**Official e8400/e8500 Retail OC Thread

[resyn]

My last E8400 C0 was batch Q750 (450 x 9 @ 1.30v loaded cpuz) - and i considered myself fortunate with that chip, specially being that it was a c0.
This new E0 is batch Q837, pack date 12/04/08.

[LennyRhys]

My last E8400 C0 was batch Q750 (450 x 9 @ 1.30v loaded cpuz) - and i considered myself fortunate with that chip, specially being that it was a c0.
This new E0 is batch Q837, pack date 12/04/08.

Yeah, 1.3v from a C0 is actually pretty good. My C0 didn't even do 4GHz at 1.4v, but fortunately I sold it.

My E0 isn't as good voltage-wise as some of the E0s on this board, but it's incredible in terms of load temps. At 4.3GHz (1.312v) my load temps are 39C. Your chip is not exactly hot at 56C, but what cooling are you using? I've never seen even 50C from my chip yet, even at 4.4GHz 1.352v.

[resyn]

Thats nuts, 39c loaded @ 4.3ghz, incredible.
As far as my cooling, i'm using a TEC cooler (peltier)

[WaterFlex]

resyn
Nice chip mate I have the same one

[LennyRhys]

I'm using my Zalman 9700. I used it for my Q6600 also but these E0 chips just take the biscuit. These low load temps are actually hilarious, even more so seeing that my chip idles at 33/39C.

[CryptiK]

My last E8400 C0 was batch Q750 (450 x 9 @ 1.30v loaded cpuz) - and i considered myself fortunate with that chip, specially being that it was a c0.
This new E0 is batch Q837, pack date 12/04/08.

resyn
Nice chip mate I have the same one

What Vtt (vFSB) are you guys using for 4500 MHz?

I'm using my Zalman 9700. I used it for my Q6600 also but these E0 chips just take the biscuit. These low load temps are actually hilarious, even more so seeing that my chip idles at 33/39C.

That is a really cool chip. Mine runs at 58*C in intel burn test and around 50*C in Orthos, cooled by TRUE.

[resyn]

Looks like I've also found 530 x 8.5 (4505) to be stable as well, at all the same voltages.
VTT @ 1.23
MCH @ 1.44

[CryptiK]

Nice work, thats a killer chip. That's on par with the best e8600's, I saw some Q822A435's doing 4500 @ 1.32v.

[WaterFlex]

[QUOTE=CryptiK;3591620]What Vtt (vFSB) are you guys using for 4500 MHz?

I`m using:
vNB=1,312 (1,3 actual)
vFSB=1,312 (1,3 actual)
PLL=1,500
Full stable. Linpack stable too.

[CryptiK]

Did you find it needed the 1.3 Vtt was it not stable with less than that?

[WaterFlex]

Did you find it needed the 1.3 Vtt was it not stable with less than that?

I have the strange behaviour of my MIIF. It needs Vtt=vNB. I am stable at Vtt=vNB=1,28V too but I like to have it in round numbers. So, 1,30 is nice

[nullface]

NB: 1.3V
VTT: 1.1
GTL: 61x
NB GTL: 58x
Vdimm: 2.05V

[shoehorned]

I'm looking to getting an e8500 from newegg, so has anyone recently bough one from there? What's the current batch and how well can it OC?

[bigwill68]

@resyn

Great Oc on a P35 board there my friend...
I guess it doe's depend on the luck of the
draw when it comes to the Batch#'s
on these Cpu's

@shoehorned

What is the rest of your setup you gonna use
for your Rig mb,memory,cooler,psu,etc?

[LennyRhys]

I'm looking to getting an e8500 from newegg, so has anyone recently bough one from there? What's the current batch and how well can it OC?

Generally speaking, all E0s overclock well (I have yet to see one that doesn't) so as long as you get the E0 stepping it shouldn't matter what batch # it is. I specifically forked out the extra £/$ to get a guaranteed E0 and I'm glad I did.

[lochabar]

4.5Ghz 1 Hr OCCT stable on air. CPU heatsink/fan is in sig.

