Amiga.org
Amiga computer related discussion => Amiga Hardware Issues and discussion => Topic started by: ikir on February 18, 2003, 08:59:43 PM
-
See this discussion on www.iksnet.it
Is it true??? :-D I hope yes! :-D
Discussion:
http://www.iksnet.it/forum/viewtopic.php?t=297 (http://www.iksnet.it/forum/viewtopic.php?t=297)
-
Really AmigaONE G4-800Mhz wins over P4 2,6Ghz, this without level3 cache?
-
The origin from this posts are the AmigaOne mailing lists.and seems to be from the latests
One XE G4 systems with already 2Mb L3 cache.
-
Ami603 wrote:
The origin from this posts are the AmigaOne mailing lists.and seems to be from the latests
One XE G4 systems with already 2Mb L3 cache.
Good. But it shoul be true bacause came from the original mailing-lis :-o
This means......
AmigaONE real kick ass!!!!!!!! :-o :-o :-o
-
Well if your read the actual post you would see that the client
DOES have altivec-support, and such benches are really good for
altivec.
There is no way a G4/800 could beat a P4/2600 in integer or single
data float.
And the 3rd-level-cache on the G4 is only needed because it
doesn't have any other posibility to use modern DDR-RAM, which
comes naturally for todays x86s.
-
Hi Kronos! How are you?
Yes. But this means that on software wich supports altivec AmigaOne800 wins over P4-2,6Ghz?
-
These are from a Finnish Amiga forum...
A1G3@733:
[Feb 18 10:34:17 UTC] OGR: using core #0 (GARSP 5.13 PPC-scalar).
[Feb 18 10:34:37 UTC] OGR: Benchmark for core #0 (GARSP 5.13 PPC-scalar)
0.00:00:16.78 [8,680,916 nodes/sec]
600MHz:
[Feb 17 14:24:36 UTC] OGR: using core #0 (GARSP 5.13 PPC-scalar).
[Feb 17 14:24:56 UTC] OGR: Benchmark for core #0 (GARSP 5.13 PPC-scalar)
0.00:00:16.55 [7,120,713 nodes/sec]
Athlon XP 1800+ native:
[Feb 17 10:46:36 UTC] OGR: using core #1 (GARSP 5.13-B).
[Feb 17 10:46:55 UTC] OGR: Benchmark for core #1 (GARSP 5.13-B)
0.00:00:16.71 [11,750,304 nodes/sec]
Athlon XP 1800+ & WinUAE:
[Feb 17 11:10:44 UTC] OGR: using core #3 (GARSP 5.13 68040).
[Feb 17 11:11:03 UTC] OGR: Benchmark for core #3 (GARSP 5.13 68040)
0.00:00:16.18 [2,201,927 nodes/sec]
But in the end, those results have very little relevance in real life apps.
-
@ Kronos
> Well if your read the actual post you would see that the client
> DOES have altivec-support, and such benches are really good
> for altivec.
Well, if *you* read the post carefully, you'll see that it says "using core #0 (GARSP 5.13 PPC-scalar)"... That means it uses the scalar, not the vector core., hence, the test ran *without* AltiVec support.
-
Mah, I have a question....The AmigaOne agains a PC have a Altivec support ?
If it don't support Altivec, I will consider that the AmigaOne G4 800 it's run like a Pentium 4 ? :-o
samo79
http://digilander.libero.it/samo79
-
@EntilZha
Please, the OS release ... quckly, please, please ... :-x :-P
-
AmigaONE rocks!!!!!!!!!!!!!!!!!!!!!! :-D 8-)
-
Not to burst any bobbles here, but the P4 is a very lousy performer in RC5...
even the posted AMD XP1800 scores should indicate that.
Other than that, i'm sure the A1 will still move some ass :-)
-
==========================================
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]
==========================================
I can probably squeeze little more performance IF I turn off the network services (e.g. bridges, internet), Norton Anti-Virus 2002, MSN messenger, Asus probe, WinXP’s GUI, OBDC services, etc and increase priority for dnetc task.
For some reason, my Athlon XP 1800+ box is faster than ksk's Athlon XP 1800+ native box.
-
Well actually, the link is to an italian site, but I posted the info on the Yahoo groups site.
But it should be pointed out that the P4 is "only" a 2.4 GHz.
My XE has the 3rd level cache.
Nice to see what others have achieved with the hardware they have available - thx, "all" I had is that P4 2.4GHz.
And yes, it has been run without Altivec support - we are working on getting the client to use the vector core (ie. Altivec core).
And, yes, I think that it rocks :-D
-
And the 3rd-level-cache on the G4 is only needed because it doesn't have any other posibility to use modern DDR-RAM, which comes naturally for todays x86s.
