FPGA Amiga

[Chucky]

Being fully compatible BRINGS the bad things aswell.. like the crashes you are talking about.. as you might know.  Amiga software is pretty often actually to be honest much badcode..  extremly much. so you never know what it expects etc etc.  so if it behaves different. some code might also execute wrong.

say they test "ok do we have a 040 or 060?"  and they test an instruction that exists on the 040 but not on the 060..   and "OK"  this instruction exists.  then we are a slow 040..  lets do this routine instead of the nicer 060 code..

and voila.  result is not as expected..

(even if there ARE better ways of detecting the 060. this was however one example how it could be)

so for me. .FPGA must be EXACT.. or not at all :-/

[guest11527]

so for me. .FPGA must be EXACT.. or not at all :-/

You surely don't what that either... I mean, as slow as the 68000 (which would be, of course, exact).

Look, it is at the very end a matter of the definition of the goals of this development, a matter of the desired use cases, and the requirements we can derive from the use cases.

You may have other requirements than Gunnar, and I certainly also have others than Gunnar. I would need a fast development machine - that does not mean "cycle exact", but "toolchain compatible". It currently isn't. Tough luck. It does not mean "it's bad". It just means "probably not the right thing for me right now".

If your requirements are "must do exactly as the Amiga does", then, I afraid, the only thing that can satisfy that is an Amiga. Then, however, you have also no rights to complain about "it being a bit slow". That's part of the "exactly as the Amiga" deal. You cannot have perfect speed without actually also loosing something.

Again, I'm not saying that this is a bad decision you have made. It is a valid decision. Problaby not the one I would have taken, but well, people are different.

IMHO: The project is great. Probably not quite for me at this point, but I don't bother too much.

[Chucky]

well if it should be a 68000 replacement.  but this is abotu the fastest.. so exact 060.

[psxphill]

Thanks for proving my point...

If you go round insulting and bullying people, then expect it back. That is not a point.

[IanP]

I wonder what people would say if a company had launched the 68080 SoC as an ASIC without any prior public discussion (with all the same features as planned for the core of the Vampire 4 standalone except the Amiga AGA backwards compatibility, ~80MHz, AMMX, FPU, 24 bit graphics, 16 bit audio, new registers etc for around $100). Would everybody be clamouring for somebody to start building Amiga compatibles using the chip or would there be people complaining that it's not identical to an existing 680x0 processor?

[Gulliver]

I wonder what people would say if a company had launched the 68080 SoC as an ASIC without any prior public discussion (with all the same features as planned for the core of the Vampire 4 standalone except the Amiga AGA backwards compatibility, ~80MHz, AMMX, FPU, 24 bit graphics, 16 bit audio, new registers etc for around $100). Would everybody be clamouring for somebody to start building Amiga compatibles using the chip or would there be people complaining that it's not identical to an existing 680x0 processor?

A company is a serious entity looking for profit, not a hobby project that is done for love, enjoyment, etc.
A company that releases any chip will document it completely, so it can be used to its full potential.
A company that releases any chip will supply all the help its developers might need.
A company will not change its chip features a lot. If an important  change is required it will create another chip for this purpose because keeping the compatibility is of supreme importance.
A company will surely posts benchmarks, but will avoid wild claims of world domination.
A company will not leave its PR to its main HW developer who is a guy that lacks the communication skills required for that job.

So a serious company would change the scenario in a huge manner. And this is just a quick list of changes, much more could be gained this way.

[IanP]

None of which has any baring on the question I asked.

[Gulliver]

It has.

Everybody wants something different in a next generation Amiga or Amiga-ish platform. So there would not be a consensus on what people would say: some will like that idea of yours and ask for someone to build something out of it. Others would argue against it for lots of reasons (they prefer their PC running WinUAE, might say that a Raspberry Pi is a faster and cheaper choice, for others MorphOS and its hardware suits best, etc.).

A serious company will probably get much more hardware developers interested and more software developers interested too. That in the end means more users and more people interested in what it offers, unlike the current state of the Apollo core.

[kolla]

A serious company will probably get much more hardware developers interested and more software developers interested too. That in the end means more users and more people interested in what it offers, unlike the current state of the Apollo core.

