***Official***Z68X/P67A-UD7 Club/Discussion/Info**

[owikh84]

Gosh that means i'm pumping ~1.63v load for 5GHz for 24/7!!

[Falkentyne]

Gosh that means i'm pumping ~1.63v load for 5GHz for 24/7!!

[sin0822]

Gosh that means i'm pumping ~1.63v load for 5GHz for 24/7!!

only a DMM will tell you, on air or water LLC level1 is fine. LLC level2 is a bit much, and i can't say its pushing 1.63v you need to check with a DMM.

LLC level1 has much better voltage regulation, and allows for a vdroop which is excellent for any 24/7 operation. you don't want teh voltage increasing under load like LLC level2 does even without the offset. its just not great for your hardware period. it is for benching. And that is what sets teh UD5 and the UD7 apart from the UD4 and below, these boards are built to handle all types of stuff that overclockers on air or water don't really encounter.

owikh you shouldn't be pushing 1.53v for 24/7 anyways!!! LOL

[sin0822]

BTW this isn't something new, LLC level2 acts this way on the B2 board as well. 1.23v set gives 1.283v idle and 1.323v load on the P67A-UD7-B2 F8X.

To be VERy honest LLC Level2 has not changed a bit, LLC level1 has changed greatly! we see MUCh better voltage regulation same goes for standard.

[owikh84]

only a DMM will tell you, on air or water LLC level1 is fine. LLC level2 is a bit much, and i can't say its pushing 1.63v you need to check with a DMM.

LLC level1 has much better voltage regulation, and allows for a vdroop which is excellent for any 24/7 operation. you don't want teh voltage increasing under load like LLC level2 does even without the offset. its just not great for your hardware period. it is for benching. And that is what sets teh UD5 and the UD7 apart from the UD4 and below, these boards are built to handle all types of stuff that overclockers on air or water don't really encounter.

owikh you shouldn't be pushing 1.53v for 24/7 anyways!!! LOL

Well i blame my chip for being a voltage suucker..
But heck latest linpack is giving me more 124 codes at this 1.53v bios.
Happy overvolting guys!

[sin0822]

what about prime? I used wprime and it seems that prime is stressing the CPu more than linpack(linx/ibt)

[owikh84]

what about prime? I used wprime and it seems that prime is stressing the CPu more than linpack(linx/ibt)

Prime has never given me BSOD except for error which i believe due to mem clockings. It actually run cooler than LinX and ~10watts lower according to realtemp. that was v26.5. Well I will switch back to MIVE B3 hopefully by this weekend and see how. Prolly my chip is degrading but i hope it's not. CPU hunting mode is on again.

On the other hand ive tried 1.63v-1.65v bios LLC1 - instant BSOD 101 and 124 in LinX, spread spectrum is always disabled

[Nostromo]

Maybe I missed it?
But will B3 bios work on B2 boards?
I do not want to tear it down, as it runs perfect, and I do not use
affected drive ports.

[sin0822]

Prime has never given me BSOD except for error which i believe due to mem clockings. It actually run cooler than LinX and ~10watts lower according to realtemp. that was v26.5. Well I will switch back to MIVE B3 hopefully by this weekend and see how. Prolly my chip is degrading but i hope it's not. CPU hunting mode is on again.

On the other hand ive tried 1.63v-1.65v bios LLC1 - instant BSOD 101 and 124 in LinX, spread spectrum is always disabled

Spread spectrum makes no difference to OC anymore man.
That is odd, might as well try what you had before see if it BDOSs. I saw voltage degradation at 1.53v and above when i did my review on D1 stepping CPUs. I found that every few days it needed 0.02 more volts after i surpassed 1.53-1.55v. Your CPu might be doing the same because I was heavily stressing my CPu as well. These CPUs just aren't made to be stressed at 1.5v+ with air cooling. Especially not with IBT or prime. I wish you luck man, but 5ghz isn't too hard to hunt for, is it? Either way i wish you good hunting.

Maybe I missed it?
But will B3 bios work on B2 boards?
I do not want to tear it down, as it runs perfect, and I do not use
affected drive ports.

They modded a few above as it seems. but normal flash wont work.

[owikh84]

Spread spectrum makes no difference to OC anymore man.
That is odd, might as well try what you had before see if it BDOSs. I saw voltage degradation at 1.53v and above when i did my review on D1 stepping CPUs. I found that every few days it needed 0.02 more volts after i surpassed 1.53-1.55v. Your CPu might be doing the same because I was heavily stressing my CPu as well. These CPUs just aren't made to be stressed at 1.5v+ with air cooling. Especially not with IBT or prime. I wish you luck man, but 5ghz isn't too hard to hunt for, is it? Either way i wish you good hunting. .

Thank man. Yes 5GHz is easy to find but vcore below 1.5v is hard for me.
Ambient is 33-35c here, water is equal to ur air lol. Again thx for effort in DMM readings.