You have it the opposite way round. x86 needs modern DDR-RAM for speed because of (bad!) legacy CPU design, while PPC is the opposite and is not sped up much by fast external RAM but is sped up greatly by extra cache. This is also why PPC is so good at cache-based calculation like RC5.
-
My brother says that we should check also the rc5 benchmarks because it is linear. He doesn`t trust very much OGR :-D
-
You have it the opposite way round. x86 needs modern DDR-RAM for speed because of legacy CPU design,
What do you mean by "legacy CPU design"?
For AMD Athlon XP’s case, the DDR architecture was inherited from DEC's Alpha AXP architecture. What do you expect from same engineers who designed the Alpha CPU?
Athlons XPs works fine enough with KT133A (that's with 133Mhz SDRAMS) chipsets.
PS; L1/L2/L3 Cache setup works fine(i.e. a speed boost for the given core) for K6 Core( for K6-III product line)....
while PPC is the opposite and is not sped up much by extra external RAM but is sped up greatly by extra cache.
One should be focusing on bandwidth throughput instead of the buzzwords i.e. what is the bandwidth throughput of PPC G4 while it talks to the outside world i.e. 64bit x 133mhz perhaps.
-
I have a 2.26 GHz P4 w 533 MHz FSB, 1 Gig ram DDR2100.
w98se
I started the program to do RC5. The OGR24, started instead (I think 24, not 25, whichever is default.) I got, with McAfee firewall and anti-virus in the background, and as near as I can tell all other unnecessary prgs off, 9.1 to 9.3 million nodes a second. So, WOW, A1!!!
RC5, stats
****************************
.....10%.....20%.....30%.....40%.....50%.....60%.....70%.....80%.....90%....100
[Feb 19 06:56:05 UTC] RC5-72: Completed CA:447DFD29:00000000 (1.00 stats units)
0.00:22:54.15 - [3,125,592 keys/s]
[Feb 19 06:56:05 UTC] RC5-72: Loaded CA:447DFDD6:00000000:1*2^32
[Feb 19 06:56:05 UTC] RC5-72: Summary: 29 packets (29.00 stats units)
0.10:49:43.92 - [3,192,544 keys/s]
[Feb 19 06:56:05 UTC] RC5-72: 16 packets (16.00 stats units) remain in
buff-in.r72
Projected ideal time to completion: 0.05:57:20.00
[Feb 19 06:56:05 UTC] RC5-72: 8 packets (8.00 stats units) are in
buff-out.r72
.....10%.....20%.....30%.....40%.....50%.....60%.....70%.....80%.....90%....100
[Feb 19 07:18:54 UTC] RC5-72: Completed CA:447DFDD6:00000000 (1.00 stats units)
0.00:22:49.01 - [3,137,318 keys/s]
[Feb 19 07:18:54 UTC] RC5-72: Loaded CA:447DFDD8:00000000:1*2^32
[Feb 19 07:18:54 UTC] RC5-72: Summary: 30 packets (30.00 stats units)
0.11:12:32.93 - [3,190,670 keys/s]
[Feb 19 07:18:54 UTC] RC5-72: 15 packets (15.00 stats units) remain in
buff-in.r72
Projected ideal time to completion: 0.05:35:00.00
[Feb 19 07:18:54 UTC] RC5-72: 9 packets (9.00 stats units) are in
buff-out.r72
.....10%.....20%.....30%.....40%.....50%...
[Feb 19 07:31:38 UTC] Paused... (user generated)
**********************
I think the highest I got was 3,240,000 keys/sec
Amiga! Why? Because I'm a power-user!
-
Atheist wrote:
I have a 2.26 GHz P4 w 533 MHz FSB, 1 Gig ram DDR2100.
w98se
I started the program to do RC5. The OGR24, started instead (I think 24, not 25, whichever is default.) I got, with McAfee firewall and anti-virus in the background, and as near as I can tell all other unnecessary prgs off, 9.1 to 9.3 million nodes a second. So, WOW, A1!!!
You're comparing apples and oranges. OGR is NOT RC5-72.
Try starting your client with the "-bench" argument to see what it 9and your machine are capable of.
-
Just to add fuel to the fire, I benchmarked my own machine a couple of minutes ago. It's an AthlonXp 2000+ (1667MHz) w/512MB of DDR 2700 RAM running Mandrake 7.1. I had XMMS playing as well as a couple of Mozilla windows open...
dnetc v2.9001-478-CTR-02112023 for Linux (Linux 2.2.15-4mdk). Please provide the *entire* version descriptor when submitting bug reports. The distributed.net bug report pages are at http://www.distributed.net/bugs/ Using email address (distributed.net ID) 'chotch@............'
