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

[1st1]

It looks like the guys making an ATARI compatible machine based on Apollo are using EmuTOS. Who cares on TOS 1.x / 2.0x for a highend ATARI machine?

TOS 2.0x does not require Line-A. It's only there for compatibility to old apps.

Please listen to other's opinion. It would be not very good to have a really fast atari computer, and not having a software working on it. Maybe the current atari software library is not much independent from hw as the amiga one. Atari community is not as big as amiga. If there is no resource to rewrite those programs, in the end, you will have a computer and the only sw that works on it is emutos (I am ironic here by intent)

If it's not good for you, then please don't care. But please don't tell other peoples what they would love to have. There is a lot of software which will run on such a fast ATARI, look at Firebee, there is already a lot, and with 68080 instead of Coldfire, there will be even more, like Calamus 2015.

[AnthonyJ]

If it's not good for you, then please don't care. But please don't tell other peoples what they would love to have.

If it's good for you, then please do care. But please don't tell other peoples what they would love to have.

Or in other, more grammatically sane words - just stop posting. You've made enough posts to make your view very clear, now leave it at that. If other people don't believe in it as much as you, that's absolutely fine, just as it's absolutely fine for you to be excited about it.

[joska]

If it's not good for you, then please don't care. But please don't tell other peoples what they would love to have. There is a lot of software which will run on such a fast ATARI, look at Firebee, there is already a lot, and with 68080 instead of Coldfire, there will be even more, like Calamus 2015.

You certainly create a lot of confusion with your posts. From your statements it sounds like the Vampire is just another clone, while in reality it will be a *lot* more compatible with ST software than even the stock Falcon. Your posts about "f**k compatibility" is exactly what the Vampire is *not* about.

[ex68k]

If it's not good for you, then please don't care. But please don't tell other peoples what they would love to have. There is a lot of software which will run on such a fast ATARI, look at Firebee, there is already a lot, and with 68080 instead of Coldfire, there will be even more, like Calamus 2015.

You certainly create a lot of confusion with your posts. From your statements it sounds like the Vampire is just another clone, while in reality it will be a *lot* more compatible with ST software than even the stock Falcon. Your posts about "f**k compatibility" is exactly what the Vampire is *not* about.

There is really a lot of switching in this thread between the "vampire-turbo-card" and the "vampire-68k-look-alike-standalone-system".
Sometimes hard to guess, which one the people are talking at the moment.

[calimero]

The point is it's not competitive with 20 year old PC chip! It's NOT modern!
Here's my suggestion for a fast and cheap accelerator. Take an off the shelf ARM SOC and interface it with the ST. Run the UAE ARM JIT on it.
It would scream along. It would probably run like a 2.x ghz 060.

well not exactly: "at the speed of a 68040@350 MHz on a RPi3 and a whooping 2 to 7 GHz on a x86 depending of the processor." recording to Faucon2001 https://sites.google.com/site/emaappsar ... the-manual

[1st1]

There is really a lot of switching in this thread between the "vampire-turbo-card" and the "vampire-68k-look-alike-standalone-system".
Sometimes hard to guess, which one the people are talking at the moment.

They do not talk about turbo card or standalone, but only about their worries.

[Frank B]

The point is it's not competitive with 20 year old PC chip! It's NOT modern!
Here's my suggestion for a fast and cheap accelerator. Take an off the shelf ARM SOC and interface it with the ST. Run the UAE ARM JIT on it.
It would scream along. It would probably run like a 2.x ghz 060.

well not exactly: "at the speed of a 68040@350 MHz on a RPi3 and a whooping 2 to 7 GHz on a x86 depending of the processor." recording to Faucon2001 https://sites.google.com/site/emaappsar ... the-manual

The RPI is not a high performance ARM chip.

Frank

[Faucon2001]

Thanks Calimero, I wouldn't have said better
Oops, I am leaving now, I don't want to put more oil on the fire
Have fun, enjoy Atari.
Cheers

[joska]

well not exactly: "at the speed of a 68040@350 MHz on a RPi3 and a whooping 2 to 7 GHz on a x86 depending of the processor." recording to Faucon2001 https://sites.google.com/site/emaappsar ... the-manual

A 350MHz 040 is still a very fast "Atari". The problem with this - atleast for me - is that ARAnyM needs JIT for this impressive speed, and JIT means no MMU. And no MMU means no memory protection in MiNT, which is a *huge* drawback. It is the #1 reason why I prefer to use my Milan060 and Falcon/Afterburner/Eclipse instead of the FireBee.

[joska]

There is really a lot of switching in this thread between the "vampire-turbo-card" and the "vampire-68k-look-alike-standalone-system".
Sometimes hard to guess, which one the people are talking at the moment.

True. Personally I'd prefer the accelerator version. The standalone version is also very interesting but it's really nothing more than a souped-up MIST.

[Frank B]

well not exactly: "at the speed of a 68040@350 MHz on a RPi3 and a whooping 2 to 7 GHz on a x86 depending of the processor." recording to Faucon2001 https://sites.google.com/site/emaappsar ... the-manual

A 350MHz 040 is still a very fast "Atari". The problem with this - atleast for me - is that ARAnyM needs JIT for this impressive speed, and JIT means no MMU. And no MMU means no memory protection in MiNT, which is a *huge* drawback. It is the #1 reason why I prefer to use my Milan060 and Falcon/Afterburner/Eclipse instead of the FireBee.

