it shows 0 50 and 100 percent including auto.

 

it shows 0 50 and 100 percent including auto.

OK, so I've made most of the changes you gave me. For some reason it wouldn't let me disable Intel Turbo Boost.

 

DRAM Frequency doesn't give an option for XMP, only various frequencies...

 

I set the BCLK to 100.1 like you said but it's running at 100.06 @ 42. Current voltage 1.2260.

 

Temperatures sitting comfortable at 65/70c, screenshot attached for CoreTemp.

 

Disabled HT as mentioned.

 

CPUz seems to be more accurate than CoreTemp. Screenshot attached.

 

With the cpu voltage offset, I set it to 0.005 which was the default after changing it from auto to offset.

 

No errors with Prime95 yet but not been running for long.

Cool. Temps are fine, up the multi to 44 and see how it holds. If it doesn't, up the offset on the vcore.

 

See if you can somewhere find an xmp setting for your ram. I have asus, so no help...

 

Post me the link to the correct manual and i'll take a look for ya.

 

 

Sent from my iPhone using Tapatalk

http://www.asrock.com/mb/manual.asp?Model=Z77%20Pro3

 

Will try 44 in a little while.

So the last Prime95 running at 4.2Ghz finished 200-220 tests in 1.5 hours with no errors at all. No sudden spikes in temperatures and voltage stayed stable. Running another prime 95 for another 1.5 hours on 4.4Ghz. Screen shots attached for current stats.

 

I may also add that my fans aren't running at 100% with these stress tests... With my AXTU (Asrock Extreme Tuning Utility) I have the target speed set to one but optimal temperature set to 45c so they adjust accordingly to reach that temperature. Pretty much off when below that temperature and speed up as it gets hotter.

 

Since I took the screenshot attached, the temperatures have sat nicely at min 65-73 max.

OK so that test ran for only an hour. All passed fine. Voltage varied between 1.120 and 1.128 and temperatures stayed at around 65-74.

 

What do you suggest for going higher or should I stay where I am?

Try FSX now and see how you like it! I think you will be a lot happier, but try and shoot for 4.6Ghz next.

Interesting... I just rebooted and left voltage offset at -0.005% and bumped the CPU ratio up to 4.6. I rebooted and it gave me a BSOD before it even got into WIndows. I rebooted back into the BIOS and raised the voltage offset to +0.100. Now, the overclock seems to be stable but the termperatures are definitely higher. On CoreTemp right now, I'm seeing the following;

 

Low: 78-84c

High: 80-87c

 

Screenshots of CoreTemp/CPU-Z attached. No errors being reported from Prime95 after around 10 minutes.

 

Voltages are varieying between 1.208 and 1.216.

 

EDIT

 

Hmm, Core 2 just failed with the following;

 

[Oct 29 16:18] Worker starting

[Oct 29 16:18] Setting affinity to run worker on logical CPU #2

[Oct 29 16:18] Beginning a continuous self-test to check your computer.

[Oct 29 16:18] Please read stress.txt. Choose Test/Stop to end this test.

[Oct 29 16:18] Test 1, 460000 Lucas-Lehmer iterations of M250519 using AVX FFT length 12K, Pass1=256, Pass2=48.

[Oct 29 16:19] Test 2, 460000 Lucas-Lehmer iterations of M245759 using AVX FFT length 12K, Pass1=256, Pass2=48.

[Oct 29 16:19] Test 3, 460000 Lucas-Lehmer iterations of M245281 using AVX FFT length 12K, Pass1=256, Pass2=48.

[Oct 29 16:20] Test 4, 460000 Lucas-Lehmer iterations of M243713 using AVX FFT length 12K, Pass1=256, Pass2=48.

[Oct 29 16:20] Test 5, 460000 Lucas-Lehmer iterations of M235519 using AVX FFT length 12K, Pass1=256, Pass2=48.

[Oct 29 16:20] Test 6, 460000 Lucas-Lehmer iterations of M233473 using AVX FFT length 12K, Pass1=256, Pass2=48.

[Oct 29 16:21] Test 7, 460000 Lucas-Lehmer iterations of M231183 using AVX FFT length 12K, Pass1=256, Pass2=48.

[Oct 29 16:21] Test 8, 460000 Lucas-Lehmer iterations of M229375 using AVX FFT length 12K, Pass1=256, Pass2=48.

[Oct 29 16:21] Test 9, 460000 Lucas-Lehmer iterations of M225281 using AVX FFT length 12K, Pass1=256, Pass2=48.

[Oct 29 16:22] Test 10, 460000 Lucas-Lehmer iterations of M221183 using AVX FFT length 12K, Pass1=256, Pass2=48.

[Oct 29 16:22] Test 11, 560000 Lucas-Lehmer iterations of M212991 using AVX FFT length 12K, Pass1=256, Pass2=48.

[Oct 29 16:22] Test 12, 560000 Lucas-Lehmer iterations of M210415 using AVX FFT length 12K, Pass1=256, Pass2=48.

[Oct 29 16:23] Test 13, 560000 Lucas-Lehmer iterations of M208897 using AVX FFT length 12K, Pass1=256, Pass2=48.

