The SpeedZone: Sure The New Dual G4/1.42 GHz Tower Is Faster Than The Others .. But By How Much?

Below we take a look at the relative performances of the current Power Mac Tower models from Apple. Aside from clock speed, the main difference between these three machines is that bottom of the line Tower comes with a single G4 processor and the two dual processor machines come with 2. All machines have 256k of L2 cache (per processor), the 1 GHz and 1.25 GHz Towers have 1 MB of L3 cache (per processor), whereas the top-of-the-line Tower has a full 2 MB. The single processor Tower has a bus speed that clocks in at 133 MHz and the dual 1.25's and dual 1.42's bus runs at 167 MHz. This is not a great speed difference and probably had little, if any, effect on the performance results listed below.

The three machines also have different graphics cards. The single processor machine an NVIDIA card and the dual processor machines an ATI card. Both come with 64 MB of video memory.

What kind of effect do these differences have on performance? In most tests that required heavy processing activity, and that could take advantage of dual processors, the dual processor machines pulled significantly ahead. However the dual processor Towers left a lot of their processing power on the table, unused by the application that was running. For example in one of our Photoshop tests only 50% of the processing capability of the dual processor 1.42 GHz machine was used. 50% of the machine's data crunching ability was left idling, waiting for 1s and 0s that never came. This situation repeated itself in many of our other single application tests. Somewhere along the data feed chain there is a bottle neck that keeps data from arriving fast enough to keep the processors fed. Since it is possible to saturate the processors, both with specialized single applications, and when running multiple processes from more than one application, our guess is that the bottle neck is not at the hardware level, but the software level .. or perhaps at the level where the two meet. But we are speculating here. In any case, in environments where you want to be running multiple processes at once, a dual processor machine becomes much more attractive, perhaps even essential.

It also appears, at least from our testing, that, depending on the game, a dual processor machine will greatly out class a single processor machine when playing games.

Difference and similarities of each machine

"Real World" Tests

For the graphs below frame rates were converted to percentages, relative to the Power Mac Dual G4/1000, which is set to 100%. For all scores, higher numbers are better.

The copy folder test not only stresses drive performance but also takes advantage of faster processing capability & caching schemes. It appears that the 1.42 GHz machine has a significantly faster drive. It also has twice the L3 cache of the 1.25 machine, which might be a performance factor in this test

In the single file test the the processor & caching capability of the machine is less of a factor.

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. (Only 50% of the processing power of the 1.42 machine was used)

Large Document & Database Type Tests

Microsoft Office X Word Macro

Machine	Power Mac Single G4/1 GHz	Power Mac Dual G4/1.25 GHz	Power Mac Dual G4/1.42 GHz
Absolute Scores - seconds	76.96	59.91	56.24

A Macro (series of complex actions) was run in Microsoft's Excel spreadsheet program. Excel is part of Office X (Only 50% of the processing power of the 1.42 machine was used)

Microsoft Office X Excel Macro

Machine	Power Mac Single G4/1 GHz	Power Mac Dual G4/1.25 GHz	Power Mac Dual G4/1.42 GHz
Absolute Scores - seconds	49.11	38.60	34.85

A Macro (series of complex actions) was run in Microsoft's Excel spreadsheet program. Excel is part of Office X (Only 50% of the processing power of the 1.42 machine was used)

AppleWorks Search And Replace Test

Machine	Power Mac Single G4/1 GHz	Power Mac Dual G4/1.25 GHz	Power Mac Dual G4/1.42 GHz
Absolute Scores - seconds	21.03	17.20	14.88

This test takes place in a large AppleWorks document. Machines with larger chip caches usually do better in this test (Only 50% of the processing power of the 1.42 machine was used)

Finder File Search

Machine	Power Mac Single G4/1 GHz	Power Mac Dual G4/1.25 GHz	Power Mac Dual G4/1.42 GHz
Absolute Scores - seconds	37.30	28.97	28.24

Mac OS X's new File Search function relies on both processor and drive performance. (Only 5% of the processing power of the 1.42 machine was used)

Number Crunching & Rendering Tests

Photoshop 7 - BareFeat's Test

Machine	Power Mac Single G4/1 GHz	Power Mac Dual G4/1.25 GHz	Power Mac Dual G4/1.42 GHz
Absolute Scores - seconds	52.21	31.85	29.42

BareFeats Photoshop Test Suite - mimics a production type workflow in Photoshop. In the single processor Tower, the processor was saturated with data. The 1.25 GHz machine had 25% left unused. Only 60% of the processing power of the 1.42 machine was used

Photoshop 7 - AltiVec Tests

Machine	Power Mac Single G4/1 GHz	Power Mac Dual G4/1.25 GHz	Power Mac Dual G4/1.42 GHz
Absolute Scores - seconds	24.46	13.53	11.47

Certain of the functions of Photoshop can take special performance advantage of the G4 processor. A dual processor machine runs rings around a the single processor one, in this test. (Only 75% of the processing power of the 1.42 machine was used)

Photoshop 7 - Normal Tests

Machine	Power Mac Single G4/1 GHz	Power Mac Dual G4/1.25 GHz	Power Mac Dual G4/1.42 GHz
Absolute Scores - seconds	110.24	83.47	73.70

The test above consists of Photoshop functions that do not take any particular advantage of the G4. The reason that the dual processor machines don't show more of a performance gap is that 50% to 55% of the processing power those machines was not being utilized. So having an extra processor was not much of an advantage ... unless you wanted to do something else at the same time.

