Comparing AMD phenom II X6 to Intel
- Thread starter vieya
- Start date
Way to go, a million benches that show ZERO clock for clock comparisons. Good job.
I5 760 or VS 1055t would be the correct comparison for clock for clock. There is no difference between the 1055t and 1090t except higher clocks, so looking at the 750 (which is still 133mhz slower clocked) compared to the 1055t in all those benches it's extremely close. Clock for clock, they are pretty much identical in performance with multithreaded apps, with intel whipping the Phenom II in all single threaded apps.
Here's a full clock for clock-
http://www.anandtech.com/bench/Product/191?vs=147
And he can always get an I7 anyway and definitely have better performance hands down.
I5 760 or VS 1055t would be the correct comparison for clock for clock. There is no difference between the 1055t and 1090t except higher clocks, so looking at the 750 (which is still 133mhz slower clocked) compared to the 1055t in all those benches it's extremely close. Clock for clock, they are pretty much identical in performance with multithreaded apps, with intel whipping the Phenom II in all single threaded apps.
Here's a full clock for clock-
http://www.anandtech.com/bench/Product/191?vs=147
And he can always get an I7 anyway and definitely have better performance hands down.
Last edited:
jasonn20
New Member
What is your score @ 4ghz in Cinebench 11.5...?
[URL=http://img337.imageshack.us/i/cinebenchu.png/]
[/URL]I don't have cinebench, plus I'm not in possession of my I7 right now as a friend is borrowing it. I'm running my I5 655k right now.
That being said, all I7's beat all X6's clock for clock with any bench.
For some reason you just don't understand architectural differences. Core count is secondary when core speed is factored in.
Once again, he can just get an I7 and be faster no matter what.
jasonn20
New Member
Once again, applications that use all six cores at once will pretty much cancel out the benefits that intel has in single threaded applications and have more multi-tasking power than the i5.. maybe even the i7 quad
What apps use 6 threads??? And please don't give me the talk about future apps, because as I already said multithreaded apps have been around for almost 10 years and a quad core is still overkill at this point. So by the time something actually uses 6 threads specifically, the 1090t will be older than dirt in PC years.
Not the I7, just look at the benches strangehold put up. I7 860 and 1055t is the only decent clock for clock comparison there.
So are you going to run cinebench on 3 cores for a comparison yet? Or have you done it and are too embarassed to post?
Not the I7, just look at the benches strangehold put up. I7 860 and 1055t is the only decent clock for clock comparison there.
So are you going to run cinebench on 3 cores for a comparison yet? Or have you done it and are too embarassed to post?
Last edited:
jasonn20
New Member
Windows 7 uses all six cores in Multi-thread... and I can find others...
lol, what? You mean the all mighty windows experience index??? mega fail there if it is.
Oh and check out my edit on the last post
jasonn20
New Member
So are you going to run cinebench on 3 cores for a comparison yet? Or have you done it and are too embarassed to post? [/QUOTE]
I'll post it but it is not a accurate comparison between the i5 quad and i7.... your cpu is using 32mn technology....
NO, I am not going by WEI .....
[/QUOTE]I'll post it but it is not a accurate comparison between the i5 quad and i7.... your cpu is using 32mn technology....
NO, I am not going by WEI .....
Last edited:
Uhh, the I7 with 2 cores/4 threads would definitely be faster. It has twice the cache size.
lower nm does not always equal higher performance, just like more cores doesn't either. Lynnfield architecture is actually faster/better than clarksdale because of double the L3 cache count. Plus, Lynnfield has a far superior memory controller. 32nm just means the die size is smaller, they did that just to fit the onboard graphics on the chip with the CPU.
Edit- I meant twice the cache SIZE not speed.
Last edited:
jasonn20
New Member
Well the problem is about 6 months ago the quad core deneb architecture was getting outclassed in every aspect by the I5 quads and I7's just destroyed them. So, since AMD can't release their bulldozer yet, for some reason, someone came up with the idea of squeezing 2 more deneb cores on the die. The problem is, it's still slow deneb architecture. While the extra cores help, it's still outclassed by the faster core speed of the Lynnfield and Bloomfield's.
What AMD should have done is added more L3 cache, it would have helped, but six 45nm cores on one die was probably hard enough to squeeze on.
What AMD should have done is added more L3 cache, it would have helped, but six 45nm cores on one die was probably hard enough to squeeze on.