P95 worker = core number in bios?

[rescuetoaster]

Hey all,

So I was playing around with folding at home yesterday, and discovered that while my setup is perfectly stable for games and benchmarks, apparently folding leads to a worker crash. I have suspected for a while that there is one core that is on the cusp of instability, since when I go to 4.0 one of the p95 workers drops out pretty quickly. Last night, F@h crashed on one of the workers as well at 3.9.

My question is - does the crashed worker in p95 mean the same thing as core number in the bios? Like, let's say worker 4 drops out consistently, would I then be able to disable core 3 (since core count starts at 0 in bios) and test for stability from there? If it's that one core that is the issue, I would potentially disable it and overclock a little more.

Just something that was rattling around in my brain. I might play around with it as a study break later.

 

[Johan45]

I would assume that the bios and P95 #'s wouold kinda line up logically but I do know that if I disable core 5 and 6 on my system P95 still reports the workes as 1-6.
You could probably fix your crashing worker with a bit more V_core.

Here is a very similar post with somone experiencing instability with F@H when system was otherwise stable.
http://www.overclockers.com/forums/showthread.php?t=728053

 

[trents]

I think so, yes.

 

[Dlaw]

An easy way to find out is to download and run Speedfan, as it has individual load graphs. Then run P95, and when the worker crashes, you know for sure which core the worker corresponds to, because it will be the only one not still under load.

Edit: I just thought of the Windows Task Manager. Nix the Speedfan, just open TM up before you run P95

 

[RGone]

I would assume that the bios and P95 #'s wouold kinda line up logically but I do know that if I disable core 5 and 6 on my system P95 still reports the workes as 1-6.
You could probably fix your crashing worker with a bit more V_core.

Here is a very similar post with somone experiencing instability with F@H when system was otherwise stable.
http://www.overclockers.com/forums/showthread.php?t=728053

You are saying that if you disable cores 5 and 6 on your 8 core processor that P95 reports the cores as 1 thru 6? If that is what you are saying, then yes that is correct. P95 does not put in place holders for the two disabled cores.
RGone...

 

[Johan45]

You are saying that if you disable cores 5 and 6 on your 8 core processor that P95 reports the cores as 1 thru 6? If that is what you are saying, then yes that is correct. P95 does not put in place holders for the two disabled cores.
RGone...

Yes that is exactly what I'm saying.

 

[RGone]

Yes that is exactly what I'm saying.

Gotcha. These threads last a very long time. Just trying for clarity. Hehehe. Except they call me one who muddies waters. No freeken clarity there.
RGone...

 

[rescuetoaster]

Alright, I played around with it.

So, worker 4 consistently failed during a few different prime tests (not blend), so I pegged core 3 as the first one I would shut off and try again. I bumped it to 4.0 ghz, and re-ran the tests with core 3 off. Prime still choked. Also, at 4.0 3dmark11 would fail on the first or second test.

Then I went through and sequentially shut off each core one by one and trying to see if I could find more stability.

The jist of it is, no, no I could not.
So everything is back on and running at 3.9. Not a big deal, it's not like I actually need the extra speed, I was just bored.

Edit: I should mention that my vcore is already fairly high, which is why I didn't just bump it up. With LLC, it tops out at 1.47. I do have some thermal headroom though.