PS: praying hard for a better CPU.

[Falkentyne]

So now that means my "good" 1.41v CPU is actually running at 1.46v since I have 1.45v in BIOS LLC1 which was showing between 1.404-1.416v at load making it an average chip instead of a good one, and my first 2600k is as bad as Owi's...yay for ego deflation...

And my 4.5 ghz 1.245v load overclock is around 1.275v instead...

So instead of having a terrible chip like the first I have a barely average one...WONDERFUL.....

But I wish someone could take a DMM to a B3 UD5 to check, cuz i can't buy one now...

[andressergio]

Maybe I missed it?
But will B3 bios work on B2 boards?
I do not want to tear it down, as it runs perfect, and I do not use
affected drive ports.

here mate i have Rev B2 and i use F8f best bios for it, but if you wanna go with B3 bioses you need this that stasio explains

http://www.xtremesystems.org/forums/...263598&page=68

cheers

[Nostromo]

here mate i have Rev B2 and i use F8f best bios for it, but if you wanna go with B3 bioses you need this that stasio explains

http://www.xtremesystems.org/forums/...263598&page=68

cheers

Ah, thank you. F8F is nice.

[sin0822]

So now that means my "good" 1.41v CPU is actually running at 1.46v since I have 1.45v in BIOS LLC1 which was showing between 1.404-1.416v at load making it an average chip instead of a good one, and my first 2600k is as bad as Owi's...yay for ego deflation...

And my 4.5 ghz 1.245v load overclock is around 1.275v instead...

So instead of having a terrible chip like the first I have a barely average one...WONDERFUL.....

But I wish someone could take a DMM to a B3 UD5 to check, cuz i can't buy one now...

i think the user C-N he has one. LLC might be very different.

I looked at the results for the UD4, and it seems when LLC is turned to enable it gives the same LLC effects as UD7's LLC Level1.

[stasio]

what about prime? I used wprime and it seems that prime is stressing the CPu more than linpack(linx/ibt)

People use legacy Prime95 instead of ver.26.6,updated for Sandy Brigde.

[Falkentyne]

I see but is my CPU bad or not good anymore since it is probably getting 1.46 (and maybe eventually 1.47v) for 5 ghz when I thought it was 1.41 till the DMM chart?

[Falkentyne]

Sin, can you do me a very urgent favor?

Find a safe and lower overclock so you won't damage your chip, and can you run a TEMP comparison with LLC1 and LLC2, with the "Target HARDWARE" (DMM) voltage for LLC2, to match the target hardware voltage for LLC1, and then compare the cpu temps at load? (Use Prime blend so the temps are easily maintained).

Good Example:
Use 4.5 ghz.
Set vcore to 1.45v bios, LLC1, and compare the temps after a few minutes of Prime small FFT (small FFT will get the temps higher faster than blend does).

Then, set the vcore to 1.30v (correction) bios LLC2, 4.5 ghz, and do the same thing.

If you're worried about chip damage, you can try 4.0 ghz LLC1 1.40v, and LLC2 1.25v (correction again).

*also, please use REALTEMP 3.67 while you're doing this, and check the power draw shown at the top right clickable tab*

^^^ THANK YOU SO MUCH, SIN !!!!!

The temps AND the power draw should be the same at load for both settings, if the DMM readout is the actual CPU voltage going into the chip, and if the chip is actually using that voltage.

I'm only asking, because I saw a very bizarre post:

I think i understand what you're saying, basiclly, lower the voltage and enable LLC level 2 so the load voltage gets what it needs, and idle droops since it doesn't require as much voltage, i did some testing and here is what iv'e learned about my rig.

It seems like it doesn't matter what Level LLC is on, as long as the BIOS has x.xx amount of voltage for that particular clock, it will stay as stable as it can/should be. For example:

5.0ghz
LLC Level 1 @ (BIOS-1.38v) Load: 1.35v - Crash
LLC Level 1 @ (BIOS-1.41v) Load: 1.38v - Crash
LLC Level 1 @ (BIOS-1.43v) Load: 1.408v - Crash
LLC Level 1 @ (BIOS-1.45v) Load: 1.41v - Passes many tests

On the other hand:
5.0ghz
LLC Level 2 @ (BIOS-1.38v) Load: 1.41v - Crash
LLC Level 2 @ (BIOS-1.41v) Load: 1.43v - Some errors and hangs, no crash
LLC Level 2 @ (BIOS-1.43v) Load: 1.45v - Some errors and hangs, no crash
LLC Level 2 @ (BIOS-1.45v) Load: 1.48v - Passed some tests but doesn't "feel" stable
LLC Level 2 @ (BIOS-1.48v) Load: 1.512v - Ran this for a long time with no errors or crashes.

