Amiga.org
Amiga computer related discussion => Amiga Hardware Issues and discussion => Topic started by: Honkybear on March 17, 2017, 04:53:33 PM
-
Please excuse my ignorance.
But is the Vampire 2 Card Capable or Could it be made Capable
of running OS 4 or Morphos.
Thanks
-
no.
-
Please excuse my ignorance.
But is the Vampire 2 Card Capable or Could it be made Capable
of running OS 4 or Morphos.
Thanks
No, even when present in legacy systems, neither MorphOS or OS4 use the 68K cpu, running solely on the PPC cpu.
So installing both a Vampire2 and a PPC is kind of pointless, unless you are dual booting.
-
No/Yes.
No to your first question, yes to the second. But as of now the developers are not interested in making that happen. I think it would be great, others think it would be horrible:)
-
No/Yes.
No to your first question, yes to the second. But as of now the developers are not interested in making that happen. I think it would be great, others think it would be horrible:)
I have an amigaone x5000 and for me will be wonderful have a ppc inside my lovly amiga 500 :P why not a vampire ppc :P i think performance will be not bad with this new fpga
-
i think performance will be not bad with this new fpga
Does the FPGA have the capacity to emulate a PPC?
-
No/Yes.
No to your first question, yes to the second. But as of now the developers are not interested in making that happen. I think it would be great, others think it would be horrible:)
I don't think Gunnar is too keen on the idea either.
But adding a cheap PPC, or a second device/card with a PPC would allow the use of NG OS'.
What would be nice is if someone devised a way to run both cpus simultaneously.
-
Does the FPGA have the capacity to emulate a PPC?
And this was pretty much my point. Was not so much does it but if it was capable or have the potential.
-
Performance of a PPC core on a low end FPGA would be poor and a waste of resources when you can easily buy a more powerful PPC instead of spending money on an FPGA. Also the Apollo team have no interest in PPC or any RISC architecture. RISC CPUs were conceived to simplify the development of new CPUs with higher clock rates compensating for the relative lack of performance compared to similarly clocked CISC processors. At that time the tools for design CPUs were very limited and the costs were very high. So Motorola thought it made sense to join a partnership to develop a RISC architecture to go up against x86 instead of trying to push there 680x0 chips to higher complexity and speeds. RISC lost the desktop computer war and PPC missed out on the "CPU/MCU/SoC of things" to ARM. The Apollo team like the 680x0 architecture and are fully focused on developing the Apollo core and accelerators/sytems based on that core so there is no chance of them ever doing a PPC core "Vampire".
-
i new it. there is a simple one word answer to the question but people will nevetheless fill this thread with endless phantasizing about a possible ppc in fpga, no matter that it would not help anything, because mos and os4 would have to be explicitely ported to that combination of hardware, which is another "no" everybody has forgotten to ask about.
-
And this was pretty much my point. Was not so much does it but if it was capable or have the potential.
what for? ppc cpus are still available.
-
I don't think Gunnar is too keen on the idea either.
But adding a cheap PPC, or a second device/card with a PPC would allow the use of NG OS'.
What would be nice is if someone devised a way to run both cpus simultaneously.
http://powerup.amigaworld.de/download.php?id=44 ;)
-
i new it. there is a simple one word answer to the question but people will nevetheless fill this thread with endless phantasizing about a possible ppc in fpga, no matter that it would not help anything, because mos and os4 would have to be explicitely ported to that combination of hardware, which is another "no" everybody has forgotten to ask about.
If put you a whole PowerUP card on an FPGA then OS4.x and older versions of MorphOS would run on it. Still agree that it would be a waste of time and resources.
-
If put you a whole PowerUP card on an FPGA then OS4.x and older versions of MorphOS would run on it.
im not so sure. there is apparently so much trouble to properly configure winuae to even run os4 that one needs either multiple pages of manual or a special application to get it right. who knows all the quirks and how such "fake" hardware would behave. what concerns morphos i have never got it to run on my apparently perfectly supported setup.
and now, what the point of this whole discussion if a sloution even if possible would only prove one point, namely that it was too complecated and unnecessary.
what strikes me about a number of posters on amiga-like sites, is that they have some fraudian problem. they dont accept what is availeble for them. they demand the impossible. as example they want amiga os on x86, but they dont want aros, which is what it could effectivly be, best case. or they dont want to run os4 on dedicated hardware, they want to run it on something completely inappropriate.
