caesparktom :
re-seated cooling block and prime95 still failed blend test and temps were unchanged at around 53C so not a seating issue. After failing blend test again, i ran sml fft p95 test and it ran for 1 hr without crashing or errors so i stopped and concluded it was a memory problem. I assumed auto detect settings in bios would correctly set the timing, frequency and voltage, but for some reason they were was off by a bit. After setting explicitly to manufacture specs it did much better: Prime95 ran 13 hours, produced no errors, but did still crash...
No bsod just sudden crash. Here are some temp readings during first hour or so.
time | c1 | c2 | c3 | c4
5:00 | 48c | 51c | 55c | 52c
5:13 | 52c | 54c | 58c | 55c
6:50 | 53c | 53c | 57c | 55c
.
.
.
temps were pretty consistent for duration of test. Actually I turned A/C off for bit to see how temps reacted and core 1,2 and 3 went up to ~ 60c and core 3 to 64c. But, it IS pretty hot here in beijing without A/C. I should note that crash occurred in middle of night when A/C was off so temps were probable around 60-64c before crash. btw, should i be concerned that core3 is always about 4c hotter than other cores?
Anyways, Stock system w/no Overclock should have no trouble burning for 24+ hours right? Not sure what do now. Seems my system is just not quite as stable as it should be and not cooling quite as well as i would like. What do you suggest? I would like overclock a bit, but if my stock system is already marginal in terms of stability i don't know that i have any room to push it.
OPTIONS:
1. Improve stability by improving cooling. can put h60 in push pull config with some premium fans.
2. Of course i can keep trying to locate weak component if infact one exists, but it is way harder to troubleshoot now that system is more or less stable. Also, I don't have spare parts to swap out and systematically retest. and anyways, I suspect after 12 hours system won't fail consistently. i.e, if i run test again I don't think it would fail at the 13 hour mark reliably. Worse yet, I had to travel after ordering parts so 30 day return periods are up.
3. maybe I can further improve stability by explicitly setting all voltage frequency settings in bios. What is good practice? auto or manually set to manufacture specification???
thanks
Let me say I am really not an expert at trouble-shooting something like this. But that being said, why don't you change out your RAM and RMA it. It is likely covered by a lifetime warranty. Also, RAM is fairly cheap on NewEgg and so you could get another set of 2 x 4 GB 1866 for $60 - $75. It seems like your RAM is too temperamental. I had absolutely no problems with mine and did not even have to set it. It automatically went the DDR 1600 which is its spec. I get that you don't have extra parts to swap out, but this is one that has a lifetime warranty and it is fairly cheap.
My CPU runs at 38 - 45C under load (I have the Noctua NH-D14). My ambient temperature here in Arizona is 82F. The H60 is approximately equal to the D-14 with fans on high. Still, even with your temps at 60C, I don't think that they are high enough to cause a problem.
I had a video card that didn't get good air circulation with my I-7 920 and that would cause BSOD on a regular basis. The card would function fine for a few hours and then BANG, BSOD would appear. The card took a while to heat up to its shutdown temp. Make sure that your video card is not overheating.
Also, could your RAM be overheating? Just a thought. I only say this because your system seems to run fine and then shut down.
Also another thought. If the H60 is not getting the job done you should email Corsair and ask them to trade for the H100. The H100 is the only cooler to date that outperforms the D-14 that is not a custom water cooler (or noisy as a jet plane air cooler). Corsair would probably swap out the H60 - they are amazing. The H100 is expensive but it is the baddest cooler on the planet right now. Keep in mind that it can be noisy. Still the H60 should be getting the job done - it is a very good cooler.
I know I said this before, but make sure all of your connections are tight and you have everything connected correctly. For example, you need to connect PCI-E power connectors to your video card not regular power connectors. The HX 750 specifies which are PCI-E specific and which are not on the cables (they are blue in color). I would assume that the special PCI-E connectors are more stable and are rated especially for that use.
The Corsair 650D is a great case for cooling, but if your GPU is not getting good air circulation it could overheat. Because your system performs fine for a long time and then fails, this is something to look at. I don't think heat is a problem with your CPU or your motherboard. A hot video card or a bad video card WILL cause a BSOD regularly.
As I mentioned before, the motherboard is most often the problem with the system builds that I have seen that had problems. I am naturally suspicious of mobos now. Note I do not think that heat is the problem for the mobo. Make sure that your mobo is not defective. I would say it is the cause in about 70% of the cases I have been involved with. Changing out the mobo cured everything. I am a little prejudiced because my first mobo was bad in my build.
You might also just hand check which components are getting hot after 24 hours of use. For example, the D-14 fins stay cool to the touch in my system but the mobo gets warm. If one component is excessively hot after you do your check, that may be the problem.
To address your question about voltages, I would go with your motherboard's defaults as they are probably conservative and are set for your 2600K. I really don't know much about the correct voltages but I doubt that you resetting them would solve the issue.
It seems that something bigger and more obvious is the problem. Do you have an old video card you could substitute? If you could substitute any of the parts, it may help you trouble shoot it.
I am sorry that you are having problems, but it is great that the situation is better.
You also may want to start a new thread and get other posters who are more experience in trouble-shooting new builds involve.
Good luck, let me know how your are doing.