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

[lp]

.

[Sorgelig]

I also would like to buy a standalone product with open spec and enough examples to show how it work. And actually i don't care whatever closed source core they will release as long as they will allow open source cores and board will be really powerful.

If it will be fully closed device like black box in style "Take what you see or pass" - then i will pass.

Currently Vampire 600 v2 has: One FPGA EP3C40, 2xRAM chips (i can't read the PN from pics, but these 2 chips are independent which is a big advantage) and 2-4 buffers. That's all. They charge 250UER for this! That's robbery. With such pricing policy and fully closed core its a full stop for me.

As for core itself, i prefer 100% compatibility than superfast 68K. There is no software for AMMX and it won't be, except some demos which could impress in 1992, but not today when you can do much more on PC (and even on ARM). I would prefer to have precise replica of 68000/020/030/040/060 for Amiga than something which won't be ever supported. Amiga(Atari) won't be closer to modern phones by speed, then why bother to make this ridiculous 68080? Just wasting the time and money. Some people already did comparisons with real 68060 80MHz and real 68060 is faster in real task intensive apps than Vampire's 68080. Of course AMMX wasn't used because all apps made for 68K don't support it. And it will remain this way because no one will write software. It's already in the past and abandoned many years ago. But FPGA is much cooler than real 68060. So it could be energy efficient modern implementation of high-end 68060.. if it could be 100% compatible and cycle exact.

[vido]

Don't think that "they" publish anything yet, see your stupid doubts, "they" have "null bock" on that kind of discussion. And that's not Apollo Team. Yes, Apollo team also does something, but they are not alone.

I am really interested in Apollo project and waiting to see the product I would like to have. But you are not objective/productive on this thread. You dont do any favor to the project. Contrary. You are doing damage to the Apollo team

[Orion_]

Currently Vampire 600 v2 has: One FPGA EP3C40, 2xRAM chips (i can't read the PN from pics, but these 2 chips are independent which is a big advantage) and 2-4 buffers. That's all. They charge 250UER for this! That's robbery.

So, you only take in consideration the price of the hardware, but not the price of the time spent creating the fpga core and the pcb schematic ?
If you want open source, it's because you know how to do fpga right ? so why don't you take an fpga board and make your own accelerator then ?
I really don't understand that spirit of wanting every source code for free, like programmer is not a job, like you should only pay the cost of the hardware.

So it could be energy efficient modern implementation of high-end 68060.. if it could be 100% compatible and cycle exact.

then, go start a 68060 core and give it us for free ! I'm waiting

[Sorgelig]

No, i wanna say, this project will have very little demand comparing to modern boards. For example Raspberry Pi has much bigger demand. I didn't ask them to put a lot of effort to develop the core which basically won't be used at the level they expect. There is open source 68K core which could be adopted to Vampire. So, basically they could make hardware only and use open source existing core. And i believe this board can boost that core development by other developers.

If o would have resources and equipment to develop such board - i would do. Although i would focus on standalone version like MiST but with more advanced specs.

Once again: neither Amiga add-on Vampire nor standalone Vampire will have commercial success in the form they want. The target audience is very very tiny. So, they set their target too high. Higher than they will get through selling. Well, they can continue to sell one by one card through ebay for 600-700EUR as they are doing now. But it's not called success. More or less successful project will be if they will sell it as a bare hardware.

[joska]

Currently Vampire 600 v2 has: One FPGA EP3C40, 2xRAM chips (i can't read the PN from pics, but these 2 chips are independent which is a big advantage) and 2-4 buffers. That's all. They charge 250UER for this! That's robbery. With such pricing policy and fully closed core its a full stop for me.

I think we all have to accept that such small scale production has it's costs. €250, what's that? Half the price of a CT60? For that price you get basically a combined CT6x+SuperVidel. Sort of.

So it could be energy efficient modern implementation of high-end 68060.. if it could be 100% compatible and cycle exact.

What is the point in a cycle exact 060? Is there a single piece of software out there that needs that?

I would prefer a "fast as fu... possible" 060.

[Orion_]

this project will have very little demand comparing to modern boards. For example Raspberry Pi has much bigger demand.

That's obvious ...
Rpi is a multi purpose cheap computer.
Vampire is made by 68k fan for 68k fan ...

neither Amiga add-on Vampire nor standalone Vampire will have commercial success in the form they want. The target audience is very very tiny.