This is on an E8400 EO batch # Q832A161.
CPU PLL 1.60
FSB term v 1.30
Dram v 2.15
NB v 1.55
Clock Overcharge 1.00
CPU Gtl .61x
NB Gtl .61x
CPU Clock Skew 1200ps Delay
NB Clock Skew 1000ps Delay
Vcore is set to 1.4125 in BIOS, you can see the load Vcore in screenshot
Highest temp during run was 70c for a few seconds here and there. Average temp was probably 65-66c.

I'm calling this 24/7 usable. Anyone think that's a bad idea?

[WaterFlex]

lochabar
Great job. But I think vNB is high. Try to clock 500x9 and decrease vNB to 1,3.

[JoeBar]

Yes vNB is high. For 530mhz sync, i need 1.34v on my Maximus II Formula.

To call it stable. Run a good Prime run (small FFTs and blend).

[shaolin95]

Hi guys!
Based on my bios settings....what do you think I should try to get more out of this?

E8500 E0 Q827A353

By shaolin95

By shaolin95
It would seem that I need a lot more voltage to move up from here.
The memory at those timings and FSB and beyond that (520).
I did try relaxing the memory and didnt help either.
It seems that I m stock at 4.2ghz.
With a Scythe Ninja first gen I am afraid of giving it too much juice. According to cpu-z under load the vcore drops to 1.336

Thanks for any recommendation.

PS Do you think I should ask here or in my mobo's specific thread?

[JoeBar]

Yeah, 1.3v from a C0 is actually pretty good. My C0 didn't even do 4GHz at 1.4v, but fortunately I sold it.

My E0 isn't as good voltage-wise as some of the E0s on this board, but it's incredible in terms of load temps. At 4.3GHz (1.312v) my load temps are 39C. Your chip is not exactly hot at 56C, but what cooling are you using? I've never seen even 50C from my chip yet, even at 4.4GHz 1.352v.

Sorry to burst your bubble but 39'C under load with your 9700 is a no way...


Unless u have <10'C room temp (living in an igloo) or u're not loading correctly your cpu.

[LennyRhys]

Sorry to burst your bubble but 39'C under load with your 9700 is a no way...


Unless u have <10'C room temp (living in an igloo) or u're not loading correctly your cpu.

An igloo at 9°C? I'd love to see that...

[DOM.]

E8500 E0 @ 4.4GHz 550*8 at 1.392v

[LennyRhys]

E8500 E0 @ 4.4GHz 550*8 at 1.392v

That's because of the low multi. With the 9.5 multi it could probably do 4.4GHz on closer to 1.3v. Taking the multi down even just half a step to 9, I need a lot more vcore. Why is your FSB so high lol?

[lochabar]

lochabar
Great job. But I think vNB is high. Try to clock 500x9 and decrease vNB to 1,3.

Alright, this is one hour OCCT stable also. All settings are the same as the 8.5x/530FSB run, with the exception of the v NB being 1.40 instead of 1.55. Turns out this made the CPU temps lower during load. Max was 66c and average was probably 60c. I kind of butchered the screen shot resizing it, lol.

One weird thing, all of a sudden, memset isn't reading my timings right. I have everything set via the EPP 500 spd you can see in the screen shot. There is no saved timings profile, so memset isn't loading these settings at startup, no idea why it is reading this. EDIT: the tRFC reading of 55 IS correct.

I ran some quick Sandra RAM bandwidth benches at both settings and the numbers are within 400 points with the 530 FSB run being the higher of the two, so am now debating which way to go for 24/7. The age old tighter timings/lower performance level, Vs. higher frequency slightly better bandwidth question. I might even be able to get 8.5x530 with the 1.40 v NB, haven't tried yet.

If this screenie is unreadable, I'll do another run and retake it. Let me know.

[CryptiK]

Nice speed lochabar, what is your batch number? If you want to know which of the 2 settings is faster, do a superpi 32M run at each setting. The higher FSB setting may win out, but if its only a little better and it requires more voltage/heat etc, maybe the slightly lower setting would be better for 24/7. It's unlikely you would notice a small difference in your day to day use of the pc.

You may be better off running say 4.4GHz & DDR2-1174 CL5 if your ram can handle it at a safe voltage.