Amiga.org
Amiga computer related discussion => General chat about Amiga topics => Topic started by: ElPolloDiabl on August 18, 2013, 12:30:48 PM
-
Hi,
Is there enough desire for kickstarter funded production run of 060's. If so would they have to be faster than 50mhz? Or is 50mhz enough?
As an alternative should we try and get a run of 50mhz 030's?
I would settle for a 50mhz 060 model with working FPU. I could afford to pay $200 for one (CPU only) plus some more for kickstarter.
Are there general purpose fabs that can do a run of 1,000-2,000 chips?
Otherwise I'll settle for the less expensive FPGA route.
-
Is there a problem to find working 060 @80+ US$?
-
50Mhz isn't close to enuff.
For me it needs to be at least 500 Mhz as a bare minimum for a new run. And have MMU + FPU.
If we sit around and do nothing we will get 500Mhz FPGA 060s in 4 years without any kickstarter.
If you want a 50Mhz 060 you can buy those right now. No need for a kickstarter.
If you want a 100Mhz 060 you can buy those right now.
Kickstarter is to make new things that don't already exist.
Another possibility is a bounty or kickstarter to make an ARM JIT to run Motorola bytecodes. You should be able to get 300Mhz 060 speed (without MMU) on a wide variety of cheap ARM cpus.
There is no 68060 JIT for iphones because Steve Jobs personally banned it.
But everyone is allowed to code 68060 JIT for ARM cpus for 10 other devices.
There are multiple 1.0+ Ghz ARM cpu computers (usually with 2-4 cores) in the $99.00 range typically with 1GB+ of RAM.
Or you can buy a top-of-the-line smartphone from Samsung or HTC or Motorola and get a 1.6Ghz to 1.8Ghz ARM cpu with 2 or 4 cores + 2GB RAM and uses hardly any electricity. They are much cheaper and more energy efficient than silly intel processors. But those smartphones have expensive 1080P or 720P screens built in so they cost around $600.00. But $600.00 for a 300Mhz 060 sounds like a damn good deal when I see ppl selling 80Mhz 060 cards with 64MB of ram for the same price!
If you are serious about a kickstarter you really need to find someone who can and WILL code an ARM JIT and focus on that. It is 1000x less risky and cheaper and more achievable than trying to get some lame cpu company to fab new high speed 68060s.
I have been trying to brainwash Mrs. Beanbag into coding it for us but I have not been successful yet. There are various ppl who could do it. Matt Hey, Rachy, whoever wrote the JIT for MorphOS, + a dozen other ppl.
I will pay $300.00 for someone to code a working FREE OPENSOURCE JIT that works on ARM. As proof it must be built into UAE and shown to work or optionally it must be built into a cheap arm computer and shown to run Amiga software. This is so if UAE source code is too hard to understand, then there is another way. For example, you could make it run on Raspberry Pi. But I am more a fan of the better ARM chips in the other competing computers such as Ouya, or any of a zillion other computers in the $99.00 to $250.00 price range. I prefer Cortex A9 atm. But if you don't like that then ok I won't discriminate against Rpi.
My offer expires in 1 year. But I reserve the right to extend it. I can't leave my bounties open ended forever and ever or I might get hit with a whole bunch of bounties all at the same time and not have enuff cash to pay them all at once. So I hafta keep things limited to something that I can GUARANTEED pay. I know $300.00 is nothing for all the hours of work you hafta put into the project. I'm sorry. I am dying and have little money. :( If I was rich I would pay $3000.00 or $30000.00 to get things moving. Grrr! I want something to get done but I have little power :*(
-
$30,000 is halfway there.
I'm sure some unemployed ex Amiga user could do the whole thing for $60,000.
I think that someone will get around to the JIT on UAE eventually so maybe no need for a kickstarter on that.
-
Amiga OS should've been here by now:
AMD Unveils Server Strategy and Roadmap - (June 18, 2013)
http://www.amd.com/us/press-releases/Pages/amd-unveils-2013june18.aspx
Best-in-class ARM CPUs, and leading x86 CPUs and APUs for enterprise and data center servers
"Berlin" chip: "...almost 8X the gigaflops per-watt compared to current AMD Opteron™ 6386SE processor."
"...built on AMD’s revolutionary Heterogeneous System Architecture (HSA), which enables uniform memory access for the CPU and GPU and makes programming as easy as C++"
AMD to launch ARM processor, Seattle, in 2014 - (June 19, 2013)
http://www.bit-tech.net/news/hardware/2013/06/19/amd-to-launch-arm-processor-seattle-in-2014/1
The SoC will be the industry's first 64-bit ARM processor.
For the desktop, AMD CPUs of 2Ghz, 3Ghz, 5Ghz etc.
That's where Amiga OS should've been by now--running on the 'common' hardware like that. :)
-
Amiga OS should've been here by now:
[/I]"Berlin" chip: AMD to launch ARM processor, Seattle, in 2014 - (June 19, 2013)
http://www.bit-tech.net/news/hardware/2013/06/19/amd-to-launch-arm-processor-seattle-in-2014/1
The SoC will be the industry's first 64-bit ARM processor.
For the desktop, AMD CPUs of 2Ghz, 3Ghz, 5Ghz etc.
That's where Amiga OS should've been by now--running on the 'common' hardware like that. :)
I agree in principle. But we don't need the 64-bit part. A normal cheap 32-bit ARM is perfect for AmigaOS.
I am surprised that MorphOS has not been ported to ARM yet.
Once MorphOS does it then Hyperion will do the same thing... 5 years later :)
-
50Mhz isn't close to enuff.
For me it needs to be at least 500 Mhz as a bare minimum for a new run. And have MMU + FPU.
If we sit around and do nothing we will get 500Mhz FPGA 060s in 4 years without any kickstarter.
If you want a 50Mhz 060 you can buy those right now. No need for a kickstarter.
If you want a 100Mhz 060 you can buy those right now.
Kickstarter is to make new things that don't already exist.
Another possibility is a bounty or kickstarter to make an ARM JIT to run Motorola bytecodes. You should be able to get 300Mhz 060 speed (without MMU) on a wide variety of cheap ARM cpus.
There is no 68060 JIT for iphones because Steve Jobs personally banned it.
But everyone is allowed to code 68060 JIT for ARM cpus for 10 other devices.
There are multiple 1.0+ Ghz ARM cpu computers (usually with 2-4 cores) in the $99.00 range typically with 1GB+ of RAM.
