~~~~~IMPORTANT SANDYBRIDGE OC INFO: Voltages & OC Guide!!~~~~~

[Stop It Rawr]

According to Stop It Rawr you can exceed 1.4V with no problems it seems and that Gibbo and Intel are clueless.

Nice try, but you may want to read a bit more thoroughly before trying to troll me and by proxy (Read the Screw Intel thread for details if anyone wants to know what this guys agenda is), the many other (frankly, better overclockers than I am because I am letting the enthusiast community do their magic before I go for silly voltages) people across the tech community who are giving Sandy Bridge hell.

You may also want to read the OP again.

Not all CPU's are as fragile as others, we have experimented upto 1.50v Vcore and 1.70v memory and had zero issues with reliability, so it seems some of fine when pushing hard.

Oh looky here, what does this say? As I stated in the other thread, Gibbo has posted the guidelines in the OP for the general community and to cover themselves from a multitude of people going OTT with their overclocks with nary a thought for the risks. As he has also stated though even if people kill their CPU's, OCUK will be honouring warranties even if people push hard, so they can't be that concerned about failure rates just yet.

 

[DBT85]

That is different to my results, everything on auto except BCLK, Multi

4ghz 1.240v auto set
4.3ghz 1.280v auto set
4.5ghz 1.3v auto set
4.6ghz 1.35 auto set
4.7ghz 1.392 auto set
4.8ghz 1.42 auto set

So for me auto volts works brilliant giving me just enough volts to be stable. Also I had installed my Akasa Venom CPU fan the wrong side of the heatpipe, its now in the proper position to the right of it near the RAM, temps have dropped 2C so now only 70C at 4.6ghz! The Akasa Venom knocked 20C off my temps!

Are those the voltages you are seeing displayed in the asus tools or in something like CPUz while under prime load? I ask because that Venom is a good cooler, and i'm getting about 10c cooler temps than you at 4.5 & 1.31v. Mine were only getting up near 70 when i was on auto.

Is it normal to have a core running much hotter than the others? I've got a core that is running like 8-10c higher than the others, though occasionally they run at the same temps. I've reapplied the paste and reset the heatsink and it's still the same.

I've noticed the same. At first i assumed i'd done the TIM wrong but even after doing it a second time the temps are the same. At idle they all match up, under load i have two that are fairly similar, one about 5c higher and one about 5c lower.

 

[Stop It Rawr]

Are those the voltages you are seeing displayed in the asus tools or in something like CPUz while under prime load? I ask because that Venom is a good cooler, and i'm getting about 10c cooler temps than you at 4.5 & 1.31v. Mine were only getting up near 70 when i was on auto.

I've noticed the same. At first i assumed i'd done the TIM wrong but even after doing it a second time the temps are the same. At idle they all match up, under load i have two that are fairly similar, one about 5c higher and one about 5c lower.

I think ambient temperatures may also be playing a part here, as is case temperatures and cooling outside of the CPU cooler alone. That said a 10degree difference for the same cooler/chip is quite dramatic, so a CPU-Z voltage assessment would be useful.

As for the temps per core. in CoreTemp my the first core is around 3degrees warmer at idle, yet is the coolest when under OCCT or IBT load. In all honesty, as long as all 4 DTS sensors are moving and are clearly not stuck past a certain level, it may be worth taking the average temperature as an indication of overall CPU temp or just eliminating the "coolest and warmest" DTS readings and use the 2 middle ones.

 

[CAT-THE-FIFTH]

Nice try, but you may want to read a bit more thoroughly before trying to troll me and by proxy (Read the Screw Intel thread for details if anyone wants to know what this guys agenda is), the many other (frankly, better overclockers than I am because I am letting the enthusiast community do their magic before I go for silly voltages) people across the tech community who are giving Sandy Bridge hell.

You may also want to read the OP again.


Oh looky here, what does this say? As I stated in the other thread, Gibbo has posted the guidelines in the OP for the general community and to cover themselves from a multitude of people going OTT with their overclocks with nary a thought for the risks. As he has also stated though even if people kill their CPU's, OCUK will be honouring warranties even if people push hard, so they can't be that concerned about failure rates just yet.

Troll?? Agenda?? I have been on OcUK longer than you have and I certainly never had an issue before.

