The Source for Mac Performance News and Reviews

Review: MetaBox's JoeCardTM Z G4/450 2MB Cache First Shipping 2MB G4 CPU Upgrade By Mike Published: 9/15/2000

*Real World* Performance Tests

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

Applications and Game Performance

This page lists 'real world' application test results in common popular programs like Photoshop 5.5, SoundJam MP (v2.5.1), iMovie 1.02 and 3D games like Quake3 and Unreal Tournament. Of the applications tested here, only SoundJam MP and Photoshop 5.5 have Altivec support, although there is said to be some optimizations in OpenGL since 1.1.2.

SoundJam MP 2.5.1 MP3 Encoding Tests

Casaday & Green's SoundJam MP features Altivec support and uses of the G4's 'Velocity Engine' (Apple's term) instruction set. Since the Apple DVD drive from my 1999 B&W G3 used in these tests seems to have less than stellar performance and could be a bottleneck, I also tested times to convert the same song to MP3 format from an AIFF file on the hard drive (removing the CDrom drive as a factor). The hard drive file was an AIFF format of the same 4 minute, 5 second song file used in the test of CD to MP3 encoding. From the results of converting from the CD, it seems pretty clear the DVD drive was limiting performance. (Soundjam did indicate it was 'using Velocity' during tests with the G4 CPU.)

Photoshop 5.5 Tests:

I ran a complete PS5Bench (21 filter test) series using Photoshop v5.5 with the Altivec extensions (604e CPU tests had the Altivec plugins removed of course). Standard PSBench settings are 1024x768, millions colors, VM off, Interpolation set to bicubic (better) and Photoshop allocated enough RAM to avoid any swap file activity from the 10MB test image filter actions. (I allocated 140MB to Photoshop 5.5 for past reviews, but 80MB for this review due to the fact the S900 had only 144MB of RAM installed.) To improve consistency of times on the filter tests (each filter is run 3 times), I set Photoshop 5.5's 'History' settings from the default 20 to 1 and unchecked the 'automatically create snapshot' option. Note: All G4 CPU tests had the 4 OS 9 Altivec extensions active and used the Adobe current Altivec Core (v5.5.1) and Lighting Effects Filter (v5.5.2). The G4/450 AGP system scores were with a rev 2.6 G4 CPU with errata. The G4/500 system scores used a rev 2.8 CPU. The MetaBox G4/450 2MB scores are in bold. Filter MetaBox G4/450 2MB Cache in B&W Apple G4/450 AGP (1 cpu) Apple G4/500 AGP (1 cpu) G3/450 in B&W Rotate 90° CW 0.2 0.3 0.2 0.7 Rotate 9° CW 2.1 2.3 2.0 2.7 Rotate .9° CW 1.9 2.2 1.9 2.5 1 pix Gaus. Blur 0.5 0.5 0.5 1.2 3.7 pix Gaus. Blur 1.7 1.6 1.4 3.0 85 pix Gaus. Blur 2.0 1.9 1.6 3.9 Unsharp Mask 50%/1pix/0 level 0.7 0.8 0.7 1.6 Unsharp Mask 50%/3.7pix/0 level 2.0 1.9 1.7 3.4 Unsharp Mask 50%/10pix/5 level 2.0 1.9 1.7 3.4 Despeckle 0.8 0.8 0.8 2.2 RGB-CYMK 4.4 4.7 4.0 4.5 Reduce 60% 0.4 0.5 0.4 1.1 Lens Flare 3.5 3.4 3.3 3.9 Color Halftone 3.1 3.4 3.0 3.5 NTSC Colors 3.7 3.9 3.4 3.9 Accented Edges 9.5 10.2 8.9 9.9 Pointillize 13.4 14.5 12.5 14.0 Water Colors 19.9 21.7 18.7 20.8 Polar Coordinates 3.0 3.3 2.9 4.8 Radial Blur 30.5 34.6 29.9 32.1 Lighting Effects 1.8 2.0 1.7 7.4 Total Time 107.1 116.8 101.2 130.5 System MetaBox G4/450 2MB Cache in B&W Apple G4/450 AGP (1 cpu) Apple G4/500 AGP (1 cpu) G3/450 in B&W FYI: Here's some other total times for comparison: Apple Dual G4/450: 79.8 seconds Newer Tech G4/400 ZIF in B&W G3: 126.3 seconds 1GHz Pentium III w/256MB PC800 RAMBUS/Win98SE: 85.8 seconds 600MHz Athlon PC w/256MB 222 SDRAM/Win98: 138.7 seconds For a complete list of the PIII 1GHz and Dual G4/450 filter scores see this page. For other G4 CPU upgrades scores (ZIFs and CPU cards for older Macs) see the G4 reviews tab on the CPU Upgrades page.

