Welcome, Guest. Please login or register.
Amiga Kit Amiga Store Iridium Banner AMIStore App Store A600 Memory

AuthorTopic: A2080 i.e. Vampire 500 V2 on an Amiga 2000  (Read 672 times)

0 Members and 1 Guest are viewing this topic.

Offline cunnpole

Re: A2080 i.e. Vampire 500 V2 on an Amiga 2000
« Reply #45 on: August 11, 2016, 01:33:06 PM »
Gunnar seems to have already done the comparison at IBM: http://www.apollo-core.com/index.htm?page=performance

"IBM compared three FPGA cores: The POWERPC 440, NIOS and APOLLO. The study was conducted in simulation and on a PLDA FPGA PCIe card using a ALTERA Stratix4 230C2."

I think those are the same results he had 2 years ago before he started adding all the new performance enhancements to the apollo core
 

Offline AJCopland

Re: A2080 i.e. Vampire 500 V2 on an Amiga 2000
« Reply #46 on: August 11, 2016, 02:08:41 PM »
Why bother with an FPGA based PPC though when these things are ~£31 individually? http://www.digikey.co.uk/product-detail/en/freescale-semiconductor-nxp/MPC5200CVR400B/MPC5200CVR400B-ND/1168085

That's what powered the Efika and is a 603e based chip, why go to the hassle of designing a PPC based FPGA system?

It made sense for the 68060 as it's incredbily expensive and the end of the road for the 68k series compatibility/availability wise. Those arguments don't hold true for PPC though.

Offline kolla

Re: A2080 i.e. Vampire 500 V2 on an Amiga 2000
« Reply #47 on: August 11, 2016, 02:47:21 PM »
Many years ago I had a router developer board with a PPC440 and FPGA on the same board, and I think maybe it even was in the same chip.
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC
 

Offline biggun

Re: A2080 i.e. Vampire 500 V2 on an Amiga 2000
« Reply #48 on: August 11, 2016, 04:12:15 PM »
Quote from: Thomas Richter;812404
whereas in PPC, it would be something approximately like
Code: [Select]
lwz r1,4(r2) add r1,r3,r1 st r1,4(r2) which are three instructions .


Thomas is 100% correct.

A RISC CPU needs significant more instruction for the same amount of work in comparison to a CISC CPU.

Offline Bennymee

Re: A2080 i.e. Vampire 500 V2 on an Amiga 2000
« Reply #49 on: August 11, 2016, 04:22:58 PM »
Quote from: AJCopland;812413
Why bother with an FPGA based PPC though when these things are ~£31 individually? http://www.digikey.co.uk/product-detail/en/freescale-semiconductor-nxp/MPC5200CVR400B/MPC5200CVR400B-ND/1168085

That's what powered the Efika and is a 603e based chip, why go to the hassle of designing a PPC based FPGA system?

It made sense for the 68060 as it's incredbily expensive and the end of the road for the 68k series compatibility/availability wise. Those arguments don't hold true for PPC though.


Indeed, lot faster then any fpga now and PPC's cpu's are for sale. Why bother making a FPGA implementation which needs several years of debugging.
Amiga 500, 1200, 4000, Amigaone, Morphos, CyberstormPPC, Blizzardppc, OS4.x
 

Offline Pentad

Re: A2080 i.e. Vampire 500 V2 on an Amiga 2000
« Reply #50 on: August 11, 2016, 04:49:56 PM »
Quote from: psxphill;812406
I'm struggling to take you seriously as you keep calling it "risk" when it is RISC.

Motorola got into the PowerPC project because IBM had persuaded Apple to switch from 680x0 to a new chip based on their POWER architecture. Apple invited Motorola to join. Apple knew that Motorola had more experience in making single chip microprocessors, but also having two sources gave them more bargaining power.


Actually, that is not true. Andy Hertzfeld gave a talk about the Mac, PowerPC, and AIM which was a fascinating look into what he fighting the rise of Wintel.  I don't have time to go into much detail (you can see his talks online anyway).

