulate... that's why you need something big and powerful like the x86 :-)

You are suggesting a symmetric design, which is expensive for this...

Not so sure. Not in CPU's anyway. 166/V2's about $20US each. 200/V4e's About $32US each. 68060 if you can find new stock....$500US. Coding bios an glue logic for a card is where the time would be spent though.

But two coldfires emulating a single 68k, at a speed, perhaps as fast as a 40Mhz 68040... really why bother at this point? Just buy a cheap PC (£250 including everything including a monitor!) and run WinUAE...

[bloodline]

HenryCase wrote:

bloodline wrote:
Have you noticed that we have gone from lots of different CPU architectures, to basically just 2... the x86 and the ARM... Yes, the PPC is clinging on too... I think you would be hard push to justify why you didn't use either the x86 or the ARM in a new design... And I think you would, probably kiss your job as a chip/system designer goodbye if you choose something other than x86, ARM or PPC in your system...

There simply isn't the investment in software to justify building a system built around the 68k...

There are plenty of different CPU architectures out there. Whilst I concur that the biggest players in the consumer CPU architecture market are x86/x64 and ARM, that doesn't stop companies using other architectures where appropriate.

I'm talking about general purpose CPU architectures... i.e. the market that the 68k was designed for...

There is basically either an x86 or an ARM that fits into any market and price point now, that needs a general purpose CPU... And you are talking to a MIPS fan here...

bloodline wrote:
Why would any company be interested in a weird, slow, incompatible and expensive design... where is the value in that? How can you sell that to anyone?

Aah, I'm starting to see where you're getting confused now...

bloodline wrote:
By which time, your Average PC will be 20 times more powerful!

I have nothing against NATAMI, but you need to be realistic in what can be achieved!

I'm trying to be realistic in what the Natami can achieve, believe me there are people much more fanatical about the Natami in the Amiga community than me.

But I doubt they are able to see the flaws in their thinking....

I think the reason we're not seeing eye to eye on this is because you think I expect the Natami to compete in the modern PC market. I do not. I do not expect to see it on the shelves of shops

Woah!!! Hang on there!!!! One step at a time... We've not got that far yet! We still have to sell this idea to an investor before we can manufacture it...

(at least not unless an outside company takes a liking to the N68070/SuperAGA chip, for mobile phones for instance).

Ok, give me one thing that N68070/SuperAGA has over an StrongARM/PowerVR chip (i.e. the chipset in the iPhone)?

 I expect it to be a successful product in the Amiga market, maybe drawing back a few old Amiga users and a few developers interested in retro hardware (demoscene coders for example) but not reviving the commercial viability of the platform.

The MiniMig was good in the Amiga Market, it was a cheap, simple design and generally compatible with most existing software.

The NATAMI is looking to be expensive and incompatible with features that are not required by the amiga software base.... How do you sell this to an investor?

The Natami is an ambitious hobby project and we have nothing to lose by giving the Natami developers our support. If the N68070 never sees the light of day at least we should have the Natami60 to play around with.

The NATAMI is a fun hobby project, really good... but I really can't be much more... the N68070 is totally pie-in-the-sky, I would suggest the originators of the idea do a lot of reading...

[bloodline]

Plaz wrote:

The 68k Addressing modes can require a huge amount of additional logic and arithmetic code to emulate and we haven't even covered Flag calculation

There's only about 12 instructions to deal with. You just never know when their going to hit the CPU though. The coldfire is a 68K core and about 99.2% compatible with 68x00. The trouble comes when one of thes dozen instructions does a diffent task on the CF than it does on the cold fire. Supervisor mode can capture and redirect invalid instructins to a lib where software takes care of it. However, these few instructions are not invalid and can't be trapped that way. You have to intercept them some how. If you have to intercept a few, you're just as well off intercepting them all with a fast JIT. IO is another problem. V2 is very pin compatible, but slow. V4 is much faster, but not IO compatible. So... to be more specific on my idea. V4 to do JIT to handle invalid instructions, and "feed" the V2 to handle buss IO and execute program code.

All good conversation, but obviouly going nowhere unless I win the lottery this week :-P

Well lets hope you win that lottery... but in the mean time...

I've spent a fair bit of time reading the tech docs for the Coldfire and it is missing quite a bit, instructions are one thing, but addressing modes are the bane of RISC designs and expunged.

Yes, a JIT is the best way to run 68k code on the Coldfire... but the Coldfire is not a instruction munching monster... it derives it's performance from not having to execute the CISC baggage of the 68k... once the 68k code has con through the JIT all that baggage is added back in...

