All Macs In-Depth Tests
 
Books & Buyer's Guides

Information Hubs




Other Product Information




Site Supporters

Apple iPhone Prices At ...
Apple Store
Canada Apple Store
Cingular Wireless
iPhone InfoZone

Using the links above supports MacReviewZone!

send this page

Send to a friend

News Feed

Mailing list ...List information...

Latest Discussion Threads
MacSpeedZone Gift Shop
The SpeedZone: Top Of The Line Tower Doesn't Live Up To Its Potential!... Yet - 35 Performance Test Results For the Dual Processor G5/2.0 GHz Tower
Check prices at: ....

Apple Store
Education Discounts
Closeout/Refurb
Amazon.com
PowerMax

MacConnection
 MacZone
MacMall
ClubMac

Additional Resources

Power Mac Info Zone
Power Mac Reviews
 Power Mac Performance
Power Mac Performance vs Others
 Power Mac Specs & Features Page
Power Mac Discussion forum
Performance Details

11/7/03

by David Engstrom

Power Mac Buyer's Guide

It seems clear that if you want to take advantage of all the power that the dual processor G4/2.0 GHz Tower makes available to you, you need to have a strategy. You either will need to be running specialized applications that will soak up all that processing capacity, (even Apple's in-house applications that we tested did not do that ... leaving up to 50% of the machine processing power on the table .. unused), or you will need to have multiple applications, running processor intensive activities, at the same time. Only under those circumstances, will you find the activity indicators in CPU Monitor, parked at the top of the scale.

Here is some of what we found about the dual processor machine:

  • In our Photoshop tests only 50% to 75% of the processing capacity was used
  • Microsoft Office X does not appear to take advantage of dual processors
  • Our iMovie tests showed that only 65% of the processing capacity was used
  • QuickTime tests showed between 50-60% usage

We had expected that, now that the System Bus had been sped up to half the processor speed, we would see the bottleneck of getting data to the processors fast enough to keep them properly fed, would be gone. Apparently not.

Since it is possible to keep the processors saturated, by either writing specialized applications, or portions of applications that can do this, or by running multiple processes at once, It doesn't appear that there is a hardware bottleneck, but that the constriction must be at the application level. Still it is a bit of a disappointment to see half your beast licking its fur.

For the professional crowd, who know their applications and what those applications can do on high-powered multiple processor fitted out machines, well you know if the dual processor G5 will work for you.

The machine will also provide much more flexibility in a production type environment. You should note that all the tests mentioned above maxed out the processors of the 1.6 and 1.8 GHz G5s.

We also saw some dramatic improvement in some gaming performance (but not all), when using the dual processor machine. Again it really seemed to depend on if the game was up to taking advantage of all the machine's resources.

So what is the conclusion? The dual processor G5 may be a bit before its time. Outside of professionals and those in a production type environment, whose workflow is capable of maxing out a dual processor machine's potential, users may want to wait on purchasing the top-end G5 ... At least until it becomes clearer that your applications have moved forward to take advantage of all that power.

We are not saying that the dual processor G5 is not faster. Across the board it is faster than the 1.8 GHz (as it should be), and in some cases it is significantly faster. It is just that it should, could and perhaps will be faster at getting work done for the average user, than it is at the moment.

In the meantime it seems that in many applications one of the processors is doing little more than developing a hair ball ... not exactly what we had expected!

Next up, a comparison of the performance of G5s to G4s ....

What machine would you buy?

Difference and similarities of each machine
 

Power Mac G5/1.6 GHz
Single Processor (#M9020LL/A)

Power Mac G5/1.8 GHz
Single Processor (#M9031LL/A)

G5/2.0 GHz
Dual Processor (#M9032LL/A )
Processors 1 x G5/1.6 GHz 1 x G5/1.8 GHz 2 x G5/2.0 GHz
L2 Cache 512K L2 @ 1.6 GHz 512K L2 @ 1.8 GHz 512K L2 @ 2.0 GHz each
L3 Cache None None None
Bus Speed 800 MHZ Frontside bus 900 MHZ Frontside bus 1 GHz Frontside bus
Memory