It's not because they will sell small quantity of their board that they don't deserve money for it.
They might want to try making a living from what they do, trying to create accelerator board for a little retro computer fan community, even if it's not lasting for life, at least they can try to live some years from there passion.

[Sorgelig]

What is the point in a cycle exact 060? Is there a single piece of software out there that needs that?

I'm talking about Amiga where some software for 060 do exist. I don't thing Atari will be supported in the near future, especially with their style of coding under the blanket.
Any turbo modes are welcome, but only as an extra, not instead. There is a lot of games(and demos) requiring cycle exact CPU.
As far as i know, there is no way to switch back to original CPU in Vampire. So, you loose 100% compatibility permanently as long as Vampire module is plugged.

Rpi is a multi purpose cheap computer.
Vampire is made by 68k fan for 68k fan ...

I'm not comparing the boards. Just wanted to show the difference in market scale. And by the way RPi software is open source. Surprise, right?

It's not because they will sell small quantity of their board that they don't deserve money for it.

Did i say that? It's a question of how much you want to invest in order to get enough back. And current investment doesn't look wise, because of small market. They could get more money by making just hardware with adapt of existing open source 68K core. Same as MiST way.
Any additional investment (money, manpower, etc) to develop cores will just spending without additional income. It will increase the cost of product, but will shrink the audience even more. Apollo team just release first Vampire, but already a lot of dramas around it from them. It means, they have problem between their expectations and what they are really getting. And problem will grow as snowball. They want to release more products, but they cannot afford such big development and finish it on-time hence long delays. And i even don't talk about support from software (apps, games). Currently on the wave of excitement, everyone forgot about absence of support from software. Later, it will be more obvious.
Do you have an idea who will write games and apps for new CPU to use the whole Vampire (or standalone) potential? Do they made agreements with publishers like Electronic Arts who will start to release new Amiga/Atari games? Or they will write high quality games by them self?

They might want to try making a living from what they do, trying to create accelerator board for a little retro computer fan community, even if it's not lasting for life, at least they can try to live some years from there passion.

This is commercial project at first. That's why they went closed source way. Commercial project should bring the money or die. It's simple rule of market. Go by pure hardware way and you will get the money. And then you can use your free time to make a core as an open source just for fun (or don't do this if you have no time or fun - you won't be obliged). As long as you sell hardware as open platform, you don't need to give much promises if HW is working. This is the only way to keep this project alive, keep interest of this small community and earn the money.
It's simple and proven formula.

[1st1]

Once again: neither Amiga add-on Vampire nor standalone Vampire will have commercial success in the form they want.

You all didn't got the point. The main product based on Apollo will NOT be a turbo card for ATARI or AMIGA computer. It also will NOT be a standalone ATARI or Amiga clone. The ASIC will be used to something where a million to produce the chip makes sense. If the ASIC finally costs 10$ and they nedd 1Million to make the ASIC, then they plan at least 100.000-200.000 of these chips. Even if all lefover ATARI/Amiga-fans will buy one that would not be enough. That thing even might not need that speed. But if you produce an ASIC today, it can run at high speed, maybe producing a slow chip based on outdated technology today is even more expensive. Those Apollo projects we are discussing here is just putting some of these chips aside and profit from mass production to have something affordable what outperforms all/most of the ATARI and Amiga tubo cards or clones. It's just a benefit for us all! Understand now?

[1st1]

You are doing damage to the Apollo team

No, that's the discussion here. Not me. Your doubts. YOu might have capabilities and skills to help them to advance the thing. But, no, instead you discuss it dead. That is a damage.

[1st1]

What is the point in a cycle exact 060? Is there a single piece of software out there that needs that?

I'm talking about Amiga where some software for 060 do exist. I don't thing Atari will be supported in the near future, especially with their style of coding under the blanket.
Any turbo modes are welcome, but only as an extra, not instead. There is a lot of games(and demos) requiring cycle exact CPU.
As far as i know, there is no way to switch back to original CPU in Vampire. So, you loose 100% compatibility permanently as long as Vampire module is plugged.

Forget about ST/STE games and demos. No turbo card can be cycle exact.