So you can see the trend almost immediatly, It seems like it will crash if the "BIOS Votlage" is below a certain threshold, regaurdless of any other settings, and in some cases the LLC Level 2 is more unstable, I have 2 other settings for LLC, [Auto] which i haven't tried, and Standard, i skipped standard because it said "Intel's Standard Load line cal. and i'm going to be money it sucks for anything over 1.4v.

(post was here http://www.overclockers.com/forums/s...77#post6825477 )
This is on a UD5. Which should have the same LLC as UD7.

[the_real_7]

Hey a little more LLC readings for you guys. Please do not trust software. I have been asked to do this for a range of voltages from 1.35v to 1.5v so I will do that in a 0.5v increments so that those of you WITHOUT a DMM or are a bit too scared to use one may be able to correctly set voltage to your CPUs. its a bit of work and i got a few other things i gotta do first, but i will do it!

Anyways please notice that software once again is WAY off, also notice LLC Level2.

wow very nice , damn my poor cpu is getting hit with 1.461volta now llc1 jeesh close to degrading vcore , with new ram take a bump to 1.470 . . .

[jacek]

did you download it? the OC hotkey program?

Yes of course. I downloaded the program.
it's EasyTune

[sin0822]

Sin, can you do me a very urgent favor?

Find a safe and lower overclock so you won't damage your chip, and can you run a TEMP comparison with LLC1 and LLC2, with the "Target HARDWARE" (DMM) voltage for LLC2, to match the target hardware voltage for LLC1, and then compare the cpu temps at load? (Use Prime blend so the temps are easily maintained).

Good Example:
Use 4.5 ghz.
Set vcore to 1.45v bios, LLC1, and compare the temps after a few minutes of Prime small FFT (small FFT will get the temps higher faster than blend does).

Then, set the vcore to 1.36v bios LLC2, 4.5 ghz, and do the same thing.

If you're worried about chip damage, you can try 4.0 ghz LLC1 1.35v, and LLC2 1.26v.

*also, please use REALTEMP 3.67 while you're doing this, and check the power draw shown at the top right clickable tab*

^^^ THANK YOU SO MUCH, SIN !!!!!

The temps AND the power draw should be the same at load for both settings, if the DMM readout is the actual CPU voltage going into the chip, and if the chip is actually using that voltage.

I'm only asking, because I saw a very bizarre post:


(post was here http://www.overclockers.com/forums/s...77#post6825477 )
This is on a UD5. Which should have the same LLC as UD7.

I wouldn't worry about that post, and LLC on the UD5 should be different as it was for B2 boards.

What do you want to see? The thing is that my temps are going to be much different, and temps and power draw especially monitored by software like that is prob going to be much different machine to machine. I can measure actual power draw fromt eh 8 pin connector through some device I bought, but unless you have that too it wouldn't matter.

You are asking if LLC level1 with teh same measurement on the DMM as LLC Level2 under load will be the same?

Honestly it you are staying under 1.5v from the trends of the DMM test I did you are going to be ok, i wouldn't worry about degradation over a months or a few months period. I don't know what you want me to do exactly? I am a bit confused lol

Yes of course. I downloaded the program.
it's EasyTune

Yea i just saw that, i guess they removed it for P67, no idea why

[the_real_7]

Gosh that means i'm pumping ~1.63v load for 5GHz for 24/7!!

wow man like that your chipping must be degrading , bring that vcore down dude

[the_real_7]

So now that means my "good" 1.41v CPU is actually running at 1.46v since I have 1.45v in BIOS LLC1 which was showing between 1.404-1.416v at load making it an average chip instead of a good one, and my first 2600k is as bad as Owi's...yay for ego deflation...

And my 4.5 ghz 1.245v load overclock is around 1.275v instead...

So instead of having a terrible chip like the first I have a barely average one...WONDERFUL.....

But I wish someone could take a DMM to a B3 UD5 to check, cuz i can't buy one now...

Falkentyne Im in the same shoes you are lol there goes our great chips to being just average

[sin0822]

the real 7 id take a bet that compared to other chips, not many people really read the DMM voltage, so maybe it still is golden. too many people take CPU-Z vcore to be correct, and setting gold standards on any chip that way is just incorrect. Its interesting to see other board not report same vcore as well.

[andressergio]

the real 7 id take a bet that compared to other chips, not many people really read the DMM voltage, so maybe it still is golden. too many people take CPU-Z vcore to be correct, and setting gold standards on any chip that way is just incorrect. Its interesting to see other board not report same vcore as well.

i always do that and GIGA is very accurate so far

[the_real_7]

the real 7 id take a bet that compared to other chips, not many people really read the DMM voltage, so maybe it still is golden. too many people take CPU-Z vcore to be correct, and setting gold standards on any chip that way is just incorrect. Its interesting to see other board not report same vcore as well.

Your right , Im pulling my dmm out tonight and see what my voltages look like on a prime load . . . thanks Sin