The supervisor mode of the Coldfire is totally different to the 68k... this means no AmigaOS... your best bet then is to use AROS...

Just buy a cheap PC (£250 including everything including a monitor!) and run WinUAE

Not any cheaper than a new card if some one can come up with one. PC emulation is a perfect idea for some things, but I can't run all my hardware. For video and music studio work I might as well just stick with my XP and all the USB hardware. As long as emulation is just amiga "running in a vacuum" it's not useful to me. It all comes back to... no hardware. Ok I'm depressed now. Time to go work on the toaster/flyer box.

Plaz

I was in a similar situation about 10 years ago... when I needed to keep up with the Music technology curve and the Amiga just wasn't cutting it... My Amiga has been relegated to old synth/effect module status... rather than the centrepiece it once was.

[bloodline]

HenryCase wrote:

I'm talking about general purpose CPU architectures... i.e. the market that the 68k was designed for...

There is basically either an x86 or an ARM that fits into any market and price point now, that needs a general purpose CPU... And you are talking to a MIPS fan here...

But 68k does fit in the Amiga market. The Natami is not aiming to be some PC killer, but rather a supercharged classic Amiga.

This isn't really an Amiga market anymore... there a retro/hobbyist market, and they aren't interested in something a bit compatible with the Amiga... They want to run their old Amiga software, they don't need or want a 9Ghz 68k...

But I doubt they are able to see the flaws in their thinking....

I don't know if that was meant to be a sly dig at me or not but I personally I think I'm fairly willing to admit when I've got something wrong.

I'm not bothering with sly digs tonight! :-)

Ok, give me one thing that N68070/SuperAGA has over an StrongARM/PowerVR chip (i.e. the chipset in the iPhone)?

It will have me coding for it. ;-)

Have you downloaded the iPhone SDK... it's pretty cool... I bet you'd love it even more than Amiga :-o

In all seriousness, it's clear that software is the thing that drives computer sales.

There is nothing else that drives computer sales... Take the iPhone for example, the hardware was pretty unimpressive... but the software was totally innovative...

From a technical perspective I think the N68070/SuperAGA chip would work well in a mobile phone.

That is delusional... it would be expensive, use too much power, get too hot, offer hardly any features of competing hardware and no way could anyone ever afford to have it fabbed at a process small enough to make it fit in a mobile.

The software is another matter. The direction of the commercial Natami will largely be determined by the Natami60 software developers.

:-?

The NATAMI is looking to be expensive and incompatible with features that are not required by the amiga software base.... How do you sell this to an investor?

With a library of Natami60 apps and games.

Your business model is technically "The Underpants Gnomes Model"

 I personally don't see any investors coming from outside the Amiga community, but who knows.

Well, go on... sell it to me...

bloodline wrote:
The NATAMI is a fun hobby project, really good... but I really can't be much more... the N68070 is totally pie-in-the-sky, I would suggest the originators of the idea do a lot of reading...

As I've said, the N68070 is a fairly new idea. I'm glad you approve of the Natami as a fun hobby project.

Well.. I am an Amiga guy... just a realistic one...

The best way we have to ensure we see this hobby project succeed is to give our advice and support to the Natami team, so what would you add/change in the current N68070 design brief?

Everything... :-(

[bloodline]

Crumb wrote:

Something as complex as a 68k CPU is never going to get even 20Hmz real speed on an FPGA... I'd be surprised if that to be honest...

Tobiflex m68k core is faster than a 20Mhz 68000 with a DE-1 board.

Yeah, and really how fast does it actually run?

[bloodline]

Zac67 wrote:

bloodline wrote:
perhaps with something like AAA filling the gap until something like NYX could be brought to market by 1992..

Err ;-)

The Nyx was the AAA prototype... link

You are correct! I am thinking of Hombre, all these stupid code names so long ago :-)

[bloodline]

Plaz wrote:

it will be a nightmare to develop and write software for unless you abstract the entire system... and if you are going to do that... why use Coldfire?

Yes, I mentioned earlier that the bios (firmware) and glue logic would be the biggest tax on such a design. Coldfire would have seemed like the logical successor to 68K for backward code compatibility with classics, but it's proving to be untrue. Too limited, too slow. A card with an embedded JIT and a 4ghz x86 chip wouldn't be easy either, but starts to make more sense.

For me that would make the most sense for an Amiga accelerator... if anyone still wanted to make one... But I think seriously, an ARM core with a JIT is the most sensible long term solution... The ARM is low power, well supported, it's not going anywhere and with a nice JIT should reach standard 680x0 performance with the current generation...

