PearPC

[macto]

If it isn't dead and isn't recovering, it just means that it is very ill.  Or perphaps the illness analogy is a bad one.  I'm simply saying that things aren't going to get much better.

[TheMagicM]

Piracy wont finish off the Amiga or kill the Pegasos..lack of software will.

[StevenJGore]

In terms of the sales and success of OS4 and future Amiga software, does it really matter whether it's running on an Intel/AMD box under emulation or running on dedicated AmigaOne or Pegasos motherboards?

You just need to look at how WinUAE has renewed and is sustaining an interest in the classic Amiga to see that emulation isn't a bad thing. Many people like myself will have purchased OS3.9 to run purely under WinUAE.

If PearPC allows me to enjoy the new AmigaOS and hopefully the software titles which will follow, without having to pay for extortionately priced hardware, then it's fine by me! :-)

Piracy wont finish off the Amiga or kill the Pegasos..lack of software will.

Exactly. And whether that software runs on purpose-built hardware or under emulation, is irrelevant. Sales will be unaffected.

Just my own opinions! :-)

Steve.

[rayt]

Exactly. And whether that software runs on purpose-built hardware or under emulation, is irrelevant. Sales will be unaffected.

The problem is that you won't be able to buy AmigaOs4 or Morphos without the Hardware, so there's no legal way of running these in emulation.

In my opinion, there's no reason for a ppc amiga emulation. If somebody wants to run classic amiga programs he can do this with uae and its 1000 times faster than this ppc emu. And you don't need a ppc amiga emu to play quake1+2 or wipeout on a x86 box.

[Matt_H]

You just need to look at how WinUAE has renewed and is sustaining an interest in the classic Amiga to see that emulation isn't a bad thing. Many people like myself will have purchased OS3.9 to run purely under WinUAE.

Unlike the classic Amiga hardware, however, The A1 and Pegasos are in production - sales matter. By the time Amiga Forever, which is licensed, hit the scene, stocks of Classic Amigas were long gone.

The problem is that you won't be able to buy AmigaOs4 or Morphos without the Hardware, so there's no legal way of running these in emulation.

Exactly!

[itix]

There is no fear of piracy and OS4/MOS never sell A1/Peg enough alone. And if it is more practical to run PPC OS in x86 box using PPC emulator rather than using real PPC, hell, why not?

[StevenJGore]

The problem is that you won't be able to buy AmigaOs4 or Morphos without the Hardware

Where's the logic in that decision? To limit sales? I can hear the nails going in the lid of OS4's coffin already.

so there's no legal way of running these in emulation.

Well... if there isn't a legal way...

rayt wrote:

Exactly. And whether that software runs on purpose-built hardware or under emulation, is irrelevant. Sales will be unaffected.

The problem is that you won't be able to buy AmigaOs4 or Morphos without the Hardware, so there's no legal way of running these in emulation.

In my opinion, there's no reason for a ppc amiga emulation. If somebody wants to run classic amiga programs he can do this with uae and its 1000 times faster than this ppc emu. And you don't need a ppc amiga emu to play quake1+2 or wipeout on a x86 box.

Apparently we will be able to buy OS4 for CSPPC/BPPC so if the UAE coders merged the PearPPC source into UAE then in theory, it would be possible to legally buy and emulate OS4 on x86.

JIT Emulation of PPC on a 3000Mhz x86 CPU would be approximately equivalent to a 1200 MHz emulated G3 CPU using the PearPC code, as it's supposed to be 40% slower than the host CPU.  This is faster than any available classic Amiga PPC accelerater.  I would definately buy OS4 if it did run under UAE on x86. Until then i'll stick with my Pegasos.

[Ilwrath]

JIT Emulation of PPC on a 3000Mhz x86 CPU would be approximately equivalent to a 1200 MHz emulated G3 CPU using the PearPC code, as it's supposed to be 40% slower than the host CPU.

WHOA!!!! Where did you get that figure??  The real author says it's about 500x slower on a cycle-for-cycle basis.  That seems to me to be quite an exaggeration.  

On my 3ghz P4, I got a somewhat usable Mac OSX emulation, yet quite slow (at least compared to THAT figure!).  Probably about what I'd expect from an early (unsupported) PowerMac.  Maybe a 75-100mhz G3 emulation.  (It took about 2.5 hours to install Mac OSX 10.3 Panther!) Still, this would put the emulation at a level with some of the slower Cyberstorm PPC and Blizzard PPC cards....  and probably have better video support.  The video and IO of PearPC seem quite fast.  It's the processor emulation core that's the bottleneck...  But still, I'm impressed for a 0.1 beta.  :-)

I'm hoping that there's some people who want to work this into UAE!  That would be awesome.  :-)  I've never actually owned a real PPC Amiga...

Ilwrath wrote:

JIT Emulation of PPC on a 3000Mhz x86 CPU would be approximately equivalent to a 1200 MHz emulated G3 CPU using the PearPC code, as it's supposed to be 40% slower than the host CPU.

WHOA!!!! Where did you get that figure??  The real author says it's about 500x slower on a cycle-for-cycle basis.  That seems to me to be quite an exaggeration.  

On my 3ghz P4, I got a somewhat usable Mac OSX emulation, yet quite slow (at least compared to THAT figure!).  Probably about what I'd expect from an early (unsupported) PowerMac.  Maybe a 75-100mhz G3 emulation.  (It took about 2.5 hours to install Mac OSX 10.3 Panther!) Still, this would put the emulation at a level with some of the slower Cyberstorm PPC and Blizzard PPC cards....  and probably have better video support.  The video and IO of PearPC seem quite fast.  It's the processor emulation core that's the bottleneck...  But still, I'm impressed for a 0.1 beta.  :-)

I'm hoping that there's some people who want to work this into UAE!  That would be awesome.  :-)  I've never actually owned a real PPC Amiga...

