Overclock.net banner
1 - 7 of 7 Posts

·
Premium Member
Joined
·
3,688 Posts
Discussion Starter · #1 ·
I see alot of people saying how great the 65nm intels are mainly with OC'ing...But how are their performance increases...
Would it be worth getting one of these compaired to an equal AMD 90nm CPU.

How much does this catch up intel to AMD when it comes to gaming performance...
 

·
Premium Member
Joined
·
5,703 Posts
This has no affect on performance at all, the only things that have improved are lower power consumption and lower heat, however you can overclock a 65nm Presler chip to 4.3GHz+ on air cooling with about 1.3-1.4v so they are alot more overclockable, your XP90c should get you 4.4-4.5GHz provided your motherboard is up to it (I don't think you current mobo supports the Preslers though, or does it?) so once you overclock one that high it should be able to match atleast an AMD 4000+ or maybe an FX55 (which is roughly 4.3GHz Intel)
 

·
Premium Member
Joined
·
3,688 Posts
Discussion Starter · #3 ·
Yeah my Mobo is Socket 478 so iw ould have to upgrade that also...thanks for clearing that up for me...I am in the process of building a new machine and did not know if I should go with AMD or Intel since they have this 65nm out now...Well I guess I will be going with AMD

Reps for you...
 

·
Banned
Joined
·
8,062 Posts
If you have the money, and dont want to wait and want a completely new rig/want to upgrade definately get the new 65nm Intel's. If you have the money and want to wait, wait for AMD 65nm to be released and see how well that does.
 

·
Premium Member
Joined
·
4,753 Posts
If there are no architectural differences between cpu's and the only change is a die shrink then there will be some differences ~ + 5 - 10% due to switching lengths and shortening the many "antenna". lengths in the interconnect circuitry of the die. Thus a die shrink will reduce costs for the producer, usually improve performance and usually enable faster clock speed on the overclock. This is the theory but the problem is that when there is a die shrink there is also usually architectural differences enacted in the silicon at the same time. Often these changes will increase the speed of the processor but not always.

The Northwood / Prescott shrink contained architectural changes that hampered the performance (namely pipeline lengthening among others) but normally there will be performance increases and overclock potential increases due to die shrinks.

R
 

·
Premium Member
Joined
·
3,688 Posts
Discussion Starter · #6 ·
Quote:


Originally Posted by Ropey

If there are no architectural differences between cpu's and the only change is a die shrink then there will be some differences ~ + 5 - 10% due to switching lengths and shortening the many “antennaâ€. lengths in the interconnect circuitry of the die. Thus a die shrink will reduce costs for the producer, usually improve performance and usually enable faster clock speed on the overclock. This is the theory but the problem is that when there is a die shrink there is also usually architectural differences enacted in the silicon at the same time. Often these changes will increase the speed of the processor but not always.

The Northwood / Prescott shrink contained architectural changes that hampered the performance (namely pipeline lengthening among others) but normally there will be performance increases and overclock potential increases due to die shrinks.

R


in English please hehe...j/k thanks
 
1 - 7 of 7 Posts
Top