[bloodline]

Kin-Hell wrote:

Zac67 wrote:

The Motorola 68060 CPU was designed that way. Its a CISC decoder in front of a RISC execution pipeline.

Err...

The '060 has no RISC core - this is garbage. RISC starts with Coldfire.

And I'd like a to see an available 5 GHz RISC CPU...

....Bullshit....

CSPPC came before ColdFire. That Power PC CPU is RISC dude!

Too Risque!

Coldfire Origins....

;-)

But Zac67 is right, the 68060 is not RISC...

You can read all about the 060 architecture here:

http://security-protocols.com/library/phreaking/68060Info.txt

It explains it all rather simply (If English is your first language).

[bloodline]

tiffers wrote:
@vloodline

The MiniMig was good in the Amiga Market, it was a cheap, simple design and generally compatible with most existing software.

The NATAMI is looking to be expensive and incompatible with features that are not required by the amiga software base.... How do you sell this to an investor?

How do you figure this?

From my understanding of technology and the available documentation regarding the NATAMI.

Sure the prototype, and the planned developer boards will be using a FPGA based board which are expensive for the features they offer.

I dissagree, an FPGA developer board is great value for money... for the hobby market, which is the market we are currently in.

However, they are investigating the coldfire option in order to lower that price barrier. Once they started looking at coldfire, and incompatabilities in the instruction set, they started looking into ways to work around those incompatabilities.

The Coldfire really isn't looking attractive...

First they suggested a .library file, or modifying code (hard to do when you don't have the source)

:-? :-? :-?

Next they discovered that they could get their SuperAGA FPGA stuff programmed into the Coldfire package, giving a low-cost ($20 / 'chip' was a price quoted) solution, which would result in a very affordable ($100 for a working board) product.

You what? $20 FPGA + a Coldfire... you have to be dreaming!?!?!

Of course then someone said, it'd be cool if you could stick a 68k into the FPGA code, then you wouldn't have to bother with coldfire issues. This is where the 68070 effort began.

I like the idea of having the 68k and the chipset on a single FPGA. I think it's agreat idea, perfect for the MiniMIG etc... But I get very annoyed when I see these crazy threads about creating a "Super68k"... IMO we should be happy to get something that is compatible with as of similar performance to the 68030... if possible, into a small enough design that can share silicon with the Chipset.

Now, how do-able this is, I don't know.
But if they could get a $20 SuperAGA + 680x0 on Coldfire solution done, and licence it out to whoever wanted to build a board to go with it, how many do you think would sell at $100 / board?

5 (with a +-2 margin of error).

The NatAmi team is claiming 100% AGA compability (not including timing)

Timing is EVERYTHING with the Amiga chipset! Listen to Jens talking about the CloneA, as soon as you are not cycle exact, you are not compatible... If you are not 100% compatible, then there is NO point using an Amiga chipset... Just use a modern one that works better.

so they say just about anything you care for (maybe not demos) should run as normal on it.

If it's not cycle exact, then only thing that might work is the OS... and anything that uses the OS in a system legal way... If that is the case, then why bother with the "Amiga Style" GFX chips?

Then they're extending it to have a larger CHIP RAM address space

Increase the Chipram, and you become incompatible...

 and that chip ram is SRAM not DRAM, so it's much faster, allowing them to run an extended instruction set in their SuperAGA 'chipset' to give better audio, better CG (including 3D) and other features, while remaining 100% compatible.

SRAM or DRAM... make no difference for something as slow as the Amiga Graphics architecture... DRAM chips are easily as fast (if not much faster) as requried. SRAM chips are just easier to interface with, thoguh quite a bit more expensive!

So it's compatible,

No it's not. I know you don't beleive me... But you have to believe Jens from Individual computers?!?!

faster,

Faster? Fast than what? A 16 year old graphics chipset... I could buy a gfx chip off the shelf for $2 that would be faster and offer more features than NATAMI...

[/quote]
 more featured,
[/quote]

More features than a 16year old chipset? See above!

and planned on being very affordable.

I plan on Dating a German supermodel with a degree in quantum mechanics... can you see where I'm going with this?

How many people out there do you think would get one? How many people bought C64DTVs?

I don't know... they were heavily discounted here (to somthing like £5) and I still didn't bother buying one... And no more were or are planned to be made... I guess it wasn't that popular...

I know I'll get a NatAmi if they can attain these goals.

If I meet that German supermodel with more than a passing interest in theoretical physics... I'll think about letting her go out with me...

