E8600 Prime Stable Overclocking Database

[Dee]

1402? There's a new one?

If Ket has done a modded version I'll give it a try.

By the way, I'm still learning about this board and it's various settings, so don't take my word for it when I say "try 1.12". All i know is it seems to be working for me with a slight nudge above default (1.10 for 45nm). I'm happy to be corrected if someone comes along and says "you should use 1.xx instead".



EDIT: 1402 is for Deluxe, not E. Booo!

Oh well. I'm quite happy with m1306, it kicks ass.

[BeastNotro]

Now Key has not done one yet ... PMP has done a 1402 for Deluxe and 1307 for P5Q .. not sure if that means P5Q-E.

Look at last post on previous page i listed his origianl thread with thsoe two bioes in it.

Some have flashed to that 1402 and they say it working fine and they using less volts then before of CPU and a few others like NB too.

Yeah i will try 1.15-1.17 and so forth, but i ain't doing 5 primes of 16hrs for em all....

I will lower once (when i have time) and try it and if its not good .... i will up it again...

[Dee]

Ah, very nice. That's what we want.

[BeastNotro]

Ah, very nice. That's what we want.

Yeah, but some have to use same volts...... lol

The 1307 is P5Q is that for your mobo?

If so you could give it a try at your own risk......

[Dee]

Naw, mine is P5Q-E. One step below the Deluxe (8-phase power instead of 16).

[BeastNotro]

Thats what i figured, but ya never know with asus...

Maybe same bios, but they say its not.. Bahahahahaaa

Either way don't try it as i doubt it is same one.

[Dee]

Indeed, I think I'll pass on flashing to the wrong BIOS.

[Devious]

Batch: Q822A435
Speed: 4500Mhz
Vcore: 1.344
Vid: 1.1625
Custom Water
Small FFT 12 hours
http://img523.imageshack.us/img523/9...mallfftxq4.jpg

[Dee]

Minor update.

Beast: you're gonna start hating me soon with these small OC updates.

Batch: Q820A670 (you put Q22 next to my entry, needs correcting)
Miles Per Hour: 4.33GHz
VID: 1.2500
vCore: 1.408V (load)
Small FFT: 3 Hours

I'm pretty sure I can drop the vCore a little, so yeah, that probably means another minor update soon!

[Dee]

Batch: Q822A435
Speed: 4500Mhz
Vcore: 1.344
Vid: 1.1625
Custom Water
Small FFT 12 hours

Hey Devious, Tjmax is 100 degrees, not 95. Your software's outta date.

[ToyTen]

Hi guys,

I know 3 hours is required but several times I heard that 1 hours of that particular small fft stability test was ok for cpu its stability.
So I was wondering if 1-2hours of small fft Prime95 was enough,kowing that Blend and Large aren't targeting cpu are they? mostly for the whole rig those 2 lasts benches.

It is not so important to me but let me know

[ToyTen]

Here is my little cpu in small fft priming:








This with the following setups straight from BIOS:

strap266,
PL9,
Twister auto,
DRAM dynamic / static auto,
Vcore 1.24375v,
CPU GTL(0/2 et 1/3) auto,
Vpll 1,5v,
Vfsb 1.3v,
Vddr 1.9v,
Vnb 1.55v,
NB Voltage Reference 0,61,
Vsb 1,05v,
COCvoltage 1,00,
LLC Performance,
CPU Spread Spectrum et PCIE "" disable,
no skew at all

No need to say that this will be tweaked at its max to become my every day setup

crazy low Vcore

[Dee]

Hi guys,

I know 3 hours is required but several times I heard that 1 hours of that particular small fft stability test was ok for cpu its stability.
So I was wondering if 1-2hours of small fft Prime95 was enough,kowing that Blend and Large aren't targeting cpu are they? mostly for the whole rig those 2 lasts benches.

It is not so important to me but let me know

Nobody stated small FFT's were necessary, but if you can do 3 hours that'll prove your chip is at least semi-hardcore!

Nice OC's, btw. I wish my VID was lower.

[BeastNotro]

I have updated the list some more with 2 more names added Devious and ToyTen ..... Also i updated DEE's OC to his new 4.3ghz and image as well.

ToyTen 3 hrs minimum is required ( i don't care if ppl used SmallFtt or Blend, as long as it is 3hrs or more if you want) because like i stated in a previous post many OC's will fail within that time frame. Example: I tried 4.4ghz at 1.40v and it failed in 56 mins running Orthos Blend test.
My 4.2ghz is 16hrs Orthos Blend Test stable (i shut her down at that time frame, it did not fail) and its stable enough for me.

I never played around yet with the GTL settings for this CPU, but i will state this .. going from 4.2ghz-4.4ghz is no major performance benefit and also many will have to decrease their memory mhz to push that high ( i said many, but not all) .... so i am happy with 4.2ghz ... sure can i do 4.5ghz Maybe with 1.45-1.47 volts as many of these CPU's can do so and many with lessor volts ... I won't spend 400.00 on a cooling setup to get an extra 300mhz...