Do you think this would be harder to solve than MMU support on the Vampire?

[joska]

Do you think this would be harder to solve than MMU support on the Vampire?

I can't tell, because I've never seen the code for the 68080 But this is the one major drawback for the "68080". However, I would - by far - prefer a Vampire without MMU over an emulator with MMU

[1st1]

68080 has a kind of memory mapping and protection unit. They call it MCU, they say it's even better than PMMU (but not compatible). Gunnar told, that they can integrate PMMU functionaility and compatibility in this MCU, but they did not see the neccessarity of that step yet (as Amiga OS does not use it). This is something ATARI peoples must explain the Apollo team. I already tryed, but my knowledge regading this is too small to have the convincing arguments. Here are guys who could do this better than me. The other way could be being happy and integrate MCU support in MiNT. The point is, instead of beeing cooperative with them amd help to advance the thing, some guys here fire on Apollo, maybe because it's not their project? Jealousy? Remainings of the Amiga-ST-bashing-era? I'm a bit helpless because of the extreme reactions against Apollo/68080/Vampire project in this discussion here.

[troed]

Jealousy?

I'm a bit helpless

This thread is pure comedy gold.

[1st1]

It's more than comedy... It's progress...

https://sourceforge.net/projects/emutos ... tos/0.9.7/

EmuTOS 0.9.7
Major changes:
(...)
- BIOS: add support for Apollo Core 68080
(...)

Thanks to the EmuTOS team! Another step towards a new ATARI computer and turbo card. While here only discussing against it, some are doing.

[joska]

68080 has a kind of memory mapping and protection unit. They call it MCU, they say it's even better than PMMU (but not compatible). Gunnar told, that they can integrate PMMU functionaility and compatibility in this MCU, but they did not see the neccessarity of that step yet

A "real" (as in "same as 68030) PMMU would be very nice, but wouldn't that require long stackframes? If that's the case then you'll loose the compatibility advantage of the current 68000-"mode".

[1st1]

I don't know, ask here: http://www.apollo-core.com/knowledge.ph ... 5&z=sDNKSQ

[joska]

I don't know, ask here: http://www.apollo-core.com/knowledge.ph ... 5&z=sDNKSQ

I have tried to register twice to that forum without success.

[BlankVector]

I have tried to register twice to that forum without success.

Actually, I thought the same...
I was waiting infinitely for the confirmation e-mail... while it was not necessary at all! You can login immediately after registering.

[joska]

I could not. I just a got message saying that my account had to be verified. Tried twice, even sent an email but never heard anything.

[1st1]

No idea, for me the forum registration also worked immediatelly. Maybe going to their IRC channel and ask there for assistance?

[joska]

A "real" (as in "same as 68030) PMMU would be very nice, but wouldn't that require long stackframes? If that's the case then you'll loose the compatibility advantage of the current 68000-"mode".

Having studied the discussions at the Apollo forum I see that I have got my facts wrong. I was under the impression that the Apollo core is an extended 68000. It's not. It's a 68040 with better backwards compatibility. That means several things:

1. TOS 1.x will not work, contrary to what I've said before.
2. Games compatibility will be poor (long stackframe, no TOS 1.x).
3. They can (and should!) implement a 040/060 PMMU.

A fast "68000" without PMMU would be OK, since the lack of PMMU is compensated by the improved compatibility and speed compared to other accelerators. But a "040" without PMMU is of little interest for me. Feeling slightly disappointed now, just hope that it will get a real PMMU and not some new invention

[AtariZoll]

A "real" (as in "same as 68030) PMMU would be very nice, but wouldn't that require long stackframes? If that's the case then you'll loose the compatibility advantage of the current 68000-"mode".

Having studied the discussions at the Apollo forum I see that I have got my facts wrong. I was under the impression that the Apollo core is an extended 68000. It's not. It's a 68040 with better backwards compatibility. That means several things:

1. TOS 1.x will not work, contrary to what I've said before.
2. Games compatibility will be poor (long stackframe, no TOS 1.x).
3. They can (and should!) implement a 040/060 PMMU.

A fast "68000" without PMMU would be OK, since the lack of PMMU is compensated by the improved compatibility and speed compared to other accelerators. But a "040" without PMMU is of little interest for me. Feeling slightly disappointed now, just hope that it will get a real PMMU and not some new invention

Huh ! I guess you need to admit that some people here saw it much earlier

[1st1]

@joska: Please tell your opinion to the Apollo team. This is the way to get them to the descicion to make the PMMU. They don't follow this discussion here.

[PeterS]

I really like my Milan, it's a pity it wasn't more popular.

I have the 68060 upgrade but you can never have too much cpu power. A Coldfire upgrade would have been nice but if an Apollo 68080 plugin cpu was available offering a good speed increase I would be interested.

I remember hearing rumours of a Motorola 68080 years ago but began to think I had imagined it.