1.25 GB PC2700
333MHz DDR SDRAM

4 GB max

1.5 GB PC3200
400MHz DDR SDRAM

8 GB max

1.5 GB PC3200
400MHz DDR SDRAM

8 GB max
Graphics Card NVIDIA GeForce FX 5200 Ultra
64MB DDR SDRAM
AGP 8X Pro graphics slot 		NVIDIA GeForce FX 5200 Ultra
64MB DDR SDRAM
AGP 8X Pro graphics slot 		ATI RADEON 9600 Pro
64MB DDR SDRAM
AGP 8X Pro graphics slot
Drive 80GB Serial ATA
7200 rpm 		160GB Serial ATA
7200 rpm 		160GB Serial ATA
7200 rpm
CD Drive

SuperDrive (DVD-R/CD-RW)

Writes DVD-R at 4x
Reads DVDs at 8x
Writes CD-R at 16x
Writes CD-RW at 10x
Reads CDs at 32x

SuperDrive (DVD-R/CD-RW)

Writes DVD-R at 4x
Reads DVDs at 8x
Writes CD-R at 16x
Writes CD-RW at 10x
Reads CDs at 32x

SuperDrive (DVD-R/CD-RW)

Writes DVD-R at 4x
Reads DVDs at 8x
Writes CD-R at 16x
Writes CD-RW at 10x
Reads CDs at 32x

PCI slots Three open full-length 33MHz, 64-bit PCI slots Three open full-length PCI-X slots: one 133MHz, 64-bit slot and two 100MHz, 64-bit slots Three open full-length PCI-X slots: one 133MHz, 64-bit slot and two 100MHz, 64-bit slots
Operating System 10.2.7 10.2.7 10.2.7
Price $1,999 $2,399 $2,999

"Real World" Tests

Basic Processor Function

CineBench 2003 Single CPU Test

Machine Power Mac G5/1.6 GHz Power Mac Single G5/1.8 GHz Power Mac Dual G5/2.0 GHz
Absolute Scores 166 187 221

CineBench 2003 Multiple CPU Test

Machine Power Mac G5/1.6 GHz Power Mac Single G5/1.8 GHz Power Mac Dual G5/2.0 GHz
Absolute Scores 166 187 393

 

CineBench 2003 consists of a series of tests. The tests above, test basic processing performance. As you can see from this processor test, (and the Fractal test below), when you have the combination of a dual processor machine and an application that can take advantage of both processors, you have a recipe for a great deal of productivity. You can run this test on your own machine by downloading the CineBench program. We did not use the beta version of CineBench, which is partially optimized for the G5, to get the results above. We did however run the optimized program on the dual processor machine and obtained the following scores: 287 in single processor mode, and 500 in dual processor mode .... a 27% to 30% performance improvement. (Dual processor usage 100%)

AltiVec Fractal (Carbon)

Machine Power Mac G5/1.6 GHz Power Mac Single G5/1.8 GHz Power Mac Dual G5/2.0 GHz
Absolute Scores - seconds 46.26 41.12 18.63

The Fractal program has been highly tuned to take advantage of the G4. It will also gobble up whatever processing capability is present. This is a good test for assessing the fundamental processing potential of each machine. The program has not yet been tailored to take advantage of the G5. If you would like to compare your machine to these results download the Fractal program and set the fractal to run at a maximum count of 65536 (Dual processor usage 100%)

Desktop Tests    

Re-Boot Test

Machine Power Mac G5/1.6 GHz Power Mac Single G5/1.8 GHz Power Mac Dual G5/2.0 GHz
Absolute Scores - seconds 48.28 42.56 43.38

Time from the twirling clock to the appearance of the hard drive on the desktop

 

Copy 900 MB Folder Containing Thousands Of Files Of Various Sizes

Machine Power Mac G5/1.6 GHz Power Mac Single G5/1.8 GHz Power Mac Dual G5/2.0 GHz
Absolute Scores - seconds 146.64 141.98 133.55

The copy folder test not only stresses drive performance but also takes advantage of faster processing capability & caching schemes. Drive performance appears to be fairly evenly matched. All three machines have the same type of drive. The 1.8 & 2.0 GHz machines' are twice as big as the 1.6 GHz's (Dual processor usage 20%)

Copy Single 1.2 GB File

Machine Power Mac G5/1.6 GHz Power Mac Single G5/1.8 GHz Power Mac Dual G5/2.0 GHz
Absolute Scores - seconds 67.04 71.44 61.97