By the late 80s, everyone knew CISC was a dead end.  Apple had developed a secret quad-core RISC chip that was very powerful (for the time) but would make new Macs incompatible with the 68k software base. Sculley was not enthusiastic about this at all (I think because it was a leftover from Jobs and the mac market was pretty soft for this kind of shock by 1988).

Anyway, IBM, Motorola, and Apple watching the rise of Wintel felt this could benefit them all to some degree. IBM had POWER but no real traction, Apple riding a dead end platform, Motorola looking to help fight off Intel. The AIM alliance was a good one.

Apple's move to PowerPC was the right decision but they woefully underestimated how much of a hole the MacOS had put them in. They were never really able to move away from underlying 68k code which hurt the PowerPC performance on the Mac.  Look how long (and messy) their nano kernel was.

Running PPC BeOS on a Mac was quite an eye opening experience and really showed how much the MacOS was crippling PowerPC performance.

You know the end, WinTel would be the winner.

If you don't know, Carl Sassenrath (wrote Amiga Exec) worked on Apple's CPU along with some other amazing people.  

I apologize, I left out a lot of details but look it up.  It is a great story of struggle against WinTel that ultimately failed to be the huge success they were all hoping for.

-P
2015 15" Macbook Pro Retina * 2.8 GHz QCore * 16 GB RAM, 1TB SSD * Windows 10 via Boot Camp * Amiga via Emulation (WinUAE in WINE Staging)
 

Offline Heiroglyph

Re: A2080 i.e. Vampire 500 V2 on an Amiga 2000
« Reply #51 on: August 11, 2016, 05:11:45 PM »
I believe very strongly that making a fast 68k compatible CPU is the right way to go for the majority of Amigans whether it is FPGA or emulation.

The majority of us have never had the opportunity to use a PPC due to cost, availability or simply a lack of interest. Most PPC software could be recompiled for 68k since assembly isn't as common.

I'll never understand why the copyright holders have abandoned the majority of their user base, the 68k users, but perhaps Apollo will give them a reason to notice us again rather than pushing platforms most of us don't own.

If not, at least we'll finally have the potential for faster 68k CPUs as FPGA technology advances and the core improves.
 

Offline psxphill

Re: A2080 i.e. Vampire 500 V2 on an Amiga 2000
« Reply #52 on: August 11, 2016, 05:42:21 PM »
Quote from: Thomas Richter;812408
Sure - as said, RISC ("risk") was a fashion statement back then, and people believed that it would be a beneficial architecture. In some sense, this is true, but I would believe that history tells now something different.

It was a beneficial architecture when it was introduced, not a fashion statement at all. What happened over time however is that the percentage of a chip that is risc vs cisc became so small that for computers the benefit is towards the chip that has market traction, which was x86 and is now x64. The phone market has proved that when there are other factors in play (like power usage and licensing the cpu to use in a SOC) that a RISC cpu like Arm can still sell in large numbers.

The Pentium Pro was Intel's first chip where the code was translated into another form, which is effectively executed by a RISC cpu. The translated code is cached, so loops are fast etc.

Quote from: Thomas Richter;812408
Again, I consider this somewhat pointless given the rather small software library for PPC on the Amiga - or possibly - the typical "applications" Amiga has found today. That's of course a completely different argument.

It's in my eyes mainly a retro system - if you want to make it fast by a modern CPU, one would pick an intel design and RTG graphics and not PPC and custom chip graphics. Wait, that's called a PC, right? (-:

I'm not talking about making it as fast as a modern CPU. I'm talking about making it fast enough to run late 90's Amiga PPC software. That sounds pretty retro to me. It wouldn't hurt if it could go quicker of course, but compatibility and price are the most important aspects. Arguing to use RTG graphics instead of custom chip graphics seems a little odd, on a thread about vampire which has it's own custom graphics.

