Thinking of a SuperCPU VIC

[groepaz]

The SuperCPU V3 is incompatible with more than only the Fastload cartridge.

According to Groepaz Gideon's implementation of the 6502 is not 100% compatible but nobody could indicate a specific piece of software it has problems with.

the various test programs that fail are not specific enough? ok then

[eslapion]

According to Groepaz Gideon's implementation of the 6502 is not 100% compatible but nobody could indicate a specific piece of software it has problems with.

the various test programs that fail are not specific enough? ok then

No. Test programs are not representative of how the CPU is used by software, especially legacy software, in the VIC-20 arena and I strongly feel you argue for the sole purpose of misleading people.

Test software look for every little nooks and crannies that software in the VIC-20 arena simply does not use.

Also, on the C64, all that's needed to activate or deactivate the built-in CPU is the pull high or low the DMA line. Not so on the VIC.

Only you consider it necessary to have a MOS 6502 on a SuperCPU for the VIC and for frivolous reasons.

I ran the rockwell CPU on my VIC and I can find NOTHING created before 1989 that will cause problems. Robotic liberation works and so does JiffyDOS.

The SuperCPU V3 is incompatible with more than only the Fastload cartridge.

Yea, I am tired... that should've been the SuperPLA V3...

But it still doesn't work with 2 different C64 mainboards. Its a serious deal.

[groepaz]

Test programs are not representative of how the CPU is used by software, especially legacy software

test programs are the one and only way to show bugs in emulators and then debug and fix them, especially emulators of legacy hardware that has tons of quirks that may or may not have been exploited by software, deliberately or not.

and obviously "100% compatible" equals "everything works, including weird test programs" - not "my favourite software from before 1989 works".

(and anyone lamenting about PLA in every thread, even when totally unrelated, should get their ass whipped)

[eslapion]

test programs are the one and only way to show bugs in emulators and then debug and fix them, especially emulators of legacy hardware that has tons of quirks that may or may not have been exploited by software, deliberately or not.

I guess you should tell that to Rockwell and WDC who sold tens of millions of X65C02 CPUs and couldn't care less about your test programs. These are no emulators, they are the real deal.

Ricoh too sold very large numbers of CMOS versions of the 6502.

You're just spreading your ultra purist mentality and telling other people what to do.

I dare you to find a SINGLE piece of VIC-20 software created before 1989 that will not work on a VIC equipped with a R65C02.

@Tokra
I consider the software you mentioned to be very nice creations but they are specific and were created in the 21st century. As far as I know they were created with the assistance of external modern computer systems.

[eslapion]

@Groepaz
Anyone lamenting about incompatibility with exotic software of interest only to coding fashionistas should get their ass whipped.

I mentioned the PLA issue because YOU 1st mentioned compatibility with Fastload was a big deal.

Once again, I dare you to find a SINGLE piece of VIC-20 software created before 1989 that will not work on a VIC equipped with a R65C02.

You're willing to argue to the death about nonsensical issues solely for the purpose of antagonising anyone who disagrees with you.

Equipping a VIC-20 CPU accelerator with a genuine MOS CPU is strictly POINTLESS.

You don't agree? Make your own and stop telling people what to do with their own projects.

[groepaz]

I guess you should tell that to Rockwell and WDC who sold tens of millions of X65C02 CPUs and couldn't care less about your test programs. These are no emulators, they are the real deal.

please stay focussed. i was referring to gideons 6502 core.

could one of the mods please remove the poinless lamenting? thanks.

[eslapion]

@Tokra
I suppose the interest of creating and running the software you mentioned is to exploit to the absolute maximum the capabilities of the original machine so I assume there is no point in adding a CPU accelerator in such a case.

Also, the software you mentioned must be very timing sensitive on the CPU instructions execution so, once more, a CPU accelerator is not only pointless, its becomes an obstacle to achieving the set objective.

[eslapion]

please stay focussed. i was referring to gideons 6502 core.

I don't care.

Any CPU acceleration on the VIC-20 involves using a CPU that has differences with the original MOS6502. These differences will probably cause less than 0.01% of software created after 1989 and none of those created before to have problems.

Yet, this is enough to get you to tell me what to do with my life and how to manage my hardware projects. I think I know what you would tell me if I ordered you to convert to NTSC all the demos you coded that only work on PAL VIC-20 and C64...

Who's lamenting?

Your posts, as usual, have only one function: antagonise

[groepaz]

I don't care.

*plop*

[pixel]

Yous two combined aren't exactly the spearhead of the homo marriage movement.

I agree with groepaz. Less than 100% compatibility could really steal time. I really fear such situations.

[groepaz]

the spearhead of the homo marriage movement

LOL

[tokra]

I remember a long discussion on Lemon64 where one couldn't get Doom to run on his VIC and he tried lots of stuff before finally figuring out he had a 65C02 in there that did not support illegal opcodes. Of course anyone can build and/or use what they like, but if you want to avoid such problems better use 100% original items. Now, if lots of people decide to use the 1541U you DO know it may not work with 100% of all software, but other people will have the same problem and the 1541U-firmware may be fixed. This is hardly possible with a CPU-replacment.

For me it's a question of economics: I can easily live with the 1541U, it works with 99.9% of the stuff I like and I don't have to carry 10 boxes of disks and a drive. A CPU-replacement does not offer me any advantages. If you have fun tinkering around with a CPU-replacement no one is stopping you. It just might not be everyone's cup of tea. That's fine.

[groepaz]

thats what i basically said

[eslapion]

IA CPU-replacement does not offer me any advantages.

That's the point. If you want 100% total compatibility then I guess you've got no need for a CPU replacement/accelerator in your VIC-20.

If you have fun tinkering around with a CPU-replacement no one is stopping you. It just might not be everyone's cup of tea. That's fine.

Same thing here...

thats what i basically said

Nope... what you said is a CPU accelerator must also have a MOS6502 CPU on it. If that's what you think then make your own.

[groepaz]

Nope... what you said is a CPU accelerator must also have a MOS6502 CPU on it. If that's what you think then make your own.

what i said is that it should be possible to switch such accelerator into "no acceleration" mode, because you cant easily remove it (like a scpu for example). and that (and only that) non accelerated mode should be 100% compatible - and the easiest (and obvious) way to achieve that is using an (additional) original (and hence 100% compatible) MOS6502. no more, no less.