Click for Wireless Adapters and Bases!
A Click shows your site support to my Sponsors
Click for Wireless Adapters and Bases!
A Click shows your site support to my Sponsors
 The Source for Mac Performance News and Reviews |
|||||||||||||||||||||||||||||||||||||||||||||||||
|
10X G3 CPU Upgrade for PCI Macs Published: 11/02/99 |
|||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||
| Applications and Game Performance | |||||||||||||||||||||||||||||||||||||||||||||||||
| Update: I've added results of Bryce 2, Infini-D and Photoshop tests with an Apple G4/450 (Sawtooth AGP system). The G4/450 system tests used Photoshop 5.5 (vs 5.0) and the supplied Altivec extensions.
This page lists test results in common Mac applications like Photoshop 5, Bryce 2, Infini-D 4.01, After Effects 3.1, Premiere 4.2 and many popular 3D games. To see additional scores with other systems and G3 Upgrades, visit my G3 Apps Tests page and the many other previous CPU upgrade reviews. For a searchable database of hundreds of entries for 3D game framerates, see my new Mac Game FPS database. Update: I'm adding scores below from my new Apple G4/450 AGP system. As you'll see, it seems the G4 system trails the MAChCarrier G4/500 in most common (non-Altivec enhanced) applications.
|
|||||||||||||||||||||||||||||||||||||||||||||||||
| 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. I didn't change the default QT movie output file options. The table below shows the results of the XLR8 MAChCarrier G3/500 as well as other systems/upgrades for comparison. Infini-D was allocated 40MB of RAM for the tests. Unless otherwise indicated, all tests used OS 8.1 and RAM was interleaved.
As shown above, Infini-D performance was excellent, outperforming even a B&W G3/400 and more than twice as fast as a stock 9600/350 604e system.
|
|||||||||||||||||||||||||||||||||||||||||||||||||
| Bryce 2 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, antialiasing was on, screen resolution was set to 1024x768, thousands colors.
Bryce performance (like all rendering apps) is primarily CPU speed dependent so an upgraded older Mac can outperform even the B&W G3 with the right CPU. |
|||||||||||||||||||||||||||||||||||||||||||||||||
| After Effects 3.1 Tests | |||||||||||||||||||||||||||||||||||||||||||||||||
| The following is a chart showing the time to render a special effects movie (10MB file size) in AE with the CarrierZIF and other systems. Resolution was set to 1024x768, thousands colors as was common on all but the Photoshop 5 tests.
|
|||||||||||||||||||||||||||||||||||||||||||||||||
| Photoshop 5.02 Tests: | |||||||||||||||||||||||||||||||||||||||||||||||||
| I also ran a complete PS5Bench (21 filter test) series. Only the total time to complete the tests are shown, but for more details on PSBench and the name of every filter used see my Photoshop 5 Performance page which lists other systems including PCs (Pentium III/Dual Pentium II). PSBench settings are 1024x768, millions colors, VM off, Interpolation set to bicubic (better) and Photoshop should be allocated enough RAM (I use 80MB) to avoid any swap file activity from the 10MB test image filter actions. Photoshop 5's default 'History' settings were used.
Note that disabling "Speculative Processing" (often called Branch Prediction) actually improved Photoshop performance (I repeated the tests to verify this). See my article on interleaved RAM for Photoshop 5 tests with and without interleaved memory. Update: I've added results of PSBench tests with my Apple G4/450 AGP system using PS 5.5 w/Altivec extensions (initial release). There was a large spread of times for the 3 repeated tests on each filter with the G4 system (I used the avg. of the 3 times). |
|||||||||||||||||||||||||||||||||||||||||||||||||
|
(Sawtooth AGP system) 450MHz G4 CPU 1MB Cache at 225MHz 10MHz Bus 258MB RAM Spec Proc ON OEM Rage128 AGP OS 8.6 VM OFF |
500MHz G3 CPU 1MB Cache at 250MHz 50MHz Bus 320MB RAM Spec Proc OFF OEM Rage128 Orion OS 8.1 VM OFF |
450MHz G3 CPU 1MB Cache at 225MHz 45MHz Bus 512MB RAM Spec Proc OFF ATI Rage128 Orion OS 8.1 VM OFF |
400MHz G3 CPU 1MB L2 at 200MHz 100MHz Bus 192MB SDRAM Rage128 16MB OS 8.5.1 VM OFF |
|
|
|
|
|
|
(Note: my Overclocked 9600 with 400MHz 604e CPU took 192 seconds a beta G4/400 upgrade in the Genesis without Altivec Software Extensions scored 149.1) |
|||
See my Photoshop5 Performance page for more results including Pentium III/Dual Pentium II
Interesting Note: a Pentium III 500Mhz (500/250/100) Windows 98 system with 256MB SDRAM took 165.7 seconds to complete this test. Adobe's Pentium III Photoshop SSE update was applied prior to the test.
| Game Performance Tests: | ||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Tests were done in Rave Quake, Quake2, Q3test 1.08 and Unreal v224b7. Since your video card is a major factor in the speed and quality of 3D games, the first gamer's purchase should be a fast 3D card such as a 3Dfx Voodoo 2, Voodoo3 or perhaps a Rage128, and plenty of RAM (Unreal and Quake 3 both need 90MB+ now). If you're well equipped with RAM (128MB or more) and have a G3 CPU already, a faster 3D video card is a far better investment for improved game performance and visuals as well. But as shown in the flat FPS scores, a card like the Voodoo3 can't seem to be saturated even with the fastest G3 CPU.
Since the 9600 had a rev 1 (slower clock speed) OEM Rage128 from a B&W G3/400, I also show the B&W G3/400 scores with that same video card (vs the faster rev 2 and Rage128 retail cards). To try and prevent the Rage128 OEM card from being a bottleneck, many tests were run at 800x600 rather than higher resolutions. In the Quake1 tests you can see the card was already saturated even at 800x600, since faster MAChCarrier G3/500 CPU speeds didn't produce any higher framerates. A retail Rage128 card would have scored about 8-10% better than the slower clock speed OEM card used in the 9600/350.
As shown in the above table, the original OEM Rage128 had a slower clock speed. The rev 2 and retail Rage128 Orions with later drivers improved performance.
Q3Test 1.08: I used Q3Test v1.08 (latest currently available) for all tests. Graphics settings in the game were :
The tables below show Q3Test performance with each of the card/system combos in frames-per-second. As you can see from the scores below, a Voodoo3 would have been a faster card in OpenGL games with a fast G3 CPU.
Unreal 224B7: The table below summarizes results from the 3rd cycle timedemo of the 'Castle flyby' scene. (All tests used the same detail settings - see my FPS Database entry page to download Unreal.ini files for RAVE and 3dfx cards.)
I now regret not using the Voodoo3 in the with the MAChCarrier G3/500, as it would have produced higher game scores with the fast CPU. |
||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Summary: Overall applications and game performance was very good, hampered in some cases by factors in the older mac logic board . |
||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Intro | Benchmarks | Appl. Tests | Software Controls | Docs/Installation | Specs/Design | Summary - or - |
||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Copyright © Mike, 1999. Users of the web site must read and are bound by the terms and conditions of use. |
||||||||||||||||||||||||||||||||||||||||||||||||||||||
