new 99 dollar 500mhz 68020?

[Bunsen]

Having hunted around the appropriate websites for a bit, it looks to me like the current state of play is this:

• It looks like they have transparent 68EC020 emulation working at ~200MHz
• They're using a software shim to handle exceptions, but I'm not sure if that is only for the 060 emulation, or for both modes.

nb: these are beta releases that are still being debugged.

Both of those imply that it's not Mac-ready. The EC020 was never used in a Mac.  And the "software shim for exceptions" is the same reason we can't use existing Amiga 060 accelerators on Macs.

The surface mount vs DIP40 issue with the 68000 is "only" a matter of altering the PCB layout to suit.

To my thinking, the ultimate scenario for Macs would be:

• Fully transparent 68000 emulation at speed, and
• Fully transparent 68040 emulation at speed.

Either of those would allow the use of the accelerator with existing Mac system software, ie, not requiring the patching that Amiga folks have already done to the various Amiga OSes to use 060s etc.

But as I have at present no way of contributing to the work, this is all wild speculation on my part.

 

[Sherry Haibara]

Well especially if you can get the 68K fast enough, you may be able to finally write a decent web rendering engine for the 68K mac and it actually be usable. 

BUT... there is a HUGE BUT... the processor can be 5Ghz, your still stuck at the 16 to 33Mhz system bus using the real hardware, this includes the RAM so its going to be slower than balls still. 

Sure the CPU can crunch the numbers as fast as lightning, but if you cant "get" it to there, then it wont run any faster. 

itll help if the L2 cache and of course the main system RAM is modern and running as fast as the processor. But again, the interrupts and all that jaz are ASIC/VIA controlled, and again, your stuck. 

That's exactly the problem I see with this kind of accelerators.

I'm not really an expert, but if I recall correctly most (if not all) 68k Macs derive all their clock signals from a single oscillator that controls the CPU, the bus speed and possibly the video signal refresh rate. You can't just expect to speed everything up, because replacing that oscillator would create all kinds of troubles with the system architecture. 

I think you could completely bypass the main oscillator by having the CPU run with its own clock, but that would have very little effect - you would still be limited by the bus and RAM speeds, so the performance gain would be negligible if present at all. 

I don't really know how accelerator cards worked back then, but if you could design an accelerator that's got its own CPU *and* RAM to bypass the stock ones, you may achieve something interesting. But that's far from being a drop-in replacement.

 

[bigmessowires]

^ This.

Initially I thought the idea was to build an entirely new model of Macintosh, which I would imagine happening one of two ways:

• New fast CPU in an FPGA, with memory controller, ASICS, I/O, sound, graphics, etc. implemented externally in hardware. This would be the closest thing to a "real" Mac, and would be very cool, but probably a tremendous amount of work. 
• The entire thing implemented in an FPGA, with connections for a VGA or HDMI display and USB peripherals. Probably an easier approach (though still very challenging), but as johnklos said, the difference between that and an emulator is subtle. It's totally different at the bottom level, but the end result looks nearly identical. Put Mini vMac on a fast ARM CPU board inside a custom enclosure, and ask people to tell the difference between that and an FPGA solution.

Talking about an accelerator card for an existing Mac sounds more promising to me. The big issue there, as Sherry and techknight and others said, is RAM. To see much benefit from a 500 MHz CPU in a computer designed for ~20 MHz, you'd need to replace all the RAM, the memory controller, and the caches. And even then, anything on the Nubus like a graphics card would still run at the same old speed. Disk I/O would be just as slow too. But it would be pretty nifty! It could be tons faster for computation-heavy work.

 

[johnklos]

The Amiga 600 accelerator has 64 megs of local, really really fast memory. Macs could use system memory to boot and add the really fast memory during boot. But even though the interface to the motherboard is 7.16 MHz, 16 bit, it still makes the old school Amiga ECS chipset shine.

 

[techknight]

Well the accelerator could bypass the decoding of the RAM addressing for the system board, entirely. And replace it with the local RAM. 

The only downside is you need to handle the Overlay function. Basically ROM/RAM swap during RESET vector. 

 

[olePigeon]

Would it be faster to do everything on the FPGA, or would it be faster to take an existing 68k compatible CPU like a Coldfire, then add the instructions needed to be compatible with a Mac?

I guess all the hard work has been done already on the Amiga side.  Probably faster for everything on FPGA.

 

[Bunsen]

It occurs to me, bigmessofwires, that your Plus Too would in many ways be the perfect development / testbed for an FPGA-based accelerator.  Keep the hardware definition of the rest of the system the same, and replace the 68000 core with Apollo or whatever.  Then uhhhhhh, fix it until it works.

