Is turbo boost stable when overclocked?

[Maximilian]

Im cranking my i7 920 up to 3.8ghz, i can get this two ways:

181 bclk with turbo ON
190 bclk with turbo OFF

The thing that im not sure about is that linX and prime will test the cpu at 3.8ghz, however when only 1 core is loaded turbo enables a 22nd multiplier which kicks that one core up to 3982mhz.

So my question is if 4 cores @ 3.8ghz are stable at a given voltage will 1 core at 3.98ghz be stable at the same voltage?

 

[Dadofamunky]

Why not? Just test it and see. Should be fine.

 

[Maximilian]

How do i just stress 1 core though? Only reason i noticed the 22x multiplier on 1 core is the brief period where linX is doing something but its not stressing all 4 cores, its like its preparing to stress them or whatever.

 

[dawp]

LinX will allow you to run as many threads as you want down to 1. just going setting and adjust the # of threads you want to run.

 

[ThatsABigOne]

Or when you start running lynx, go into task manager, processes, right click on process that is stress testing, and click on affinity. Select one core, boom, you are stressing on one core.

 

[Rifter]

turbo did not work for me over 3.8Ghz. Was not stable. Had to go turboless to get to my 4.2Ghz overclock on my 930.

 

[Maximilian]

Well i managed 150 runs of linX at 3.8ghz with turbo. But it crashed in seconds when stressing only 1 core. So yeah, im gonna stick with 190x20, its not stable due to the 22x multiplier applied on 1 core when it is stressed.

 

[Shmee]

ya, i think bloomsfield prefers odd multis. I had my 920 D0 running at about 4.2, or 200x21. I found it to be pretty stable, as long as the turbo was locked as a 21 multi and the multi was odd.

 

[Danitondar]

Yes of course is Stable!
in 4.3(2500K)