[Oct 29 16:23] Test 14, 560000 Lucas-Lehmer iterations of M204799 using AVX FFT length 12K, Pass1=256, Pass2=48.

[Oct 29 16:24] Test 15, 560000 Lucas-Lehmer iterations of M200705 using AVX FFT length 12K, Pass1=256, Pass2=48.

[Oct 29 16:24] Test 16, 560000 Lucas-Lehmer iterations of M196607 using AVX FFT length 12K, Pass1=256, Pass2=48.

[Oct 29 16:24] Test 17, 560000 Lucas-Lehmer iterations of M194561 using AVX FFT length 12K, Pass1=256, Pass2=48.

[Oct 29 16:25] Test 18, 560000 Lucas-Lehmer iterations of M188415 using AVX FFT length 12K, Pass1=256, Pass2=48.

[Oct 29 16:25] Test 19, 560000 Lucas-Lehmer iterations of M186369 using AVX FFT length 12K, Pass1=256, Pass2=48.

[Oct 29 16:26] Test 20, 560000 Lucas-Lehmer iterations of M184319 using AVX FFT length 12K, Pass1=256, Pass2=48.

[Oct 29 16:26] Test 21, 800000 Lucas-Lehmer iterations of M172031 using AVX FFT length 12K, Pass1=256, Pass2=48.

[Oct 29 16:27] Test 22, 800000 Lucas-Lehmer iterations of M163839 using AVX FFT length 12K, Pass1=256, Pass2=48.

[Oct 29 16:27] Test 23, 800000 Lucas-Lehmer iterations of M159745 using AVX FFT length 12K, Pass1=256, Pass2=48.

[Oct 29 16:28] Test 24, 800000 Lucas-Lehmer iterations of M157695 using AVX FFT length 12K, Pass1=256, Pass2=48.

[Oct 29 16:28] Test 25, 800000 Lucas-Lehmer iterations of M155649 using AVX FFT length 12K, Pass1=256, Pass2=48.

[Oct 29 16:29] Test 26, 800000 Lucas-Lehmer iterations of M153599 using AVX FFT length 12K, Pass1=256, Pass2=48.

[Oct 29 16:30] Test 27, 800000 Lucas-Lehmer iterations of M147455 using AVX FFT length 12K, Pass1=256, Pass2=48.

[Oct 29 16:30] Test 28, 800000 Lucas-Lehmer iterations of M143361 using AVX FFT length 12K, Pass1=256, Pass2=48.

[Oct 29 16:31] FATAL ERROR: Rounding was 0.4997551888, expected less than 0.4

[Oct 29 16:31] Hardware failure detected, consult stress.txt file.

[Oct 29 16:31] Torture Test completed 27 tests in 12 minutes - 1 errors, 0 warnings.

[Oct 29 16:31] Worker stopped.

 

Any advice?

Edited October 29, 201213 yr by MikeP

Well, that seems about it for your CPU and the current cooling. Lower to 4400 and see what is the lowest vcore you can set without P95 crashing. Go in steps, not like you wrote before 0.005 to 0.1, lol. Leave now at 0.1, see if stable at 4400, if stable lower to 0.09 etc...when start crashing increase by 0.005, and see if stable. If not go up again by 0.005, see if stable. This is how a stable overclock is made. I usually just go 0.01 above the last stable, but hey, I got watercooling, so temperatures aren't really my worry... I can barely get mine to run at 5.0 because of VERY high voltage settings (read 1.56 or about).

I have it at 44 with -0.005 which seems to be stable @ 1.120-8v

 

I'm going to save up for a new corsair case with a h100 setup hopefully.

-0.005

 

What the hell is that? How do you get 1.200V with offset of -0.005 beats me... but hey, if it's working.

Be sure to run the Prime95 stable for at least 10hours to declare it stable and not heating up too much. Though your temps aren't going to go much up after some half an hour or so. It's when the case heats up. But also do count the room warming, that also plays a role in air cooling.

 

Good case and H100 are good ideas.

 

http://www.asrock.co...?Model=Z77 Pro3

 

XMP:

OC Tweaker Screen

scroll down to DRAM Timing Configuration

there should be something called XMP Profile DDR3...

and then there should be something called

"Load XMP Setting": Auto, Default, Profile 1, Profile 2... default is Auto.

 

Best is, open CPU-Z, go to the page SPD, find the slot that contains your RAM, there are 4 slots, some will show data, and see what is the XMP configuration - what is the supposed RAM XMP speed. It is the speed you should be running your RAM at.

So in the BIOS, I guess it will be Profile1. You gotta see how it works on your computer.

I don't know lol... On my voltage offset it has -0.001 and +0.001 and it raises each way in 5's...

 

any suggestions on a corsair case?

Sure: 650D or 800D, depending what size you need.

I managed to get it quite stable last night at 4.7 with an offset of +0.070 but temperatures reached 100 so I think with the right case and cooling I could get 4.7.

<p>What's the sizes differences between the cases?

 

The 800D seems a bit much for a case whereas the 650D is a bit better on the wallet, supports a dual radiator and that's what the H100 is...</p>