 

[bigmessowires]

There's also a Macintosh FPGA core for the MiniMig board, which was developed by Benjamin Herrenschmidt. http://www.minimig.net/viewtopic.php?f=6&t=411&start=10

 

[eraser]

it is as fast as a 500mhz 020 cause it is superscalar,has two execution units,fast fpu and many other improvements

That's another problem, actually.  The '020 was/is capable of multiprocessing but this was never implemented on the 68K-era Mac.  It doesn't matter if your system has 1 core or 256 cores if the software was only written to utilize 1 of them.

In PC land there are even cases where running old OSes (like Windows NT 3.1) in VMs on multi-core systems will make the system run slower or become unstable because they don't know how to handle the shift between cores.  To get them working properly the hypervisor has to be locked to a single core (affinity).  

 

[Gorgonops]

An "execution unit" isn't a "core". I'll save you the "TL;DR" but in short a CPU that presents itself externally as a single "core" is of course actually made of a bunch of separate little bits and bobs (instruction decoders, ALUs, etc) and faster CPUs sometimes have more than one of each of these bits and bobs wired in parallel so the CPU can execute (parts of) multiple instructions at once. CPUs capable of doing this (transparently, or partially so) are broadly termed "Superscaler"; the first superscaler x86 CPU was the original Pentium, while the only superscalar "classic" 680x0 CPU was the 68060. So... yeah, apparently they designed a Superscalar 68020-compatible; assuming they did things properly there shouldn't be any need for software changes, although it should be noted that in many cases to take full advantage of the "two or more instructions at once" capability it's best to recompile your code so it can order things to take optimal advantage of the hardware.

(There are usually limitations as to what sorts of instructions can execute simultaneously; obviously this applies if the next instruction directly relies on the results from the last one, but there are other cases where instructions can block each other. But these changes are not of the same order as rewriting for SMP support.)

 

[eraser]

Ah yes, I had a dumb moment.  You are correct, I was thinking in terms of cores.  There are some issues in terms of pipelining and the order of execution but yes, it's not as dramatic as a redesign for SMP.

Although some work has been completed in getting this to run as a Mac I am still skeptical at this phase.  Even the very nice Basilisk II emulator, which has been in development for years, is still not perfectly stable nor compatible with all applications.  This FPGA seems like an even taller order at the moment.  

 

[Unknown_K]

I don't mess around with 68K Macs for speed. Making accelerators for 68K macs (ones with PDS slots) is not economical. You could run software on a FPGA board I guess but then so much for ADB and Nubus hardware.

 

[Schmoburger]

I personally can't see any real-world justification for the expense and/or effort of development on a largeish scale. Cool idea on paper but it really does sound like a case of travelling the world for the sake of sixpence. Not that I am in any way intending to trivialise the efforts of the developer in question whom is working upon the project at present, I just really can't see that there is significant benefits to really motivate enthusiasts to want to get behind it... Cool just isnt enough in a lot of cases.

 

[Gorgonops]

Don't worry, it's not just you who doesn't get it. The market for the utterly gonzo Amiga upgrades is apparently driven by the dreams of a core of super-enthusiasts (which I can't imagine could *possibly* be very many people at this point?) who still believe the Amiga is somehow still a living platform and the existence of something like a "500mhz 68020" will somehow bring developers back to make new... Amiga-things. What's particularly amusing/pointless about this thing is it takes this already tiny market, which for the most part has bought into the whole "Amiga PPC" thing, and tries to split off the chunk who this whole time have been pining for faster native 680x0 CPUs instead. Are there *really* enough people out there that even *think* they need such a thing to justify its existence? Obviously this thing is going to be of zero use running classic games, as those were already written to target the Amiga platform as-shipped so... it really seems like a solution to a problem that doesn't really exist. The only people who "need this" really need a new computer, hate to break it to them.

And following that same line of reasoning it makes even less sense as a Mac upgrade for the simple reason that unlike the Amiga Apple *had* a well-executed and nearly seamless transition to PowerPC so there already scads of machines out there that should run classic Mac software as fast or faster than a crazy-fast accelerator will. It's probably not the best thing to look at but, for instance, if we play with Speedometer 4.0 benchmark numbers taken when running in "68k code mode" a 500mhz-ish G3 or G4 should be comparable to a Mac II running at about 500mhz. (And of course the machine will be scads faster when running native code.) And that's assuming that a Mac II with one of these upgrade cards would really be as fast as what you'd get if you could just magically turn the bus clock in a Mac II up to 500mhz, which is a pretty laughable idea.

So... yeah, not getting it. It is, however, "cool".

 

[bigmessowires]

You may be looking at this backwards. It's not:

"I really want a much faster 68000 Amiga. Hey, I could build a 500 MHz superscalar '020 with an FPGA!"

but more like:

"I really want to tinker high-speed FPGA processors. Hey, I could build a much faster Amiga!"

and from that point of view, it makes as much sense to me as anything else around here.