[Feb 19 08:37:57 UTC] Automatic processor type detection found
an AMD K7-6 (Athlon XP/MP/-4) processor.
[Feb 19 08:37:57 UTC] OGR: using core #0 (GARSP 5.13-A).
[Feb 19 08:38:16 UTC] OGR: Benchmark for core #0 (GARSP 5.13-A)
0.00:00:16.68 [11,873,670 nodes/sec]
[Feb 19 08:38:16 UTC] OGR: using core #1 (GARSP 5.13-B).
[Feb 19 08:38:38 UTC] OGR: Benchmark for core #1 (GARSP 5.13-B)
0.00:00:19.60 [11,491,880 nodes/sec]
So the Pentium used in those tests doesn't look too good at all.
-
Yeah. Like a G4 800 would ever beat the fast Intel or AMD CPU's in non-synthetic benchmarks, be it price or performance. I'm living in the real world - which one are you living in? :evil:
-
Hay volmer clam down!! have a :pint: :-D
I am more interested in the benchmarks of the AMDs compaired to the Intels.
Besides, this is only a disscussion no need to get to hostile!
-
>I'm living in the real world - which one are you living in?
Loosen up a bit. This is just FUN stuff, nothing more.
Have a red&white pill. :-D
-
I for one hope my xe kicks the arse of the p4 wouldent it be great not only the best computor in the world but fast as well ow please release os4 the Amiga will be once again king of the computor world
-
I think its interesting how poorly all of these chips do, both AmigaOne and x86, compared to the venerable old CS-PPC...
http://amigapro.com/Images/280.gif
The old CS-PPC (mine is slightly oc'd there, to 66/280MHz) is just under half as fast in OGR as the AmigaOne G4/800, and it completely buries the Athlon per MHz. The 604e has a better FPU than both chips as well.
-
Juzz wrote:
Well actually, the link is to an italian site, but I posted the info on the Yahoo groups site.
But it should be pointed out that the P4 is "only" a 2.4 GHz.
My XE has the 3rd level cache.
Nice to see what others have achieved with the hardware they have available - thx, "all" I had is that P4 2.4GHz.
And yes, it has been run without Altivec support - we are working on getting the client to use the vector core (ie. Altivec core).
And, yes, I think that it rocks :-D
Cool! :-D
-
its an Amiga theres no doubt it bound to rock you wouldent expect anything less
WERE BACK AND WE KICK ASS
-
poweramiga2002 wrote:
WERE BACK AND WE KICK ASS
Yeah! 8-)
-
@KennyR
No I am correct !!
The perfect CPU-RAM setup was what we had in an A2000 with Fast-RAM.
The CPU had access to the full 8mb every cycle.
Problem is that while CPUs got faster and faster the RAM wasn't keeping
up which made caches necessary. A cache is always a bottleneck,
and the more stages you got the worse it gets. Having a huge gap
between to stages is also a problem.
So starting with a slow main-RAM is not a good idea.
The use of a fast cache is much bigger in popular benchmarks
than it is in real-world SW, and that is why PPC is hanging behind
x86 when both run the same SW on the same OS.
-
Does any of this x86 vs PPC matter?
It always seems to come up.
What matters is, what is the A1 / Pegasos going to do for you?
Is your MorphOS / OS4 experience, compared to OS3.x running on your trusty 680x0, going to kick arse?
With a minimum G3 600MHz behind it, which lets face it is far more powerful than any existing Amiga hardware, what sort of apps can we look forward to?
Will the audiophiles get realtime soft synths?
Will the arty types get more pixel mutating power?
Will the gamers get to blow new stuff up?
All this fretting over what the x86 can do faster and vice versa is just silly.
Lets get smart, have fun and bash out some nice shiny new apps for these machines, rather than running synthetic benchmarks and whingeing ;-)
-
You have it the opposite way round. x86 needs modern DDR-RAM for speed because of (bad!) legacy CPU design, while PPC is the opposite and is not sped up much by fast external RAM but is sped up greatly by extra cache. This is also why PPC is so good at cache-based calculation like RC5.
All CPUs need a fast interface to the rest of the system. Bad legacy CPU design? Uh, no. x86 CPUs run at a relatively high speed compared to main system RAM and you want to keep the CPU fed with instructions. Fast caches help with this problem.
Do you know why the G4 is sped up more by extra integrated cache as opposed to faster main system RAM? It's simple--Motorola has crippled the main system bus to 166 MHz SDR. So while the Athlon is at, what, 166 MHz DDR (333 MHz effective) and the P4 is a, what, 133 MHz QDR (533 MHz effective), the lowly G4 is forced to use 166 MHz or 133 MHz.