Quote from: Pentad;812419
Apple had developed a secret quad-core RISC chip that was very powerful (for the time) but would make new Macs incompatible with the 68k software base. Sculley was not enthusiastic about this at all (I think because it was a leftover from Jobs and the mac market was pretty soft for this kind of shock by 1988).

Project Aquarius was going nowhere.

http://lowendmac.com/2006/growing-apple-with-the-macintosh-the-sculley-years/

Sculley loved it, he hated that it was going nowhere and the chip designer they had hired thought it was impossible.

Quote from: Pentad;812419
Anyway, IBM, Motorola, and Apple watching the rise of Wintel felt this could benefit them all to some degree. IBM had POWER but no real traction, Apple riding a dead end platform, Motorola looking to help fight off Intel. The AIM alliance was a good one.

Motorola wasn't there to help fight off Intel, they were there to make money. Apple had decided to ditch the 680x0 cpu's, but aquarius and all the other internal projects had failed. Motorola's own RISC CPU (the 88000) was a disaster, so after IBM contacted Apple and got them excited about POWER then joining up with IBM was Motorola's only chance to hold onto some of the pie.

Apple even had an OS running on x86 before they had it running on PPC, but when that project collapsed they quickly hacked something together to run on PPC and then bought NextStep. They knew that MacOS7/8/9 were a problem, but all their efforts to move away had failed. OSX has always run on x86, even though they never sold it until they started selling x86 hardware. I suspect they regretted choosing PPC for a long time, it was only the Pentium 4 failing that gave them any cause for celebration. When Intel realised they had to do something serious and went back to the Pentium 3 design and improved it to make Intel Core, then there really was no stopping them. The PPC didn't recover and even the PS3/Xbox360 cpu cores aren't that good.

https://en.wikipedia.org/wiki/Copland_(operating_system)

But we seem to be going off topic, PPC wasn't a great choice. But it was a choice that Phase 5 made, so it would be nice to be able to run PPC software as well as taking advantage of the new software for vampire.
« Last Edit: August 11, 2016, 06:50:10 PM by psxphill »
 

Offline Thomas Richter

Re: A2080 i.e. Vampire 500 V2 on an Amiga 2000
« Reply #53 on: August 11, 2016, 07:40:07 PM »
Quote from: psxphill;812425
It was a beneficial architecture when it was introduced.
So it seemed, at least. There is nothing completely wrong with the observation that a simpler CPU could run faster, but what Mot and IBM underestimated is that bandwidth did not grew proportional to the clock speed of the CPU, and that compiler performance and optimization did not improve as much as they hoped for. Intel's IA64 is a failure as well, for related reasons.  Nowadays, the CPU - due to increased complexity - can perform the "compilation" of "abstract code" (say, x86 assembly) to its "raw form" much better (due to code and execution statistics) than a static compiler analysis can, and shorter instructions also help to cut down bandwidth requirements.  All, of course, if you have the power to drive this complex machinery.  
Quote from: psxphill;812425
not a fashion statement at all.
Oh, 'mon. Back in those days, it was "RISC" here, "RISC" there, all around. If you believe that engineering does not have fashion movements, you've probably not yet observed one. Currently, it's the "IoT" business and "Cloud everything" all around. Back then, "RISC" was the thing to do.  
Quote from: psxphill;812425
What happened over time however is that the percentage of a chip that is risc vs cisc became so small that for computers the benefit is towards the chip that has market traction, which was x86 and is now x64.  
Which is, deep inside, also a RISC design, but (wisely) with a backwards compatible high-level just-in-time CISC compiler. (-: Yes, that means added complexity.  
Quote from: psxphill;812425
The phone market has proved that when there are other factors in play (like power usage and licensing the cpu to use in a SOC) that a RISC cpu like Arm can still sell in large numbers.
Oh, sure. But there is also a reason why ARM has thumb code, you know? (-;  
Quote from: psxphill;812425
The Pentium Pro was Intel's first chip where the code was translated into another form, which is effectively executed by a RISC cpu. The translated code is cached, so loops are fast etc.