 

[johnklos]

Amiga people like to do things other people say they can't. So while the rest of the world was saying that Quake and other game engines would never run on Amigas, people went and made them run on Amigas.

The attraction for the Amiga 600 is that it's the smallest Amiga and the only accelerators it ever got were m68030 accelerators. Therefore, something faster than an m68030 is nice. That it happens to be faster than an m68040 and will eventually be faster than an m68060 is just bonus.

Remember - there are thousands of games for Amigas! With a floppy emulator, you can boot and play any of thousands of games quickly and easily.

 

[trag]

but more like:

"I really want to tinker high-speed FPGA processors. Hey, I could build a much faster Amiga!"

and from that point of view, it makes as much sense to me as anything else around here.

What he said.    No one is going to make money building 68K Mac upgrades -- or not much money.    But the hardware development can certainly be fun.

I still want to build a video card for the SE/30 with really fast memory on board -- at which point, why not replace the system RAM with some of the RAM on the "video" card, etc.

 

[Gorgonops]

Amiga people like to do things other people say they can't. So while the rest of the world was saying that Quake and other game engines would never run on Amigas, people went and made them run on Amigas.

Well, so... looking it up it looks like the Quake port recommends a 68060, which is a good, what, 3x times faster than the best CPU a real Amiga ever shipped with? DOS Quake would run sorta-mostly-okay on 486DX4s and its target platform was the original Pentium 60 so... yeah, I'm not sure running on a 68060 is that impressive a feat. How does it run on actual *unmodified* Amiga?

 Remember - there are thousands of games for Amigas! With a floppy emulator, you can boot and play any of thousands of games quickly and easily.

Well, so this is where I have to admit ignorance: with "most" Amiga games is the experience actually improved by a more powerful CPU? I was sort of under the impression that outside a relative handful of titles released after the platform's "official" death most of the classic Amiga games were targeted at a bone-stock A500, with a lesser number aimed at machines approximating the A1200's specs. Sure there *are* these games like the various FPS clones/ports with scalable rendering that can suck up all the CPU cycles they can get, but, again, I was sort of under the impression that the bulk of the games available for the platform treated it more like a gaming console? (Again, I could be completely off-base as nobody I knew had an Amiga; a few computer shops in the area had an A500 sitting in the corner in the late 80's running demos but I don't think they were exactly flying off the shelves.)

In any case, perhaps I gave the wrong impression... I certainly think it's *neat* what these guys are doing. I'm actually very impressed that we've reached to point where a small band of hobbyists can engineer a superscalar CPU and be able to just flash the design onto an off-the-shelf chunk of silicon silly putty. Perhaps I'm just wondering out loud if an "accelerator card" is really the best goal for such a thing given all the compromises that come along with going that route vs. a complete FPGA re-creation. If nothing else those A600s aren't getting any younger.

 

[Cory5412]

Perhaps I'm just wondering out loud if an "accelerator card" is really the best goal for such a thing given all the compromises that come along with going that route vs. a complete FPGA re-creation. If nothing else those A600s aren't getting any younger.

That's almost exactly literally what I was going to say.

If somebody came along and thought that they could reimplement a fairly basic but well appointed Mac, such as the Quadra 650 or 800, I'd be extremely interested in the project, and would possibly be interested in helping out (financially.)

It doesn't have to be the 650 or 800 specifically, but I like that idea because even if you can't get anything faster than what that system already was, you've got a pretty healthy system already.

To be honest, rather than "really fast" 68k Mac, I would love to see a "really durable" one. I would like either a really good, really accurate emulator for 68k Macs (I don't care if it brings my quad i7 to its knees) or some kind of re-engineered 68k system.

Either pack it all into something the shape and size of a NUC or a Mac mini with, say, video (HDMI?), ADB, Serial, Ethernet, Audio, and an outboard-facing SD reader, or build a micro ATX board (or a new box, so you have room to properly mount NuBus hardware? Who knows) with all the trimmings of a "regular" Quadra 650/800, plus a few extras -- maybe make it a 40 or 50MHz '040, build 128 or 256 megs of RAM directly onto the board, one or two SD or CF slots for mass storage, and either build on a newer GPU or figure out how to add a whole raftload of V-RAM to the existing one, so we can just use it with regular LCD monitors.

It would cost a lot and it'll never happen, but I can dream, right?

 

[TheWhiteFalcon]

Even running full out I have trouble imagining any 68k emulator maxing out an i7.

Could someone still get NuBus slots? Or would you end up scavenging dead boards?

The cool thing would be to build Quadra 840AV boards. Useless, but cool nontheless.