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 664 times)

0 Members and 1 Guest are viewing this topic.

Offline Rob

Re: A2080 i.e. Vampire 500 V2 on an Amiga 2000
« Reply #30 on: August 10, 2016, 11:34:02 PM »
Quote from: Oldsmobile_Mike;812383
This has been mentioned before as well. I think it was something along the lines of "the PPC chip they chose was incompatible with any of the other Amiga PPC hardware or software". Useless. :(  I could be wrong, though! ;)


It uses the same core as Tabor so we'll wait and see how that performs although they did say they were considered using a different processor after receiving feedback from the community.  There are much better choices available now too.
 

Offline QuikSanz

Re: A2080 i.e. Vampire 500 V2 on an Amiga 2000
« Reply #31 on: August 11, 2016, 01:29:41 AM »
Quote from: Oldsmobile_Mike;812371
Just remembered. I recall hearing a while back that the Vampires weren't compatible with Zorro slot expansions. Installing one would render all your other cards inoperable. I wonder if this has been addressed yet?


If you loose the use of Zorro slots this does not sound very desirable. I need to use some of those for other things.
 

Offline Sparky

Re: A2080 i.e. Vampire 500 V2 on an Amiga 2000
« Reply #32 on: August 11, 2016, 03:29:29 AM »
Quote from: jdryyz;812363
It occurred to me that if you can provide 68k and "AGA" in an FPGA, wouldn't PowerPC also be possible?


I think I read a post in the apollo forums nixing the PowerPC idea, they want to keep it classic and make the fastest 68k systems .. or something along those lines.
 

Offline Steady

Re: A2080 i.e. Vampire 500 V2 on an Amiga 2000
« Reply #33 on: August 11, 2016, 03:52:47 AM »
@Srdjan

That's a weird kickstart number for an A2000. I would've thought it would be 37.175.
 

Offline Heiroglyph

Re: A2080 i.e. Vampire 500 V2 on an Amiga 2000
« Reply #34 on: August 11, 2016, 03:53:54 AM »
Quote from: QuikSanz;812389
If you loose the use of Zorro slots this does not sound very desirable. I need to use some of those for other things.


That's not correct. People are using them with A500's with attachments on the side and with A2000's with zorro network cards.
 

Offline QuikSanz

Re: A2080 i.e. Vampire 500 V2 on an Amiga 2000
« Reply #35 on: August 11, 2016, 04:59:25 AM »
Quote from: Heiroglyph;812393
That's not correct. People are using them with A500's with attachments on the side and with A2000's with zorro network cards.


Well, it looks to have a USB port. If a hub can be used I won't need Algor. However I need network and high speed Ser/Par card for printing and other stuff.
 

Offline Heiroglyph

Re: A2080 i.e. Vampire 500 V2 on an Amiga 2000
« Reply #36 on: August 11, 2016, 07:19:33 AM »
That's not USB, you're probably seeing the JTAG port.

This isn't the hardware that will be for sale, only for development until the hardware is final.

This one has HDMI, IDE and an SD card but it will probably change before they go out for sale to the public. I've heard of onboard ethernet for example, but not with this early development board.
 

Offline psxphill

Re: A2080 i.e. Vampire 500 V2 on an Amiga 2000
« Reply #37 on: August 11, 2016, 09:35:30 AM »
Quote from: Thomas Richter;812378
Not really. You should keep in mind that the PPC has longer opcodes, and performs less operations per opcode as it is a risk processor.


PPC & 68060 appear to have quite similar instruction rates. RISC processors were designed for the type of operations that C compilers needed. So there may be times you can code something more efficiently on a 68060, if you're writing in a compiled language then the difference won't be anyway near as pronounced as you say.

The problem is that getting an FPGA to run at a speed equivalent to a 100mhz 68060 has taken a long time and is likely the top end of performance for that FPGA no matter what CPU you are simulating. It doesn't make sense to him because you would have to dedicate the same amount of time to create a PPC and it would only have the performance of a 100mhz 68060. You can beat that with ancient PPC that nobody wants to buy anymore because they are too slow.
 