In the single file test the the processor & caching capability of the machine is less of a factor. The performance differences here are almost within the margin of error (Dual processor usage 7%)

Import QuickTime DV Into iMovie

Machine Power Mac G5/1.6 GHz Power Mac Single G5/1.8 GHz Power Mac Dual G5/2.0 GHz
Absolute Scores - seconds 18.45 16.78 16.11

We imported a 175 MB DV file into iMovie 3. (Dual processor usage 60%)

Burn CD

Machine Power Mac G5/1.6 GHz Power Mac Single G5/1.8 GHz Power Mac Dual G5/2.0 GHz
Absolute Scores - seconds 73.82 75.54 78.04

CD burning only uses a fraction of the processor to get the work done. All three machines have the same SuperDrive, so it is no surprise that they turn in essentially the same score (Dual processor usage 5%)

Let 1000 Windows Bloom

Machine Power Mac G5/1.6 GHz Power Mac Single G5/1.8 GHz Power Mac Dual G5/2.0 GHz
Absolute Scores - seconds 27.20 24.14 21.81

Let 1K Windows Bloom is a simple carbon application that opens and closes 1,000 windows. This gauges how fast screen redraws are taking place. The scores above track the percentage of clock speed improvement exactly. You can download the small program to run on your machine, and compare your scores to the ones here. (Dual processor usage 45%)

Launch Photoshop

Machine Power Mac G5/1.6 GHz Power Mac Single G5/1.8 GHz Power Mac Dual G5/2.0 GHz
Absolute Scores - seconds 13.15 13.50 10.31

We used a 21 MB files to open Photoshop 7 by dropping it on the Photoshop icon. A trial version of Photoshop 7 can be downloaded if you want to run your own test. It appears to have helped a little having the dual processors ... but the few seconds difference are sort of inconclusive.

Finder File Search

Machine Power Mac G5/1.6 GHz Power Mac Single G5/1.8 GHz Power Mac Dual G5/2.0 GHz
Absolute Scores - seconds 36.11 31.31 27.01

Mac OS X's new File Search function relies on both processor and drive performance. In this test we are searching the file names of thousands of items, contained in two folders. About 30% of the processing potential of the 1.6 & 1.8 GHz machine's was utilized during the test. (Dual processor usage 10%)


Large Document & Database Type Tests

Microsoft Office X Word Macro

Machine Power Mac G5/1.6 GHz Power Mac Single G5/1.8 GHz Power Mac Dual G5/2.0 GHz
Absolute Scores - seconds 59.44 53.40 44.84

A Macro (series of complex actions) was run in Microsoft's Word processing program on a 1.6 MB document. Word is part of Office X. For the most part Office cannot take advantage of dual processors. (Dual processor usage 50%)

Microsoft Office X Excel Macro

Machine Power Mac G5/1.6 GHz Power Mac Single G5/1.8 GHz Power Mac Dual G5/2.0 GHz
Absolute Scores - seconds 50.41 44.13 34.91

A Macro (series of complex actions) was run in Microsoft's Excel spreadsheet program, on a 8 MB file. Excel is part of Office X. (Dual processor usage 50%)

AppleWorks Search And Replace Test

Machine Power Mac G5/1.6 GHz Power Mac Single G5/1.8 GHz Power Mac Dual G5/2.0 GHz
Absolute Scores - seconds 39.19 37.28 29.29

This test takes place in a 3 MB AppleWorks document. Machines with larger chip caches usually do better in this test. The 1.8 GHz machine's score seems a little lower than it should be. (Dual processor usage 50%)

Graphics Test

CineBench 2003 4D Test

Machine Power Mac G5/1.6 GHz Power Mac Single G5/1.8 GHz Power Mac Dual G5/2.0 GHz
Absolute Scores 198 226 265

This part of the CineBench program relies on the CPU for graphics rendering (Dual processor usage 45%)

CineBench 2003 Software GL Test

Machine Power Mac G5/1.6 GHz Power Mac Single G5/1.8 GHz Power Mac Dual G5/2.0 GHz
Absolute Scores 468 473 640

This part of the CineBench program relies on the software for graphics rendering (Dual processor usage 45%)


CineBench 2003 Hardware GL Test

Machine Power Mac G5/1.6 GHz Power Mac Single G5/1.8 GHz Power Mac Dual G5/2.0 GHz
Absolute Scores 838 854 1057