Or you can buy a top-of-the-line smartphone from Samsung or HTC or Motorola and get a 1.6Ghz to 1.8Ghz ARM cpu with 2 or 4 cores + 2GB RAM and uses hardly any electricity. They are much cheaper and more energy efficient than silly intel processors. But those smartphones have expensive 1080P or 720P screens built in so they cost around $600.00. But $600.00 for a 300Mhz 060 sounds like a damn good deal when I see ppl selling 80Mhz 060 cards with 64MB of ram for the same price!
If you are serious about a kickstarter you really need to find someone who can and WILL code an ARM JIT and focus on that. It is 1000x less risky and cheaper and more achievable than trying to get some lame cpu company to fab new high speed 68060s.
I have been trying to brainwash Mrs. Beanbag into coding it for us but I have not been successful yet. There are various ppl who could do it. Matt Hey, Rachy, whoever wrote the JIT for MorphOS, + a dozen other ppl.
I will pay $300.00 for someone to code a working FREE OPENSOURCE JIT that works on ARM. As proof it must be built into UAE and shown to work or optionally it must be built into a cheap arm computer and shown to run Amiga software. This is so if UAE source code is too hard to understand, then there is another way. For example, you could make it run on Raspberry Pi. But I am more a fan of the better ARM chips in the other competing computers such as Ouya, or any of a zillion other computers in the $99.00 to $250.00 price range. I prefer Cortex A9 atm. But if you don't like that then ok I won't discriminate against Rpi.
My offer expires in 1 year. But I reserve the right to extend it. I can't leave my bounties open ended forever and ever or I might get hit with a whole bunch of bounties all at the same time and not have enuff cash to pay them all at once. So I hafta keep things limited to something that I can GUARANTEED pay. I know $300.00 is nothing for all the hours of work you hafta put into the project. I'm sorry. I am dying and have little money. :( If I was rich I would pay $3000.00 or $30000.00 to get things moving. Grrr! I want something to get done but I have little power :*(
Doesn't UAE4All2 have a JIT core now?
I could have sworn it got one a few months back.
Maybe i'm thinking of AGA support.
-
UAE4all2 68020 CPU emu is coded in dreadfully slow C.
-
UAE4all2 68020 CPU emu is coded in dreadfully slow C.
Well there's always Cyclone which while not a JIT is written in ARM assembly and heavily optimized for speed.
http://notaz.gp2x.de/cyclone.php
-
@ElPolloDiabl
If you're interested, utsource.net have them at $100 USD (last price known). They have 10-100 of them, minimum quantity 1 piece. I use them for all sorts of weird things (A2386SX bridgeboard RAM, AmigaOneXE sound chips, that sort of thing).
http://www.utsource.net/mc68060rc50.html
-
@ElPolloDiabl
If you're interested, utsource.net have them at $100 USD (last price known). They have 10-100 of them, minimum quantity 1 piece. I use them for all sorts of weird things (A2386SX bridgeboard RAM, AmigaOneXE sound chips, that sort of thing).
http://www.utsource.net/mc68060rc50.html
I notice they sell the mysterious MC68060FE133.
Has anyone ever been able to confirm that these are genuine parts?
All I've ever seen is rumours that they were manufactured by Freescale for secret military projects long after they had offically stopped making the 060.
-
Hi,
Is there enough desire for kickstarter funded production run of 060's. If so would they have to be faster than 50mhz? Or is 50mhz enough?
As an alternative should we try and get a run of 50mhz 030's?
I would settle for a 50mhz 060 model with working FPU. I could afford to pay $200 for one (CPU only) plus some more for kickstarter.
Are there general purpose fabs that can do a run of 1,000-2,000 chips?
Otherwise I'll settle for the less expensive FPGA route.
We could dream but a realistic production run should be 80--133 mhz with fpu and mmu. These chips could be made without redesign the chip.
-
We could dream but a realistic production run should be 80--133 mhz with fpu and mmu. These chips could be made without redesign the chip.
Freescale wouldn't even entertain the notion of a new production run unless a few million dollars worth of guaranteed orders were involved and as they own the IP no one else could make them without buying a licence and then we are back to the millions of dollars problem again.
-
I notice they sell the mysterious MC68060FE133.
Has anyone ever been able to confirm that these are genuine parts?
All I've ever seen is rumours that they were manufactured by Freescale for secret military projects long after they had offically stopped making the 060.
The problem with such speeds isn't the CPU - it's that the boards that use the 68060 aren't ready for such high speeds. The RAM access and the support chips are just too slow to be of much use above 50MHz (though I believe up to 80MHz is still an appreciable difference, it's getting to be quite a headache to get that high because of the amount of overclocking).
-
If you are going to make a new CPU, don't make an 060... it has horrible issues that you have to deal with, making a LOT of software either partially or fully incompatible with it. I would much prefer a 100MHz+ 68040. No compatibility issues really, and a long tested and proven CPU.
I had to write a LOT of 68060 specific code and patch the crap out of the Mac OS to get it to work at all on the 060 (there was never a 060 Mac, and the OS was never going to be 060 friendly without a major re-write). In the end, the overall speed for most everything end up being slower than a 33MHz 040.
If you're doing things that are static (ray trace and number crunching), the 060 is probably fine. But, I hate that thing. Give me an 040 any day!
-
I would settle for a 50mhz 060 model with working FPU. I could afford to pay $200 for one (CPU only) plus some more for kickstarter.
If you can afford to pay so much, i would certainly sell one from my stock to you :D
Seriously, a working full 060 should be much less than 50 USD (usually a bit more for rev6).
If Freescale want to make new one, i hope that would cost no more per CPU.
-
I notice they sell the mysterious MC68060FE133.
Has anyone ever been able to confirm that these are genuine parts?
Yes. They were tried in the Natami 060 board. They work but lack FPU and/or MMU.
Freescale wouldn't even entertain the notion of a new production run unless a few million dollars worth of guaranteed orders were involved and as they own the IP no one else could make them without buying a licence and then we are back to the millions of dollars problem again.
And no one is likely to want a large number of the 68060 in it's current state (except the U.S. military). It needs some modern enhancements to be appealing today. It's a great processor but it's a low clock speed, no memory controller, small caches, etc.
-
There is no 68060 JIT for iphones because Steve Jobs personally banned it.
Is there any credible source that can confirm this?
-
Is there any credible source that can confirm this?