Also your CPU batch number is a good one so take a look at the list and see what Devious and KitFit1 has achieved with the same batch number as yours ... KitFit1 is using a Mach GT II and is at -23c.. as for Devious he is using a Custom Water Cooling setup (its in his Signature) so i assume he is using that.

[ToyTen]

Nobody stated small FFT's were necessary, but if you can do 3 hours that'll prove your chip is at least semi-hardcore!

Hi,
Do you mean I should stress it 8hours or so ?

[ToyTen]

I have updated the list some more with 2 more names added Devious and ToyTen ..... Also i updated DEE's OC to his new 4.3ghz and image as well.

ToyTen 3 hrs minimum is required ( i don't care if ppl used SmallFtt or Blend, as long as it is 3hrs or more if you want) because like i stated in a previous post many OC's will fail within that time frame. Example: I tried 4.4ghz at 1.40v and it failed in 56 mins running Orthos Blend test.
My 4.2ghz is 16hrs Orthos Blend Test stable (i shut her down at that time frame, it did not fail) and its stable enough for me.

I never played around yet with the GTL settings for this CPU, but i will state this .. going from 4.2ghz-4.4ghz is no major performance benefit and also many will have to decrease their memory mhz to push that high ( i said many, but not all) .... so i am happy with 4.2ghz ... sure can i do 4.5ghz Maybe with 1.45-1.47 volts as many of these CPU's can do so and many with lessor volts ... I won't spend 400.00 on a cooling setup to get an extra 300mhz...

Also you CPU batch number is a good one so take a look at the list and see what Devious and KitFit1 has achieved with the same batch number as yours ... KitFit1 is using a Mach GT II and is at -23c.. as for Devious he is using a Custom Water Cooling setup (its in his Signature) so i assume he is using that.

Thanks for the database update BeastNotro

Oki, I rely on your word as they can't be false at all, I have a little experience in that matter too confirming it to me I was theorically wondering about small fft 1h bench would be enough but ok, that's it 'got the point

So, I know I can push it furtherer, till 1,39 real Vcore load/idle.
But it is not what I want to achieve: best bandwidth with harmonious settings, and the less volts as possible too naturally.
I'm not too far anymore

Good continuation to you guys

[BeastNotro]

Hi,
Do you mean I should stress it 8hours or so ?

Many will do 8 hours SmallFtt (CPU) and 8 Hours LargeFTT (memory)

I always run Blend for a minimum of 12 hours.

Sicne you have a decent/good batch number why stop at 4.2ghz as that batch number so far many can do 4.5ghz with decent volts... thats what you should shoot for and PRIME for long hours ... its up to the person on what they decide what is stable so run what you like.... . but for this thread 3 hours minimum.

Edit:

Look at Devious OC 1.34v for 4.5ghz on similar batch number as yours so try it... .. make sure you have decent/good cooling (like always for OCing. )

One other thing to note many ppl's OC can fail in 16hrs or 10hrs or 5 hours or whenever hours ... its up to the person to decide whats stable for them .. like i stated as for me ... Minimum of 12hrs Orthos Blend is what i like or more.
-------

[Dee]

Beast, I posted another update. 4.33GHz, see edited post up there ^^^

[ToyTen]

Yes everybody has it's own conception of stability, I think that furtherer then 3 hours at such a stress is going to more stable than almost all computers bought like at Dell or Acer, ...
So I will be very happy with that

One thing concerning prime's category of benches: according to what I read on the log itself Blend it the best targetted one to test ram's stability right ? Large is in between small fft's and blend and must run longer. Am I correct on this ?

Thx

[BeastNotro]

UPDATE

ToyTen what was the Full Load of your OC when priming as 1.22 is the Idle when prime was shut down?

I must PM another Two in the list as i missed theirs also....

PPL make sure prime is running still when you take the screen shot of the Full Load Vcore.

Edit:

Ok the other two i stated in their posts the Full Load and provided screen shots ... RCG_Bex and Nugzo ..

Ok just remember ppl requirement is the FULL LOAD Vcore and yes i like to see the Screen Shot of it and best thing is to do ... is do so while prime is running still.. Thks

[ToyTen]

UPDATE

ToyTen what was the Full Load of your OC when priming as 1.22 is the Idle when prime was shut down?

I must PM another one in the list too as i missed his also....

PPL make sure prime is running still when you take the screen shot of the Full Load Vcore.

There you go, I misunderstood what was required BUT ... I took this a little before "stopping" the test:


There is no droop at all on this awsome Premium X48

[BeastNotro]

Ok ToyTen i will accept it ... just make sure for the future that you take screenshot of full load even if it fluctuates Upwards .. most mobo's do so, fluctuate that is . K .... thks.

[BeastNotro]

DEE i corrected my type of your batch number ...

Also updated your OC again and link to picture of it .... now if you can lower Vcore more and OC higher .. post away....

What is your GTL settings at?

[ToyTen]

no thx, you asked it to fulfill the chart

A question that JohnnyV asked on the Premium mobo topic: what would be too much concerning the vFSB ?
Max value is not good ? (I don't know myself)

[jaredpace]

nice DB beastnostro. Keep the results coming!