Apollo Team announces developing of Vampire standalone version to run as AMIGA and ATARI ST

[joska]

I'm not even sure if there's a single piece of Amiga software that requires a cycle exact 68000.

I think some of the older demos might.. and I know there can be loads of problems with running a PAL demo on an NTSC machine so there some timing sensitivities with the older A500 era stuff.

That is a completely different thing. If you try to run a PAL (50Hz) demo on an NTSC (60Hz) machine then there's a good chance that there isn't enough CPU time left to finish a frame before the next frame is displayed.

[joska]

I would be very surprised if there's a single piece of Amiga 060 demo/game that requires a cycle accurate 68060...

There is another problem - they've added new instructions into 68K which will confuse and may break apps/games/demos expecting original behavior.

Really? And exactly how can this happen? How can the presence of e.g. a 68060 instruction break an app that is compiled for a 68000? Yes, if there are *missing* 68000 instructions, or the instructions behave differently than on a real 68000 there will be problems. But as I understand it this is not the case with the Vampire.

It looks like the games/demo people does not understand the purpose of this accelerator. It is *not* intended to run games and demos from 1988. They don't claim that it will either. This accelerator is for people who either wants to run "productive" stuff really fast, write or port new stuff for it or simply explore it's possibilities.

[vido]

It looks like the games/demo people does not understand the purpose of this accelerator. It is *not* intended to run games and demos from 1988. They don't claim that it will either. This accelerator is for people who either wants to run "productive" stuff really fast, write or port new stuff for it or simply explore it's possibilities.

Exactly!
I never understood why I would need so fast accelerator to play old games on it. You can still play them under emulation if you port old version of hatari. It should run fast enough.
Such CPU power needs new or productive software which can use more power.

[ericgus]

I'm not even sure if there's a single piece of Amiga software that requires a cycle exact 68000.

I think some of the older demos might.. and I know there can be loads of problems with running a PAL demo on an NTSC machine so there some timing sensitivities with the older A500 era stuff.

That is a completely different thing. If you try to run a PAL (50Hz) demo on an NTSC (60Hz) machine then there's a good chance that there isn't enough CPU time left to finish a frame before the next frame is displayed.

yes but my point was there are things that ARE timing sensitive .. I was just using that as a quick example, not a specific one.

[ericgus]

I would be very surprised if there's a single piece of Amiga 060 demo/game that requires a cycle accurate 68060...

There is another problem - they've added new instructions into 68K which will confuse and may break apps/games/demos expecting original behavior.

Really? And exactly how can this happen? How can the presence of e.g. a 68060 instruction break an app that is compiled for a 68000? Yes, if there are *missing* 68000 instructions, or the instructions behave differently than on a real 68000 there will be problems. But as I understand it this is not the case with the Vampire.

It looks like the games/demo people does not understand the purpose of this accelerator. It is *not* intended to run games and demos from 1988. They don't claim that it will either. This accelerator is for people who either wants to run "productive" stuff really fast, write or port new stuff for it or simply explore it's possibilities.

It really is two wholly different markets.. retro gamers and modern users .. the proposed "standalone" version I think would be more appropriate for people who as you say want to use productive stuff and/or use it as replacement/substitute for a modern computer. However I agree with what others have said that that market would most likely be pretty small .. at least when compared (to the already small) retro gamer community that uses original hardware and not emulators or things like MiST. This "modern user" community would be an even smaller subset of that.

[1st1]

I would be very surprised if there's a single piece of Amiga 060 demo/game that requires a cycle accurate 68060...

There is another problem - they've added new instructions into 68K which will confuse and may break apps/games/demos expecting original behavior.

This is bullshit. As long new instructions use new opcodes (means: don't replace old instructions) they don't disturb applications which are only using the old istaructions. The only thing which can disturb is if the new CPU is (much) faster (or slower) than the old one, when it is about timing critical software. But that has nothing to do with new instructions.

