Click Here
home features news forums classifieds faqs links search
6071 members 
Amiga Q&A /  Free for All /  Emulation /  Gaming / (Latest Posts)



Lost Password?

Don't have an account yet?
Register now!

Your support is needed and is appreciated as is primarily dependent upon the support of its users.

Main sections
OS4 Zone
IRC Network
AmigaWorld Radio
Top Members
Amiga Dealers
About Us
Terms of Service

IRC Channel
Ports: 1024,5555, 6665-6669
SSL port: 6697
Channel: #Amigaworld
Channel Policy and Guidelines

Who's Online
4 crawler(s) on-line.
 30 guest(s) on-line.
 1 member(s) on-line.


You are an anonymous user.
Register Now!
 Kronos:  3 mins ago
 matthey:  6 mins ago
 DiskDoctor:  10 mins ago
 zipper:  27 mins ago
 Karlos:  55 mins ago
 amigakit:  2 hrs 2 mins ago
 pixie:  2 hrs 3 mins ago
 vox:  2 hrs 48 mins ago
 BigD:  3 hrs 3 mins ago
 Hypex:  3 hrs 10 mins ago

/  Forum Index
   /  Amiga OS4 Hardware
      /  32-bit PPC on FPGA
Register To Post

Goto page ( Previous Page 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 )
Re: 32-bit PPC on FPGA
Posted on 26-Feb-2024 4:57:23
#401 ]
Elite Member
Joined: 9-Mar-2003
Posts: 5559
From: Australia



ppcamiga1 wrote:

It doesn't help that NG AmigaOS businesses have ignored and practically blocked 68k AmigaOS FPGA development that has a better chance to be NG than PPC PC hardware.

This is BS. I like my vampire but it is still many times slower than any ppc.

Try the DevilutionX benchmark.

Ryzen 9 7900X, DDR5-6000 64 GB RAM, GeForce RTX 4080 16 GB
Amiga 1200 (Rev 1D1, KS 3.2, PiStorm32lite/RPi 4B 4GB/Emu68)
Amiga 500 (Rev 6A ECS, KS 3.2, PiStorm/RPi 3A+/Emu68)

 Status: Offline
Profile     Report this post  
Re: 32-bit PPC on FPGA
Posted on 26-Feb-2024 6:39:09
#402 ]
Cult Member
Joined: 25-Sep-2022
Posts: 512
From: Unknown


This is BS. I like my vampire but it is still many times slower than any ppc.

Try the DevilutionX benchmark.

I think it does depend on the task.

The APOLLO 68080 CPU is clock by clock generally more performant than most PowerPC CPUs.
We explained that PowerPC often needs more instructions for doing the same job.
This can be 3 PowerPC instructions for doing what a single 68K instruction does.
But sometimes it can also be 4/5/6/ or 7 PowerPC instructions for doing what a single 68K instruction does.

Of course the FPGA does limit the clockrate.
While you get 100MHz in the FPGA putting the same core in an ASIC might give you 1 Gigaherz.

If you compare Vampire and PowerPC systems
then Apollo 68080 has better performance per clock
But PowerPC Systems have higher clock - as they use an ASIC.

If you look not only at the CPU clock but also at the memory performance
than the Apollo 68080 does in fact beat many PowerPC systems.

The memory performance of many PowerPC systems is pretty bad
and the Data-Cache if many PowerPC system can not automatically prefetch.
The G5 PowerPC 970 can - and the Apollo 68080 can both do this.
And this gives both a major performance boost.

If you look at some applications or games like Diablo port
then you can see that Diablo runs on the Vampire a lot smoother
and faster than on AmigaOne with 1000 MHz PowerPC.
The reason is for a big part the better Data-Cache of the Apollo CPU and its better memory performance.

 Status: Offline
Profile     Report this post  
Re: 32-bit PPC on FPGA
Posted on 26-Feb-2024 7:00:05
#403 ]
Elite Member
Joined: 29-Oct-2012
Posts: 3709
From: Germany



OneTimer1 wrote:


cdimauro wrote:

Again, Red Herring. You're trying again to change the topic.

The topic is "32-bit PPC on FPGA" but people stopped writing about it, after it was explained why a slow PPC is a bad idea.

Why you're again trying to change this part of the discussion? YOU talked about the Amiga OS and YOU made PRECISE statements about it, which were FALSE.

Why couldn't you be intellectually honest to admit it, instead of trying to change the topic?



matthey wrote:
cdimauro Quote:

So, there's no real opportunity 'til now. And you can't expect any source code opening, after so long: it's clear evidente that they want to keep closed their goose that laid the golden eggs.

However, you were also in touch with Jens: you can contact him to see if there's some space for a different project.

The Natami project led by Thomas Hirsch was very open and transparent with a clear goal to revive the 68k Amiga. Quote:

Thomas Hirsch, the creator, wanted to revive the Amiga long ago. He worked for many years rebuilding the OCS and AGA chipsets.

Gunnar von Boehn and Peter Kaltstein were working on their daily jobs in the same department of a hardware development lab as Thomas. They teamed up with Thomas to help him to get the NatAmi out, and were very motivated by seeing the 1st prototype in action at an Amiga-meeting in 2008. Soon after, more people joined the team.

The original prototype was designed like an A4000 using 50MHz memory. But starting in 2008, Thomas has been redesigning the Amiga DMA engines to make full use of modern memory. This means that memory bandwidth is greatly enhanced.

Jens had a similar philosophy as Thomas and tried to open the N68050 CPU core created for the Natami project. Is there any question where the closed hardware secretive cult philosophy came from?

The Natami project was too hard lacking a high performance 68k CPU core and being blocked by AmigaNOne IP squatters. While AmigaNOne hardware was selling in the low thousands, the open and transparent Natami project generated huge interest without advertising. The Natami "MX Bringup Thread" had 761,487 views at one point and likely went significantly higher.

Natami "MX Bringup Thread" Quote:


The more open and transparent Natami project felt to me like it generated more excitement than the AC/Vamp project. I posted the Natami "MX Bringup Thread" views to various Amiga forums claiming this was an indication of a large 68k Amiga retro market but it was largely discounted. It wasn't until later that THEA500 Mini success backed up my claims and suggested that there may be a large enough market for mass production and creating competitive affordable hardware. I would have also liked to see a professionally developed "revived" retro 68k Amiga but some things are too hard in Amiga Neverland.

That was a big missing opportunity for the Amiga revival which I think that it's very difficult, if not even impossible, to be again.

That's because the THEA500 Mini project has already took the chance to "revival" the platform for plenty of people which used our beloved platform and bought it to revive the good old times.

Now you can't do the same again, because people will not buy a duplicated of what they've already bought.

That's why also the A600GS project has no chances to succeed: time over. It will sell, for sure, because there are passionate which like to have better platforms compared to what the THEA500 Mini offers, but the average guy which has already bought the latter will not buy the new one.

 Status: Offline
Profile     Report this post  
Goto page ( Previous Page 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 )

[ home ][ about us ][ privacy ] [ forums ][ classifieds ] [ links ][ news archive ] [ link to us ][ user account ]
Copyright (C) 2000 - 2019 was originally founded by David Doyle