This part of the CineBench program relies on the graphics card for graphics rendering. The 1.6 GHz & 1.8 GHz machines have the same graphics card. The dual processor machine has the ATI RADEON 9600 Pro. (download the CineBench program to run on your own machine and compare scores) (Dual processor usage 45%)

Number Crunching & Rendering Tests

Photoshop 7 - BareFeat's Test

Machine Power Mac G5/1.6 GHz Power Mac Single G5/1.8 GHz Power Mac Dual G5/2.0 GHz
Absolute Scores - seconds 36.81 32.28 21.81

BareFeats Photoshop Test Suite - mimics a production type workflow in Photoshop. If you would like to run this test on your machine, and you don't have a copy of Photoshop, download it from Adobe (only advisable if you have a high-speed connection). Then download the action file and drop it on the Photoshop icon (Dual processor usage 75%)

Photoshop 7 - AltiVec Tests

Machine Power Mac G5/1.6 GHz Power Mac Single G5/1.8 GHz Power Mac Dual G5/2.0 GHz
Absolute Scores - seconds 41.71 37.03 24.89

Certain of the functions of Photoshop can take special performance advantage of the G4/G5 processors. This suite of filters was run on a 21 MB file. We did not use the G5 plug-in Adobe has developed but BareFeats tested with and without it and found very little difference in performance (Dual processor usage 75%)

Photoshop 7 - Normal Tests

Machine Power Mac G5/1.6 GHz Power Mac Single G5/1.8 GHz Power Mac Dual G5/2.0 GHz
Absolute Scores - seconds 68.20 59.71 50

The test above consists of Photoshop filters and functions that do not take any particular advantage of the G4/G5. Again we used the 21 MB file. (Dual processor usage 50%)

iMovie - Effect Render

Machine Power Mac G5/1.6 GHz Power Mac Single G5/1.8 GHz Power Mac Dual G5/2.0 GHz
Absolute Scores - seconds 96.11 84.74 63.18

In this test, an Aged Effect is run on a 175 MB DV file using iMovie 3.03, with the Effect left at its default setting This is a processor intensive task and will take advantage of dual processors, if they are present. (Dual processor usage 65%)

Encoding/Decoding Tests

QuickTime 6.3 Encode

Machine Power Mac G5/1.6 GHz Power Mac Single G5/1.8 GHz Power Mac Dual G5/2.0 GHz
Absolute Scores - seconds 23.38 20.20 17.67

A 5MB movie is converted to a QuickTime movie using default setting (Dual processor usage 50%)

QuickTime 6.3 Convert To DV Stream

Machine Power Mac G5/1.6 GHz Power Mac Single G5/1.8 GHz Power Mac Dual G5/2.0 GHz
Absolute Scores - seconds 112.37 99.04 70.59

The 5 MB Movie is converted to a DV Stream ready for import into iMovie. Default setting was used (Dual processor usage 57%)

iTunes 4.0.1 - AAC Encode From CD

Machine Power Mac G5/1.6 GHz Power Mac Single G5/1.8 GHz Power Mac Dual G5/2.0 GHz
Absolute Scores - seconds 93.16 92.92 76.70

All three machines read CDs at 32X. Four songs totaling 16.5 minutes in length were converted in this test. (Dual processor usage 50%)

iDVD Import

Machine Power Mac G5/1.6 GHz Power Mac Single G5/1.8 GHz Power Mac Dual G5/2.0 GHz
Absolute Scores - seconds 39.09 36.29 26.68

175 MB DV file is imported into iDVD ready for burning to a DVD disk (Dual processor usage 75%)

StuffIt Expand

Machine Power Mac G5/1.6 GHz Power Mac Single G5/1.8 GHz Power Mac Dual G5/2.0 GHz
Absolute Scores - seconds 46.56 40.66 33.60

This is straight processor intensive work and does not take any advantage of the G5 processor... faster clock speed means faster performance. This test was performed on a 23 MB StuffIt file that contained thousands of individual files (Dual processor usage 50%)

Multitasking

iTunes AAC Encode + AppleWorks Search & Replace + Finder Copy (Multitasking)

Machine Power Mac G5/1.6 GHz Power Mac Single G5/1.8 GHz Power Mac Dual G5/2.0 GHz
Absolute Scores - seconds 41.93 41.12
-