Do you have a Falcon and a software collection for it? I have, and there are some things which demands a 68060, even for supervidel there is already something, some demos and games. They don't demand "cycle exact", they demand "fast". I would love to play Doom on my Falcon. And do you know Calamus 2015? I have, and I have to use it on PC (inside an embedded ATARI emulation), because on my TT it is just slow. Have you ever tryed to paint in a JPG with Photoline? It needs already 10 minutes to load on my TT (which is having a graphics card).

To run your old demos and games, just stay with your original ST. But don't talk a superfast ST/Amiga project dead.

[EvilFranky]

You can already play Doom on a standard Falcon

[ericgus]

So, they set their target too high. Higher than they will get through selling. Well, they can continue to sell one by one card through ebay for 600-700EUR as they are doing now. But it's not called success.

I agree .. if it was half that price i'd consider buying a Vampire 500 for my a500 but at 600e .. nope .. thats more than my 500 cost me .. and I am sure I am not alone.. but you are right .. the hobbyist market is very small..

[farvardin]

I really don't understand that spirit of wanting every source code for free, like programmer is not a job, like you should only pay the cost of the hardware.
/.../
then, go start a 68060 core and give it us for free ! I'm waiting

Sorgelig is passionate enough to have improved the ZX spectrum core on the MiST (and he also made several other improvements) for free!

I've bought the MiST fpga board especially because it is open source and most cores are open source.

I also support many open source projects by sending money to the developpers. It's what I enjoy in the open source.

[Eero Tamminen]

You can already play Doom on a standard Falcon

http://doom.wikia.com/wiki/BAD_MOOD

On a non-accelerated Falcon. One just needs 14MB of RAM and enough disk space.

(Game supported also playing Doom II WADs at some point, but I'm not sure whether the final version still works with them. They need quite a bit more processing power than the Doom I WADs.)

[christos]

To summarise: the vampire team have an idea for a use case outside of the Amiga/ST/Mac that we can't think of, even though many of the forum users are in the field. That's highly unlikely, but it's possible since a few businesses have managed to reuse old, outdated products in a new market. Apparently they want the Atari st community to help them in two ways. Provide insights on the 68k that they don't have, since the atari st implementation is different to that of the amiga, and provide an additional source of income for further development of their product. However they seemed to have reached a bit of a dead end, since the atari community wants a fast machine that is also backwards compatible, meaning that most use cases involve, do something really fast (e.g. compiling) then go back to see how it works on a standard machine. Also, high end users, have a few other outlets such as the Firebee or something like a Falcon + CT60 + Supervidel.
What is not going to happen under any circumstances. Falcon compatibility. The Firebee is a long term project, worked on by very knowledgeable people, but it's not even close to it. However, I am not at all against to a new GEM compatible computer, like the Milan and the Hades.
Anyway, what I think 1st1 has missed is that unlike the Amiga community, almost noone in the Atari ST side thinks that the 68K can become a viable everyday solution, with the possible exception of some music artists or a couple of businesses that do not wish to change their atari based systems because of habit.

[vido]

@hristos

I am using my Milan and FireBee as everyday solution. But it is true I dont require much. I need MiNT and terminal with ssh, Papyrus X and for most of my browsing NetSurf is enough.
But you are right. Atari community has a lot of experiences with clones and fast accelerators compared to Amia scene and we know what we can expect. Most of existing users dont dream Atari was and still are the best computers. We like it and use it from different reasons but is cant compare to today PC and Mac computers. Not because of hardware. That is quite easy problem to solve compared to software support of the platform. Still there is plenty software which can be ported to meet my needs, but that wont bring new Atari users. At least not enough ...
While Amiga community is dreaming, most od the Atari community is living in real world ...

Still I hope I will be able to get fast Atari clone based on Apollo core. Till then I support FireBee. FireBee (CT6X,MiST,MIlan, Hades,..) is real and allready useful for my needs

[calimero]

@christos thanks for summarizing

@vido "Most of existing users dont dream Atari was and still are the best computers." - EXACTLY !

...one note regarding Amiga OS: Amiga OS 4 got "live" icons in "taskbar" even before Mac OS X ("live" icons - e.g. Calendar application icon show real date instead of one fixed icon). RiscOS from Acorn has this even in late 80s though...

[christos]

@hristos

I am using my Milan and FireBee as everyday solution. But it is true I dont require much. I need MiNT and terminal with ssh, Papyrus X and for most of my browsing NetSurf is enough.