There's no JIT for Apps on iPhone, don't think it's allowed on Android, Blackberry or Windows Phone 8. The underlying App execution system does usually use a JIT engine but you can't embed your own to run LUA/Python/etc within your own Apps.
That information might be old as I haven't involved in mobile development for a while. That's the basic gist of why you can't have a for of JIT for emulators etc.
-
If you are going to make a new CPU, don't make an 060... it has horrible issues that you have to deal with, making a LOT of software either partially or fully incompatible with it. I would much prefer a 100MHz+ 68040. No compatibility issues really, and a long tested and proven CPU.
The Rev 6 68060 is very stable and all know bugs are fixed. It runs very cool compared to the 68040 and is a much better design. Both CPU designs made mistakes though. The 68040 dropped the FINT/FINTRZ FPU instruction which is very important for FPU support and the 68060 dropped the 64 bit versions of integer instructions with MULS/MULU being very important.
I had to write a LOT of 68060 specific code and patch the crap out of the Mac OS to get it to work at all on the 060 (there was never a 060 Mac, and the OS was never going to be 060 friendly without a major re-write). In the end, the overall speed for most everything end up being slower than a 33MHz 040.
Apple tried to make the MacOS 68040 compatible but they also tried to keep it from being 68060 compatible, especially after a 68060 Amiga became the fastest Mac. Did you notice how the older MacOS 6-7.5 versions were more compatible with the 68060 than the later ones?
If you're doing things that are static (ray trace and number crunching), the 060 is probably fine. But, I hate that thing. Give me an 040 any day!
The 68060 is much more than a number crunching can't branch DSP. Branches and loops became much faster not that the 68040 was bad (the 68060 is good even compared to modern processors). You may have disabled the branch cache, turned off superscalar execution and used 1/2 I/D caches for maximum compatibility though. I would expect a 68060@50MHz to still run faster than a 68040@33MHz with all this disabled.
-
Is there any credible source that can confirm this?
If you get caught with a JIT you get banned from the Apple AppStore.
You can confirm it with thousands of ppl. You can start by asking every single person who needs a JIT, such as emulator coders.
I'm a credible source and I confirmed it.
Its actually worse than "no JIT" it is/was "no scripting of any kind" and "no dynamically executed code".
Apple randomly makes up new rules and changes old ones all the time so I have no idea if these rules are still in effect or not.
-
If you are going to make a new CPU, don't make an 060... it has horrible issues that you have to deal with, making a LOT of software either partially or fully incompatible with it. I would much prefer a 100MHz+ 68040. No compatibility issues really, and a long tested and proven CPU.
I had to write a LOT of 68060 specific code and patch the crap out of the Mac OS to get it to work at all on the 060 (there was never a 060 Mac, and the OS was never going to be 060 friendly without a major re-write). In the end, the overall speed for most everything end up being slower than a 33MHz 040.
If you're doing things that are static (ray trace and number crunching), the 060 is probably fine. But, I hate that thing. Give me an 040 any day!
It is 2013 and you are still letting Apple brainwash you. This makes me very sad. :(
-
It is 2013 and you are still letting Apple brainwash you. This makes me very sad. :(
:laugh1:
He does have a point though ... Making really fast new 68k cpu that would be pin compatible to 040 would get a lot of interest in Mac68K and Atari camps and who knows where else ...
060 wasn't used that much in computers ...
I don't think Freescale would go after you unless it causes them big financial damage and retro market is not what they consider a market at all ...
And I hope you're not _really_ dying ??? :confused:
-
It is 2013 and you are still letting Apple brainwash you. This makes me very sad. :(
Jim started working on Fusionx86 long ago.
Codename: PCx :D
OS8 on 060 is running fine.
OS8.1 is very unstable ...
Are there any (unreleased) Fusion-Updates?
I like OS8 on my o6o/9o .. but 8.1 would be nice to have.
-
One thing that always made me wonder about 040 was that frequency doubling inside the core ...
So if the externally 040 works at 25 Mhz it works at 50 Mhz internally which is probably why it gets so hot (that and the integrated FPU which NOBODY really needs :p) ...
Technology of chip production at that time was the reason they couldn't push it beyond that but I think that if someone made them today they would run cold ?
Also , would it be possible instead of doubling the frequency inside the core to quadruple it or even go 8 x 16 x 32 x :biglaugh:
-
Apple tried to make the MacOS 68040 compatible but they also tried to keep it from being 68060 compatible, especially after a 68060 Amiga became the fastest Mac. Did you notice how the older MacOS 6-7.5 versions were more compatible with the 68060 than the later ones?
There was no difference with OS8 in 060 compatibility. The patching was pretty much the same. Apple didn't "try" to make the Mac OS non-compatible with the 060. There were several Mac accelerator companies chatting with me about patches needed to make the 060 work, and it was agreed by everyone that there was no deliberate act to make sure the 060 didn't work. It just turned out that Apple did a lot things that killed the 060's expanded architecture.
The 68060 is much more than a number crunching can't branch DSP. Branches and loops became much faster not that the 68040 was bad (the 68060 is good even compared to modern processors). You may have disabled the branch cache, turned off superscalar execution and used 1/2 I/D caches for maximum compatibility though. I would expect a 68060@50MHz to still run faster than a 68040@33MHz with all this disabled.
Unfortunately, that was not the case. The various Mac benchmarking programs showed only minor improvements in certain benchmarks with the 060. SuperScalar always had to be off, and there was a limited amount of branch caching allowed in certain portions of the OS code, and the instruction and data caches were toggled off and on without anyone realizing it. Surprisingly, memory functions were quite a bit slower with the 060. We could compare the 040 speed vs. 060 speed using the same Phase 5 setup, just swapping the CPU card. So, the memory was the same.
Keep in mind that the FPU was the Mac's biggest asset for the OS. This is why you didn't see many LC (or any EC) CPUs going into Macs. The MMU was needed of course for virtual memory. The FPU was used by EVERYTHING in the OS! The position of where to draw a pixel on the display was calculated by the FPU, not the CPU because it was faster to do it this way. When Joe and I re-wrote Apple's PACK4 and PACK5 in full assembly (like everything else we did), we actually broke most current benchmark programs in the FPU tests and we made the Mac insanely fast - to the point where production studios like Amblin Entertainment were using Amigas with my Mac emulation to run Avid video editing suites because that setup would run circles around real Macs... and they could also use Lightwave for rendering too.
Apple didn't "brainwash" me. I reverse engineered their entire OS and custom hardware. I know how their stuff worked better than anyone.