iMovie - Effect Render

Machine	Power Mac Single G4/1 GHz	Power Mac Dual G4/1.25 GHz	Power Mac Dual G4/1.42 GHz
Absolute Scores - seconds	294.33	160.49	144.46

An Effect Render is run on an movie using iMovie. This tests takes advantage of dual processors. ( 80% of the processing power of the 1.42 machine was used)

AltiVec Fractal (Carbon)

Machine	Power Mac Single G4/1 GHz	Power Mac Dual G4/1.25 GHz	Power Mac Dual G4/1.42 GHz
Absolute Scores - seconds	52.94	21.21	18.92

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 machine

Encoding/Decoding Tests

QuickTime Encode

Machine	Power Mac Single G4/1 GHz	Power Mac Dual G4/1.25 GHz	Power Mac Dual G4/1.42 GHz
Absolute Scores - seconds	32	25.11	20.88

A QuickTime encode compresses a QuickTime movie for streaming on the Web. (Only 50% of the processing power of the 1.42 machine was used)

QuickTime Convert To DV Stream

Machine	Power Mac Single G4/1 GHz	Power Mac Dual G4/1.25 GHz	Power Mac Dual G4/1.42 GHz
Absolute Scores - seconds	152.70	103.06	93.24

Converting QuickTime movies to DV allows you to import them into iMovie. (Only 60% of the processing power of the 1.42 machine was used)

iTunes - MP3 Encode From CD

Machine	Power Mac Single G4/1 GHz	Power Mac Dual G4/1.25 GHz	Power Mac Dual G4/1.42 GHz
Absolute Scores - seconds	75.61	71.58	76.71

During the encode only 60% of the processing capacity of the dual processor machines was used. The speed of the CD drive appears to be a bottle neck ... not getting data to the processor quickly enough.

StuffIt Expand

Machine	Power Mac Single G4/1 GHz	Power Mac Dual G4/1.25 GHz	Power Mac Dual G4/1.42 GHz
Absolute Scores - seconds	9.86	7.32	6.28

StuffIt does not take advantage of dual processors

Multitasking

iTunes MP3 Encode + AppleWorks Search & Replace + Finder Copy (Multitasking)

Machine	Power Mac Single G4/1 GHz	Power Mac Dual G4/1.25 GHz	Power Mac Dual G4/1.42 GHz
Absolute Scores - seconds	69.44	44.70	34.85

MP3 Encode, AppleWorks search & replace and folder copy are all carried out at the same time. In a production type environment the Dual Processor Towers make a lot of sense ... if you do some planning.

QuickTime Encode + Fractal (Multitasking)

Machine	Power Mac Single G4/1 GHz	Power Mac Dual G4/1.25 GHz	Power Mac Dual G4/1.42 GHz
Absolute Scores - seconds	84.75	38.79	32.34

QuickTime Encode and Fractal render are both carried out at the same time.

QuickTime DV + iMovie Effect Render (Multitasking)

Machine	Power Mac Single G4/1 GHz	Power Mac Dual G4/1.25 GHz	Power Mac Dual G4/1.42 GHz
Absolute Scores - seconds	624.92	284.24	256.01

QuickTime DV Encode and iMovie Effect are performed at the same time. Dual Processor machines rule the day

MP3 Encode + Photoshop Normal + Word Macro (Multitasking)

Machine	Power Mac Single G4/1 GHz	Power Mac Dual G4/1.25 GHz	Power Mac Dual G4/1.42 GHz
Absolute Scores - seconds	167.95	138.85	118.99

The MP3 Encode, Photoshop Normal and Word Macro tests are all run at the same time

Gaming

Giants Highest Frame Rates

Machine	Power Mac Single G4/1 GHz	Power Mac Dual G4/1.25 GHz	Power Mac Dual G4/1.42 GHz
Absolute Scores - fps	65 fps	133 fps	155 fps

Both the 1.25 and 1.42 Towers have the ATI Radeon 9000 Pro graphics card with 64MB DDR video memory. The Game Giants responds well to greater processing capability

Giants Average Frame Rates

Machine	Power Mac Single G4/1 GHz	Power Mac Dual G4/1.25 GHz	Power Mac Dual G4/1.42 GHz
Absolute Scores - fps	25.2 fps	47 fps	57.5 fps

Giants Lowest Frame Rates

Machine	Power Mac Single G4/1 GHz	Power Mac Dual G4/1.25 GHz	Power Mac Dual G4/1.42 GHz
Absolute Scores - fps	13.9 fps	25 fps	26.5 fps

4x4 EVO 2 Average Frame Rates- High Quality Setting

Machine	Power Mac Single G4/1 GHz	Power Mac Dual G4/1.25 GHz	Power Mac Dual G4/1.42 GHz
Absolute Scores - fps	17 fps	25 fps	27 fps

4x4 EVO 2 Average Frame Rates- Medium Quality Setting

Machine	Power Mac Single G4/1 GHz	Power Mac Dual G4/1.25 GHz	Power Mac Dual G4/1.42 GHz
Absolute Scores - fps	26 fps	39 fps	40 fps

4x4 EVO 2 Average Frame Rates- Low Quality Setting

Machine	Power Mac Single G4/1 GHz	Power Mac Dual G4/1.25 GHz	Power Mac Dual G4/1.42 GHz
Absolute Scores - fps	36 fps	52 fps	55 fps

For very demanding games you can't beat a Tower .. at least not yet