2 Very Different E8400s

[Dostoyevsky77]

My first E8400 (Q748A116) needed a lot more vCore for sigged 4.23GHz than others were reporting (1.45v), so I got another one to see what's going on.

The new chip (Q745A731) doesn't need as much vCore (only 1.40v) for the same OC, but it runs 10(c) hotter.

To confirm it wasn't the AS5 breaking-in, I remounted the first chip and it was still running 10(c) cooler.

Q748A116 @ 4.32 35 / 55 / 1.45v

Q745A731 @ 4.32 45 / 65 / 1.40v

Your opinions are appreciated:
1) Which chip should I keep?
2) Is it safe to go beyond 1.45v if the chip is running cooler?
3) Is it safe to go beyond 65(c) if the vCore is lower?

[MoreGooder]

I've read that Intel botched the temp sensors on these processors. Given that you can't trust the temp data for either of your processors, perhaps it's best to keep the one that overclocks at the lower vcore. This means it may have more headroom, and is likely actually cooler than your first one to begin with. Just my theory.

[Funky]

Whats the vid on both chips? I bet the VID of the hotter chip is much lower.

[G80]

I wouldent personaly go beyond 1.45v not when the temp readings are faulty.

[KiD0M4N]

Whats the vid on both chips? I bet the VID of the hotter chip is much lower.

+1

Please report back with the VIDs of the two CPUs. Use latest CoreTemp to detect it.

[Spyrus]

Q745A731 keep that. Low vcore for same speeds. Temp is not a problem as far
as you touch the hsf and feeling it just warm.

[CERO]

As we cant "trust" on those temp sensors, for sure I would pick the one with low vcore requirement, this could also explain they run hotter, maybe also that one got a lower VID?

[Dostoyevsky77]

Ack, sorry. VID is identical on both. 1.125v, IIRC (at work ATM).

I will go with consensus. It just makes me sad to have bungled temperature reporting (hoping that's what it is). I felt a lot better when 4.23 meant nothing hotter than 55(c).

Thanks so much for your opinions so far!

[CERO]

I see that u are using a lapped Tuniq, with such cooling I dont think U need to worry about the temps at all, as long u use "safe volts" and as long you are stable, your are totally fine, mine is even worse. I got a fixed 38c on both cores, but as soon I use 1.368vcore during prime core0: goes 45c and core1: 43, no way it could be that "cool" but on the other hand the air comming from the back of my TRUE is way cooler then my Q6600 dissipated, and that was 52-55c during prime.

[Tofu Guy]

Just got Q748A116 E8400 from a friend. Still need to test it out. Don't have a board for it right now, just couldn,t pass up an e8400 for $100.

On the topic side. when placing 1.45V, or more with a 45nm die, one will have to start worrying about an electron breaking through a potenial well and jumping between transistors...which could then ruin the whole core. I am not expereinced in OCing like most here, but that is the greatest threat when you reduce the size of a manufacturing die for a cpu, Slow 1.45V may have been perfectly safe with 65nm, but now with 45nm, the possible extra potential could cause a case like that to happen.

[Dostoyevsky77]

Yes, agreed. Let me know how your experience is with that chip!