iMovie 1.02 Tests

I used the iMovie tutorial project as a test for performance, simply assembling all the clips back-to-back (no transitions). This way it's easier for the reader to duplicate my test (since transitions would introduce an almost infinite number of possible variations). I exported the file using the CDROM medium size option, recording the time it took to complete the movie. The default codec is Sorenson. The graph below shows times to complete the export to a 'CDROM medium size' movie with the same B&W G3 with the MetaBox G4/450 2MB cache CPU upgrade and a G3/450 1MB cache CPU. As you can see, even at the same clock speed the G4 upgrade literally cut rendering times in half.

Game Performance Tests:

Unlike many 'serious' applications, 3D games are one of the best tests to show a system's overall performance. Very few applications stress the total system more than modern 3D/OpenGL games. The CPU, Memory Bus and Video subsystems are pushed very hard in these games, therefore they are a great way to show the benefits of more modern PCI/Memory controllers, faster memory bus and more advanced motherboard designs. Although I show results with a Rage128 (rev2), Voodoo5 and Rage128 Pro AGP, this is a CPU review not a video card review so I limited resolutions to 640x480 to avoid the fill rate limits at higher resolutions with the ATI Rage128 cards (the Radeon due soon should be much better in this respect however). Tests were done in Quake3 (retail) v117 and Unreal Tournament v425a (latest available at the time of posting this). All systems had OS 9.04 and OpenGL 1.1.3 installed. The Rage128 cards used the latest Apple drivers as of this date (incl. the Multimedia 1.0 update). The B&W G3 66MHz PCI slot was used for all graphics card tests in that system. The G4/AGP used the Rage128 PRO AGP card. Quake 3 Tests: I used the following game settings: 32-bit color, 640x480 mode Lightmap lighting Geometric detail high Texture quality 32-bit Texture detail 1 notch from max Texture filter - bilinear No reduction in game option settings (Dynamic lighting on, high quality sky, ejecting brass, etc. all on). No Quake3 config file tweaks. Quake 3 Arena Tests 32Bit 640x480 Timedemo 1 FPS G4/450 2MB (Rage128) 36.9 G3/450 1MB (Rage128) 35.5   G4/450 2MB (Voodoo5) 43.1 G3/450 1MB (Voodoo5) 38.1   G4/500 AGP (Rage128 PRO AGP) 59.2 For comparisons results with other systems/video cards search the Mac Game/FPS Database. Unreal Tournament Tests: The 'cityintro' timedemo was run in Unreal Tournament full version with the 425a update applied. All tests were run at 640x480 (to minimize the effect of video card fill rate limits, since this is a test to see what the benefit of the CPU upgrade was, not a video card test). OpenGL and Rave tests were run at 32bit mode with medium detail, and low audio quality, with a "min desired framerate" set to 0. Dynamic lighting was enabled. All results are in frames-per-second, higher is better. Since the UT's timedemo stats reports min, max and average framerates, all are listed below. The Cityintro test is basically a flyby with no firing, etc. so I also included the Wicked400 demo torture test as well. Scores with Wicked400 have always been very low even with the fastest Macs I have. One important note I wanted to make about the OpenGL scores. I had