Already the vaunted Altivec units are hampered by being starved for instructions. This has been a complaint by coders on the Mac platform for a while. It's funny, the Mac has DDR DRAM but it only gets accessed through an SDR bus... the CPU is crippled.
That's the only reason that programs that can basically fit in cache (data and instructions for the critical loops) perform so well with the PPC... and heck, that's an ideal case for any CPU.
and it completely buries the Athlon per MHz. The 604e has a better FPU than both chips as well.
So? I don't see any 1.2 GHz 604e CPUs laying around, who cares? The CPU couldn't scale much higher and the manufacturers moved on. The relative efficiency of a 280 MHz CPU compared to modern CPUs of different architectures, even, has little to no practical relevence. I mean, a 68040 might have even better per-MHz FPU performance than even the new PPC970 but who would use that over the new CPU? ;)
Anyway, the whole "x86 needs more bandwidth because it's poorly-designed" is a false assertion. ALL modern CPUs benefit from higher bandwidth to the main system memory. With the G4 there is no option for increasing that speed which is why y'all should be happy that there's a huge L3 cache included.
I like PPC but there are some shortcomings that can't be ignored for the current CPUs. That should all go away with the PPC970, a CPU I'm excited about. :D
-
BTW, I'm not trying to bash the G4 and say, "omg, x86 is better, ur so dum!" Heck, I'd LOVE to have a dual-G4 system! It's possible to like something but also be aware of its shortcomings. :)
The Pentium 4 and the Athlons have their own shortcomings, as well, they just aren't really relevent to a discussion about overall G4 performance since it is more of a worst-condition exercise of the mind. All I wanted to do was clear up some misconceptions and whatnot.
-
AMIGA all the way.
AMIGA rulez!
Anything else sucks.
-
Hei Ikir ma come faranno sti poveretti a leggere i topic italiani nei nostri siti ?? :-D :-D ....
Mi sa che dobbiamo cominciare a tradurre le pagine! :-? :-?
Ciao
-
Hello! I am not much of a power user, nor do I consider myself a knowledgeable computer user, but I found this intersting bit of information on Digita's Organizer for PC, on an article entitled "Evolution of the CPU" 'The Pentium ... 'interestingly, the awkward and old architecture is such a barrier to improvement, that most of the Pentium-compatible CPU's (NextGen Nx586, AMD K5, Cyrix M1, and even the Pentium's replacement, a 2-chip-14-stage CPU/Cache module named 'P6', do not even clone the Pentium, but emulate it with specialized hardware decoders which convert Pentium instructions to RISC-like instructions which are executed on specially-designed superscalar RISC cores, actually faster than the Pentium itself. One rumour has it that IBM is developing hardware to translate Pentium instructions to run on a PowerPC CPU called the 615.' So maybe all the "modern" enhancements to x86 are just workarounds to the CPU's limitations.
-
samo79 wrote:
Hei Ikir ma come faranno sti poveretti a leggere i topic italiani nei nostri siti ?? :-D :-D ....
Mi sa che dobbiamo cominciare a tradurre le pagine! :-? :-?
Ciao
Troppo lavoro :-)
-
AmigaPro wrote:
I think its interesting how poorly all of these chips do, both AmigaOne and x86, compared to the venerable old CS-PPC...
http://amigapro.com/Images/280.gif
The old CS-PPC (mine is slightly oc'd there, to 66/280MHz) is just under half as fast in OGR as the AmigaOne G4/800, and it completely buries the Athlon per MHz. The 604e has a better FPU than both chips as well.
IPC(instruction per cycle) alone doesn’t rescue a CPU’s fortunes…
One should be focuing on IPC x Mhz = Performance.
PS; Athlon XP with the Barton Core should improve the IPC side of things.
-
>Does any of this x86 vs PPC matter?
Well, somebody made a comparison with Pentium 4 2.4Ghz and X86 in general.
-
Focal_Loco wrote:
Hello! I am not much of a power user, nor do I consider myself a knowledgeable computer user, but I found this intersting bit of information on Digita's Organizer for PC, on an article entitled "Evolution of the CPU" 'The Pentium ... 'interestingly, the awkward and old architecture is such a barrier to improvement, that most of the Pentium-compatible CPU's (NextGen Nx586, AMD K5, Cyrix M1, and even the Pentium's replacement, a 2-chip-14-stage CPU/Cache module named 'P6', do not even clone the Pentium, but emulate it with specialized hardware decoders which convert Pentium instructions to RISC-like instructions which are executed on specially-designed superscalar RISC cores, actually faster than the Pentium itself. One rumour has it that IBM is developing hardware to translate Pentium instructions to run on a PowerPC CPU called the 615.' So maybe all the "modern" enhancements to x86 are just workarounds to the CPU's limitations.
