Currently stress testing my 6600k in P95. First try at 4.3ghz

>>56069599
One core got as hot as 65C for an isntant, but dont think I'm going to worry until 75C. Wish I had spent more on my gpu, had the money. Oh well, sell it a year down the road. Does everything I want at 1080p right now.

Those temps are great. What are you using for a cooler? A 1 hour burn is more than sufficient. Keep in mind nothing will ever push your cpu quite as hard as p95. Even maxing out all the cores doing a transcode or render.

File: confused.png (131KB, 1739x944px) Image search: [Google]

131KB, 1739x944px

>>56069748
I am using 27.9 because I heard 28.x was extremely stressful to the point that it may be harmful, if that makes a difference. Should I try for 4.4 or more on stock vcore, or am I being greedy? Would it increase temps if I didnt add more juice?

The temps did get a lot higher for a bit there, but fans sound even quieter than originally now with lower temps.

File: speccy.png (23KB, 533x578px) Image search: [Google]

23KB, 533x578px

>>56069748
Oops, using Cryorig H7 with tuniq tx-2 or 3, cant remember.

Here's a pic at stock clocks in the winter. Think my room was about 65 degrees F, but I can't be sure.

>>56069859
Ditched the 7850 as well, though wishing I had dished out a bit more, seeing my 6600k OC.

>>56069834
28 uses avx, which pushes haswell and later chips much harder. It's an even more unrealistic workload because not a lot of programs even use that instruction set. Voltage is the only risk here. If you push the clock too high it will simple freeze/crash/reboot. I'd see how high you can get the clock on 1.2v and call it a day. Never go over 1.3.

>>56069923
While I wait out the last bit of this hour, might as well ask. What should I realistically expect at 1.2v? I really wasn't even planning to up voltage, as I only have a msi 4gb 470 Oced to 1360 on core and 7300 on memory, probably more with some more tinkering. Probably in between a stock 480 and 1060, if I had to guess.

>>56069987
I actually think 1450/7600 may be available to me if I put in the time. But still, seems so far behind the cpu I dont think i can justify upping the vcore. Unless its still speeding up notorious amd overhead for the 470.

>>56069987
It all depends on your chip. Some are better overclockers than others. Honestly 4.3 at the stock voltage is really good if you want to stay there. You might get 4.5 if you up the voltage, but that's not much of an improvement compared to the added heat and power you'll be wasting.

>>56070012
Probably just sell it in 6 months to a year for vega or new nvidia, got it for a steal. Sorry, I'm done being a blog bitch. Test almost done.

Thanks for the advice anon.

File: 7EkCP9i3Pkzplki5BfgBsGF0o1_500.jpg (72KB, 500x375px) Image search: [Google]

72KB, 500x375px

>>56070046

>>56069599
Use IBT. Linpack is much harder on the CPU.

Also don't worry about temperatures at TJunction until they go over TJmax with Linpack.

> Figure I'll let it ride for 3-5 hours and be satisfied.

I've had P95 crash after 25 hours.
I've had CPU BSOD in Battleifled 4 after 48 hours in P95.
I've had what appeared to be a 100% stable OC that passed Prime 95 for more than 48 hours crash in Intel Burn Test.
Prime95 temperatures are ice cold compared to Intel Burn Test.


Prime 95 is almost impossible to pass.

>>56069599
Not very impressive. Try 1.3 for 4.5. You might have to up to 1.35 for 4.6.

>>56069599
Why are you using Speccy for monitoring? At least use a tool that's designed for that, like HWiNFO64 or HARDWAREMONITOR.