problems applying the 425a update and ended up performing a clean install and then applying the patches. I didn't realize until preparing to run the G4/AGP system tests that I had not reset UT to my normal 150MB allocated RAM setting (Westlake notes OpenGL mode requires more RAM). Had I noticed this sooner I would have re-tested (I thought I had reset UT to 150MB). The 107MB default UT allocation may have affected OpenGL scores somewhat with the Rage128 cards but a repeat of the Cityintro and Wicked400 tests on the G4/500 AGP with 150MB allocated showed scores were literally the same, with a max of 0.5 fps gained from the additional RAM allocation with the Cityintro test and only 0.2 FPS avg gain on the Wicked400 OpenGL mode test (at least on that system/video card combo). The Voodoo5 PCI card however, would lock up quickly in UT with only 107MB RAM allocated, so scores in the table for the Voodoo5 were with 150MB allocated. (Glide mode ran fine with 107MB allocated to UT, but since that's 16bit mode the results are not show in the table (but are farther down in the text). Unreal Tournament CityIntro Tests 32Bit 640x480 AVG FPS MIN FPS MAX FPS G4/450 2MB (Rage128 Rave) 33.51 18.7 86.8 G3/450 1MB (Rage128 Rave) 30.6 19.2 63.3   G4/450 2MB (Rage128 OpenGL) 34.0 15.1 88.9 G3/450 1MB (Rage128 OpenGL) 28.9 11.0 87.9   G4/450 2MB (Voodoo5 OpenGL) 33.1 11.5 92.3 G3/450 1MB (Voodoo5 OpenGL) 28.9 10.3 91.4   G4/500 AGP Rage128 PRO AGP (Rave mode) 40.2 24.6 81.7 G4/500 AGP Rage128 PRO AGP (OpenGL mode) 35.6 16.5 88.5 Wicked400 is a recorded demo with intense action, weapons firing, multiple players, etc. and is used to show perhaps worst case framerates during actual play. Unreal Tournament Wicked400 Tests 32Bit 640x480 AVG FPS MIN FPS MAX FPS G4/450 2MB (Rage128 Rave) 20.4 12.3 30.5 G3/450 1MB (Rage128 Rave) 18.8 12.4 30.3   G4/450 2MB (Rage128 OpenGL) 18.9 10.6 31.1 G3/450 1MB (Rage128 OpenGL) 16.99 9.42 28.79   G4/450 2MB (Voodoo5 OpenGL) 17.9 10.5 30.3 G3/450 1MB (Voodoo5 OpenGL) 16.5 10.2 26.7   G4/500 AGP Rage128 PRO AGP (Rave mode) 21.8 13.9 32.3 G4/500 AGP Rage128 AGP (OpenGL mode) 19.2 11.2 29.9 Just like all the other results in my original Voodoo5 review, performance was disappointing with the current drivers. As you can see above, in some cases a Rage128 was faster at 640x480 modes than the Voodoo5. (Granted at higher resolutions the fill rate limits of the Rage128 would cause FPS to drop, where the Voodoo5 card would stay relatively flat from 640x480 to 1024x768). Since it wouldn't be fair to include 16-bit Glide mode tests in comparison to the 32Bit modes above but for those that are interested Glide mode results with the V5 and the G4/450 CPU upgrade, here they are: Voodoo5 Glide (16Bit) Results w/G4/450 2MB Upgrade: Cityintro (min/avg/max) 25.2/41.8/61.4 Wicked400 (min/avg/max) 12.8/20.71/35.7 As I commented on in the Voodoo5 review, UT scores seem very similar regardless of video card, at least at 640x480 modes.

Metabox has pages of their own benchmarks in Photoshop, Final Cut Pro and MacBench. The next page describes the software cache control supplied with the MetaBox upgrade. Or you may use the links below to jump to a specific page.

Index of MetaBox's JoeCard TM Z G4/450 2MB Cache Review Intro | Benchmarks | Appl. 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.