Corsair XMS Pro PC3200c2 2GB Review - Benckmarking 3D Mark and Doom 3
(Page 3 of 5 )
Benchmarks
Our testing for the Corsair XMS Pro PC3200c2 2GB Dual Channel Kit includes the benchmarks: 3D Mark 2003/2005, SuperPi, Doom 3, and Far Cry. Here is our test system.
Test system:
Motherboard: Foxconn NF4SK8AA-8EKRS
Processor: AMD 64 3500+ Venice
Video Card: Powercolor X800GTO 128MB 256-bit
Memory: Corsair—Corsair XMS Pro PC3200c2 2GB (2 x 1 GB) Dual Channel Kit
Generic—Kingston Value RAM 1 GB (2 x 512 MB) Single Channel
We didn’t have a comparable kit available for the test, but were able to compare some budget memory. The difference in price between the memory sets is over $110 and the Kingston is lacking a lot of the capacity and memory bandwidth, so these are not usually competing products. This should just show what you get by choosing much higher quality memory instead of going just for what is inexpensive. It won’t show the Corsair kit running anything in its actual price range (stay tuned to Dev Hardware for future memory tests, where we'll stack up comparable modules).
First we’ll try our Futuremark tests: 3D Mark 2003 and 3D Mark 2005. These, as you probably know, simulate games. Since they are not real games though, they are really artificial benchmarks. Futuremark still gives us a good idea of how much games will like hardware setups. I ran each test a few times and used the median result.
One thing that I did different in this test was include the scores from default BIOS memory timings. With the memory set at 3-3-3-8 and 2.60V, I recorded the data as Corsair XMS Pro Stock. After correcting the latencies to 2-3-3-6 per the specifications, I recorded the scores as Corsair XMS Pro Spec. This should show how differently the memory will perform if you don't adjust the latencies.


As you can see, the Corsair gives us a 500 point increases in speed over the really cheap alternative. In 3D Mark 2003, adjusting the timings up to specifications gave us a solid and noticeable improvement. In the 2005 test, it didn’t do much for our score.
Next, we’ll test a real game, Doom 3. This test and the later ones only have the memory set at only the 2-3-3-6 latency. These tests are in high detail with no AA or AF. This game really pushes our memory.

Doom 3 | Generic | Corsair XMS Pro | % Change |
640x480 | 71.3 | 92.7 | 30.0% |
800x600 | 64.5 | 85.1 | 31.9% |
1024x768 | 58.6 | 71.7 | 22.4% |
1280x1024 | 44.1 | 49.8 | 12.9% |
It looks like the Corsair XMS Pro is running the game up to 32% faster than the value RAM alternative. You can definitely see how powerful the kit is. Let’s also take a look at Doom 3 with AA. This test was run at 1280x1024.

Doom 3 AA | Generic | Corsair XMS Pro | % Change |
None | 44.1 | 49.8 | 12.9% |
2x AA | 35.8 | 40.4 | 12.8% |
4x AA | 24.2 | 26.5 | 9.5% |
Admittedly, antialiasing puts a lot more stress on the video card than anything else, so I didn’t expect the results to be as profound as the last ones. The X800GTO kind of chokes in this test, but the Corsair still manages to give the system a solid 10% increase in FPS.
Next: Benchmarking Far Cry and SuperPi >>
More Memory Articles
More By Developer Shed
| Recommended by Dev Hardware |
|---|
|