MP3 Encode, AppleWorks search & replace and folder copy are all carried out at the same time

QuickTime Encode + Fractal (Multitasking)

Machine Power Mac G5/1.6 GHz Power Mac Single G5/1.8 GHz Power Mac Dual G5/2.0 GHz
Absolute Scores - seconds 67.50 61.07 28.74

QuickTime Encode and Fractal render are both carried out at the same time. This test involves heavy processor activity. As you can see if you organize your work to keep processor saturated with data, a dual processor machine can really get a lot more work done than a single processor one. (Dual processor usage 100%)

QuickTime DV + iMovie Effect Render (Multitasking)

Machine Power Mac G5/1.6 GHz Power Mac Single G5/1.8 GHz Power Mac Dual G5/2.0 GHz
Absolute Scores - seconds 209.64 184.80 91.20

QuickTime DV Encode and iMovie Effect are performed at the same time (Dual processor usage 100%)

AAC Encode + Photoshop Normal + Word Macro (Multitasking)

Machine Power Mac G5/1.6 GHz Power Mac Single G5/1.8 GHz Power Mac Dual G5/2.0 GHz
Absolute Scores - seconds 184.72 163.95 80.15

The AAC Encode, Photoshop Normal and Word Macro tests are all run at the same time (Dual processor usage 100%)

Gaming

Giants Highest Frame Rates

Machine Power Mac G5/1.6 GHz Power Mac Single G5/1.8 GHz Power Mac Dual G5/2.0 GHz
Absolute Scores - frames per second 88 106 220

Giants Average Frame Rates

Machine Power Mac G5/1.6 GHz Power Mac Single G5/1.8 GHz Power Mac Dual G5/2.0 GHz
Absolute Scores - frames per second 34 38 82

Giants Lowest Frame Rates

Machine Power Mac G5/1.6 GHz Power Mac Single G5/1.8 GHz Power Mac Dual G5/2.0 GHz
Absolute Scores - frames per second 21 21.5 41

It is the processing capability which is making the difference here. Our testing with the Giants game has shown that it responds well to faster processors, and will take advantage of dual processors if they are present. To obtain frame rates we take the character out for a jog around a predetermined route and measure the frame rates at the same intervals.

4x4 EVO 2 Average Frame Rates- High Quality Setting

Machine Power Mac G5/1.6 GHz Power Mac Single G5/1.8 GHz Power Mac Dual G5/2.0 GHz
Absolute Scores - frames per second 28 29 40

4x4 EVO 2 Average Frame Rates- Medium Quality Setting

Machine Power Mac G5/1.6 GHz Power Mac Single G5/1.8 GHz Power Mac Dual G5/2.0 GHz
Absolute Scores - frames per second 57 58 60

4x4 EVO 2 Average Frame Rates- Low Quality Setting

Machine Power Mac G5/1.6 GHz Power Mac Single G5/1.8 GHz Power Mac Dual G5/2.0 GHz
Absolute Scores - frames per second 61 60 60

4x4 EVO 2 is a very demanding game. It would appear that the graphics card is doing most of the work here

It seems clear from our testing that different games, take various levels of advantage, of the hardware they are running on.

OK you have seen the results .... What machine would you buy?

Top of page

Recent Additions

Dashboard Icon Check out our new dashboard widget tracking site updates and providing quick access to key areas of MacSpeedZone and MacReviewZone!

Reader Specials

firefox Firefox Search Plugins - search this site and others from within Firefox!

HandHelditems.com - Personalize your iPod with us. Shop hundreds of unique iPod accessories and save up to 80%.

Apple Store Apple Store - The size of a pack of gum, iPod shuffle weighs less than a car key. Which means there’s nowhere your skip-free iPod shuffle can’t go. Click Here

Home New & Old Macs Upgrades Performance Machine Specs Opinions Buyer's Guides MacReviewZone

Copyright 1996-2007 by Cider Press Publishing LLC all rights reserved. MacSpeedZone is not authorized, sponsored, or otherwise approved by Apple Computer. Apple, the Apple logo, Macintosh, iPod, iBook, iMac, eMac, and PowerBook are registered trademarks of Apple Computer, Inc. Additional company and product names may be trademarks or registered trademarks and are hereby acknowledged.