Offline wawrzon

Re: A2080 i.e. Vampire 500 V2 on an Amiga 2000
« Reply #38 on: August 11, 2016, 11:46:30 AM »
wasting time with ppc whatsoever is not what apollo team is going to do. they told it over and over. and they are right to do so. ppc was an attempted (and failed) solution for lack of further development on 68k front. exactly the gap apollo core is about to fill in. what sense would that make to give up on that (apparently successful approach) and try to repeat some old failed hackery attempt? especially software titles that could take advantage of it is too few even to mention..
 

Offline wawrzon

Re: A2080 i.e. Vampire 500 V2 on an Amiga 2000
« Reply #39 on: August 11, 2016, 12:05:53 PM »
Quote from: jdryyz;812380
Oh well. Maybe this might be made some day:

http://ultimateppc.nl/specifications.php

It's for big box Amigas, though. I think Gideon is way too busy with the 1541 Ultimate right now though.  :D


if you want ppc, this is your choice:
https://github.com/Sakura-IT/SonnetAmiga/
its better and more compatible with warpup ultimateppc would likely ever be. and its real, working and being worked on.
 

Offline Thomas Richter

Re: A2080 i.e. Vampire 500 V2 on an Amiga 2000
« Reply #40 on: August 11, 2016, 12:10:54 PM »
Quote from: psxphill;812399
PPC & 68060 appear to have quite similar instruction rates. RISC processors were designed for the type of operations that C compilers needed.
I would rather say, Coldfire was as it only allowed 32-bit operands (mostly), which corresponds to the implicit conversion to "int" the C language performs. PPC was Motorola's attempt to jump on the bandwagon of the latest fashion at that time, which was "risk".   Clearly, "risk" makes a lot of things simpler for the CPU designer because none of the instructions that affect memory have any other secondary side effect - unlike the CISC cores. But it also means that everything needs to go through registers in some way. Thus, a C code such as  
Code: [Select]
rect.y += delta_y  would translate, in 68K, to something  
Code: [Select]
add.l d2,4(a0) which is something your average C compiler can do, 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 of 32 bit instead of one instruction of 16 bit. Clearly, if you then later on use r1 or the result of the addition for any other computation, one of the instruction "amortizes", but on average, the PPC requires more instructions, and longer instructions, and more registers.  It's a horse of a different color. The hope was that PPC - due to a simpler execution unit - would allow an easier upscaling of the CPU to higher frequencies. Unfortunately, the memory bandwidth bottleneck was probably not yet relevant enough at that time.    
Quote from: psxphill;812399
So there may be times you can code something more efficiently on a 68060, if you're writing in a compiled language then the difference won't be anyway near as pronounced as you say.
Wasn't that Apple back then when porting MacOs to PPC that they came up with about a code expansion of this factor of two? I seem to remember some statement like this. Given the instruction sizes and the instruction performance, it seems about to fit.  
Quote from: psxphill;812399
The problem is that getting an FPGA to run at a speed equivalent to a 100mhz 68060 has taken a long time and is likely the top end of performance for that FPGA no matter what CPU you are simulating. It doesn't make sense to him because you would have to dedicate the same amount of time to create a PPC and it would only have the performance of a 100mhz 68060. You can beat that with ancient PPC that nobody wants to buy anymore because they are too slow.

That's a different story, of course. If you ask me whether it would make sense to emulate a PPC on an FPGA for amiga applications then my answer would also be "clearly no", basically because the amount of applications for which this would be an advantage is too small, and you could buy a somewhat "modern" (ehem) PPC system anyhow if you would want to. But that's a different market.
 

Offline psxphill

Re: A2080 i.e. Vampire 500 V2 on an Amiga 2000
« Reply #41 on: August 11, 2016, 12:52:59 PM »
Quote from: Thomas Richter;812404
PPC was Motorola's attempt to jump on the bandwagon of the latest fashion at that time, which was "risk".

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.

I'm sure there are examples of 1 68060 instruction needing to be expanded to 3 PPC instructions. But the question I would ask is how often those types of expansions occur & whether it's mitigated by any situations where the opposite occurs and whether the PPC can run multiple instructions more often than the 68060.

I'm reasonably confident that if someone put the same effort in then they could create a PPC that has a similar real world performance to a 100mhz 68060. Finding that person is the hard part. Especially as it's far slower than the original PowerUP boards, let alone the SAM/X1000 etc.

What would make sense is for there to be a board that allows you to fit an FPGA and an off the shelf PowerPC, so that people don't have to choose between vampire or a phase 5 PPC.
« Last Edit: August 11, 2016, 01:01:01 PM by psxphill »
 

Offline Dandy

Re: A2080 i.e. Vampire 500 V2 on an Amiga 2000
« Reply #42 on: August 11, 2016, 01:11:15 PM »
Quote from: Bodie;812335
Is this version available for pre-order yet?

Also...hmmm I have a spare A2000 with one of the dreaded German Motherboards...



Why "dreaded"?
:confused:
All the best,

Dandy

Website maintained by me

If someone enjoys marching to military music, then I already despise him. He got his brain accidently - the bone marrow in his back would have been sufficient for him! (Albert Einstein)
 

Offline Thomas Richter

Re: A2080 i.e. Vampire 500 V2 on an Amiga 2000
« Reply #43 on: August 11, 2016, 01:22:07 PM »
Quote from: psxphill;812406
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.
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.  
Quote from: psxphill;812406
I'm sure there are examples of 1 68060 instruction needing to be expanded to 3 PPC instructions. But the question I would ask is how often those types of expansions occur & whether it's mitigated by any situations where the opposite occurs and whether the PPC can run multiple instructions more often than the 68060.
At the expense of higher memory throughput, sure. So let's see some facts. If we believe Wikipedia, then the 68060 has 110Mips at 75Mhz, the PPC601 157 Mips at 80Mhz, which is 1.46Mips/MHz for the 68K and 1.96 for the PPC, thus indeed better.  

This improvement is mostly, as I believe, because the IPC of the PPC is higher, due to the simpler instruction set. IPC of the 68060 is only 1.3, of the PPC 1.9. Of course, the PPC is also a later design, so it's a bit apples vs. oranges, as most of these tests.

At least this source:  

http://www.microapl.co.uk/Porting/ColdFire/cf_68k_diffs.html

also critically remarks the lower code density of the PPC, which seems quite natural given the simplicity of the instruction set and the design. Unfortunately, no hard numbers there on particular algorithms.  
Quote from: psxphill;812406
I'm reasonably confident that if someone put the same effort in then they could create a PPC that has a similar real world performance to a 100mhz 68060.
Well, maybe, I'm not a technical expert in this field, though one should note that such a design would still be far behind existing PPCs available on the market, so the attempt would be a bit pointless.  
Quote from: psxphill;812406
What would make sense is for there to be a board that allows you to fit an FPGA and an off the shelf PowerPC, so that people don't have to choose between vampire or a phase 5 PPC.

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? (-:
 

Offline Thomas Richter

Re: A2080 i.e. Vampire 500 V2 on an Amiga 2000
« Reply #44 on: August 11, 2016, 01:24:56 PM »
Quote from: Dandy;812407
Why "dreaded"?
:confused:

Well, because the CPU connector is not exactly working as in the "B" models, and hence, it is not exactly obvious that a card created for the "B" would also work in the "A". Typically, you would have to remove the 68K from the "Braunschweig" ("A") models to get turbo boards working. Whether that's sufficient for any possible Turbo card is something I cannot answer.