The Source for Mac Performance News and Reviews

Review: Newer Tech's iMAXpowr G4/433 CPU Upgrade G4 CPU Upgrade for iMac 233 - 333MHz models By Mike Published: 7/19/2000

*Real World* Performance Tests

Intro | Benchmarks | Appl/Game Tests | Software Controls | Installation | Specs/Design | Summary

Applications and Game Performance

This page lists test results in common Mac applications like Photoshop 5.5, SoundJam MP, Bryce, Infini-D, After Effects, and many popular 3D games like Quake1, Quake2, Quake3 and Unreal Tournament. As always, a clean reboot was performed before running each test. All results on this page used the application's (or game's) built-in timing function to eliminate human error variations possible when starting and stopping a timer/stopwatch.

SoundJam MP 2.11 MP3 Encoding Tests

Casaday & Green's SoundJam MP features Altivec support that shows a sample of what gains can be had from use of the G4's unique instruction set. The iMac rev A's stock CDrom drive is not a great performer however in audio extraction. I could have sworn I saw some indication of Altivec being enabled (in some dialog or panel) during past G4 tests with Soundjam, but didn't see it when running these tests (I copied over the Soundjam folder from a G3 system rather than reinstall from scratch with my SoundJam 1.0 CD - perhaps that mattered).

Photoshop 5 Tests:

The table below shows the results of PSBench (21 filter test) series between the iMac rev A G3/233 and the same system with the iMac iMAXpowr G4 upgrade. Beige G3/233/512K cache (stock CPU) and a Beige G3 w/466MHz/1MB cache upgrade. The total time to complete the 21 filters is shown. PSBench settings are 1024x768, millions colors, VM off, Interpolation set to bicubic (better) and Photoshop should be allocated enough RAM (100MB for this review) to avoid any swap file activity from the 10MB test image filter actions. The latest versions of Adobe's Photoshop 5.5 Altivec addons (as of the review date) were used. Note that as shown in all the past G4 upgrade/system reviews tests - Photoshop's Altivec support only benefits certain functions (some dramatically so however). Lighting Effects seems to be the most accelerated, as is shown in some past reviews that included tests with a G4 with and without the Altivec core plugin and Lighting Effects (Altivec version) installed. PS5Bench Test (10MB Image File) iMac rev A w/iMAXpowr G4/433 433MHz CPU Speed 1MB Cache at 217MHz 66MHz Bus 192MB RAM ATI RageII (6MB) iMac G3/233 (Rev A/Original CPU) 233MHz CPU Speed 512K Cache at 117MHz 66MHz Bus 192MB RAM Onboard ATI RageII (6MB) Gain Rotate 90 0.4 1.4 3.5x Rotate 9 2.7 5.8 2.1x Rotate .9 2.5 5.3 2.1x Gaussian Blur 1 0.8 2.5 3.1x Gaussian Blur 3.7 2.3 5.9 2.3x Gaussian Blur 85 2.7 8.0 3x Unsharp 50/1/0 1.1 3.3 3x Unsharp 50/3/7/0 2.7 6.6 2.4x Unsharp 50/10/5 2.7 6.7 2.5x Despeckle 1.0 4.6 4.6x RGB-CMYK 4.8 9.7 2x Reduce Size 60% 0.6 2.3 3.8x Lens Flare 4.3 8.0 1.9x Color Halftone 3.5 7.3 2.1x NTSC Colors 4.1 8.5 2.1x Accented Edges 10.5 21.2 2x Pointillize 15.1 28.9 1.9x Water Color 22.0 44.1 2x Polar Coordinates 3.8 10.7 2.8x Radial Blur 35.0 69.6 2x Lighting Effects 2.1 17.2 8.2x PS5Bench Index 124.7 Sec. 277.0 Sec. 2.2x System iMac rev A w/iMAXpowr G4/433 433MHz CPU Speed 1MB Cache at 217MHz 66MHz Bus 192MB RAM ATI RageII (6MB) iMac G3/233 (Rev A/Original CPU) 233MHz CPU Speed 512K Cache at 117MHz 66MHz Bus 192MB RAM Onboard ATI RageII (6MB) Gain Note: For more comparisons of Photoshop performance, see the Systems page and previous G4 CPU upgrade reviews. When looking at the results above, consider which filters you use most often when determining how much a G4/Altivec upgrade will benefit you. Many of the filters in this test don't use Altivec. Weigh the gains from those that you use most often.