The K7 and K8 families does that trick (i.e. breaking down the X86 instructions into RISC like instructions before feeding into the main pipeline stages).
IBM's PPC 970 also has decoding/crush stage for PPC instructions.
-
The old CS-PPC (mine is slightly oc'd there, to 66/280MHz) is just under half as fast in OGR as the AmigaOne G4/ 800, and it completely buries the Athlon per MHz. The 604e has a better FPU than both chips as well.
What matters in the end is that the G4 800mhz, is 2.6 times faster than your 604 280mhz.
better IPC dont always equal better overall performence.
-
Karlos wrote:
All this fretting over what the x86 can do faster and vice versa is just silly.
Lets get smart, have fun and bash out some nice shiny new apps for these machines, rather than running synthetic benchmarks and whingeing ;-)
Maybe it's just me, but this makes a lot of sense!
Anyhow, the new AOS4.0 kernal is going to be bigger, because it has more capability AND more instructions, due to being RISC, instead of CISC, So, we really have NO idea of real world performance. Even that benchmark doesn't necessarily reveal anything, because it was using Linux after all. BUT, RC5-72, at a binary level, is probably almost the same in Linux or AOS4.0. Only, the external actions of the OS's determine how many cycles are taken away from the OGR's or RC5s calculated. So we wait.....
Amiga! "Last one out, turn off the monitor," says Amy, the mascot.
-
This should be interesting:
http://www.math.wm.edu/~cahunter/NASA_G4_Study.pdf (http://www.math.wm.edu/~cahunter/NASA_G4_Study.pdf)
-
Gobsmacked ! :-o
I did wonder when I saw the indifferent scalar results, but when I saw the Altivec enhanced results, me jaw dropped !!
I am now even more excited about my G4-800 now !
I expect 3D-games will bloody well FLY on AmigaOne now ! (when it is an amiga that is - ie: OS4 on AmigaOne) I imagine any G4 equipped Pegasos'll fly too, but thats a little OT here ! ;-)
I cant wait to run a Mandelbrot generator on it ! (small hobby of mine :-P )
-
Sorry to be a downer, but not everything can be Altivec optimised.
OTOH, software 3D geometry calculation may be a good canditade, but remember that a good 3D hardware geometry accelerator will probably beat Altivec hands down for this kind of stuff.
Hopefully, Warp3D Nova will address this and we can all look forward to highly accelerated games.
Personally, I'm quite excited at the thought of Altivec optimised softsynths / streaming codecs ;-)
-
AmigaOne really kicks asses!!! :-) :-) :-D :-D :-o :-o
-
Has anyone yet run SPECmark, Bytemark or any other little tougher tests?
How does the L2 and L3 cache show up in those benchmarks?
Amiga 68k benchmarks in PPCLinux UAE would be also nice...
-
Karlos wrote:
Hopefully, Warp3D Nova will address this and we can all look forward to highly accelerated games.
Personally, I'm quite excited at the thought of Altivec optimised softsynths / streaming codecs ;-)
Yeahhhhhhhhhhhhh 8-)
-
@Helgis75 I like the way you put ass in plural. Im likely to agree. AOne is gonna kick a very wide spectra of ass :-D
While hackling about the difference between x86 and g4 is somewhat fun, I think, in the light of recent events, that a comparison between G3 and G4, AmigaOne and Pegasos is a topic with much more meat on
-
The glory of AmigaOne will only be realised when moto get off of their buts and make the 1.4 GHz available.
Or rather, to heck with them, the 970 from IBM @1.8 GHz!!!
Unless the producers of our neurons give us something we can run with, we'll glow rather than shine!!
Amiga! Has 2 legs in a tree legged race. :-(
-
It´s the os that LIMITS the speed of the computer not the other way around, winXP is just as slow on my duron 128mb as win95 was on my p60 32mb.
what good is a faster processor if the os keeps getting slower?
It´s the speeds of AmigaOS4 or Morphos versus windows or linux or macos thats important.
-
Really AmigaONE G4-800Mhz wins over P4 2,6Ghz, this without level3 cache?
if it where only true...
lets get real for a change... 133mhz bus .v. 533mhz bus....
PunyPC cant beat a paper bag... it's crap... my P4 2.4/533fsb stomps dual G4 1.4's in Lightwave render benchmarks accross the boards...
-
I dont know why pee cee,s of crap are even mentioned here there not worthy of being associated with the almighty Amiga we kick there ass in everything pee cees is just like having a mini then putting a 30,000 hourse power engine in and using a rubberband as the drive shaft still go no where !!!!!!!!!