Clearly. But it avoids the problem of the PPC RISC design of overly long instructions and low code density, but uses a "abbreviated high-level" assembler syntax.  
Quote from: psxphill;812425
I'm not talking about making it as fast as a modern CPU. I'm talking about making it fast enough to run late 90's Amiga PPC software. That sounds pretty retro to me.
But, unlike the 68K, you *can* buy fast PPC chips. So, again, what's the point?  Or, what's the point with PPC on Amiga anyhow? As said, the software library is not exactly "huge".  
Quote from: psxphill;812425
Arguing to use RTG graphics instead of custom chip graphics seems a little odd, on a thread about vampire which has it's own custom graphics.
Who argues against RTG graphics?  
Quote from: psxphill;812425
Motorola wasn't there to help fight off Intel, they were there to make money. Apple had decided to ditch the 680x0 cpu's, but aquarius and all the other internal projects had failed. Motorola's own RISC CPU (the 88000) was a disaster, so after IBM contacted Apple and got them excited about POWER then joining up with IBM was Motorola's only chance to hold onto some of the pie.  
At the time back then, the choice made of course sense. I certainly don't argue against it. CISC seemed to run into a dead end, RISC was a fashionable new toy that made huge promises - and all the arguments were also quite reasonable, no question.  The problem is just that the development did not quite work out as expected. Memory speed fell behind raw CPU power, bandwidth and compatibility with legacy applications became more important than simplicity of the CPU design, so RISC did not met its expectations.  
Quote from: psxphill;812425
I suspect they regretted choosing PPC for a long time, it was only the Pentium 4 failing that gave them any cause for celebration.
Not only Apple. Also AMD. The Pentium 4 was really a big disaster, more designed by the needs of the marketing department than by smart engineering, driven by the need to sell CPUs by the GHz number on them.  This broke down when execution speed hit the brick wall at 4GHz, a speed at which a signal takes several clock cycles to run from one edge of a chip to another... Actually, it was not *that* unexpected as the physical limits were known. I'm not clear which miracle intel was actually hoping for.  
Quote from: psxphill;812425
When Intel realised they had to do something serious and went back to the Pentium 3 design and improved it to make Intel Core, then there really was no stopping them. The PPC didn't recover and even the PS3/Xbox360 cpu cores aren't that good.
The market was too small to allow improvement, Apple had to pay $$$ for them, and finally, the CPU design also showed its limitations, see above. If bandwidth is limited, many long and simple instructions are not the best choice. ARM targets a completely different market, where raw performance is not important, but performance per Watt is. *There*, a simpler design helps that can be upscaled to the requirements of the design. That's something ARM is really good at - customizing CPU cores for specific needs.  
Quote from: psxphill;812425
But we seem to be going off topic, PPC wasn't a great choice. But it was a choice that Phase 5 made, so it would be nice to be able to run PPC software as well as taking advantage of the new software for vampire.

No, PPC wasn't a great choice, indeed. X86 would have been a much better choice, but a choice that wouldn't have been accepted by users that are driven more by ideology than technology. The x86 chips are probably an unorthogonal mess, but they are still high performing, powerful chips.
 

Offline Rob

Re: A2080 i.e. Vampire 500 V2 on an Amiga 2000
« Reply #54 on: August 11, 2016, 08:57:42 PM »
Quote from: wawrzon;812402
ppc was an attempted (and failed) solution for lack of further development on 68k front.


I hardly call PPC a failed attempt.  It was held its own against Intel hardware for quite some time and it was used in mainstream computing hardware up until 2006.
 

Offline psxphill

Re: A2080 i.e. Vampire 500 V2 on an Amiga 2000
« Reply #55 on: August 11, 2016, 09:13:04 PM »
Quote from: Thomas Richter;812432
Oh, 'mon. Back in those days, it was "RISC" here, "RISC" there, all around.