-
I am a tech at Intel in Arizona and work on a series of tool sets in the fab that produces modern CPUs. I can tell you 100% without a doubt that you will not get a FAB to produce a small run of chips. For one thing the wafers in use now are 12" and the old 8" wafers are hard to come by and not cheap. The lithography tools that produce the die are million dollar tools alone and are no longer in use to produce the chip. You would have to get a new set of die made and those are hundreds of thousands for a complete set to make a chip. The man power to run the tool sets, the routes for automation and the cutters, packagers and assembly is all out of production. This would all have to be set back up again and thats not going to happen for a small run.
Just buy a used one, its the best your ever going to get. Get creative in looking for them. The AMiga is far from the only thing that used an 060. LOTS of older embedded systems used them. Heck while I was in the Navy I found a ton of old single board computers that had 040 and 060 chips in them. If you look, you will find them.
Some examples:
http://mediaserver.voxtechnologies.com/FileCache/BVME6000.htm
http://en.wikipedia.org/wiki/Motorola_Single_Board_Computers
http://www.cpuboards.com/01b005-03
Also here is a nice set of auctions for some single bare processors.
http://www.ebay.com/sch/i.html?_odkw=motorola+VME&_osacat=0&_from=R40&_trksid=p2045573.m570.l1313.TR0.TRC0.Xmotorola+MC68060&_nkw=motorola+MC68060&_sacat=0
-
I am a tech at Intel in Arizona and work on a series of tool sets in the fab that produces modern CPUs. I can tell you 100% without a doubt that you will not get a FAB to produce a small run of chips. For one thing the wafers in use now are 12" and the old 8" wafers are hard to come by and not cheap. The lithography tools that produce the die are million dollar tools alone and are no longer in use to produce the chip. You would have to get a new set of die made and those are hundreds of thousands for a complete set to make a chip. The man power to run the tool sets, the routes for automation and the cutters, packagers and assembly is all out of production. This would all have to be set back up again and thats not going to happen for a small run.
Just buy a used one, its the best your ever going to get. Get creative in looking for them. The AMiga is far from the only thing that used an 060. LOTS of older embedded systems used them. Heck while I was in the Navy I found a ton of old single board computers that had 040 and 060 chips in them. If you look, you will find them.
I was thinking more like using FPGA instead of a real chip and making it pin compatible by soldering it to a small PCB with 040 pins ...
I think it would still cost hundreds of thousand but if you got interest from say 5000 people on kickstarter (so only Amiga project would not do here , it would have to be Mac and Atari too) you could probably make it for something around 100-200 $ per cpu ...
-
If you get caught with a JIT you get banned from the Apple AppStore.
You can confirm it with thousands of ppl. You can start by asking every single person who needs a JIT, such as emulator coders.
I'm not contesting any of this. The specific part I wanted you to confirm is that "Steve Jobs personally banned" JITs.
I'm a credible source and I confirmed it.
No, man, you talk from the wrong end of your body a lot. :)
Its actually worse than "no JIT" it is/was "no scripting of any kind" and "no dynamically executed code".
Apple randomly makes up new rules and changes old ones all the time so I have no idea if these rules are still in effect or not.
You may run scripting engines that execute arbitrary code. Not natively, though, so probably no JIT.
-
Another possibility is a bounty or kickstarter to make an ARM JIT to run Motorola bytecodes. You should be able to get 300Mhz 060 speed (without MMU) on a wide variety of cheap ARM cpus.
There is no 68060 JIT for iphones because Steve Jobs personally banned it.
But everyone is allowed to code 68060 JIT for ARM cpus for 10 other devices.
There are multiple 1.0+ Ghz ARM cpu computers (usually with 2-4 cores) in the $99.00 range typically with 1GB+ of RAM.
(
I looked into that ... Better option than ARM would be x86 compatible tiny passive cooling single core boards ...
http://www.via.com.tw/en/resources/pressroom/pressrelease.jsp?press_release_no=1547
It's 1.8 Ghz and has 1 Gb and is cheaper than Raspberry Pi on EBay ... (EDIT : Still in production for a price of 250$)
Making FPGA card with Amiga chipset and combining it with this would give you *fast* Amiga ... :hammer:
-
If there is still a supply left then that is good enough.
As you add more clock speed the performance increases become less. So that means redesign of the CPU and more cost.
There will be a heap of modern software you will want to run that requires at least 1Ghz.
FPGA has none of the prohibitive cost in designing a cpu.
-
I'm not contesting any of this. The specific part I wanted you to confirm is that "Steve Jobs personally banned" JITs.
No, man, you talk from the wrong end of your body a lot. :)
I
You may run scripting engines that execute arbitrary code. Not natively, though, so probably no JIT.
It's unlikely Steve Jobs had any hand in the decision.
For security you can't execute any code that is generated at run time, which pretty much stops a JIT in it's tracks. I believe the only scripting language you're allowed to use is javascript and the script has to be deployed from the app store.
-
100Mhz or more, then I'd have interest.
-
make it a dual 060 for A1200 and quad 060 for big box :)
Actually, on the more serious side, a 1260 with built in SATA or even a replacement Blizzard SCSI daughter board with SATA. Maybe integrated USB controller as well.
-
Freescale wouldn't restart 060 production, never going to happen I'm afraid.
I think the only possible future for new accelerators for Classic's is FPGA. It's unlikely that anybody will implement an 040/060 as an FPGA but it looks like we'll soon have the a fully compatible 020/030 option. If this design could be clocked much higher than a real 020 and have faster memory access then it might actually be able to perform almost fast as a real 060 when running Classic 68k software. It might even be possible to add some 040/060 instructions to the 020 core in the future to enable it to run some 060 software.
The other option is for some to make an unpopulated board which would be nice but I don't see it happening :(
It's a shame really, if you applied Jen's ACA design to an 060 you would have a seriously fast Amiga. Something with a 105Mhz clock speed, 512MB SDRAM and boosted cache read/write would be awesome. The ATARI boys managed it with their CT 060 but for some reason nobody has done this for Amiga Classics. I think the closest we have to a 'modern' 060 Amiga board is the FPGA arcade daughter board.
-
Unfortunately, that was not the case. The various Mac benchmarking programs showed only minor improvements in certain benchmarks with the 060. SuperScalar always had to be off, and there was a limited amount of branch caching allowed in certain portions of the OS code, and the instruction and data caches were toggled off and on without anyone realizing it. Surprisingly, memory functions were quite a bit slower with the 060. We could compare the 040 speed vs. 060 speed using the same Phase 5 setup, just swapping the CPU card. So, the memory was the same.