AMIGA WE ARE THE BEST !!!!!!!!!!!!!!
-
mips_proc wrote:
Really AmigaONE G4-800Mhz wins over P4 2,6Ghz, this without level3 cache?
if it where only true...
...
But it IS true. On those tiny little tings, like OGR/RC5. ;-)
-
Bytemark test on the A1G3SE by some happy A1 user:
http://personal.inet.fi/cool/pekosbil/a1benchmarks.htm (http://personal.inet.fi/cool/pekosbil/a1benchmarks.htm)
-
Useless Information below, I know, but worth a look.
AthlonXP 2400+ (2Ghz), Windblows XP Pro, normal startup stuff running, browsers etc.
dnetc v2.9001-477-GTR-02111118 for Win32 (WindowsNT 5.1).
Please provide the *entire* version descriptor when submitting bug reports.
The distributed.net bug report pages are at http://www.distributed.net/bugs/
Using email address (distributed.net ID) 'krashamiga@austarmetro.com.au'
[Feb 24 07:00:55 UTC] Automatic processor detection found 1 processor.
[Feb 24 07:00:55 UTC] OGR: Loaded 25/25-14-9-4-15-1 (1.75 Gnodes done)
[Feb 24 07:00:55 UTC] OGR: 23 packets remain in buff-in.ogr
[Feb 24 07:00:56 UTC] OGR: 0 packets are in buff-out.ogr
[Feb 24 07:00:56 UTC] 1 cruncher has been started.
[Feb 24 07:01:04 UTC] #1: OGR:25/25-14-9-4-15-1+6-54-21-45 [1,871,285,442]
[Feb 24 07:01:04 UTC] OGR: using core #0 (GARSP 5.13-A).
[Feb 24 07:01:23 UTC] OGR: Benchmark for core #0 (GARSP 5.13-A)
0.00:00:16.96 [15,349,112 nodes/sec]
[Feb 24 07:01:23 UTC] OGR: using core #1 (GARSP 5.13-B).
[Feb 24 07:01:43 UTC] OGR: Benchmark for core #1 (GARSP 5.13-B)
0.00:00:16.92 [15,452,267 nodes/sec]
But, with a G4 @ 800Mhz getting 11,000,000+ nodes is pretty impressive indeed. That surely ends the argument of "More Mhz means faster/better" ;)
-
But, with a G4 @ 800Mhz getting 11,000,000+ nodes is pretty impressive indeed. That surely ends the argument of "More Mhz means faster/better" ;)
AMD has stated instructions per cycle x clock speed = performance.
"More Mhz means faster/better" is dependant on the targeted market.
For example, Pentium 4’s target market is well suited for the entertainment market i.e. play games with very high FPS such as in Quake III, Unreal Tournament and 'etc'. AthlonXP/NForce2 combo should be competitive in that regard.
One should be focusing on end product’s performance and price. “Instructions per cycle” by itself is pretty point less. It’s just an academic exercise.
-
ksk wrote:
Has anyone yet run SPECmark, Bytemark or any other little tougher tests?
How does the L2 and L3 cache show up in those benchmarks?
Just get some late model Power Macintosh’s benchmarks.
-
ksk wrote:
mips_proc wrote:
Really AmigaONE G4-800Mhz wins over P4 2,6Ghz, this without level3 cache?
if it where only true...
...
But it IS true. On those tiny little tings, like OGR/RC5. ;-)
On the other tiny little tings e.g. OpenSSL, the outcome is quite different for PPC.
The ideal benchmark should be a multi-combo benchmarks.
For OpenSSL benchmarks refer to
http://www.macosxhints.com/article.php?story=20020113045343563
-
Hammer wrote:
ksk wrote:
Has anyone yet run SPECmark, Bytemark or any other little tougher tests?
How does the L2 and L3 cache show up in those benchmarks?
Just get some late model Power Macintosh’s benchmarks.
After bbrv's recent hints about A1G4XE's pathetic performance, I want to see benchmarks run on the A1. (but also benchmarks run on MOL should be ok)
BTW: I think this benchmark digs out some possible memory bus bottlenecs: http://www.netlib.org/linpack/ (http://www.netlib.org/linpack/)
The JAVA version: http://www.netlib.org/benchmark/linpackjava/ (http://www.netlib.org/benchmark/linpackjava/)
(some a bit offtopic PPC 970 talk that I just saw) (http://www.geek.com/news/geeknews/2002Dec/bch20021226017922.htm)
-
Hammer wrote:
...
For OpenSSL benchmarks refer to
http://www.macosxhints.com/article.php?story=20020113045343563
:-o :-o :-o :-o :-?
