 |
|
|
|
In the Forums... |
Posted: September 15, 1999 Written by: David "Spunk" Grampa Estimated retail price: $199.99 Benchmarks
Test Conditions: Tests Run: Quake II For benchmarks in Quake II, I performed some of the most frequently used demos, including ID Software's Demo1.dm2, Demo2.dm2, and 3Finger's Crusher.dm2. These were done on Timedemo 1 with v-sync and sound enabled, along with 16-bit color. Here are the benchmarks:  As you can see, OpenGL performance is close to par with such cards as the TNT2 and Voodoo3. This keeps Matrox in the race for 3D card king. Keep in mind that this is the lesser version with the G400 chip, howeverf, and I am comparing it to such peers as the TNT2 (not Ultra) and Voodoo3 2000/3000 (not 3500). Another thing to take notice of is the stableness in framerates between higher resolutions and lower ones. The petty Savage4 outperforms the G400 at a low resolution of 640x480, but when you jump any higher the G400 nearly doubles the Savage4's performance. Embodying the attribute of being able to render at higher framerates, without a huge performance hit, is a good quality. One that the G400 easily achieves. 3DMark 99 Max Let's see if the same holds true when comparing performance at 16-bit color to performance at 32-bit color. Here are some benchmarks run with 3D Mark 99 Max. (Note: benchmarks taken with 16-bit z-buffer, v-sync enabled, triple buffering, and 16/32 bit color.)  As you can see, performance while running at 16-bit color is only slightly higher than that of performance at 32-bit color. Once again, this is a good feature when sporting no significant loss in frames while running at a higher color depth. Also, take into consideration that the benchmarks kept up with the Voodoo3 and TNT2 at higher resolutions. This serves as even more evidence proving that the Matrox G400 is the best all-around board in this generation of 3D accelerators. Did the G400 choke when performing almost all of the strenuous features it offers? Leave it to Quake III to find out...   |
||
|
| |||
|---|---|---|---|