Berkley RISC-I CPU outperformed every other single chip microprocessor in 1982. People talk about things that are good.
 

Offline biggun

Re: A2080 i.e. Vampire 500 V2 on an Amiga 2000
« Reply #56 on: August 11, 2016, 09:24:27 PM »
Quote from: psxphill;812438
Berkley RISC-I CPU outperformed every other single chip microprocessor in 1982. People talk about things that are good.


When a RISC CPU needs in avg 2 instruction to do the same work as the CISC CPU.
And the RISC CPU can execute 1 instruction per clock -
and the CISC CPU (like 68000) needs in average 8 clocks per instruction.

Then the RISC CPU is faster.. in avg by factor x4
This is your story of the 80th.

Now if the CISC CPU upgrades and does 4 instructions in a single cycle then the CISC CPU does the amount of work of 8 RISC instructions per cycle.
This means now the 68080 CISC CPU is many times faster then the RISC.

Very easy to understand.

Offline wawrzon

Re: A2080 i.e. Vampire 500 V2 on an Amiga 2000
« Reply #57 on: August 11, 2016, 09:25:40 PM »
Quote from: Heiroglyph;812423
I'll never understand why the copyright holders have abandoned the majority of their user base, the 68k users, but perhaps Apollo will give them a reason to notice us again rather than pushing platforms most of us don't own.


lets pray not, we have been prey to those long enough..
 

Offline wawrzon

Re: A2080 i.e. Vampire 500 V2 on an Amiga 2000
« Reply #58 on: August 11, 2016, 09:36:48 PM »
Quote from: Rob;812437
I hardly call PPC a failed attempt.  It was held its own against Intel hardware for quite some time and it was used in mainstream computing hardware up until 2006.


with apple, perhaps. for the (past) time being, but with amiga it has been a crook and a hack all along, first as a powerup/warpos nonsense, and then as if this hasnt failed enough, with so called ng ppc systems. among those morphos have proven more sensible and competent, with choosing and reaching their goals, and especially with stepping away from trying to design and establish (lol) their own ppc hardware platform. heck, it was their devs, who started it once. nevertheless even them, they have left most of the interested audience behind..
 

Offline psxphill

Re: A2080 i.e. Vampire 500 V2 on an Amiga 2000
« Reply #59 on: August 12, 2016, 06:38:02 AM »
Quote from: biggun;812439
Now if the CISC CPU upgrades and does 4 instructions in a single cycle then the CISC CPU does the amount of work of 8 RISC instructions per cycle.
This means now the 68080 CISC CPU is many times faster then the RISC.

Very easy to understand.

It is easy to understand. However the 68060 didn't do 4 instructions in a single cycle. So it sounds like you are comparing optimised vampire to an existing PPC chip, which wasn't relevant to the decision made to switch from 680x0 to PPC in the 90's.

It's also not relevant now, unless you can software emulate the PPC on vampire quicker than a phase 5 PPC board.

I'm interested in how maintainable that 4 instructions in a single cycle is. I assume that doesn't involve touching ram in any way, i.e. both code and data all fits in caches. Also how complex are the instructions?

Quote from: Thomas Richter;812432
X86 would have been a much better choice, but a choice that wouldn't have been accepted by users that are driven more by ideology than technology. The x86 chips are probably an unorthogonal mess, but they are still high performing, powerful chips.

I don't think enough Mac owners would have cared what CPU it had, as it would still have been marketed as better than a PC. Apple didn't even wait for x64 to jump to Intel and they have ditched support of x86 now. If Microsoft dropped support for x86 then there would be mass outrage, but it's socially accepted that it's ok to hate Microsoft. If you disrespect Apple then you damage the brand and then people might question why you paid extra for it in the first place.
« Last Edit: August 12, 2016, 07:51:33 AM by psxphill »