The 68060 speed would drop to less than half with superscalar off. The CPU would be scalar with all the limitations of superscalar. Turning branch caching off makes the branching performance about the same as the 040. The 040 does outperform the 060 working in memory and has a larger cache fetch. This allows larger instructions but the 68060 can handle mixed instructions well, is faster with more complex addressing modes and is faster at shifting and multiplying. The 040 also has the 64 bit integer instructions and optimizations for bit field instructions in registers that would help it. The 060 is a clear winner with the FPU and has a clock speed advantage. It probably comes down to the code for the MAC. Code that is optimized for an 040 is not going to be optimal in an 060. I don't know if optimizing code for an 060 with superscalar disabled and many of the caches turned off would even be possible. It would be quite handicapped but still faster than a 68030 (the 68060 resembles in some ways a superscalar 030).
Keep in mind that the FPU was the Mac's biggest asset for the OS. This is why you didn't see many LC (or any EC) CPUs going into Macs. The MMU was needed of course for virtual memory. The FPU was used by EVERYTHING in the OS! The position of where to draw a pixel on the display was calculated by the FPU, not the CPU because it was faster to do it this way. When Joe and I re-wrote Apple's PACK4 and PACK5 in full assembly (like everything else we did), we actually broke most current benchmark programs in the FPU tests and we made the Mac insanely fast - to the point where production studios like Amblin Entertainment were using Amigas with my Mac emulation to run Avid video editing suites because that setup would run circles around real Macs... and they could also use Lightwave for rendering too.
The 040 FPU runs in parallel to the integer units but is still quite slow compared to integer. It's not very easy to go back and forth between the FPU and integer either with the lack of FINT/FINTRZ and no fp<-> unsigned integer. I'm kind of surprised you were using the FPU for the display. Are you sure it wasn't the MMU? There are drivers for Fusion/Shapeshifter that are faster with the MMU and the way the MAC renders the screen.
I think the only possible future for new accelerators for Classic's is FPGA. It's unlikely that anybody will implement an 040/060 as an FPGA but it looks like we'll soon have the a fully compatible 020/030 option. If this design could be clocked much higher than a real 020 and have faster memory access then it might actually be able to perform almost fast as a real 060 when running Classic 68k software. It might even be possible to add some 040/060 instructions to the 020 core in the future to enable it to run some 060 software.
Remember the crazy fpga hardware I was talking about the other day with a 150MHz enhanced 68k in fpga soon and a possible 500MHz+ 68k CPU in fpga in about a year? It is possible but I don't want any "announcements" or vapor ware claims. There may be some interesting reading on http://www.amigacoding.de/ if you haven't been over there recently. I have seen the I/O expansion board early schematic ;).
-
Remember the crazy fpga hardware I was talking about the other day with a 150MHz enhanced 68k in fpga soon and a possible 500MHz+ 68k CPU in fpga in about a year? It is possible but I don't want any "announcements" or vapor ware claims. There may be some interesting reading on http://www.amigacoding.de/ if you haven't been over there recently. I have seen the I/O expansion board early schematic ;).
In the form of accelerator cards for existing Amigas or complete systems like FPGA Arcade/MiST/Minimig?
-
In the form of accelerator cards for existing Amigas or complete systems like FPGA Arcade/MiST/Minimig?
Sort of complete system. It uses standard "commodity" fpga boards with a custom I/O expansion board (for common Amiga joystick, keyboard, mouse ports and such but also SATA/PCIe/HDMIsh etc).
-
The ATARI boys managed it with their CT 060 but for some reason nobody has done this for Amiga Classics. I think the closest we have to a 'modern' 060 Amiga board is the FPGA arcade daughter board.
And a new board was planned using two 68060 :
http://rodolphe.czuba.free.fr/Phenix/overview.htm
-
@Cosmos
There is no mention of Amiga compatibility. Was it to run emulated? 68k Linux would have ran on it.
-
Read better : was for the Atari world...
-
I think the only possible future for new accelerators for Classic's is FPGA.
I think it's less likely to happen than freescale producing 060's again.
As FPGA's get more powerful the benefit of shoehorning one into an old Amiga as just a processor upgrade makes less and less sense. Someone could make a motherboard that fitted an old Amiga case, but even that has disadvantages.
-
I think it's less likely to happen than freescale producing 060's again.
As FPGA's get more powerful the benefit of shoehorning one into an old Amiga as just a processor upgrade makes less and less sense. Someone could make a motherboard that fitted an old Amiga case, but even that has disadvantages.
Where have you been the last months? in a cave?
http://www.amiga.org/forums/showthread.php?t=65653
-
Where have you been the last months? in a cave?
http://www.amiga.org/forums/showthread.php?t=65653
The thread is about a 68060, not a partially emulated 68020. Come back when the Vampire 600 can beat a 80mhz 68060. I haven't got anything against that project, it's just not what we're discussing.
-
The thread is about a 68060, not a partially emulated 68020. Come back when the Vampire 600 can beat a 80mhz 68060. I haven't got anything against that project, it's just not what we're discussing.
And that thread is about an accelerator board with an FPGA that plugs into an A600 which can be flashed without whatever CPU you like. It's already been shown running at pretty high clockspeed and it's perfectly answer the quesiton of whether or not someone will ever make and FPAG based accelerator board.
-
And that thread is about an accelerator board with an FPGA that plugs into an A600 which can be flashed without whatever CPU you like. It's already been shown running at pretty high clockspeed and it's perfectly answer the quesiton of whether or not someone will ever make and FPAG based accelerator board.
Yes that's right, this is the future :)
The 68060 was a great chip but because it's hard to find and very expensive, an FPGA design is more attractive these days.
-
As FPGA's get more powerful the benefit of shoehorning one into an old Amiga as just a processor upgrade makes less and less sense. Someone could make a motherboard that fitted an old Amiga case, but even that has disadvantages.
Eh? I'd have thought it makes more and more sense. If it is faster and cheaper than existing "real" 680x0 accelerators, I don't see any disadvantage.
-
If you get caught with a JIT you get banned from the Apple AppStore.
You can confirm it with thousands of ppl. You can start by asking every single person who needs a JIT, such as emulator coders.
I'm a credible source and I confirmed it.