Infini-D 4.01 Tests

Like most of today's applications, Infini-D 4.01 does not use Altivec extensions, but is a common application I have used for comparisons of CPU/FPU performance. I used the same 'Chapter 7 completed' tutorial scene file from my past reviews. Rendering quality was set to Ray Trace, medium anti-aliasing, shadows on, patch detail low. I didn't change the default QT movie output file options. The graph below shows times to complete the 150 frame movie (10.6MB) rendering with an iMac DV SE (G3/400MHz) previously reviewed but tested again with OS 9.04, my rev A iMac G3/233 (OS 9.04) with the stock CPU and with the iMAXpowr G4 CPU upgrade. Infini-D was allocated 40MB of RAM for the tests. Note the G3/400 is slightly faster than the G4 CPUs of the same speed in this test. Another interesting note was that the G4/450 AGP Sawtooth system finished this same test more than 2 minutes faster with OS 8.6 than with OS 9. Note - my rev 1 Apple G4/450 AGP system (v2.6 CPU) took 38 Min+ to complete this test. (Spec. Processing is disabled in the v2.6 G4 systems)

Bryce 2 Tests

Bryce 2 testing used the samples scenes in the KAI folder called "Magical Knight I" and "Alexandria II". Bryce 2 was left at the default memory allocation of 16980k, antialiasing was on, screen resolution was set to 1024x768, thousands colors. Since Bryce doesn't use Altivec, a G3 CPU of the same speed does just as well as a G4 CPU.

After Effects 3.1 Tests

The following is a chart showing the time to render a special effects movie (appx. 8MB file size) in After Effects 3.1. Resolution was set to 1024x768, thousands colors as was common on all the application tests except for Photoshop 5.5 (which used millions colors). Note: I do not own After Effects v4.1, but there is an Altivec Plugin for that version (see the FAQ's Apple G4 section for a full list of applications/links to Altivec program updates).

Game Performance Tests:

Tests were done in Quake2, Quake3 and Unreal Tournament. All systems were running OS 9.04 and OpenGL 1.1.3. The iMAXpowr G4 CPU had OS 9's Altivec extensions active. Quake 2: All tests used the latest Quake2 version as of this date (v1.03) as well as OpenGL 1.1.3. I edited the config file to force 16bit and disable multitexturing (since it chokes the ATI cards). Graphics settings in the game were : 16-bit color Max Texture Quality No Sync Every Frame No CD Audio Full Screen The Crusher results seemed low with Voodoo2. I'll be retesting with the iMAXpowr G4 when I return from Macworld. Quake III v117 Results: I used the latest version available at the time of the review for all tests. Graphics settings for the standard Timedemo 1 tests were: 16-bit color Lightmap lighting Geometric detail high Texture quality 16-bit Texture detail 1 notch from max Texture filter - bilinear Game Options - disabled only Marks on Walls and HQ Sky Note - what is not shown in the results below is the poor image quality of the VR128 chip in the iMac DV SE in 16bit color mode (32bit looks great, but is slower). For sample screenshots of 16bit vs 32bit Quake3 with the iMac DV SE - see the game tests page of that system review. I also ran tests with the FPS boosting Locki config file which dramatically reduces image quality and game details to provide higher framerates. The results are shown below. Unreal Tournament: I tested using both the builtin (retail version only) 'cityintro' test and also the user created demo file called 'wicked400' (which is a recording from actual play). (For results of other systems search the Mac Game FPS database. For info on how to run these tests see the FPS Database entry page.) The Wicked400 is recording of a multiplayer game with many players firing, etc. and is a much better indicator of performance than the Cityintro playback which has no player action, etc. For links to the Wicked400 file, see the entry page of the Mac Game/Video card FPS database.

The next page describes the software cache control supplied with the card. Or you may use the links below to jump to a specific page.

Index of Newer Tech iMAXpowr G4 433MHz Review Intro | Benchmarks | Appl/Game Tests | Software Controls | Installation | Specs/Design | Summary - or - Back to www.XLR8YOURMAC.com

Copyright © Mike, 2000. All Rights Reserved. All brand or product names mentioned here are properties of their respective companies. Users of the web site must read and are bound by the terms and conditions of use.