[bloodline]

amigaksi wrote:
>>    wolfchild wrote:
>>    Whenever Natami is mentioned, it seems inevitable that the project team members will eventually be ridiculed.

>In order for them to be ridiculed, first they must write something ridiculous surely?

Not really, some people like Galileo

Galileo was not ridiculed... he was persecuted by the Church, an organisation with hardly a good reputation regarding scientific matters...

, Newton, etc.

Newton was a highly respected scientist... never ridiculed... Do you even know history?

got ridiculed although they were right.  Some people have more experience/knowledge of specific subjects than others and if it does not fit into the "norm" of what everyone else does/follows, it makes others ridicule them.

Err... no... we don't live in the middle ages... we have the scientific method. If you suggest something and are able to back it up with evidence, repeatable results and testable observations, people will test and except your results.

 Although, most people picked the slower-processor based Amiga at 7.16Mhz over the Atari ST (@8Mhz), that was a good choice in the 80s.  

I doubt even 1% of the people who bought computers in the 80s even know what speed their CPU was... or even that CPUs had speed ratings.

You bought the computer that had the most software, or the one that had the better software...

Now if you use an Amiga over a 4Ghz PC, it's not the "norm" since only the processor speed counts in most people's minds.

If you use an Amiga over a 4Ghz PC (does anyone have a 4Ghz PC?!?!), then you don't need the modern features that modern hardware and software can provide.

I can no longer use my Amiga as either my work machine... my hobby machine or even my everyday machine...

[bloodline]

wawrzon wrote:
bloodline, you are overly negative, just as biggun is too enthusiastic over the matter.

Overly neagative say you. But I am not happy with people basically making a joke out of the Amiga. It was the best computer of its time... But all these absurd claims do nothing but drag the memory of this milesone in computer history through the mud.

look at this thread, some 50% of it are your posts, trying to convince people that something is impossible, i thing u could easily use this energy in some more constructive way.

I'm simply trying to show people the flaws in their thinking. I offer my advice based on the research that I do and the experience I have, I don't not pull numbers out of the air, and make insane claims, that are not technically feasable!

i believe, most people involved in amiga scene have got a clou about natami, but just keep quiet as long as anything is proven either way (at least me do).

as for selling rates of a complete amiga compatible system for  any 100eur not to mention 100$ (lol) it would surely exceed 10 pieces.

But not, if it was not 100% Amiga compatible. The NATAMI team can waste time adding a 1000 new features but for every incompatiblity it introduces... the market is halved.

just notice how much you pay for any old amiga hardware today.

Hadware that is proven to work with the Amiga software base.

and dont mention minimig should have been a bigger succes as nobody sane actually needs replacement for original a500 i believe. well i have no need at least.

What Amiga Compatible do you have need for?

as for jens schoenfeld arguments, do not forget that he is working on a concurent design. the amiga-hardware designers scene here in germany and poland (e3b, elbox, ic) seems to takes no risks by supporting each other products, due to the little market i believe.

j.s. always underlines that clone_a would be cycle exact replacement for the amiga, but is an a4k (with a cyberstorm060/ppc) cycle exact with a a500?

An A4000 is cycle exact with the Amiga... it is an Amiga...

if it is probably natami aga chipset replacement is compatible too

But with the AGA chipset timing was never really an issue, it was the 2megs Chip and register gaps filled with new functions that caused the problems... switch these off (in the early boot) and all but one of my A500 software would work...

since there it was shown to public and there are snapshots of it running genuine amiga apps on the natami page.

I've seen screen shots of Workbench running... That is all... And that could run on Amithlon with no Amiga hardware (and very minimal CIA emualtion).

[bloodline]

alexh wrote:

bloodline wrote:
But with the AGA chipset timing was never really an issue

Not sure that is 100% true.

I did say "really", but I can't off the top of my head think of any place where I had an issue with the timings... Blitter and DMAs and stuff all seems to be the same... (Though I remember the AGA fetch modes all being messed up...) Though I'm not going to go back and look at 14year old source code to confirm :-) I didn't have an A1200 compatible ActionReplay either so I never hit the AGA hardware as hard as the OCS...

bloodline wrote:
it was the 2megs Chip and register gaps filled with new functions...

Yeah, and more importantly the new kickstart ROM and cpu caches.

I was going to mention them... but really we are talking about the chipset...

bloodline wrote:
switch these off (in the early boot) and all but one of my A500 software would work...