Your comments on people who don't spend loads on CPUs as not being true enthusiasts(in the other thread) are just great.

If you are so confident with pushing 1.5V through your new CPU then why don't you do so instead of waiting for others to take the risk??

 

[silvercut]

Which particular Dominators ?

These ones: http://www.overclockers.co.uk/showproduct.php?prodid=MY-288-CS&groupid=701&catid=8&subcat=1517

 

[JonnyGLC]

Well, my chip literally will not go over 4.6GHz.

I tried loads of stuff last night all the way upto 1.4V to hit 4.7, to no avail - windows just freezes on the boot screen.

After reading all the stuff about Auto in this thread, out of curiosity I just set everything to Auto and set 47 multiplier and the same thing happens

Enabled the CPU PLL Overvolt? Mine wouldn't boot until that was enabled at 4.6ghz

EDIT: Never mind, just saw that you've updated since!

 

[Max3]

Has anyone actually finished ocing and started enjoying the fruits of their labour with increased fps in their favourite games/apps or anything?

I'm watching the borne trilogy whilst linx is doing a 50run

 

[Stop It Rawr]

Has anyone actually finished ocing and started enjoying the fruits of their labour with increased fps in their favourite games/apps or anything?

I'm watching the borne trilogy whilst linx is doing a 50run

I have tested a few games that were causing me issues with my old rig and all, so far, have gained significant FPS due to the upgrade so I am happy.

One thing particularly important to me is F1 2010, and the ability to run the game at 60FPS online and record using Fraps at 1920 x 1080 without FPS reduction is really good as I am going to be part of a little championship and would like to record our races.

 

[rchads89]

Got mine stable on the UD7. Although the UD7 is going back to ocuk maybe tomorrow as it came to me tampered with and scratched.


But anyway i have the cpu clocked @ 4.8 turbo off all energy features off, on 1.4vcore, not the best but not bad at all! ( i actually think i can lower the vcore ).

 

[ChrisGTL]

Just got my Titan Fenrir Evo with Akasa Viper fan installed so started messing with a better OC tonight.

44x - 1.30v
Vdimm - 1.35V
Calibration disabled

Running Prime95 small FFT's as we speak......

CoreTemps reporting:-

Core0 = 49
Core1 = 53
Core2 = 51
Core3 = 50

Looking good so far........still not sure why CPU-Z is reporting my core voltage of 1.068V.

I'm running the Gigabyte UD3 with latest BIOS.

 

[rustEswan]

Well the 1.8b6 MSI bios seems to be pretty decent. I can do 4.7ghz @ 1.392v (CPU-z) load but taking heed of the advice clocked it back to 4.6ghz which only needs 1.344v to be 100% stable. Ill stick here and start to get some proper day to day experiance with this thing, a week of dicking around (mainly with a sub-par Asus board) im ready to have a crack and let the bioses mature.

The previous 1.7 bios that mysteriously appeared on the MSI site then dissappeared within hours was also far better than the 1.6 and with each new revision vcore required for a particular clock seems to be steadily dropping so my advice currently would be stick to guidelines, in the next few weeks higher clocks could be available at safer voltages.

 

[rchads89]

Well the 1.8b6 MSI bios seems to be pretty decent. I can do 4.7ghz @ 1.392v (CPU-z) load but taking heed of the advice clocked it back to 4.6ghz which only needs 1.344v to be 100% stable. Ill stick here and start to get some proper day to day experiance with this thing, a week of dicking around (mainly with a sub-par Asus board) im ready to have a crack and let the bioses mature.

The previous 1.7 bios that mysteriously appeared on the MSI site then dissappeared within hours was also far better than the 1.6 and with each new revision vcore required for a particular clock seems to be steadily dropping so my advice currently would be stick to guidelines, in the next few weeks higher clocks could be available at safer voltages.

So you had a asus before and now gone to MSI? IF i have read that right 0o

 

[MikeHiow]

Well, 4.8GHz at 1.39V failed on Prime, but I've no idea when. I left at 20.30, and came back a moment ago to find the machine frozen 20% through a page dump on a blue screen. Something about a failed instruction on secondary processor.

I've gone back to 4.6GHz and 1.34v.

