Amiga.org

Amiga News and Community Announcements => Amiga News and Community Announcements => Amiga Hardware News => Topic started by: zee4 on May 29, 2003, 05:25:51 PM

Title: 1Ghz AmigaOne benchmarked
Post by: zee4 on May 29, 2003, 05:25:51 PM
Everybody loves benchmarks, so here's some pitting several AmigaOne models, the Pegasos board and an AMD Athlon XP 1800+. The AmigaOnes where running Linux and none of the tests where optimized for AltiVec.

AmigaOne benchmarks (http://personal.inet.fi/cool/pekosbil/a1benchmarks.htm)

Not surprisingly the Athlon comes out on top in most tests, though AmigaOne G3/1Ghz  does better in several tests. Down at the bottom the A1 G3/600Mhz and the Pegasos duke it out for last place, it's odd how  there seem to be small variations between what should be the same kit.

The real test will come when these tests are run under OS4- I think losing the Linux baggage might give it a bit of an extra push.  And maybe using AltiVec too on the G4...

...then again you know what they say about statistics and benchmarks...
Title: Re: 1Ghz AmigaOne benchmarked
Post by: KennyR on May 29, 2003, 06:31:21 PM
I would be more interested in seeing memory speed than CPU speed. CPU speed isn't very important when you can just replace the CPU. Memory speed won't ever improve.

So can someone do a comparison between Pegasos and A1XE using the same quality of DIMMs, using a tool like memtest? I would be very interested in seeing it.
Title: Re: 1Ghz AmigaOne benchmarked
Post by: uncharted on May 29, 2003, 08:27:37 PM
TBH I really don't see the point in this.  It's just another (unrealistic) pissing contest.  Niether Borad is going to give you the overall performance of a modern X86 board.  These benchmarks (or any other for that matter) aren't going to give any indication of what the system will feel like and how well it suits and indivdual.  

If you are after raw power then don't look to this market to provide it.  The positives of these new systems cannot be measured by some obscure test.
Title: Re: 1Ghz AmigaOne benchmarked
Post by: createcoms on May 29, 2003, 09:59:53 PM
Erm Raw power was never the focus of the project.  We will ownz x86 purely by superior coding.  I advise to watch closely key points like boot time, memory requirements and fps on latest generation cards where code is what matters.  The amiga WILL have the last laugh........
Title: Re: 1Ghz AmigaOne benchmarked
Post by: jahc on May 30, 2003, 02:35:34 AM
How did they get a 1GHz G3? Overclocking?
Title: Re: 1Ghz AmigaOne benchmarked
Post by: Sloxa on May 30, 2003, 09:25:20 AM
this bencmark tells only cpu speed.. nothing more!!

this really dosent tell system performance...
like every body know, even athlon xp 2400+
cant run games well... when harddisk load..
then games start jerking!!!  nykii ja pätkii!!!
if you have fast cpu and rubbish other system...
then you have not  so fast computer.! :-D
Title: Re: 1Ghz AmigaOne benchmarked
Post by: Rogue on May 30, 2003, 11:57:45 AM
Quote
this really dosent tell system performance...


Yeah, how can anybody be so naive as to think that CPU performance would have anything to do with the system's speed
:-)
Title: Re: 1Ghz AmigaOne benchmarked
Post by: minator on May 30, 2003, 03:19:22 PM
Quote
this bencmark tells only cpu speed.. nothing more!!


Not true, from the descriptions of the tests they are also indirectly testing memory access.  Anything which reads non-sequnetially is going to have serious latency hits.

Evidently the Athlon has a rather better memory controller than the rest.

There was a time when PowerPCs used to *rule* in ByteMarks.