Power Mac G5 Quad Benchmarks

melgross · Nov 10, 2005

p0intblank said:
How do you know all this? Do you have a Quad? If these results are true.. my God.

Yeah, but take dual Athlons and you can cut those numbers almost in half as well.

2.00 =1:10
1:15 =0:45

What would a Quad Opteron do with 4 channels to memory from the built-in controller?

GregA2 · Nov 10, 2005

My Mac Mini 1.42 scored a 37.75 on Xbench

Pie · Nov 10, 2005

Wrong Wrong Wrong

risc said:
Using which version of XBench? There is no way you got over 100 using the current version when the baseline 100 points is a dual 2 GHz G5 running Tiger.

http://www.xbench.com/Xbench_1.2.dmg

Hello everybody... these are my benchmarks. The machine arrived tuesday 8th november 5.00pm UK time.

I DID USE THE LATEST VERSION OF XBENCH (1.2)

How does this thing feel? Well, it's relative. At home i've been using a 15" 1.25 powerbook and in work i use a dual 1.25 G4 powermac.

This thing feels lightening. there's 3BG Ram in it. I will be using it mainly for ProTools LE usage. I'm waiting for the upgrade to 7 (maybe weekend before I can give it a proper shake down)

BakedBeans · Nov 10, 2005

Pie said:
Hello everybody... these are my benchmarks. The machine arrived tuesday 8th november 5.00pm UK time.

I DID USE THE LATEST VERSION OF XBENCH (1.2)

How does this thing feel? Well, it's relative. At home i've been using a 15" 1.25 powerbook and in work i use a dual 1.25 G4 powermac.

This thing feels lightening. there's 3BG Ram in it. I will be using it mainly for ProTools LE usage. I'm waiting for the upgrade to 7 (maybe weekend before I can give it a proper shake down)

ay chance of a cinebench test bud?

Pie · Nov 10, 2005

BakedBeans said:
ay chance of a cinebench test bud?

Sure...

CINEBENCH 2003 v1
****************************************************

Tester : Pie

Processor : Quad G5
MHz : 2.5 GHZ
Number of CPUs : 4
Operating System : 10.4.2

Graphics Card : GeForce 6600
Resolution : <fill this out>
Color Depth : <fill this out>

****************************************************

Rendering (Single CPU): 359 CB-CPU
Rendering (Multiple CPU): 1016 CB-CPU

Multiprocessor Speedup: 2.83

Shading (CINEMA 4D) : 353 CB-GFX
Shading (OpenGL Software Lighting) : 1051 CB-GFX
Shading (OpenGL Hardware Lighting) : 1871 CB-GFX

OpenGL Speedup: 5.29

****************************************************

Does this make any sense to anybody?

pknz · Nov 10, 2005

I just did a Xbench test on my Rev B iMac and got 70.93, so I'm guessing the PowerMac Quad is quite a substantial increase in XBench Performance not just on paper.

Although how credible Xbench tests are is another story.
Edit: Just redid the test with Safari open and got a differing score, although I gues that can be expected

jhu · Nov 10, 2005

AidenShaw said:
LINPACKD doesn't use AltiVec...

if you go by the ground rules they've laid out on top500.org, you can't change the source code, but you can use compiler optimizations to achieve vectorization (i don't know how vectorizable the code is). the second test, 1000d, seems to be more flexible, so perhaps explicit vector optimizations can be coded in.

50548 · Nov 10, 2005

Seems pretty darn fast, even though I don't trust in XBench...

Just as an idea, my Rev. B iMac G5 (see signature) in Automatic speed setting and running Safari and Network Utility gave a score of 67.93...

obione · Nov 10, 2005

PowerMac 2.3 dc

Cinbench. xCPU =580
Photoshop test =0:45 sec
xBench =110

nxent · Nov 10, 2005

b.k.jackson said:
For comparison, my PowerBook (15" Alu) 1.25 G4 got a 102.80

Brian

102.8 is based on the old xbench system... they recalibrated everything so a dual 2Ghz G5 was about 100 (something like that), so 151.8 should be pretty fast-

Mr. Anderson · Nov 10, 2005

Pie said:
Sure...

CINEBENCH 2003 v1
****************************************************

Tester : Pie

Processor : Quad G5
MHz : 2.5 GHZ
Number of CPUs : 4
Operating System : 10.4.2

Graphics Card : GeForce 6600
Resolution : <fill this out>
Color Depth : <fill this out>

****************************************************

Rendering (Single CPU): 359 CB-CPU
Rendering (Multiple CPU): 1016 CB-CPU

Multiprocessor Speedup: 2.83

Shading (CINEMA 4D) : 353 CB-GFX
Shading (OpenGL Software Lighting) : 1051 CB-GFX
Shading (OpenGL Hardware Lighting) : 1871 CB-GFX

OpenGL Speedup: 5.29

****************************************************

Does this make any sense to anybody?