This is like you have a car without air conditioning. And one day you have to use a different car which has ac. As long you don't use the ac (and is switched off) that car will heat or cool you the same way as in your normal car.

[exxos]

Come on people, lets keep it civil please!

[joska]

yes but my point was there are things that ARE timing sensitive .. I was just using that as a quick example, not a specific one.

The only time you get CPU-related timing issues on an Amiga would be if the CPU is too slow. That does not seem to be a problem with the Vampire

[ericgus]

For the record the Vampire board does interest me as an interesting way to "update" my old A500 to be closer to an A1200 or even an A4000 .. the aspects of the Vampire that interest me most have little to do with the cpu feature of the vampire, the SD card/IDE, HDMI output and the potential to have RTG/AGA chipset functionality in my old A500.. Honestly as far as the CPU goes as long as it can competently work akin to the 68EC020 that was found in a1200 or the cpu found in an a4000 {68EC030 or 68040 CPU} then that's good enough, and I am sure I am typical of most people who would buy the vampire. I suspect the Atari ST crowd is probably very similar.. Before this thread started I had heard of the Vampire but didn't pay much attention to it, I will say after this long discussion I have done more research and I do find it very interesting and I hope they are successful bringing all their boards to the market.. (assuming they can do it at a reasonable price and not what I am seeing on ebay)

[hubersn]

This is bullshit. As long new instructions use new opcodes (means: don't replace old instructions) they don't disturb applications which are only using the old istaructions.

Don't know about 680x0, but many other processors have an undefined instruction trap which e.g. raises an exception or branches to a vector or whatever. So if a previously undefined instruction becomes a defined instruction, behaviour has changed.

hubersn

[1st1]

THat's true, but who uses undefined opcodes to just run in a trap? Two exceptions: Line-A and Line-F on TOS. But... Check Apollo instrcution list on their page, they are not used! Additionally, maybe only game and demo programmers use such tricks, and I think we already agreee that such games and demos will not run on Apoloo as it is too fast.

Summarize: Instead of being excited that you can get a high speed turbo card with integrated full hd graphics you still search the hair in the soup. What's going on with the ATARI community???

[mfro]

... Instead of being excited that you can get a high speed turbo card with integrated full hd graphics ...

I don't know if you know anything more than I do, but for now, I can't see I can.

My understanding is that there is still not enough supply to even serve the Amiga community's demands and regarding Atari, we don't have anything more than just vague announcements. Even if there would be enough supply, to my knowledge there is nothing available to drive the HD graphics in an Atari and its not even clear yet if the card would work in it at all. There were pretty powerful accelerators available to Amigas before (admitted, much more expensive) and none of them ever made it into the Atari world.

Wake me up when all this is available and working, then there will be still enough time for excitement.

[edit: on second thought, this might need additional explanation as it might be read as criticism to the Apollo guys - it's not. They are a hobbyist project with limited resources like many others and they are pretty clear on what they think they will be able to deliver and what not (and I agree what they have appears to be promising - but not more. Yet). So its not them I accuse for trying to rise false expectations - it's you, 1st1. And this will probably won't do any good on the project. I can only repeat myself: read more carefully...]

[hubersn]

Summarize: Instead of being excited that you can get a high speed turbo card with integrated full hd graphics you still search the hair in the soup. What's going on with the ATARI community???

Since I am not part of the ATARI community (I'm only here because I happen to have a MIST), you might forgive me my lack of enthusiasm.

However, I understand the lack of enthusiasm for the generic "we build a really fast 680x0 clone" case. Faster and cheaper ARM SoCs are readily available, and most people here seem to lack the imagination to see where such a fast clone could be successful. So that leaves the retro stuff. I think we all agree that the really fast 680x0 clone will not help the retro games/demos department. Which leaves a very small group to get potentially excited.

Add to that the fact that many of us have been burnt before (I am member of the Acorn RISC OS scene, you wouldn't believe how many broken promises I witnessed over the years), and you might have an idea why there is a lack of excitement.