-
If put you a whole PowerUP card on an FPGA then OS4.x and older versions of MorphOS would run on it. Still agree that it would be a waste of time and resources.
.. or grab an old G4 Mac from the garbage retail shop aka ebay for less than peanuts and install a current version of MorphOS on it. Maybe not as freaky as on a 80ies computer pimped with an FPGA, but way cheaper, straight forward and more powerful.
I really like Vampire and am considering to equip my A600 with one, but definitely not for MorphOS when there are actually fast machines available (which cost quite less than the very nice and fair priced Vampires).
-
Actually, the RISC vs CISC argument is a bit more complicated than has been mentioned.
Each approach has its advantages, for instance, CISC can create denser code with a larger number of available instructions.
But frankly, I don't favor either approach right now, as long as it works, who cares how?
And some cpus are a hybrid of both concepts.
For some time now, AMD cpus have run X86/X64 instructions with a RISC core.
An Amiga with both an FPGA based 68K accelerator and a PPC processor would be ideal as it would allow us to run both legacy and NG code.
-
.. or grab an old G4 Mac from the garbage retail shop aka ebay for less than peanuts and install a current version of MorphOS on it. Maybe not as freaky as on a 80ies computer pimped with an FPGA, but way cheaper, straight forward and more powerful.
I really like Vampire and am considering to equip my A600 with one, but definitely not for MorphOS when there are actually fast machines available (which cost quite less than the very nice and fair priced Vampires).
Mmm, well that is a sound point until you get to the X5000. That is going to be costly whether it runs MorphOS or OS4.
But I don't see anything wrong with supporting both NG hardware and legacy hardware with FPGA accelerators.
The performance boost for legacy hardware will help make it relevant again.
-
An Amiga with both an FPGA based 68K accelerator and a PPC processor would be ideal as it would allow us to run both legacy and NG code.
then how about an amiga with 68k to run amiga software, ppc to run your linux ports you call ng software and x64 to run aros or whatever system your productivity demands with today appropriate resources? sigh..
-
But I don't see anything wrong with supporting both NG hardware and legacy hardware with FPGA accelerators.
why always this duty to support something? simply choose what you want to use.
-
why always this duty to support something? simply choose what you want to use.
Why choose? I have multiple systems.
And I am already running some 68K software on the PPC platforms.
As to Linux ports, I neither use nor support them.
The entire Libre Office idea seems foolish to me.
I DO support the idea of further development of Final Writer though.
And why move to more advanced hardware?
I've heard this argument before with RTG cards.
Because it works better.
If Amiga development had continued, we wouldn't question the idea.
-
RISC CPUs were conceived to simplify the development of new CPUs with higher clock rates compensating for the relative lack of performance compared to similarly clocked CISC processors.
The idea was to simplify the design as much as possible to
1. make room on the chip for larger register and cache storage
2. get rid of microcode
3. introduce pipe-lining
By the 90's moore's law allowed CISC to use some of the RISC techniques & RISC became more complex. It would be great if you could spin up a 200mhz 603e in an FPGA, but 1. it's probably going to need a better FPGA, 2 its going to need someone who cares about PPC enough to dedicate that much of their life to doing it.
-
PPC 440 series CPUs like the one in the SAM 440 were developed in FPGAs originally. They were glacially slower than the Vampire 2 is when running the 68080. Consequently, if the 68080 were remade as an ASIC (regular masss-produced fast chip) it would probably be faster than a SAM 440 of the same clock speed. It might even be fast enough to run a PPC emulator to run the newer PPC stuff in much the same way as the Intels run OS 4 on WinUAE.
-
PPC 440 series CPUs like the one in the SAM 440 were developed in FPGAs originally. They were glacially slower than the Vampire 2 is when running the 68080. Consequently, if the 68080 were remade as an ASIC (regular masss-produced fast chip) it would probably be faster than a SAM 440 of the same clock speed. It might even be fast enough to run a PPC emulator to run the newer PPC stuff in much the same way as the Intels run OS 4 on WinUAE.
Thanks for the information. My main reason for asking the question was that these days. It's near impossible to get hold of a PPC Card at a reasonable price and I thought could it be done with the Vampire 2 Card and therefore allow Amiga users to run OS4 and Morphos on an Amiga. I know there are several schools of thought on both of these Operating Systems. I thought it would be a good question to put to the Amiga community. Thanks Again