Is the Coldfire project dead?

[bloodline]

Well Oli is still around, he is posting on the Vulcan website about that one that got sold on Ebay.

[bloodline]

lordv wrote:
Coldfire V5 has an ability to fully emulate 68k opcodes... But only AT THE FIRST SIGHT. No provisions made to emulate 32-bit muls and mulu instructions (coldfire just won't trap on them and will make rubbish), so virtually every piece of real amiga code will fail on coldfire.

The overall JIT emualtion as well as other methods of workaround for this bug will make coldfire totally uneffective when compared even with 060.

So the coldfire accelerators "suck and must die" =)

It would probably work out better to build the emulator around an XScale, and then run a 68k JIT on that.

[bloodline]

billt wrote:
>As far as I understand, there is also a software-bit to the
>ColdFusion accelerators. There will be somekind of emulation
>made in software (on a ROM on the board) that will take care
>of this.

The "emulation" would be very very similar in concept to 68040.library or 68060.library. In fact, it's the exact same concept, but will probably need to trap and emulate more instructions than those libraries did. If they put it on the board and not in a file on disk, I sure hope they use a nice FLASH that can easily be rewritten with bugfixes and performance improvements.

If the Coldfire doesn't trap 32-bit muls and mulu instructions then there is no way it could ever run Amiga software without something like a JIT... in which case using a modern RISC would be a better idea.

[bloodline]

But karlos... the coldfire isn't going to trap the 32bit mul, it's just going to return the wrong result.

Also, if one were to go the JIT route... then I would prefer an XScale rather than a G4, as the ARM uses less power and generates less heat and is much cheaper!

[bloodline]

Yeah I know about dynamo :-)

I see what you mean, the JIT can catch the erroneous instructions rather than the CPU... but personally... if we are going the JIT route... I wanna use an XScale :-p

[bloodline]

Karlos wrote:

@Whabang

Dont worry, most of it is academic argument  - unless someone does release a G4/G5 card for the classic :-D

Or an XScale card :-D

Imagine a Trapdoor connector with an FPGA to convert the ZII (A1200 trapdoor) bus signals (and generate an interupt) to one of the hi-speed serial interfaces (Hirose DF12C(3.0)60DS0.5V80 or NSSP?) for a gumstix computer...

Gumstix Board

That would be a cool A1200 Acelerator...

[bloodline]

Karlos wrote:

bloodline wrote:
Or an XScale card :-D

Or an XScale card.

What about an Athlon64 card running a JIT emulation? For tower systems only of course. You get the benefit of decent off the shelf controllers for memory, AGP, IDE etc then :-)

I've thought about that actually... since Free HyperTransport core are available for FPGAs...

The Athlon64 has 16 General purpose registers, which makes it ideal for 68k emulation... it's general purpose regs are 64bit, so JIT book keeping could be kept in the upper 32bits of the registers (sort of treat the 16 64bit regs as 32 32bit regs)... the CPU is REALLY fast and very cheap... it's 64bit (always a good thing :-D), it got strong Vector support (MMX, SSE, 3DNow!), SMP is possible... oh and AROS support is there ;-).

But, the CPU is big, it's hot, it draws a huge amount of power.. it needs one hell of a heatsink... while I agree that it would be perfectly acceptable for someone with a Tower system... these people would probably be better off buying an A1/Peg or a modern PC and Running WinUAE...

The XScale on the other hand could go into a stock A1200 (or even other Amiga Models) and run with no heat, power or space issues. It should be able to far outperform the 060 , and be far more affordable...

Hell if we're using an FPGA one could even put a few nice features in there too... a Chunky buffer (+ Blitter) with a DMA to Lisa... which would allow the CPU to write chunky pixels and would convert the pixels to Planar  during the screen refresh :-)

[bloodline]

Karlos wrote:
One reason I suggested the althon 64 was largely because it apparently dissipates a lot less heat than previous generations.

That's true... 32Watts normal, 65Watts underload, 95 Watts max (IIRC)... but my BlizzPPC melted... and I'm sure that didn't pump out 35Watts of heat... I think that the small "British" RISC ARM CPU is better suited to the task :-D

Hell... I want an Athlon64 Accel for my A1200 now :lol:

[bloodline]

Karlos wrote:
How the hell did you melt your blizzppc?

I have the 040 version and it doesnt get that hot!

IIRC I have the same board as you... 25Mhz Full 040 + 240Mhz PPC... I kept it in the desktop case, with the trapdood lid off... then the fan melted... then wires melted and other things melted... I found an old 486 and used the cooler from that to cool the 040... but it was a pain in the bottie as I was moving around a lot at the tim... also taking the A1200 to gigs and stuff so I swaped my Bliz1230-IV in for increased reliability. I've not put the PPC back in since :-/

[bloodline]

Dan wrote:
400Mhz, well then a Coldfire would be faster than what my pc does in UAE(6 times faster than a 060-50Mhz( or  PPC-JIT on 1Ghz MOS/OS4)).

You forget about the IPC... it's not that great on the Coldfire... certainly not as good as the 060 (or the PPC or the Xscale or the Althon64 etc...)

[bloodline]

Dan wrote:

bloodline wrote:

Dan wrote:
400Mhz, well then a Coldfire would be faster than what my pc does in UAE(6 times faster than a 060-50Mhz( or  PPC-JIT on 1Ghz MOS/OS4)).

You forget about the IPC... it's not that great on the Coldfire... certainly not as good as the 060 (or the PPC or the Xscale or the Althon64 etc...)

I don´t know anything about that. first term java and c-programming only here.:-P
IPC=interprocess communcation?

Instructions per Cycle :-)