It says in the docs that it's about 40% slower. 40% of 3000 is 1200.

Even at 75-100 Mhz it would still be usuable for Warp/PowerUP stuff I would imagine.

Maybe if it was compiled for AMD64 native, it would be a lot faster?

[drHirudo]

JIT Emulation of PPC on a 3000Mhz x86 CPU would be approximately equivalent to a 1200 MHz emulated G3 CPU using the PearPC code, as it's supposed to be 40% slower than the host CPU.

This can be achieved when you don't need to emulate the CPU i.e. emulator like ShapeShifter for PPC. 1 GHz Amiga will emulate 600MHz Mac, which is quite acceptable me thinks.

[Ilwrath]

[...]using the PearPC code, as it's supposed to be 40% slower than the host CPU.

This can be achieved when you don't need to emulate the CPU [...]

This makes sense...  Like I said, the video and IO portion seems to work fairly quickly.  So, an unsupported PowerPC platform can run OSX at 40% slower than a real Mac of the same clock runs it.  That's good to know.  So, an A1 or Pegasos could use this and get quite a usable Mac emulation.  (Usable is a word I've never heard MoL described as.  I can't pass judgement on it myself, though, as I don't have a fast enough PPC to try it.)

But, back to PearPC....  If you use the processor emulation, you certainly get no where near losing only 40%.  In fact, I don't think a PPC emulation is even possible to that tight of performance...  I mean, think about it.  That's a lot tighter than any 680x0 emulation routines, and how much simpler of a processor is the 680x0 compared to a G3....

Anyhow, I had only read the x86 docs on PearPC, which was why I questioned the origins of that 40% figure.  ;-)

[bloodline]

PearPC's long term speed problems will probably lie with the PPC MMU and the prophensity for PPC programs to use Big Endien Data.

The MMU problem wil be solved in time, but learning how PPC software tens to use the MMU, the Byte ordering problem won't be solved until the x86 grows a big endien mode/instructions.

All in all I expect this Emulator to get quite a bit faster, I see no reason to think that 70% of native speed is an unobtainable figure :-D

bloodline wrote:
PearPC's long term speed problems will probably lie with the PPC MMU and the prophensity for PPC programs to use Big Endien Data.

The MMU problem wil be solved in time, but learning how PPC software tens to use the MMU, the Byte ordering problem won't be solved until the x86 grows a big endien mode/instructions.

All in all I expect this Emulator to get quite a bit faster, I see no reason to think that 70% of native speed is an unobtainable figure :-D

And when it does hit 70%, please someone bang the code into AROS! :-)

[mdwh2]

Matt_H wrote:
I think this is terrible news for the Amiga community. The Amiga community is so small that the AmigaOne and Pegasos need every sale they can get; every sale matters. By eliminating the need for such hardware to run AmigaOS or MorphOS, it simply opens up the market to piracy from outside the community - hardware sales decrease, driving up prices, and CD images of OS4 and MorphOS find their way to P2P clients.

And the Pegasos is bad because it stops people buying AmigaOnes, and the AmigaOne is bad because it stops people buying A4000s and other Amiga hardware. Why is someone using a PC to run AmigaOS "outside the community", and someone using a Pegasos or AmigaOne "inside the community"? All you're doing is arbitrarily dividing up the market so that your preferred solution counts as being a "True Amigan", and everyone else are evil pirates who don't add anything to the Amiga market or community.

Let's look at the issues. Whilst it is true that it is useful if Hyperion can make money from hardware (as Apple do), firstly it's likely that emulation will be far below native performance. Anyone serious about spending hundreds of pounds on a G4 Amiga isn't going to settle for much slower performance. Do you think this emulator will be terrible news for the Mac community? On the other hand, people like me who have an interest, but no desire to buy new hardware, will buy an extra copy of OS4.

One thing I have wondered - will Hyperion or KMOS or whoever receive money from the AmigaOne profits? If not, then unlike with Apple, the hardware argument is a non-issue. I don't care for Eyetech anymore than I'd listen to the argument that we should worry about sales of classic Amiga hardware. My only worry is that Hyperion are limiting themselves to running on a single motherboard, so there are obvious problems if Eyetech run into trouble.

You've also got to take into account the knock on effect of having extra Amiga users in the market - that's a bigger market for software developers, and also possible more developers who are using the emulators.

And oh look, it's the "only PC users are pirates" tired old myth again.

Soon enough, there could be people looking for AmIGa ROmZ - not a good thing.

Aside from being annoying on some emulation forums, why is this "terrible news"?

I'm not saying whether the sorts of people who download old A500 games are right or wrong to do so, but this has nothing to do with the Amiga today.

[mdwh2]

rayt wrote:
The problem is that you won't be able to buy AmigaOs4 or Morphos without the Hardware, so there's no legal way of running these in emulation.

What happened to the plans for users with PPC A1200/4000s? Have these been dropped?

In my opinion, there's no reason for a ppc amiga emulation. If somebody wants to run classic amiga programs he can do this with uae and its 1000 times faster than this ppc emu.

There's not a great need right now, but AmigaOS 4 and MorphOS aren't available on 68k, so if these end up having any future, then emulation would be worthwhile.