Originally posted by: Denithor
Secondly - FSB more important than raw clockspeed - do not know where in the world you got that impression but it's totally dead wrong. FSB typically accounts for <1% difference, even for quad-core chips where it has the most impact. Read a few reviews before posting comments like that.
Dead wrong? Are you kidding me?
It is what separates the Q8200 and the Q6600 in terms of performance. Less than 1% difference? Ask yourself this, what is the ONLY thing that makes the Q8200 faster then the Q6600?
Clock speed? No.
Cache? No.
FSB? Obviously.
Other members on this forum we also quick to point this out. All around benchmarks show this. Here is a post from the NuclearMC CPU Benchmark thread.
Originally posted by: rogue1979
E6400@2816MHz - 8786
E6600@3300MHz - 10,345
E2180@3300MHz - 10,505
Brisbane 4000+ @2.65GHz 6850
Now check that out. An E2180 has the same clock speed, but only 1/4 of the cache. Going from 1MB to 4MB is a huge leap, and yet it hangs right in there.
Originally posted by: Idontcare
I ran a test, 200MHz FSB versus 400MHz FSB. This is with QX6700. Same 2.4GHz for both tests (12x200 versus 6x400):
2.4GHz (6x400 Mhz) = 12670
2.4GHz (12x200 MHz) = 11739
So that's an 8% increase in the Nuclearus score for a doubling (2x) of the FSB.
This also shows the performance increase of a raised FSB.
Yes, I read articles. I also find in much more meaningful to compare real life performance as well. Its one of the main reasons everyone was so excited over 45nm quads (and the lower power consumption).