Amiga.org
Amiga computer related discussion => Amiga Hardware Issues and discussion => Topic started by: Alkemyst on March 08, 2003, 01:32:30 PM
-
http://personal.inet.fi/cool/pekosbil/a1benchmarks.htm
-
Lets just start by saying the same as Ole E. have already said several times on ANN.lu
Don't just say that these benchmarks are useless, but do come up with some better app's to use for benchmarking.
Do bear in mind though, that they must not be Hardware specific, as the test setups don't have similar Harddrives and gfx adaptors. (and CPU in the AMD case)
be helpfull, not hatefull :-)
-
More useful benchmarks like lmbench and x11perf would be useful.
As the discussion on ann.lu points out this is very preliminary and some more volunteers could help shed some light on the differences.
I still don't know why, for example, the faster G4 is slower at FPU fourier than the slower G3.
-
interesing results.
-
strobe wrote:
More useful benchmarks like lmbench and x11perf would be useful.
As the discussion on ann.lu points out this is very preliminary and some more volunteers could help shed some light on the differences.
I still don't know why, for example, the faster G4 is slower at FPU fourier than the slower G3.
That G4 /G3 difference may be beacuse people may have used different set of link libraries while building up those binaries.
It's not stated on page but I think that those Pegasos numbers are from April-1 machine.. April-2 was supposed to make some diference, but I can't be sure.
-
Worthless benchmark, as it gives no information on the setups used. Maybe the Pegasos had a different form of RAM than the A1 and Athlon. Maybe they had different kernel revisions or configurations. What cards are in each machine? What about the hard drives, that can throw off results too.
Without this data, those benchmarks are worth less than the bandwidth used to download them.
-
Just about the only things you can read into those figures are:
(1) raising the clock rate of a system usually increases its computing speed;
(2) a Pegasus G3 @ 600 MHz performs about the same as an A1 G3 @ 600 MHz.
Not very surprising, really.
tony
-
The problem is, no data about the actual hardware used. What video card, HD, RAM, etc. Any and all of these can schew benchmarks off by up to 30% in any direction, depending on the benchmark.
Heck, I remember a benchmark where the hardware was chosen specifically to cripple a particular OS for an OS 2 OS comparison, causing an artificial "boost" in the sponsoring company. That's why public disclosure of the hardware is so critical.
-
JoannaK wrote:
strobe wrote:
More useful benchmarks like lmbench and x11perf would be useful.
As the discussion on ann.lu points out this is very preliminary and some more volunteers could help shed some light on the differences.
I still don't know why, for example, the faster G4 is slower at FPU fourier than the slower G3.
That G4 /G3 difference may be beacuse people may have used different set of link libraries while building up those binaries.
It's not stated on page but I think that those Pegasos numbers are from April-1 machine.. April-2 was supposed to make some diference, but I can't be sure.
Those test (all AmigaOnes and Pegaos) were done with exactly the same executable. Yes, the Pegasos used on those test is a April-1 one...
-
That a G3 600 machine performs similar to a G3 600 is nothing surprising methinks. However, something strikes me: AmigaONE-s ran Debian, only the Pegasos had Suse, while Genesi delivers Pegasoses with Debian and Eyetech delivers AONEs with Suse...
Huhh?
-
@Alkemyst
My Athlon XP 1800+ with ASUS nForce 2 (MCP-T) board delivers about +11,800,000 OGR: (Nodes/sec).
http://www.amiga.org/forums/showthread.php?t=1287
==========================================
Athlon XP 1800+, ASUS nForce 2, WinXP-SP1
[Feb 18 22:25:23 UTC] Automatic processor type detection found
an AMD K7-6 (Athlon XP/MP/-4) processor.
...
[Feb 18 22:25:44 UTC] OGR: using core #1 (GARSP 5.13-B).
[Feb 18 22:26:03 UTC] OGR: Benchmark for core #1 (GARSP 5.13-B)
0.00:00:16.31 [11,800,528 nodes/sec]
==========================================
http://www.amiga.org/forums/showthread.php?t=1234
Athlon XP 1800+, ASUS nForce 2, WinXP-SP1
==========================================
[Feb 19 01:15:35 UTC] Automatic processor type detection found
an AMD K7-6 (Athlon XP/MP/-4) processor.
[Feb 19 01:15:35 UTC] OGR: using core #0 (GARSP 5.13-A).