Yes, I've used my ct60 as main computer for a few months,back in 2008 and it was actually useable but the software wasn't quite there. My main problem was the lack of UTF-8 support when it came to typing. Basically I couldn't type Greek in a way that would be viewable on other computers. I could read just fine though. I really wonder if the amiga people (in classic OS mode) have solved this.

But you are right. Atari community has a lot of experiences with clones and fast accelerators compared to Amia scene and we know what we can expect. Most of existing users dont dream Atari was and still are the best computers. We like it and use it from different reasons but is cant compare to today PC and Mac computers. Not because of hardware. That is quite easy problem to solve compared to software support of the platform. Still there is plenty software which can be ported to meet my needs, but that wont bring new Atari users. At least not enough ...
While Amiga community is dreaming, most od the Atari community is living in real world ...

Bringing new users is kind of an improbability. The ST platform is basically a hobby platform for people who are nostalgic. Still it's not impossible that someone might become involved just to see what can be done. If an inexpensive board is released, then people might try to tinker with it and since TOS/GEM is more hardware independent than AmigaOS it might even be a better platform for it.

Still I hope I will be able to get fast Atari clone based on Apollo core. Till then I support FireBee. FireBee (CT6X,MiST,MIlan, Hades,..) is real and allready useful for my needs

Exactly, if I could afford a Firebee I would have had one on my desk.

[joska]

What is the point in a cycle exact 060? Is there a single piece of software out there that needs that?

I'm talking about Amiga where some software for 060 do exist. I don't thing Atari will be supported in the near future, especially with their style of coding under the blanket.
Any turbo modes are welcome, but only as an extra, not instead. There is a lot of games(and demos) requiring cycle exact CPU.

I would be very surprised if there's a single piece of Amiga 060 demo/game that requires a cycle accurate 68060... I'm not even sure if there's a single piece of Amiga software that requires a cycle exact 68000.

A cycle exact re-implementation is only of interest on the ST(e), and then it would have to run at 8MHz too. That would be a limiting factor for someone who only has a single ST and use it to watch or make demos, but they would hardly buy an accelerator in the first place.

[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.

[Sorgelig]

There is one of big problems: once you use this board (add-on or standalone) you loose original 68K CPU. So, if you install it into A600, then your only choice is this emulated CPU. While developing cores for MiST, almost every retro system i've touched has been depend on exact CPU and chips cycles. You can use non-accurate model, and you will get a lot of small glitches and incompatibilities making you want to forget about this replica and get a real system. If you don't have a 100% replica with cycle exact simulation, then why bother to get this FPGA implementation at all? There is WinUAE for Amiga emulation, which you can run on a fastest Intel CPU and get powerful Amiga, faster than you ever will need. I don't know if there is a good Atari emulator since i never interested in Atari (sorry). Most people have powerful enough PC to run the emulator and it's free.

So, here is the question: if there is a standalone board pretending to be a, let's say, Amiga retro system replica, but intentionally is not really replica, then why you would like to use it instead of an emulator? Can add these AMMX instructions and 1 cycle per instruction into emulator with significantly less effort than make a FPGA core. There is absolutely no FPGA requirements for such experiments. It will come and it will go. The classic Amiga will remain

[Sorgelig]

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.

[calimero]

There is one of big problems: once you use this board (add-on or standalone) you loose original 68K CPU. So, if you install it into A600, then your only choice is this emulated CPU.

http://www.kipper2k.com/accel600.html : "The Vampire II will sit on top of your original CPU..."
so you plug Vampire 2 in Amiga over existing 68000 but you have no option to use it?

[1st1]

There is one of big problems: once you use this board (add-on or standalone) you loose original 68K CPU. So, if you install it into A600, then your only choice is this emulated CPU.

http://www.kipper2k.com/accel600.html : "The Vampire II will sit on top of your original CPU..."
so you plug Vampire 2 in Amiga over existing 68000 but you have no option to use it?

Phuck on original 68000 in a vampirzed ST. Don't care, irrelevant. TT and Falcon are also not cycle exact to standard ST demos. Apollo makes it fast as hell and it is no more compatible to old demos and games. But Calamus and Photoline, Troll, Netsurf and such aplications will run fast as hell in MiNT.

For those old applications I have another more than 20 different working ST/STE in original condition.


You also will not buy a Ferrari to drive it slow just in town.