 

[rustEswan]

So you had a asus before and now gone to MSI? IF i have read that right 0o

Yeah see this thread...
http://forums.overclockers.co.uk/showthread.php?t=18228203

...Not mine but posted lots in there, to cut a long story short - no end of issues, lovely EFI interface but poor bios code imo, so bad I dont think it should be on sale yet. I dont doubt the hardware is fantastic and will be awesome when the bios is sorted but right now I wouldnt touch it with a barge pole.

To keep this on topic, with a lot of testing different settings on the Asus its voltage behaviour was just horribly unpredictable, when it did POST, which was rarely, it would always seem to way over compensate for vdroop making it a nightmare to remain within these recommendations.

Having said all that, im really happy with the MSI but dont think I can bring myself to send the Asus back - Im very tempted to hang onto it and wait for the bios to get to a usable state.... but its a lot of money to be sat in the box when its not really required.

 

[JonnyGLC]

Just to say... My 2500 would run IBT and Linx all day and night and would even run Prime small FFT overnight as well.

20 minutes into throwing Prime blend mode at it, BSOD

 

[rustEswan]

Just to say... My 2500 would run IBT and Linx all day and night and would even run Prime small FFT overnight as well.

20 minutes into throwing Prime blend mode at it, BSOD

Given the differences between the tests id suggest mem voltage could be the culprit depending on your settings.

I tested my lowest vmem possible for XMS3 1600 9-9-9-24, which was a 1.525 with various mem test programs before attacking the CPU and then stuck to prime95 blend as a stability/temperature test always checking my vcore on CPU-z as per earlier posts.

 

[rchads89]

Yeah see this thread...
http://forums.overclockers.co.uk/showthread.php?t=18228203

...Not mine but posted lots in there, to cut a long story short - no end of issues, lovely EFI interface but poor bios code imo, so bad I dont think it should be on sale yet. I dont doubt the hardware is fantastic and will be awesome when the bios is sorted but right now I wouldnt touch it with a barge pole.

To keep this on topic, with a lot of testing different settings on the Asus its voltage behaviour was just horribly unpredictable, when it did POST, which was rarely, it would always seem to way over compensate for vdroop making it a nightmare to remain within these recommendations.

Having said all that, im really happy with the MSI but dont think I can bring myself to send the Asus back - Im very tempted to hang onto it and wait for the bios to get to a usable state.... but its a lot of money to be sat in the box when its not really required.

Ahh yeah heard bad things about ASUS, nothing they can't fix though.

Well to be honest Asus overs nothing more than what the MSI offers what you have now so i would send it back and buy a SSD or something

Heard very good things about that msi gd65, apprently the quality is amazing which i can't understand because dont Asus own MSI now? ud think asus would skimp on the parts with it being cheaper than there boards?

 

[TrojanWhore]

@ Gibbo:
I have a 2500k, asus PRO, XMS3, so basically like the allosaur bundle.
Can I ask which BIOS version you are using for the pre-clocked bundles?

didn't get an answer before, so in essence, I'm bumping my own question....sorry if that's frowned upon.

or to put it another way, what issues do each of the BIOS' available cause and/or solve.

I'm on 0402 here, which not only is not on the asus site (pulled due to problems?), is a lower version number than the so called "first released" on the site (0804)

Im not planning to clock for a while yet (at least not till this starts to feel slow), so im only really bothered about stability for now, but I'd rather only flash once if possible.

Trojan

 

[rchads89]

didn't get an answer before, so in essence, I'm bumping my own question....sorry if that's frowned upon.

or to put it another way, what issues do each of the BIOS' available cause and/or solve.

I'm on 0402 here, which not only is not on the asus site (pulled due to problems?), is a lower version number than the so called "first released" on the site (0804)

Im not planning to clock for a while yet (at least not till this starts to feel slow), so im only really bothered about stability for now, but I'd rather only flash once if possible.

Trojan

Why only once? Anyone buying a p67 board should flash the bios straight away.

 

[rustEswan]

^^+1 that 0402 bios is god awful, with the asus board all I was looking for was stability didnt even get around to properly playing with OC's it was that bad.

It did improve stability marginally with each release i tried, the best (but still terrible) was the beta 1053 version. Just pay extra attention to those voltages I swear the damn thing will give you something you're not expecting even on auto v at stock clocks I had some unexpected, if not destructive/worrying voltages.