[Feb 19 01:15:55 UTC] OGR: Benchmark for core #0 (GARSP 5.13-A)
0.00:00:16.71 [12,000,449 nodes/sec]
...
==========================================
Running 24 processes in the background (WinXP).
Why just those benchmarks programs?
What about “OpenSSL”?
-
@by Hammer
LOL Dont ask me i just posted the link.
-
@tonyw
Don't confuse the CPU with the overall system
-
Pecosbil wrote:
Those test (all AmigaOnes and Pegaos) were done with exactly the same executable. Yes, the Pegasos used on those test is a April-1 one...
Ok. Didn't see that mentioned on Saku foorum, it's good to know.
IIRC many of those old bytemarks are so smal tests they fit into L1 cache of C3 CPU... only couple of them were affected at all of absense of L2 cache as shown on those first numbers from Gunne.
One that would show performance diffies might be real life size tests. Like doing batch of 50 2megapixel 1600*1200 sized digital images and make some basic effects (smooth?) to them and store back to disk (in png?). Main idea is that even if software itself would it to CPU caches the data definitely would not. Jpeg would be about 500k but it expanded to 32-bit image for manipulation takes bit over 7 megabytes or Ram each.
That imagesize is nothing biggie these days (actually 5 megapixel is more top of line and batch conversions (scaling etc) are everyday items with anyone taking pictures and using them.
I think this kind of test could be made quite simply with standard Unix tools (PPM etc..) and some shellscripting?
-
Posted to ANN where this discussion started:
Amigaone v Peg Benchmark : Comment 142 of 142
Posted by bbrv (212.198.0.93) on 09-Mar-2003 06:27:27
In Reply to Comment 141:
Good Morning...just had the chance to catch up on this thread. The issue was NEVER G3 vs. G4. The issue was application oriented performance -- what you could do and how well you could do it. Processing speeds as an end in themselves can only find amusement with a crowd of integrated circuit engineers. Detrackers and malcontents have twisted the original post to mean something different than what was stated. Strobe's post just above is in the proper vein for a "real" discussion that begins with system performance while really "doing" something. :-) Gunne too has posted well here.
Please read our original comment again and if you are so inclined, please feel free to continue this discussion by email with us if you so desire.
Thanks and best regards,
Raquel Velasco and Bill Buck
Genesi
;-)
Note: Stobe's post here is correct too!
-
Who realy cares about the peg comes from dce so dont expect any warenty or if you do expect to wait a least 2 years b4 you get your comp back :-P :-P :-P :-P :-P :-P :-P :-P :-P :-P :-P :-P :-P
-
@bbrv
Quote from amiga.org (http://amiga.org/forums/showthread.php?t=44024):
---
The April2 brings the Articia fully to the performance levels it claims to achieve. We have purchased the remaining worldwide supply of the old Articias and will make 400 more Pegasos main boards with the G3 CPU. Our G3 Pegasos performs better than the G4 Teron with the Articia. This will be common knowledge soon (let the others understand this on their own our word will stand
as testimony to our knowledge and skill)
-
Hi Alkemyst...no need to do double work...;-)
Keep reading the rest of the thread on ANN!
:-D
Best regards,
Raquel and Bill :-)
-
:lol:
Where have you been poweramiga?
We have swapped out more than 100 Pegasos Betatesters already...;-)
Actually, we will be at DCE in a couple of weeks and we will see what we can work out with Thomas for old Phase V/DCE boards that have been returned.
Have a great day!
Raquel and Bill :-)
-
by bbrv on 2003/3/9 12:57:29
Hi Alkemyst...no need to do double work...
Keep reading the rest of the thread on ANN!
Alot of ppl dont read ANN.
-
I know a lot of people wanted some benchmarks from both sides(including BBRV), but if you ask me--the whole thing is an exercise in futility. Why? Well, the only fair way to test these systems is with AmiGOD.
Then, you have the real "Amiga vs. Amiga" benchmarks. Until such time as AmigaOS4 shows up, A1 should be with-held from such tests. If it can't run Amiga apps--it's not an Amiga.
I am not ragging on the A1 hardware, but at this point you might as well test the speed of a UniVac for all the good testing an A1 would do. Give Hyperion some time to get AmigaOS4 out, then test them all with amigod, syspeed, or whatever speed tool can be found on the aminet.
