How much can a driver affect a benchmark.

[Nozoned]

So, I decided to mess around with OCing my video card.

MSI R6850 Cyclone PE/OC Radeon HD 6850 1GB 256-bit...

First thing I did was run a Heaven benchmark to see what the numbers were stock. After running Heaven I thought it seemed very low and looked back at the one I ran when I first built the computer. I was surprised at the difference in numbers.



The only thing that I have done since building this rig is updated the drivers. Could that be why the score is so much lower now?

Should I try the older driver?

If so, where is the best place to find older drivers?

 

[Widjaja]

Sometimes drivers can give benchmarks false readings unfortunately yet not affect the performance of the card in games.

I stopped running benchmarks a long time ago due to finding in most cases there will be a small difference but the performance is not noticeable in gaming.

 

[2DividedbyZero]

drivers and benchmarks can show a vast difference between versions, then again, benchmarks are designed to "show" the raw performance abilities of the card - then try and find a game optimized to give that same performance.....

driver makers tend to put benchmarks results before gaming performance in aid to selling the card/technology. Improvement can be found between drivers and games yes, but bigger improvements will be found between drivers and benchmarks

 

[Nozoned]

Alright, thanks guys. I'll play around with it a bit and see what happens. I haven't had any issues as far as games go, been staying around 60fps. The games aren't that graphic intensive though.