The SpeedZone - New Generation Of Towers Performance Compared To The Last Generation: 16 Processor, Graphics, Game, Hard Drive & Application Performance Tests
Below you will find performance results for most of the currently shipping Apple Tower computers. This will, hopefully, give you an feel for the relative performances. We will be adding new machines on an on-going basis ... so check back once in a while.
New machines added to the list below include the following:
Power Mac Dual G5/2.0 GHz (Summer, 2003)
Power Mac Dual G5/1.8 GHz (Summer, 2003)
Power Mac Single G5/1.8 GHz (Summer, 2003)
Power Mac Single G5/1.6 GHz (Summer, 2003)
Power Mac Single G4/1.25 GHz (Summer, 2003)
Don't see performance information on the Mac you are interested in? Send a request for a custom performance comparison between any two Macs (example).
Note: Most of the machines below were tested when running 10.2.x (Jaguar). However a couple of the machines were running 10.3 (Panther). We have noted below where running the newer OS made a difference. We test all machines with the operating system they ship with when we receive them.
The tests below are from our suite of real world application tests. These tests feature a diverse selection of applications commonly used by the Mac community. The test suite was designed to render an accurate and well rounded picture of a machine's performance. All of the tests below, except for the game tests and the CineBench tests, were timed with a stopwatch. For timed scores, lower numbers are better. For the CineBench & Game scores higher numbers are better.
IMPORTANT: We have changed how we present the data below. In the past we expressed the performance differences as a percentage of a 'Base' machine. We are now showing the actual time it took for a machine to complete the test. In cases where the test was not timed, we give you the actual score.
Processor Intensive Tests
CineBench 2003 consists of a series of tests. This one, tests basic single CPU processing performance. . You can download the CineBench program to run on your own machine and compare scores. We used the version of CineBench that is not optimized for the G5 for this test, and the one below. The scores above are actual scores ... not timed tests. Higher numbers are better
This is the same test as above, however it will take advantage of dual processors if they are present. The scores above are actual scores ... not timed tests. Higher numbers are better
The Fractal program has been highly tuned to take advantage of the G4 and is precisely the type of work that the G4 was made for. It will also gobble up whatever processing capability is present. This is a good test for assessing the fundamental processing potential of each machine. Really highlights the processing advantage of a dual processor machines, and the G4 processor, when applications are tuned to take advantage of these features. You can download the Fractal program and run it on your own machine to compare scores. You will need to set the fractal to maximum count of 65536, to be consistent with our testing. The version of the AltiVec Fractal program we used, was not optimized to run on the G5, and it runs comparatively slower on G5 machines. However the current version of the Altivec Fractal is optimized for the G5.
A 5MB movie is converted to a QuickTime movie using default setting. At the moment this function of QuickTime does not take good advantage of dual processors.
Graphics Performance
This part of the CineBench program relies on the graphics card for graphics rendering. Download the program and run your own test. The scores above are actual scores ... not timed tests. Higher numbers are better
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. You can download the small program to run on your machine and compare your scores, to the ones here.
Note test results for this test are different depending on whether you are running OS 10.2.x or 10.3. The Dual 1.8 GHz, Stock Dual G4/1.0 GHz and the Stock Blue & White G3/350 were the only machines running 10.3. You can safely assume that, were all machines running the same OS, the Dual G5/1.8 would fall somewhere between the Dual G5/2.0 and Single G5/1.8. The performance of the G5/1.6 GHz is probably hurt in this test by the lack of a L3 cache
Game Performance
Giants is tuned to take advantage of the processor(s). Indeed, when run on a dual processor machine, you will see Giants fly ahead of the same game running on a single processor machine. It also helps to have a better graphics card.. The scores above are actual scores (frames per second) ... not timed tests. Higher numbers are better
4x4 EVO2 is a demanding racing game. It appears that this game is much more graphics card dependent than processor dependent. The scores above are actual scores (frames per second) ... not timed tests. Higher numbers are better All the G5s have the same graphics card except for the Dual G5/2.0
Drive Performance
The copy folder test not only stresses drive performance but also takes advantage of faster processing capability & caching schemes. The Dual G5/1.8 Tower was released well after the other G5s, and it appears that it came with a better drive
In the single file test the the processor & caching capability of the machine is less of a factor.
Application Performance
BareFeats Photoshop Test Suite - mimics a production type workflow in Photoshop. If you would like to run this test on your machine download the action file and drop it on the Photoshop icon.
A Macro (series of complex actions) was run in Microsoft's Word on a 500 page document. Word is part of Office X.
CD-ROM drive speed and CPU processing power are the factors influencing the scores above. Four songs totaling 16.5 minutes in length were converted. Note: converting files off the hard drive (rather than from the CD), took only half as long.
This test takes place in a large AppleWorks document.
In this test, an Aged Effect is run on the 175 MB DV file with the Effect left at its default setting This is a processor intensive task. This function of iMovie does not take good advantage of dual processors. Though both the Dual 1.8 GHz and Dual 2.0 GHz were running the same version of iMovie, the 1.8 GHz machine was running OS 10.3 and the 2.0 was running 10.2.7. Perhaps this has skewed the relative results of these two machines. There is no technical reason that the 1.8 GHz machine would come out on top.
Multitasking Performance
MultiTasking: The MP3 Encode, Photoshop Normal and Word Macro tests are all run at the same time. (1.6 GHz G5 is hurt by its slower iTunes encode speed)
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.
Additional Resources 
For Great Prices On Upgrades 