Its actually worse than "no JIT" it is/was "no scripting of any kind" and "no dynamically executed code".
You can sell emulators on the Apple AppStore, but you are not allowed to sideload any software on to them... So it either has to come with all the software it's ever going to run... Or you need to sell extra software via an "in app purchase".
There are three problems with running emulators on an iOS device:
1. If the emulator is free open source, it must be free on the AppStore.
2. Most of the software you want to run on the emulator will still be in copyright (or you already own and don't want to buy again).
3. (This is the biggy) The old games just aren't suited to running on a touch screen devices and are often quite awkward to play/use.
-ON TOPIC-
A production run of any 68k is never going to happen, technology has moved on and you are never going to get rights to the IP.
Your options are an FPGA, the A600 Vampire board is already here or (my favourite) an ARM board with an emulator/JIT, with an interface for Plugging into Amiga CPU slots :)
-
You can sell emulators on the Apple AppStore, but you are not allowed to sideload any software on to them... So it either has to come with all the software it's ever going to run... Or you need to sell extra software via an "in app purchase".
There are three problems with running emulators on an iOS device:
1. If the emulator is free open source, it must be free on the AppStore.
2. Most of the software you want to run on the emulator will still be in copyright (or you already own and don't want to buy again).
3. (This is the biggy) The old games just aren't suited to running on a touch screen devices and are often quite awkward to play/use.
-ON TOPIC-
A production run of any 68k is never going to happen, technology has moved on and you are never going to get rights to the IP.
Your options are an FPGA, the A600 Vampire board is already here or (my favourite) an ARM board with an emulator/JIT, with an interface for Plugging into Amiga CPU slots :)
I'd like the same but with an x86 rather than ARM just for the raw power and the latest x86 chips probably run cooler than an 040 does. :)
Regarding the App Store, isn't there a legal problem with GPL apps? I seem to remember VLC having issues.
-
I'd like the same but with an x86 rather than ARM just for the raw power and the latest x86 chips probably run cooler than an 040 does. :)
Regarding the App Store, isn't there a legal problem with GPL apps? I seem to remember VLC having issues.
My money is on ARM long term... But yeah, a x86-64 based JIT accelerator would be three levels of awesome!
Yeah, the AppStore licence changed a couple of years back to allow GPL, I already downloaded VLC so I wasn't affected by its removal, very useful for watching WMVs on my iOS devices. But it is returning now the Licence allows GPL.
-
My money is on ARM long term... But yeah, a x86-64 based JIT accelerator would be three levels of awesome!
We need a genius wizard to port the existing 32-bit JIT to AMD64 first though. Jason maybe?
Yeah, the AppStore licence changed a couple of years back to allow GPL, I already downloaded VLC so I wasn't affected by its removal, very useful for watching WMVs on my iOS devices. But it is returning now the Licence allows GPL.
VLC is the first thing I install on any Mac or Linux box. :)
I wonder if OS4 will receive a port of it now they have Qt4?
-
We need a genius wizard to port the existing 32-bit JIT to AMD64 first though. Jason maybe?
if it was up to me, I'd just port the WinUAE hardware emulation to QEMU, and use QEMU's m68k dynamic-translation (JIT like) which works on any architecture.
-
if it was up to me, I'd just port the WinUAE hardware emulation to QEMU, and use QEMU's m68k dynamic-translation (JIT like) which works on any architecture.
Hi Jason! Sounds like a nice project for you to fit into that 25th hour of each day. ;)
-
if it was up to me, I'd just port the WinUAE hardware emulation to QEMU, and use QEMU's m68k dynamic-translation (JIT like) which works on any architecture.
Let's focus on AROS first ;)
Hahahah, that does sound like a very cool idea!
-
100Mhz or more, then I'd have interest.
100+Mhz I'm in.
Rich
-
It's already been shown running at pretty high clockspeed
"Performance: More than 6Mips"
Eh? I'd have thought it makes more and more sense.
If there was an FPGA that was fast enough to do 500mhz 68060, which do you think would sell more.
1. An A600 accelerator
2. A minimig sized board that can be a 68060/PPC Amiga/Mac, PC, a PS1/PS2, Jaguar, Archimedes, etc etc with wireless peripherals
I can't see why anyone would want a 500mhz ECS machine, even AGA limitations would hold it back.
The form factor doesn't lend itself to being used on the main TV either. Although converting an A600 to a battery operated wireless keyboard might work.
-
"Performance: More than 6Mips"
Remember that the vampire board is meant for a 600, he's already had it running up to 12 MIPS which is higher than any physical A600 has ever managed.
With a bigger FPGA and some clever coders, it's feasible that a reasonably fast board could be made. Also, it could be made for a lot less than a real 060 board, that's more important than speed anyway. What's the point of developing a new physical 060 based board if it costs $500, only about 3 people would buy one.
-
I would think that everyone who wanted a 060 or 030 would have one by now.
The only thing to do is to go for more speed. That would not be cheap by doing the CPU first.
FPGA gives you the ability to keep optimising the CPU.
-
What's the point of developing a new physical 060 based board if it costs $500, only about 3 people would buy one.
That is still likely cheaper than an FPGA board that can beat a 68060.
I don't see there being a market for either of them.
-
@ above
The thrill of having the classic hardware/software perform at an amazing speed. When an emulator does it, its not a big deal.
I'm not pushing for this, but it will be awesome (when it happens) and 100% compatible as oppose to when we got PowerPC accelerators at 200-300mhz.
-
@ above
The thrill of having the classic hardware/software perform at an amazing speed. When an emulator does it, its not a big deal.
It's an expensive thrill. A reconfigurable FPGA computer would be much better, you'd get a thrill every time someone created a new personality for it.
-
It's an expensive thrill. A reconfigurable FPGA computer would be much better, you'd get a thrill every time someone created a new personality for it.
Oh, I thought that you could have both. You could fit heaps of functions on an accelerator card as well.
I agree. It might be a big bottleneck when accessing the classic hardware. Sought of like the switch between 16-bit and 32-bit RAM.
-
Oh, I thought that you could have both. You could fit heaps of functions on an accelerator card as well.
I agree. It might be a big bottleneck when accessing the classic hardware. Sought of like the switch between 16-bit and 32-bit RAM.
Once you've added sound and graphics output to your accelerator card, you're unlikely going to want to switch your cables between both. So you'll add ECS/AGA to the FPGA and throw in 8mb of chip ram. Making the A600 a rather inconvenient keyboard.