Nowadays, you can easily create excitement by "just delivering stuff". So just do it. Talk is cheap.

hubersn

[Sorgelig]

Faster and cheaper ARM SoCs are readily available, and most people here seem to lack the imagination to see where such a fast clone could be successful. So that leaves the retro stuff. I think we all agree that the really fast 680x0 clone will not help the retro games/demos department. Which leaves a very small group to get potentially excited.

Absolutely correct! That's what I'm trying to tell.
It's more important to emulate/simulate retro platform as close to real as possible on modern chips. Turbo is welcome, but ONLY after 100% compatibility.
Real hardware sooner or later will die, thus having 100% replica is more useful than childish scores in benchmarks.

[Sorgelig]

THat's true, but who uses undefined opcodes to just run in a trap?

This only shows that you don't understand the programming on low level.

[joska]

It's more important to emulate/simulate retro platform as close to real as possible on modern chips.

More important to whom? They don't seem to have any problems selling Vampires for the Amiga...

[1st1]

THat's true, but who uses undefined opcodes to just run in a trap?

This only shows that you don't understand the programming on low level.

So go here and check for your trap tricks. http://www.apollo-core.com/index.htm?page=instructions This is the complete list of instructions of 68080.

[Frank B]

THat's true, but who uses undefined opcodes to just run in a trap?

This only shows that you don't understand the programming on low level.

So go here and check for your trap tricks. http://www.apollo-core.com/index.htm?page=instructions This is the complete list of instructions of 68080.

All the FPU ones are listed yet 404. Looks like that list is a superset.

[fury23]

It's more important to emulate/simulate retro platform as close to real as possible on modern chips.

More important to whom? They don't seem to have any problems selling Vampires for the Amiga...

More important to retro comunity.

[Rajah Lone]

More important to retro comunity.

I wonder why retro community would buy something that move itself out of retro space.

From my point of vue, Compatibility and Enhancement is oxymoron. If you want both, then you'd better search for real unicorns.

[joska]

More important to retro comunity.

Another one that does not realize that the retro community is a lot more diverse than playing games and wathcing demos from the 80's

[ex68k]

More important to retro comunity.

I wonder why retro community would buy something that move itself out of retro space.
From my point of vue, Compatibility and Enhancement is oxymoron. If you want both, then you'd better search for real unicorns.

Oxymoron? Not really. Look at the MIST we have already, you can just start whichever core you like to use.
So it "could" be the 110% atari/amiga/etc. compatible core, or the 1e99 times faster 68k, with some enhancements.
Your choice, on the same hardware.

Apollo is only interested (as it seems!) in the fast, faster, even faster core. So somebody else would have to take
care of the compatible, 100% core and chipsets ....

[Rajah Lone]

Oxymoron? Not really. Look at the MIST we have already, you can just start whichever core you like to use.
So it "could" be the 110% atari/amiga/etc. compatible core, or the 1e99 times faster 68k, with some enhancements.
Your choice, on the same hardware.

Thus you have to choose between Compatibility AND Enhancement. You can't have both at the same time. I wonder how demos behave in the STEroids mode, or if Degas Elite and some old GEM applications have correct display in enhanced VDI modes.

Apollo is only interested (as it seems!) in the fast, faster, even faster core.

I thought the majority would be also interested in the graphics part of this FPGA card, but there I'm mistaken. One wants overclocked CPU to feel power and manhood, but not be at ease with a colorful and large screen?

[ex68k]

I thought the majority would be also interested in the graphics part of this FPGA card, but there I'm mistaken. One wants overclocked CPU to feel power and manhood, but not be at ease with a colorful and large screen?

Problem here is the frame rate, and all this PAL/NTSC switching on the 100% compatible cores.

You can have a 4K display, but then you would have to go for EmuTOS, or, whatever, leaving the compatibility route ...

[solskogen]

For what it is worth: the demo systematic error by Imagina does almost work in STe-reoids. It does not work in normal STe mode.