Originally Posted by exyia
try going to the link and actually reading. I only posted the picture to spark interest
I did - thanks for the link. I also read the 3770k analysis in BF3. These are the sort of benchmarks that I have been looking for for a while and confirm my experience that clock speed has a big impact on BF3 multiplayer.
The conclusions I drew from the articles are that, for BF3 at least:
- hyperthreading is bad and should be turned off in BF3
- more physical cores are better for more fps
- higher clock speeds produce more fps.
The last is no surprise, but the first two are a little surprising, particularly since I had thought that BF3 had been optimised for hyperthreading. It also highlights that virtual cores behave very differently to physical cores.
The disappointment from the articles is that the 3770k and 3930k were tested on different maps, so it was not possible to compare the results. It would be interesting to compare the 3770k and 3930k at identical clocks with HT off in the same multiplayer map.
It would also be interesting to know how these results would change (if at all) in other games like Crysis 3, Bioshock Infinite or Far Cry 3, and whether repeating the tests with a single gpu would produce similar results.