It sounds like a lot of work to go to for people who want to upgrade an A600, so you'll want to build a standalone card as well. This is the one that everyone will buy anyway, making you wonder why you bothered with trying to cram it into an A600 case in the first place.
Sure there is room for a cheap one today, but anything more complex is kinda pointless.
-
I can't see why anyone would want a 500mhz ECS machine, even AGA limitations would hold it back.
I'd love a 500MHz 060 in my A3000/PIV, I never use any ECS screenmodes.
-
In the end, the overall speed for most everything end up being slower than a 33MHz 040.
The various Mac benchmarking programs showed only minor improvements in certain benchmarks with the 060.
Maybe on Fusion which I never ran... I had the BlizzPPC 240/060-50
these benchmarks are from my current Blizz1260
-
060 is quite a lot better than 040. After I got my 060 I gave my 040 machine away.
-
I'd love a 500MHz 060 in my A3000/PIV, I never use any ECS screenmodes.
It's still being crippled by the Zorro bus. An FPGA that could do a 500mhz 68060 could also emulate a PIV with much higher memory bandwidth. I don't have an A3000 but according to the internet, buster 11 hits 10mb/s with a PIV. DDR2 is orders of magnitude faster than that.
Throw in a floppy port that can use standard PC high density floppy drives, Ethernet, 96khz 32bit sound, sata, usb and I'm not sure there is much out of the A3000 that you'd still want to be using.
-
It's still being crippled by the Zorro bus. An FPGA that could do a 500mhz 68060 could also emulate a PIV with much higher memory bandwidth.
Throw in a floppy port that can use standard PC high density floppy drives, Ethernet, 96khz 32bit sound, sata, usb and I'm not sure there is much out of the A3000 that you'd still want to be using.
Something akin to the Chameleon 64 would be great. Works as an add-on to real hardware but can also be used standalone.
-
Something akin to the Chameleon 64 would be great. Works as an add-on to real hardware but can also be used standalone.
I understand the concept, it just makes less sense to do it when you've got eight different Amiga models instead of one model of C64.
It will be cheaper, easier, quicker etc to just produce something standalone. By the time it happens the number of people who say they want one that fits in a real Amiga will decline & I believe that if a cheaper standalone card was offered first then that will clear out most of the rest of the potential customers.
-
I understand the concept, it just makes less sense to do it when you've got eight different Amiga models instead of one model of C64.
It will be cheaper, easier, quicker etc to just produce something standalone. By the time it happens the number of people who say they want one that fits in a real Amiga will decline & I believe that if a cheaper standalone card was offered first then that will clear out most of the rest of the potential customers.
Quite possibly but we are talking imaginary hardware here.
-
It's still being crippled by the Zorro bus. An FPGA that could do a 500mhz 68060 could also emulate a PIV with much higher memory bandwidth. I don't have an A3000 but according to the internet, buster 11 hits 10mb/s with a PIV. DDR2 is orders of magnitude faster than that.
Throw in a floppy port that can use standard PC high density floppy drives, Ethernet, 96khz 32bit sound, sata, usb and I'm not sure there is much out of the A3000 that you'd still want to be using.
Yeah and then you'd need to re-write major chunks of the OS to support the new hardware, plus you'd want some of the things that modern OS's have, so that there'll be very little of OS 3 you'd still be using....wait a minute..wait a minute...I think we've gone down this path already....
-
Don't throw the thread off on a tangent.
There are only two subjects: A CPU faster than an 060. It's in the works already.
Also the standalone FPGA based Amiga.
It will really be based on demand. Nobody is going to make a card unless there is demand for one.
-
Quite possibly but we are talking imaginary hardware here.
New standalone hardware is going to happen because it's too easy. Many will fail but there are several fpga projects in the works. Standalone makes the most sense. If there is a PS2 port for using a real Amiga keyboard and 9 pin joystick ports, what is there to miss? SATA, usb, PCIe, memory cards, most of this new stuff is fast and cheap.
The TG68k is reaching 68020-68030 speeds and is not fully pipelined but it may reach low 68040 speeds. A fully pipelined CPU in similar speed but larger fpga with smart caching should be 68040-68060 speed. A bigger fpga is a real plus for a faster CPU and leaves room for an FPU and MMU/MPU. That's why the Apollo aims for a Cyclone 5 minimum (that and ALMs are an advantage over LEs). The fpgaArcade is going to be hard pressed to fit an FPU in it's fpga. This is fine for what the fpgaArcade is targeting though. They would do better to focus on retro console emulation like Genesis and X68000 after the Amiga and Atari ST, IMO.
-
Yeah and then you'd need to re-write major chunks of the OS to support the new hardware, plus you'd want some of the things that modern OS's have, so that there'll be very little of OS 3 you'd still be using....wait a minute..wait a minute...I think we've gone down this path already....
You'd just need drivers, not rewrite chunks of the OS. Ideally it would also be able to run in legacy mode where SATA appears like gayle ide or you could mount a hard drive image from a file on a USB stick and have it appear as a IDE/SCSI hard drive/cdrom etc.
The purpose of using SATA is more about getting cheap hardware than utilising all it's functionality, although if there was the option for that by replacing the scsi.device driver in kickstart then that would be great too. Even if it was just making it emulate IDE fix express.
I'm also more interested in a USB gamepad appearing like a 9 pin D joystick than being able to run a USB stack on the Amiga. Although if you could make the USB controller compatible with one of the existing Amiga solutions then that would be a bonus.
I'm not sure I'd bother with PS2, a USB to Amiga keyboard convertor would be better. http://ezhid.sourceforge.net/amikbd.html http://www.amiga.org/forums/showthread.php?t=64931
The simplicity of PS2 draws you in, but it's actually just hassle if you're also going to support a USB keyboard. People will expect to be able to use a PS2 or Amiga keyboard, which are different and you'd need to configure it (without using a keyboard). Only supporting USB is much easier as you just detect a keyboard and use it.
-
if it was up to me, I'd just port the WinUAE hardware emulation to QEMU, and use QEMU's m68k dynamic-translation (JIT like) which works on any architecture.
Hang on... It's taken me a while but wouldn't this allow you to creat some "interesting" virtual hardware setups... Like an Amiga with an x86 or ARM as a CPU!? -edit- or even something really weird like a PPC!? ;)
-
or even something really weird like a PPC!? ;)
Nobody has created a Phase5 PPC emulator because they can't be bothered, not because it's particularly hard.