Now we need to get someone with a dual 2.7 to run the same thing - anyone?

D

dontmatter · Nov 10, 2005

iGary said:
For example:

Gary Reich's iMac G5 156.42 iMac (Flat Panel) 2004-11-02 14:55:43.0
Gary Reich's iMac G5 152.28 iMac (Flat Panel) 2004-12-27 13:45:22.0
Gary Reich's iMac G5 155.70 iMac (CRT) 2005-03-04 10:32:45.0
Gary Reichs Computer 69.78 PowerMac G4 (Sawtooth) 2003-11-20 11:25:21.0
Gary Reichs Computer 135.86 PowerMac G4 (MDD) 2003-11-20 15:04:31.0
Gary Reichs Computer 134.24 PowerMac G4 (MDD) 2003-11-20 15:08:53.0
Gary Reichs Computer 135.84 PowerMac G4 (MDD) 2003-11-20 15:59:39.0
Gary Reichs Computer 135.84 PowerMac G4 (MDD) 2003-11-20 16:03:13.0
Gary Reichs Computer 137.33 PowerMac G4 (MDD) 2003-11-20 16:38:36.0
Gary Reichs Computer 130.30 PowerMac G4 (MDD) 2003-11-24 20:11:24.0
Gary Reichs Computer 69.13 PowerMac G4 (Sawtooth) 2003-12-04 12:44:59.0
Gary Reichs Computer 141.95 PowerMac G4 (MDD) 2004-03-16 18:32:06.0
Gary Reichs Computer 141.16 PowerMac G4 (MDD) 2004-09-15 17:06:19.0
Gary Reichs Computer 139.72 PowerMac G4 (MDD) 2004-10-04 17:37:37.0
Gary Reichs Computer 140.87 PowerMac G4 (MDD) 2004-10-07 14:17:39.0
Gary Reichs Computer 101.30 iBook (White) 2004-10-29 12:20:46.0
Gary Reichs Computer 101.82 iBook (White) 2004-10-29 12:29:32.0
Gary Reichs Computer 101.94 iBook (White) 2004-11-02 12:25:53.0
Gary Reichs Computer 135.05 PowerMac G4 (MDD) 2004-12-15 19:47:10.0
Gary Reichs iMac G5 156.98 iMac (Flat Panel) 2005-05-09 12:41:24.0
Gary Reichs Power Mac G4 137.04 PowerMac G4 (MDD) 2005-05-02 20:18:02.0
Gary Reichs Power Mac G5 267.48 PowerMac G5 (Orig) 2005-09-01 17:50:31.0
Gary Reichs Power Mac G5 268.28 PowerMac G5 (Orig) 2005-08-13 12:52:26.0
Gary Reichs Power Mac G5 269.37 PowerMac G5 (Orig) 2005-07-14 18:22:15.0
Gary Reichs Power Mac G5 271.43 PowerMac G5 (Orig) 2005-07-16 03:49:40.0
Gary Reichs Power Mac G5 260.94 PowerMac G5 (Orig) 2005-07-29 16:09:26.0
Gary Reichs Power Mac G5 124.87 PowerMac G5 (June 2004) 2005-11-10 12:02:37.0

Holy crap, you need a less computer centric world. That's a LOT of computers. Even if you bought at a good deal and sold at a good price.

Pie · Nov 10, 2005

Horse Photohop Speed Test

Right, I've just run the Horse Photohop Radial Blur Speed Test.

Result...

19.2 seconds

AidenShaw · Nov 10, 2005

Kairy said:
Apple has always adopted new technology , and dragged the entire market after them. (scsi, firewire, pci-e, ...)

SATA? PCs first. DDR? PCs first. DDR2? PCs first. USB? PCs first. USB2? PCs first. ADC? Nobody else used it.

PCIe? Apple is last with PCI Express, it's been common in PCs for a year or more. (Apple did drop PCI, though, much to the chagrin of people with an investment in PCI cards and no PCIe replacements around.)

1394? Never really caught on in the PC market, and now even Apple is giving it a cool shoulder.

AidenShaw · Nov 10, 2005

jhu said:
if you go by the ground rules they've laid out on top500.org, you can't change the source code, but you can use compiler optimizations to achieve vectorization (i don't know how vectorizable the code is). the second test, 1000d, seems to be more flexible, so perhaps explicit vector optimizations can be coded in.