-
After bbrv's recent hints about A1G4XE's pathetic performance,
Any numbers?
I want to see benchmarks run on the A1. (but also benchmarks run on MOL should be ok)
What for? Does Pegy run "AmigaOS 4.0"?
I have enough Linux boxes inside my house..
“AmigaOS 3.9” compatibility can be obtain via the cheaper X86 based solutions (still waiting for Berniethlon).
The JAVA version: http://www.netlib.org/benchmark/linpackjava/
Dependant on the Java virtual machine. Good for benchmarking between Java VMs btw.
-
ksk wrote:
Hammer wrote:
...
For OpenSSL benchmarks refer to
http://www.macosxhints.com/article.php?story=20020113045343563
:-o :-o :-o :-o :-?
Its just an illustration on how some benchmarks can change a particular outcome. OpenSSL benchmark doesn’t tell the real picture except for OpenSSL based tasks.
Ideally, one should use a combo benchmarks.
Oracle or MySQL benchmarks would be nice for my case.
-
Hammer wrote:
After bbrv's recent hints about A1G4XE's pathetic performance,
Any numbers?
No.
Just that Peg1G3@600 "performs better than" A1G4XE.
I want to see benchmarks run on the A1. (but also benchmarks run on MOL should be ok)
What for? Does Pegy run "AmigaOS 4.0"?
Nope.
But I love using AmigaOS and I have load of AOS3.x apps.
I want to continue using AmigaOS like OS.
Peg1 with MOS is one of the options.
“AmigaOS 3.9” compatibility can be obtain via the cheaper X86 based solutions (still waiting for Berniethlon).
I woud like to get rid of my noisy x86 kludge HW.
But yes, Umilator would be also one option, if it existed.
The JAVA version: http://www.netlib.org/benchmark/linpackjava/
Dependant on the Java virtual machine. Good for benchmarking between Java VMs btw.
But when using identical JVM, it's also good for benchmarking system performance, right?
**************
Update:
Example of Linpack results on x86:
http://www.tech-report.com/reviews/2002q4/athlonxp-2800/index.x?pg=3
I would love to see A1G4XE somewhere between those graphs...
http://www.tech-report.com/reviews/2001q2/pentium4-1.7/index.x?pg=4
-
Just that Peg1G3@600 "performs better than" A1G4XE.
Need more than that i.e. refer AMD's benchmark disclosures as an example.
I woud like to get rid of my noisy x86 kludge HW.
Would that change the fortune of AmigaOS significantly?
But when using identical JVM, it's also good for benchmarking system performance, right?
Some of benchmarks doesn't reveal Java VM type.
Refer to http://www17.tomshardware.com/graphic/20021218/vgacharts-05.html
Athlon XP/nForce VS Pentium 4 in regards to entertainment market segment.
-
ksk wrote:
BTW: I think this benchmark digs out some possible memory bus bottlenecs: http://www.netlib.org/linpack/ (http://www.netlib.org/linpack/)
The JAVA version: http://www.netlib.org/benchmark/linpackjava/ (http://www.netlib.org/benchmark/linpackjava/)
Using IE 6.0 SP1(via MS Java VM), WinXP Pro SP1, Athlon XP 1800+(1.5Ghz), ASUS A7N8XD-D. This test box delivers 127.795 Mflops.
(Point scoring can be increased by switching to ram's bandwidth to PC2700 (with suitable 333Mhz FSB CPU), I’m currently using PC2700 DDR RAMs @ PC2100 speeds).
PC3200 could be use for 400Mhz FSB Athlon XP (o/c)for the maximum potential of EV6 bus).
This test box has beaten Ron Buss's "124 Mflop/s; AMD XP 2700+; 2/10/03" results.
I don't know what has happen to Ron Buss's Athlon XP 2700+ box... It should be faster compared to 266Mhz FSB class 1.5Ghz Athlon XP CPUs .
I may try Sun's Java VM later.
PS; I'm well aware of IBM's PPC 970 (release @ 2003 H2). I don't know why you have posted this? This product has yet to materialized in the market place.
Secondly, the IBM PPC 970 has competitions from Intel’s IA-64, Intel’s Pentium 4 (Prescott core) and AMD’s X86-64.
Noted that IBM PPC 970's has an estimated of 55 million transistors (similar to Pentium 4 and Athlon XP (Barton core). So much for peace and quiet.
-
post gone to bitheaven...
-
My C64 runs OS4 natively...:P
:)
-
I must not get sucked into this thread....I must not get sucked into this thread...
D'Oh!
ksk wrote:
mips_proc wrote:
Really AmigaONE G4-800Mhz wins over P4 2,6Ghz, this without level3 cache?
if it where only true...
