Game Benchmarks:There is no sense in running game benchmarks at resolutions nobody will be playing at; if you are still stuck at 640x480 or 800x600 you might consider upgrading your video card. We think that 1024x768 is the absolute minimum “gaming” resolution on the PC, we’ll also include benchmarks for 1280x1024, but the general rule of thumb is that the workload will shift more towards the video card when increasing the resolution.
Let’s take a look at the frame rate in FarCry with the following settings:
Maximum quality option, Direct3D renderer
Level: Pier, demo: 1.tmd
Pixel shader: default model
Antialising: None
Anisotrophic filtering: 1×
HDR: disabled
Geometry Instancing: disabled
Normal-maps compression: disabled
We used this benchmark utility from HardwareOC.
At 1024x768 you gain 6% going from high latency PC3200 to low latency memory, overclocking your memory to 280Mhz offers you an increase in FPS of 1%! This can hardly be called impressive.
At 1280x1024 we only gain 5% by using more expensive PC3200 memory; Overclocking does not produce an impressive increase either.
Let’s see if Doom 3 has any advantage of an increased HTT speed; detail was set to “High Quality”
At 1024x768 we only see an increase of ~1% going to tight timings at 200Mhz and another ~1% gained by overclocking to 280Mhz.
At 1280x1024 there is almost no difference at PC3200 speeds, overclocking the HTT bus offers us ~3% higher FPS
Multimedia Benchmark:
There were comments from our readers that "Gamers" do other things as well beside gaming, one of these activities was re-encoding DVDs, so we recorded the time it took DVD Shrink to re-encode “Teigertje” to make it fit on a DVD+R:
We see a ~4% increase in encoding performance... if you like to see other none-gaming benchmarks, please drop us a line at the forums.
Conclusion:
After completing the game benchmarks it’s clear that that overclocking the HTT/FSB to increase memory bandwidth on the Athlon 64 does not deliver the performance boost expected.
The synthetic benchmarks show a large difference, but when it comes down to games, which rely on the graphics card quite a bit, you will see that they hardly run any faster, +1-3% at best is all you are going to get. So if you plan on upgrading your system, you might want to look towards buying a new graphics card, faster hard drive, better soundcard, larger monitor, or anything else to enrich your gaming experience. Upgrading to expensive, fast running memory should be one of the last items on your shopping list if you are a gamer.
If you have questions or comments, don’t hesitate to let us know what you think at our forums
would be interesting to see if there would be differences in other apps like encoding, and also have a pp4 comparison.
Great job on the test though