Not to mention, a roadmap one can relate to, proper documentation and support in toolchains.

People have paid good money for 68k support in glibc and gcc before (not even so long ago, I think it was 7 years ago that CodeSourcery, now Mentor Graphics, owned by Siemens, updated the kernel and the GNU toolchain for Linux/m68k with threaded library support).

[grond]

say they test "ok do we have a 040 or 060?"  and they test an instruction that exists on the 040 but not on the 060..   and "OK"  this instruction exists.  then we are a slow 040..  lets do this routine instead of the nicer 060 code..

and voila.  result is not as expected

And what damage is there? The 080 will execute both 040 and 060 optimised code faster than any 040 or 060.

[grond]

Oh, and yes, it is not a company doing this so expecting the project to be run like Motorola would have done it is a bit silly.

Obviously you have no idea how much work has went into it but all you can do is demand it should have been more like documentation, compiler support, MMU blabla. Must be really nice to sit in your comfy chair by the fireplace and explain to the world how things should be done...

[ppcamiga1]

I wonder what people would say if a company had launched the 68080 SoC as an ASIC ...

Correct answser on this question is:
 if natami/apollo/vampire will be as it was announced in 2008 - 2D/3D ps2 graphics performance, cpu at least as fast as slowest NG hardware and fully compatible, for 100 E only, people will be happy with it.
   In 2018 vampire has 68060 50 MHz integer performance, 12 times worse floating point performance than 68060 50 MHz, has no MMU, has no 3D support.  
Advantages over old 68060 cards for amiga are: faster RAM and faster 2D graphics.
  At current price, performance, compatibility vampire is not attractive to amiga ng users, and advanced classic users.
It is simple, nobody want to change hardware to worse and pay for it.

[wawrzon]

It is simple, nobody want to change hardware to worse and pay for it.

yeah. and now kneel down in a dark corner facing the wall. cover your ears, start rocking your head forth and back and repeat your mantra to yourself for the rest of your life.

[adonay]

Oh, and yes, it is not a company doing this so expecting the project to be run like Motorola would have done it is a bit silly.

Obviously you have no idea how much work has went into it but all you can do is demand it should have been more like documentation, compiler support, MMU blabla. Must be really nice to sit in your comfy chair by the fireplace and explain to the world how things should be done...

I agree a 100%

I have never been in any online community that has more idiots\trolls  than the Amiga community.
Question is why would anyone develop anything for such a useless group of nagging 5 year olds ? I have never seen any new product that has not been slaughtered verbally by someone for lacking this or that here.

Funny thing is that it is always like this . End-user demand allot and never contributes with anything.  

There are allot of "smart" people here and if they don't like something i think its about high time they start producing something better them self. Rather than constantly talking %&$#?@!%&$#?@!%&$#?@!%&$#?@! about things they are not part of developing .  

Attacking the vampire core is idiotic as it is not a done deal yet , same goes for lack of SDK or whatever, it may eventually be done . I find the vampire core interesting and its constant development means nothing but good news for the end user. If it does not have what you need now why buy it ?

[Kremlar]

I agree a 100%

I have never been in any online community that has more idiots\trolls  than the Amiga community.
Question is why would anyone develop anything for such a useless group of nagging 5 year olds ? I have never seen any new product that has not been slaughtered verbally by someone for lacking this or that here.

Funny thing is that it is always like this . End-user demand allot and never contributes with anything.  

There are allot of "smart" people here and if they don't like something i think its about high time they start producing something better them self. Rather than constantly talking %&$#?@!%&$#?@!%&$#?@!%&$#?@! about things they are not part of developing .  

Attacking the vampire core is idiotic as it is not a done deal yet , same goes for lack of SDK or whatever, it may eventually be done . I find the vampire core interesting and its constant development means nothing but good news for the end user. If it does not have what you need now why buy it ?

Exactly!

[ppcamiga1]

There are no attacks on the vampire.
Just some unsatisfied customers express their opinion about this crap.
Instead of announced second coming of Jay Miner, there is cheaper 68060, with faster RAM, but slower FPU, no MMU and 3D.