slightlyhuman
[H]ard|Gawd
- Joined
- Jan 20, 2003
- Messages
- 1,186
So I'm looking into a quad core for my laptop. It currently has a 2.8ghz C2D(6mb cache). I was talking to a guy I play bf3 with about it and he said I should get an x9100(3.06ghz 6mb) c2d and would get better performance. Mainly wanting to upgrade my bf3 experience on my laptop.
I don't understand how a c2d can out perform a c2q with 12mb of cache.
Then he wrote this:
99% of the time it does. I have gone over this many times. The primary reasons, or gist of the situation, is when a thread is not run as afffinity set, which is most of them. This mean the thread has to be doble loaded to each core on the CPU and data off loaded as well. This causes a huge cycle per cycle perfornamce hit between the CQD and C2D. Then there is where they ae affinity set at stock on non quad thread opps the C2D is clocked way higher and overclocks higher as well...............
Edit; the iCore handles threads and cache differently. This is why with the CQD intel never pushed it on the masses but has no issue in doing so with the iCores......................
Can someone educate me and tell me if I'm wasting my time?
I don't understand how a c2d can out perform a c2q with 12mb of cache.
Then he wrote this:
99% of the time it does. I have gone over this many times. The primary reasons, or gist of the situation, is when a thread is not run as afffinity set, which is most of them. This mean the thread has to be doble loaded to each core on the CPU and data off loaded as well. This causes a huge cycle per cycle perfornamce hit between the CQD and C2D. Then there is where they ae affinity set at stock on non quad thread opps the C2D is clocked way higher and overclocks higher as well...............
Edit; the iCore handles threads and cache differently. This is why with the CQD intel never pushed it on the masses but has no issue in doing so with the iCores......................
Can someone educate me and tell me if I'm wasting my time?