-
Alkemyst wrote:
Alot of ppl dont read ANN.
So, give them a Link to follow. Copying a single post (or part of it) out of context is not alays best opinion as it can cause misunderstandings and unnecessary flamewars.
And what about those who can't understand what they are reading? In these times those seem to be the biggest problem around :-P
-
by JoannaK on 2003/3/9 19:36:15
Quote:
Alkemyst wrote:
Alot of ppl dont read ANN.
So, give them a Link to follow. Copying a single post (or part of it) out of context is not alays best opinion as it can cause misunderstandings and unnecessary flamewars.
And what about those who can't understand what they are reading? In these times those seem to be the biggest problem around
It was in context to what BBRV just said here.
-
It was in context to what BBRV just said here.
But you were the one that started it with a forum topic not more than 2 weeks ago.
-
by downix on 2003/3/9 22:43:26
Quote:
It was in context to what BBRV just said here.
But you were the one that started it with a forum topic not more than 2 weeks ago.
So what your saying is from that time BB has not been responsible for his own comments about the benchmarks.
Maybe you havent noticed but this is another thread.
He is now saying that he didnt say what he did say.
the proof was in the link.
But infact i didnt start it, BB Claims did.
-
@Alkemyst
Nobody is claiming any of that, I pointed out that you made the original claim, pulling text out of context to push an agenda without providing a link to the original post in order for people to read that the comments posted were a responce to Ben Hermans. That means, frankly, that you did the post just to mislead people and to generally troll. And now you'll respond in like here, attempting to act all clean while blaming others for the problems of the world. So go ahead, do it, I'm prepared.
-
I just got the lowdown on the A1-G3-600 and Pegasos used for the tests:
The A1-G3-SE had a 133Mhz FSB
the Pegasos had a 100Mhz FSB
And the Pegasos kept up with a machine that had a 33% higher clockspeed for memory access. Means that the tests were predominantly held in the L2 cache, and therefore do not deny nor support bbrv's original claims.
-
-
Why didn?t they set both motherboards to 133 Mhz (or 100 Mhz) FSB?
Neither board was set up to go against the other. They were benchmarked seperately under different conditions.
IF BBRV?s claims are generalized in terms of scope then it would be countered by this specific benchmark scenario.
But they were not, they were focused on a specific section, responsiveness. So, a CPU-bound benchmark will not reveal how well the I/O tasks are handled on each board, which is where responsiveness is measured.
-
Nobody is claiming any of that, I pointed out that you made the original claim, pulling text out of context to push an agenda without providing a link to the original post in order for people to read that the comments posted were a responce to Ben Hermans. That means, frankly, that you did the post just to mislead people and to generally troll. And now you'll respond in like here, attempting to act all clean while blaming others for the problems of the world. So go ahead, do it, I'm prepared.
Wrong! i did post the link to where it came from but you have forgotten.
Here http://www.amiga.org/forums/showthread.php?t=1438
The link to where it came from is just before my Sig
When have i said i blame others for the problem of the world ?
Your really putting a hell of a lot of words into my mouth.
-
@downix
The test conditions were set as closely identical as possible. And those test can be easily repeated. And yes, so far they are just CPU+L2 cache specific benchmarks but still they prove that bbrv's claim is not true for April1 in every area (and I think it has been clear that bbrv meant April2).
bbrv said that pegasos G3 (artisia2 version) "performs better" than the best of A1. There was no mention about responsivenes etc. When you say it performs better, without any detail, you (deliberately) let people think that it's better in every aspect (very close to the definition of FUD, if it's not true).
And as far as I know, peg1s without artisia 2 have all been sold with 100Mhz FSB.
To overclock peg1 to 133Mhz, jumper soldering is required...
@all
Those tests are more than just about what bbrv claimed.
IMO: those test are a superb start of future Amiga HW comparissons, while others just whine and whine...
What we soon have is a set of tests that dig out every possible performance issue of those HWs... those test will be repeated ASAP when april2 pegs are ready, etc...
I'm sure people will be interested in knowing if A1G4XE@800Mhz is worse performer than peg1april2G3@600mhz, before they choose what to buy.... same for the A1lite, peg2, etc and it's interesting as well. :-)
-
Another Benchmark Fight
Yawn!
I run programs not benchmarks on my computer! :-D
Go fight over Intel vs AMD instead :-)