The Source for Mac Performance News and Reviews

Review: Formac Pro G3 450 ZIF CPU upgrade Published: 7/6/2000

Applications performance results and comments

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

Applications/General use comments

This page lists test results in common Mac applications like Bryce 2, Infini-D 4.01, After Effects 3.1, Premiere 4.2 and also results of the PSBench tests in Photoshop 5. These apps were used since they are primarily CPU/FPU bound and are a good test of gain from adding a faster CPU upgrade. To see additional scores with dozens of other G3 and 604e upgrades, read the other CPU upgrade reviews (including G4 upgrades). Older system results (including 604e CPUs) are listed at my G3 Apps Tests page.

Infini-D Tests

I used the same Infini-D 4.0 tutorial file rendering test as I've done in past reviews using the 'Chapter 7 completed' scene. Rendering quality was set to Ray Trace, medium anti-aliasing, shadows on, patch detail low.

Bryce Tests

I also ran tests in Bryce 2, using 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, anti-aliasing was on, screen resolution was set to 1024x768, thousands colors.

After Effects 3.1 Tests

The following is a graph showing the time to render a special effects movie (10MB file size) in After Effects. Resolution was set to 1024x768, thousands colors as was common on all but the Photoshop 5 tests.

Photoshop 5 Tests:

The table below shows the results of PSBench (21 filter test) series between a 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 (108MB for this review) to avoid any swap file activity from the 10MB test image filter actions. Note that Photoshop performance does seem to scale almost linearly with CPU speed (the G3 233/512KB times were almost double those from a 466/1MB). Since v5.02 was used for the original Apple G3/233 tests, I used the same version for the G3/466 upgrade tests. The fact the G3/466 used a VR128 video card is a minor impact on the scores I believe, since this is not a moving image test and is primarily CPU/FPU bound.

Beige G3 w/Formac ProG3 466MHz CPU Speed 1MB Cache at 233MHz 66MHz Bus 160MB RAM ATI VR128 (16MB)	Beige G3/233 (Original Apple CPU) 233MHz CPU Speed 512K Cache at 117MHz 66MHz Bus 160MB RAM Onboard ATI RagePro (6MB)
130.1 Seconds	254.0 Seconds
Total Time for 21 Filter Test (Seconds) (smaller is better)    Ref: B&W G3/400 = 148.6

Game Performance Tests:

Games are one area where performance can be as much related to your video card as your CPU. As you'll see below, the VR128 graphics card in the review system was fill rate limited to the point where at 1024x768 a faster CPU can't squeeze out more frames per second. Think of your video card as a bucket with a certain sized hole in the bottom. Regardless of how fast the bucket is filled (faster CPU), the size of the output hole (fill rate/output capability) is a limiting factor to higher game performance. For those with onboard video or weak 3D graphics cards, the first gamers purchase should be a faster video card (see my video cards page for reviews) , then plenty of RAM (best to have at least 128MB with the latest 3D/openGL games, and VM will still have to be enabled for some of the latest 3D games). For those wondering why I used a VR128 card instead of a faster gaming card like the Voodoo3. The reason is this machine has a Toshiba DVD drive and I need the Rage128 for DVD playback. (Another Mac has the Wired4DVD card installed.) I'm including Quake1 performance only since it's one of the least demanding of real 3D games but still shows that at 1024x768, the VR128 doesn't deliver higher framerates even with a faster CPU. Quake 2 v1.03 Performance Not as demanding as Quake3, running Quake2 in 16bit mod doesn't choke the VR128 as much, but you can see it's still the bottleneck at 1024x768, where even a faster CPU doesn't help (you can't squeeze more out of the card...) Nothing can help the 1024x768 fill rate limits of the VR128 card. A Voodoo3 with OpenGL games usually delivers near twice the FPS at 1024x768 as the retail Rage128 cards. The other downside to the Rage128 is currently all 16bit games look badly on it (see iMac DV SE review game/apps page for screenshots) Quake 3 v117 Performance I used the latest Quake3 v117 for the Mac available at the date of this review. All tests were run with the standard, high quality settings (lightmap, high geometry detail, etc. - only color depth and resolution were changes from the default, as-installed Quake3 config. (Attempts to use a reduced quality/higher FPS config file like Loki's resulted in a unusable display with this VR128 at least.) Now 32Bit Results: To see how other (faster) Mac graphics cards performed, see the reviews at the Video Cards page.

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 Formac Pro G3 ZIF Review Intro | Benchmarks  | Appl. Tests | Software Controls  | Docs/Installation | Specs/Design | Summary - or - Back to www.XLR8YOURMAC.com

Copyright © Mike, 2000. No part of this sites content or images are to be reproduced or distributed in any form without written permission. 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.