OC i72600 Sandy Bridge (mystery when resetting bios).

[hide291]

Hello,

I will try to be brief and if you need me to delve deeper into the matter, just let me know and I will attach screenshots or whatever...

The thing is that I have been running a stable OC at 4.2ghz (+24%) clock + 105.0 (+5%) bclk + memory at 2.2 for YEARS.

But because of a damn power outage, I had to reset the bios and put all the values back with the difference that everything is still ok and stable but I noticed during stress tests for example the stress of Aida64 that shows if there is throttling etc...and everything is ok both temperatures, etc...BUT I noticed the clocks and without doing a stress test the clock is at 4.2 with the typical variation caused by EIST and C1, etc...but with stress ON or games where the cpu is working the clocks automatically drop to 4095 without any variation. I am going crazy as I don't know what it is due to, it doesn't reflect any kind of throttling, the maximum temperature is 60 degrees...

I have been touching the bios for hours and the only thing I see that interferes with it are the EIST, c1, 2...etc...so I said to myself, ok then I will disable all savings and that's it...Well no, if I disable them and track the clocks directly they stay stable at 4095 without variation when the OC is set to 4.2 without any error. It's like playing with activating and deactivating savings and I get the 4.2 again but with the problem that I have already mentioned that they drop to 4095 when I put them to the test and/or gaming, etc...Instead during windows tasks it stays at 4.2.

Does anyone have a slight idea of what could have happened?

Thanks and regards!

[Fassou]

Ignoring the possibility that before you didn't notice it, and now that you looked at it more closely, you discovered this variation, it occurs to me that perhaps it is integrated in some way, by voltage consumed, heat dissipated or simply by matching the FSB, forces a kind of downclock to the micro.

But if the micro is 100% stable, it's not something to lose sleep over.

Salu2!

[Espinetenbolas]

I agree with Fassou… I already owed you this but you didn't notice.

It's also possible that this OC is not stable for whatever reason and the board lowers the Mhz.

AIDA64 is a pain at the stress test level.

Try LinX: LinX-0.6.5.7_11.3.1.002.rar (3,87 MB) - uploaded.net

Just run it, load a problem that fills up practically the RAM and give it 3 or 6 passes or at least 30 minutes.

You can monitor with HWiNFO… which I recommend.

[hide291]

Hello!

It didn't happen before, that's why I've noticed it and it's not that it will be noticed much with that small variation but I'm curious to know what it could be.

It's more about letting me, it doesn't even let me set the clock to a fixed value. If I disable all the saving systems the pc consumes 4.2 but within win it directly sets it to 4095 neither going up nor down (as it should be but at 4.2) instead if I activate the systems it goes to 4.2 with its typical variation for saving but it's about putting stress on it and going down to 4095. I've used Aida64, passmark, prime95, cpuid stress test... it always happens the same thing. :wall: