OC limitations of this build??

[kiloslov]

so i have her stable now. test going good. but question.. If im getting BSOD after adjustments is that my limitation or my configuration?

http://valid.x86.fr/q0gt6i

 

[Mandrake4565]

Just so were on the same page, you're saying its stable but how long have you stress tested it for and was it using Aida?

If you're getting BSOD and are only changing the multiplier then it's most likely, lack of CPU voltage

 

[kiloslov]

First off, thank you all for the help you've provided. I find this community very interesting and welcoming!

I stressed her for about a hour (+/-) on aida64 extreme
This is what i have currently. i got her to 4.6 on 1.45, not proud of that voltage to be honest. i really only wanted to get there around 4.8 or 4.9 any ideas on what i can fiddle with?


- - - Updated - - -

I will add that my mobo does not support LLC (**** you msi) so im sagging down to 1.4 - 1.42 under load.

- - - Updated - - -

Another addition while your reading this since were kinda having some limbo time between post.

bus clock adjustments. when and where should that even take place? I have played with it briefly though could not find a stable setting above 100. Ive seen others run around 103.00 validated. is there something i could check into to help me learn how to stabilize that?

 

[EarthDog]

That's a huge jump for 100mhz... wow.

Nothing you can do as far as adjusting other voltages to lower vcore... seems like 4.6ghz is it.

I wouldn't worry about bclk.. it has no bearing on performamce really so just leave it.

 

[Mandrake4565]

The only time I really play with the Bclk is when I'm looking for those last extra Mhz when benching my stuff. You're not going to notice any difference in everyday usage with playing with it. As far as your Oc goes, it's a silicon lottery you never know what you're going to get. 4.5/6 Ghz is still a respectable Oc, just make sure it's stable. Let Aida run at least 2 hours before you call it stable.

 

[trents]

QFT!!

I can't think of a CPU that was limited to 1.3v for overcocking. The older you go the more they could take really. But 1.4-1.45v has been the limit for Haswell, IB and SB for daily. 1.3v is nearly stock for some SB!

Really? Last year when I was overclocking my 4790k Haswell I was certain folks were saying don't exceed 1.3 by much for 24/7. Makes me feel real good that mine is rock stable at 4.9 on 1.25 vcore. This one must have been more of a cherry than I first thought. Now you know why I have no desire to upgrade to Skylake. It would probably be a performance downgrade for me.

 

[EarthDog]

Really. Those that may have told you that were being very conservative. You have a great chip, no doubt!

Edit: http://www.overclockers.com/forums/showthread.php/760902-Safe-24-7-ring-voltage-for-4790k

Looks like, somehow, you never received an answer here. Haswell stock was around 1.1 to 1.2 out of the gate. .1 or so is nothing.

 

[Mandrake4565]

Really? Last year when I was overclocking my 4790k Haswell I was certain folks were saying don't exceed 1.3 by much for 24/7. Makes me feel real good that mine is rock stable at 4.9 on 1.25 vcore. This one must have been more of a cherry than I first thought. Now you know why I have no desire to upgrade to Skylake. It would probably be a performance downgrade for me.

That is a really good chip Mr T. and no there really isn't any reason to upgrade. I have both the 4770k and 6700k and yes the 6700k is slightly faster clock for clock if you're not benching no reason to change. I will note, as you may know I have a particularly hot 4770k (delid I know ED ) and I topped out at 4.4 @ 1.23 24/7 2 hours prime, due to temps in the 90's. My 6700k I have at 4.8 2 hours Prime with 1.40 V and temps topped out at high 70 c on the same full loop.

 

[trents]

One thing kind of anecdotally that I have noticed as a difference between AMD and Intel overclocking is that Intel's seem to produce more heat than there AMD counterparts just by jacking up the clock speed without increasing the core voltage. Seems like when I was overclocking AMD chips the clock speed was a moe minor player in the amount of heat increase. Wonder if the HT technology of the intels has something to do with that? Or maybe there are more automatic voltage adjustments going on behind the scenes or something.

 

[kiloslov]

Really? Last year when I was overclocking my 4790k Haswell I was certain folks were saying don't exceed 1.3 by much for 24/7. Makes me feel real good that mine is rock stable at 4.9 on 1.25 vcore. This one must have been more of a cherry than I first thought. Now you know why I have no desire to upgrade to Skylake. It would probably be a performance downgrade for me.

yeah i really wanted to just bench it to 5. looks like im going to need to just bite the bullet and build a benching setup so i dont cook burgers on this daily build. lol

BUT thank you all. im stable at 4.5 as of this morning 8 hours aida64.


So lets get on to it, now that you have me addicted to this lol where do i start to hop on this benching train??

 

[EarthDog]

One thing kind of anecdotally that I have noticed as a difference between AMD and Intel overclocking is that Intel's seem to produce more heat than there AMD counterparts just by jacking up the clock speed without increasing the core voltage. Seems like when I was overclocking AMD chips the clock speed was a moe minor player in the amount of heat increase. Wonder if the HT technology of the intels has something to do with that? Or maybe there are more automatic voltage adjustments going on behind the scenes or something.

can't say I noticed that. The same laws of physics applyrics to both chips. My temps barely change when I raise clocks and leave voltage alone. That said, the major voltages are set manually on my systems so as not to have auto raise voltages unnecessarily.

@ kilo, check out the benching section sticky threads!!! Join the madness! Also look for the current competition we are running.

 

[kiloslov]

can't say I noticed that. The same laws of physics applyrics to both chips. My temps barely change when I raise clocks and leave voltage alone. That said, the major voltages are set manually on my systems so as not to have auto raise voltages unnecessarily.

@ kilo, check out the benching section sticky threads!!! Join the madness! Also look for the current competition we are running.

will do my man! thanks for the addiction fellas!

 

[Mandrake4565]

One thing kind of anecdotally that I have noticed as a difference between AMD and Intel overclocking is that Intel's seem to produce more heat than there AMD counterparts just by jacking up the clock speed without increasing the core voltage. Seems like when I was overclocking AMD chips the clock speed was a moe minor player in the amount of heat increase. Wonder if the HT technology of the intels has something to do with that? Or maybe there are more automatic voltage adjustments going on behind the scenes or something.

Can't say I noticed an increase in temps when I changed only the clock speed Mr. T. What I did notice specifically with the 4770k I have is it certainly reminded me of my Fx 8350 when it comes to heat. This may also be because the Haswells have the VRM on the die, in addition to the poor TIM and IHS mount on them.


kiloslov

This months competition, come join us...we have cookies.