...
But it IS true. On those tiny little tings, like OGR/RC5. ;-)
The thing is, for whatever reason, the P4 does not implement certian (or maybe all) integer rotate instructions. Such things end up implemented as shifs and or'ing etc.
The RC5 client makes heavy use of these and as such, the PPC which is top dog at shift/rotate anyway, will cope rather better. Im not sure for certian, but if altivec is able to perform these instructions on sets of integer data in parallel, you can expect a big performance increase.
The thing is that these are synthetic tests. The P4 at the same clockspeed as a G4 will probably outperform it in other areas, especially in memory intensive code.
I really do think it's in IBM / Motorola's interest to haul ass and start supporting technology like DDR memory. Before anybody says that isn't a performance requirement (which frankly I doubt), just think about the price alone - DDR memory is getting cheaper than PC133 memory, simply because the PC market moves ahead so quickly and there is more demand.
What's more, the DDR capable PPC's would have less dependency on large L3 caches which would lower overall system prices further.
-
Now A1SE and A1XE tested side by side:
http://personal.inet.fi/cool/pekosbil/a1benchmarks.htm (http://personal.inet.fi/cool/pekosbil/a1benchmarks.htm)
Soon also on pegasos....
(please notice that those test are not done by me (so, all credits to Pekka, please ;-) ) and that they are no way scientificly accurate, but they should give some rough hints about the performance)
( after recompiling with motorolla math libraries, it seems that PPC performance jumped up 300% in some cases) :-)
-
Just for reference..
I made my own math benchmark and compared the 68060 with the creme de la creme of Intel and AMD.
The result was that the 68060 could only run with the pedal to the metal (1 IPC) in one case (tested 1+1, 1+0, X+1,(X+Y)*z, etc). I also tested FP, Integer and fast FP emu. There was also a test where random data in three buffers (far larger than any of the processors cache) was processed against eachother. Both the AMD and the Intel performed flat out in all cases (same result no matter how advanced the exercise was). The 68060 was left far, far behind in the IPC race. Make no mistake: neither Intel nor AMD produce crap.
I'm back by the way. Well, sort of.
-
Like anyone actually cared. Perhaps I should give it the day, or so.
Just spotted my own tagline (silent scream). That's what I get for being away all this time :-(
-
While Pekka is waiting for the results of peg1, is there anywhere a pegasos benchmark page?
-
I have a few bench marks of my own A1200 crashes 6 to 7 times a year ibook about 2 times a week pc 6 to 7 times a day !!!!! :-D :-D :-D :-D :-D :-D :-D :-D :-D :-D :-D :-D :-D :-D :-D :-D :-D :-D :-o :-o :-o :-o :-o :-o ;-) ;-) ;-) ;-)
-
Mine crashed last night :-(
Was programming at the time and lost some work :-x
I wouldn't mind, but it was AmigaAmp that trashed the exec memory list (no idea why) - I hadn't even compiled my code.
Still, first time in 5 months :-)
-
poweramiga2002 wrote:
I have a few bench marks of my own A1200 crashes 6 to 7 times a year ibook about 2 times a week pc 6 to 7 times a day !!!!! :-D :-D :-D :-D :-D :-D :-D :-D :-D :-D :-D :-D :-D :-D :-D :-D :-D :-D :-o :-o :-o :-o :-o :-o ;-) ;-) ;-) ;-)
Today I received a rather bizarre crash message on my Windoze machine. In the dialog box it stated that Save had crashed? What the hell is Save? LOL, and no I was not trying to save anything it just flashed before my eyes! Weird :crazy: .
-
More results:
http://personal.inet.fi/cool/pekosbil/a1benchmarks.htm (http://personal.inet.fi/cool/pekosbil/a1benchmarks.htm)
(do not give me the credit of those benchmarks, please, I'm not the author)
Also pegasos results exist, but some of them will not be published untill a few unclear bits are sorted out.
( UPDATE: Also Pegasos results exist now. More tests are in development (most likely also with Altivec). )
regards
ksk
-
Adobe's Great Render Race (PIV vs Dual-G4).
=======================================================================
Dell's Pentium IV 3.06Ghz VS Apple's 1.25Ghz Dual-Proprocessor Power Macintosh G4.
Adobe's applications used in benchmarking are;
+ After Effects®
+ Illustrator®,
+ Photoshop®
Click here for more info. (http://www.adobe.com/motion/pcpreferred.html)
Just a side note; Intel seems to have embedded X86-64 features into the Prescott/Nocona core.
Click here for more information. (http://www.chip-architect.com/news/2003_03_26_Prescott_clues_for_Yamhill.html)