It has nothing to do with the rules - AltiVec simply doesn't support 64-bit floating point. (SSE does, however)

sephirot · Nov 10, 2005

Other cinebench marks:

http://www.barefeats.com/g5c.html

Dual G5 2.5 Ghz -- 633
Dual Opteron 2 Ghz -- 523

See yourself.

Andy

PS: Man, where is my Quad G5? Still shows a shipping date of 30. Nov. I want it now!!

steve_hill4 · Nov 10, 2005

Not that this is too important in the grand scheme of things, but I just tested my AMD on CineBench and got the following:

CINEBENCH 2003 v1
****************************************************

Tester : Stevie

Processor : AMD Athlon 2600+
MHz : 2.09GHz
Number of CPUs : 1
Operating System : Windows XP Professional

Graphics Card : GeForce 4600 128MB
Resolution : 1280x1024
Color Depth : 32-bit

****************************************************

Rendering (Single CPU): 210 CB-CPU
Rendering (Multiple CPU): --- CB-CPU

Shading (CINEMA 4D) : 241 CB-GFX
Shading (OpenGL Software Lighting) : 165 CB-GFX
Shading (OpenGL Hardware Lighting) : 102 CB-GFX

OpenGL Speedup: 0.68

****************************************************

Not sure what people would think of this 2.5 year old AMD system I was running it on, especially as I was actually running more than I would normally, but I am still impressed to this day what it can cope with. Comparing the rendering alone with the 2.13GHz Pentium-M, (which scored 283), I don't think AMD fare too badly, do they? I seriously hope Apple consider them in a few years if intel don't meet the standard we hope they do.

Peace · Nov 10, 2005

Don't know if it's already been said because I skipped to the last page but.
Keep in mind XBench 1.2 is a universal binary.
It's built to run on Intel and Mac so there "could be" some errors.

macpro2000 · Nov 10, 2005

I know that the whole Xbench thing surely has to be messed up, otherwise it's basically close to 50% increase in speed over a Dual 2 (4GHz). Now linearly only a Dual 3 (6GHz) would be 50% faster than the Dual 2...so surely Quad 2.5s (10GHz) would be more than 50% faster.

melgross · Nov 10, 2005

Peace said:
Don't know if it's already been said because I skipped to the last page but.
Keep in mind XBench 1.2 is a universal binary.
It's built to run on Intel and Mac so there "could be" some errors.

I think that there were *always* errors.

Peace · Nov 10, 2005

melgross said:
I think that there were *always* errors.

Hey! Just tryin to be nice

shewhorn · Nov 10, 2005

PS CS2 benchmark for a PC

Hi folks, I'm excited to see if the architecture of Photoshop is such that it can take advantage of more than two processors. So far it would appear that with this one filter the answer is definitely yes. I would also be interested in knowing what the performance is for single threaded filters? Does the increased scheduling load of 4 processors actually slow down a single threaded filter relative to a dual 2.5 PM (I doubt it, scheduling is a problem as you add more processors but 4 processors shouldn't introduce much of a burden at all), will there be a huge difference in performance (I guessing no for single threaded filters) or will it be slightly faster (this is my expectation due to the faster memory bus, you'll see slightly faster results for single threaded apps)?

Anyhow for those interested here's another data point.

Pentium 4 @ 3.2 GHz (Prescott D stepping), clocks in at 61.4 seconds.

Cheers, Joe

Fiveos22 · Nov 10, 2005

Pie said:
Right, I've just run the Horse Photohop Radial Blur Speed Test.

Result...

19.2 seconds

Damn.

rickvanr · Nov 10, 2005

For reference, my Dual 2.7 scores 130+ (http://ladd.dyndns.org/xbench/merge.xhtml?doc2=137613)

Also note the Quad that was tested had 8GB of memory, and memory speeds up Xbench tests.

Mr. Anderson said:
Now we need to get someone with a dual 2.7 to run the same thing - anyone?

D

I'll do it..

Lacero · Nov 10, 2005

Fiveos22 said:
Damn.

Yeah. Looks like once we get quad 3.0Ghz machines, I'm upgrading my current Power Mac. That I promised myself.

However, with the intels on their way, my plans for upgrading have been thrown a monkey wrench. What to do? What to do.

Power Mac G5 Quad Benchmarks

macrumors 6502

macrumors member

macrumors newbie

macrumors 68040

macrumors newbie

macrumors 68020

macrumors 6502a

Guest

macrumors newbie

macrumors 6502

Moderator emeritus

macrumors 6502a

macrumors newbie

macrumors P6

macrumors P6

macrumors newbie

macrumors 68000

Cancelled

macrumors 65816

macrumors 6502

Cancelled

macrumors newbie

macrumors 65816

macrumors 68040

macrumors 604

Our Staff