You must not have had much software then ;-)

Hhahaha, I was poor back then! Seriously, I think it was only MegaLoMania that really put up a fuss... Though I "know"* I got it working... Perhaps something else too... Damn UAE is so much easier :-)

*as in; quite sure...

[bloodline]

Crumb wrote:
@JJ

Do you have graphic card in that PPC 1200? A BlizzardVisionPPC would be perfect for running MorphOS or AmigaOS4

I have a BVision in my A1200... but can't run OS4 as I only have 32Meg Fast.

[bloodline]

wawrzon wrote:
@bloodline: what kind of amiga compatible i need, you ask?
accidentally just that kind the natami specs state, even up to genlocking (overlay) possibility, because that is what i tehnically use them mostly for.

I did not ask what you wanted. I asked what you actually needed from your Amiga. Do you need an Amiga Compatible or some kind of SuperAmiga that might not be so compatible.

i wouldnt complain about some extra computing speed but thats not the primary issue for me.

So what you need is an Amiga compatible... So something like a bugfixed MiniMig or CloneA would suit your needs?

i can use pc for most computing tasks. up till now i mostly used amigas as simple, reliable, cheap, idiot-proof (no shutdown needed, just pull the power cord) controllers for my "video" installations.

Right, so you have software that you currently use on Amiga systems that you need to use in your video installations. That's fine, and when all the original Amiga hardware dies, having a supply of new Amiga Compatible hardware would be good for you... but your software can't use and doesn't need the extra features of NATAMI... So why add complexity (increasing cost and debugging time) to the Amiga Clone by having SuperAGA features?

i do not estimate too that every extra feature introduced to the amiga compatible will narrow the interested user base as long as it doesnt contradict an existing one. it even possibly  will become part of some future standard.

The Amiga chipset has already defined its standard... 16 years of unchanged hardware have defined that standard. Any changes are not compatible and never will be.

i sense you see no future for anything connected to amiga either way since you have already placed it in sentimental past and want to conserve this as such. im fine with it.

No, I want effort to be directed in directions that make sense. The Amiga Hardware has no future, it never had a future, it was a product of the 80's computing environment. It solved problems that technology has surpassed and which simply don't exist any more.

We all here have software that we use, for some reson (it doesn't matter why), that needs an Amiga to run... Amiga hadware has a limted life an we need to develop solutions to account for that... CloneA, Minimig and TobiFlex68k (I personally also include AROS in this list) are all vital in preserving the Amiga... but NATAMI seems like feature bloat and without good reason, except as a private hoby project where it is a good learning tool for someone... and N68070 is just a waste of everyone's time, it can never be as powerful or as complex as a modern CPU, and frankly any software that need a 68k does not need anything better than 68030... really...
 

[bloodline]

tiffers wrote:
@bloodline

I was just quoting what Gunnar/biggun has said in other threads, on other forums.

You (and others here) certainly seem to have much greater knowledge than I, about the specific technologies being discussed.

The technlogies are publically documented, a little bit of time spent of Google and a few Emails to the companies that sell this technology will give you a good idea what is possible and how much it costs... Also there are plenty of people on these forums who work with this stuff everyday who can explain this stuff.

First thing I do when someone makes a claim that seems too good to be true... Google... :-)

I was just trying to show that NatAmi (according to Gunnar) is really placed to be cheaper than the MiniMig (in the CPU + chipset + Coldfire all-in-one configuration) and more compatible than the MiniMig in it's current state, and thus shouldn't be slammed too hard on the 'incompatible and expensive' claims.

But I really can't see:

a) How it can be cheaper than the MiniMIG with the feature bloat of the SuperAGA and even a CPU core on there, we are talking about a HUGE FGA.

b) How it can be as compatible as the MiniMIG, which is already a very mature beta product, and still full of bugs. The added complexity of SuperAGA is going to increase Debug time. The added features of the SuperAGA Chipset will decrease compatibility, unless they are kept away from the AGA emualtion (i.e. moded out)... and then what is the point of them, no existing Amiga software can use them, and if you are writing new software... use a new machine.

c) How can the coldfire really make much sense in this situation... Hell, even I as someone who dislikes the PPC, can see it's a better fit in this situation for wide range of technical and cost reasons...

d) that all the problem which plague the idea of SuperAGA apply to the N68070... Who wants a not very 68k compatible CPU that runs faster than a 68k at several orders of magnitude the cost of an modern CPU?

Sub-$100 NatAmi And another

The proof, as I have heard said, will be in the pudding.

tiffers.

That's true... but at the moment we don't have any pudding... just some vague recipies and expensive ingredients...