I though about doing it, but I have too much other stuff going on. It doesn't help that I've never owned one so I have no software for it or even an idea of what I'd use the emulation for.
-
Nobody has created a Phase5 PPC emulator because they can't be bothered, not because it's particularly hard.
I though about doing it, but I have too much other stuff going on. It doesn't help that I've never owned one so I have no software for it or even an idea of what I'd use the emulation for.
For running the B/CSPPC version of OS4 on hardware that isn't crap.
That's my reason for lusting after such a feature in UAE anyway. :)
But Bloodline's idea of having virtual 68k and PPC cores in an (A)SMP variant of AROS would be even better. ;)
-
Hang on... It's taken me a while but wouldn't this allow you to creat some "interesting" virtual hardware setups... Like an Amiga with an x86 or ARM as a CPU!? -edit- or even something really weird like a PPC!? ;)
+Eleventy-one to that! ;)
-
I'm with the qemu improved thing too!.
:)
-
For running the B/CSPPC version of OS4 on hardware that isn't crap.
That's my reason for lusting after such a feature in UAE anyway. :)
Are there any killer apps on OS4?
-
Are there any killer apps on OS4?
Absolutely none whatsoever but I still lust after such an emulator. :)
-
Absolutely none whatsoever but I still lust after such an emulator. :)
Would it be possible to adapt PearPC? I remember successfully running OSX on that thing but it was really slow. I don't remember what kind of hardware had back then, but it would be interesting to let my i7 have a crack at it now :)
-
060 is quite a lot better than 040. After I got my 060 I gave my 040 machine away.
Well I just recently got my first 040@40 (which I got to upgrade to 060@66) and can tell you it's 3X as fast as my 030@50 and have seen no compatability errors in WHDload that I've been hearing about for years.
I've had it in my desktop running for days with no issues and the CPU barely gets warm to the touch (mind you it has a small fan on it & my case has another fan).
-
Would it be possible to adapt PearPC? I remember successfully running OSX on that thing but it was really slow. I don't remember what kind of hardware had back then, but it would be interesting to let my i7 have a crack at it now :)
PearPC emulates a Mac, so if the AOS4 devs want to let their OS run on it it can. I wouldn't be surprised if MOS can already run on it :)
-
Would it be possible to adapt PearPC? I remember successfully running OSX on that thing but it was really slow. I don't remember what kind of hardware had back then, but it would be interesting to let my i7 have a crack at it now :)
It's theoretically possible, same as using the GXEmul or QEMU dynamic translation cores but nobody seems interested to do the work required to integrate it with UAE to appear as a Phase5 board to the emulated Amiga.
Looking at the code of all three, I say using GXEmul would be the least amount of work but that's only my guess as doing such a thing is way above my abilities but i'd very much like to see it happen.
Maybe a hefty "Phase5 PPC board emulation for UAE" bounty might tempt someone?
Then again I reckon more people would be interested in a 68k-JIT for UAE on PPC and last time I checked that bounty prize wasn't particularly large in relation to how much work Álmos is putting into it.
http://euaejit.blogspot.co.uk/
http://www.amigabounty.net/?function=viewproject&projectid=35
http://gxemul.sourceforge.net/
-
PearPC emulates a Mac, so if the AOS4 devs want to let their OS run on it it can. I wouldn't be surprised if MOS can already run on it :)
MOS is hardwired to only boot on specific machine ID's so PearPC would have to be modified to present itself as one of the supported models.
Bigfoot already demonstrated MorphOS running on a modfied QEMU on Linux a few years ago.
(http://bigfoot.morphos-team.net/test/qemu.png)
-
Well I just recently got my first 040@40 (which I got to upgrade to 060@66) and can tell you it's 3X as fast as my 030@50 and have seen no compatability errors in WHDload that I've been hearing about for years.
I've had it in my desktop running for days with no issues and the CPU barely gets warm to the touch (mind you it has a small fan on it & my case has another fan).
060's run a lot cooler than 040's do and as you've noticed they are so much faster too. :)
Do yourself a big favour and buy a copy of the HSMathLibs 060 version. The speedup is very noticable.
http://www.hsmathlibs.de/index_e.html
Make sure you install either Cyberpatcher from Phase5, Oxypatcher or MuRedox+Thor's 68060 libs.
http://phase5.a1k.org/index.php?tools
And Matt Hey's CopyMem 060 is an amazing piece of work that is a must if you want to get the best performance out of your machine.
http://aminet.net/package/util/boot/CopyMem
It's not 060 specific but I recommend installing TLSFMem too for another dramatic speed up.
http://www.platon42.de/files/util/TLSFMem.lha
-
@ nicholas (http://www.platon42.de/files/util/TLSFMem.lha)
thanks for the links as they'll come in in handy when I do the 040 to 060 upgrade; however, my previous post was based on just using my 040@40 and I must admit I'm surprised as I've read so many negative comments about the 040 boards.
-
@ nicholas (http://www.platon42.de/files/util/TLSFMem.lha)
thanks for the links as they'll come in in handy when I do the 040 to 060 upgrade; however, my previous post was based on just using my 040@40 and I must admit I'm surprised as I've read so many negative comments about the 040 boards.
You're welcome.
Even with the 040 installed you can install Thor's 68040.library, mmu.library, MuRedox, CopyMem and TLSFMem.
Then when you put the 060 in you just need to install Thor's 68060.library, power off, install the 060 and power on. Job done! :)
There is also an 040 version of the HSMathLibs. I had them when I was using a WarpEngine 040@40MHz and just bought another licence of the 060 version when i upgraded to a CSMK2 060@50MHz as they are quite cheap and really worth the money.
-
You're welcome.
Even with the 040 installed you can install Thor's 68040.library, mmu.library, MuRedox, CopyMem and TLSFMem.
Then when you put the 060 in you just need to install Thor's 68060.library, power off, install the 060 and power on. Job done! :)
There is also an 040 version of the HSMathLibs. I had them when I was using a WarpEngine 040@40MHz and just bought another licence of the 060 version when i upgraded to a CSMK2 060@50MHz as they are quite cheap and really worth the money.
great stuff! I have only installed Thor's 040 library & mmu.library so far but will try the others soon:drink:
-
great stuff! I have only installed Thor's 040 library & mmu.library so far but will try the others soon:drink:
Prepare for a very noticable speed up! :)
-
Absolutely none whatsoever but I still lust after such an emulator. :)
I get it, there is always something shinier though.