**Official e8400/e8500 Retail OC Thread

My 8400 only wants to run with 500FSB. Anything lower or higher, and it goes unstable. My memory can eat 533x2 (Corsair 8500C2D) so it's not a problem. Motherboard is a P5K Premium.
So far it's running at 1.400V, 500x8, stable, low temps (watercooled, so temps are not limiting factor either). I am disappointed at the voltage needed to reach that point... And puzzled at the unstable 533, 445, 450 FSB settings (i put the proper divider for memory every time to keep it under 1066).
I am positive it's a CPU problem, i tried lowering the CPU multi, the motherboard would do 533MHz FSB without any glitch.

Can anyone advise on what GTL multipliers I should be using on my P5K-E Wifi? I have no idea what to set those to.

Quote:

---

Can anyone advise on what GTL multipliers I should be using on my P5K-E Wifi? I have no idea what to set those to.

---

cpu .57x, nb .61x or .61x .67x or who the hell knows...

Quote:

---

So far it's running at 1.400V, 500x8, stable, low temps (watercooled, so temps are not limiting factor either). I am disappointed at the voltage needed to reach that point...

---

mine needs 1.41v for 4ghz - it only bums me out when i am dinking around on these forums and see all the s.o.b.s doing it with 1.3v+/-. just to spite my stinker of a 8400, i am now running it at 4.2ghz with 1.5v!

Welp, here's my results for now. I know Superpi 1m isn't really a stability test but for me gaming is where stability matters for me. I've been playing Age of Conan beta, CS Source and Crysis for the past 3 hours without a crash. Maybe theres still some more left in the chip but frankly I am a complete noob when it comes to C2D overclocking (just came off the boat from s939 world).

http://img.photobucket.com/albums/v428/Tibu/4gspi1m.jpg

This isn't bad really :)


I have an important question please: can I set vfsb to 1.35 or 1.4 for everyday usage ?
Thanks in advance

After reading Anandtech's disaster, I wouldn't recommend it... I notice no difference here at 450FSB with 1.3 or even 1.4... And plester I hope ya CPU doesn't run into trouble as 1.5Vcore compares to about &.6Vcore daily for these 45nms... I would go again back to 4ghz 1.4Vcore... or sell this sucker and get another one...

Quote:

---

Originally Posted by Leeghoofd

After reading Anandtech's disaster, I wouldn't recommend it... I notice no difference here at 450FSB with 1.3 or even 1.4... And plester I hope ya CPU doesn't run into trouble as 1.5Vcore compares to about &.6Vcore daily for these 45nms... I would go again back to 4ghz 1.4Vcore... or sell this sucker and get another one...

---

well thank you. It is for h24 at 525+ fsb, at least I try. but I'll certainly get it back to 1.3 then.
Bedankt hoor ;)

Am i the only one here with Q810A076?

Only googled out another guy from some odd forums.

Volts wise i'm pretty satisfied, can get it 4.5Ghz pi 32mil stable below 1.5V for Vcore. 1.22V will take it up to 3.85Ghz orthos stable atm. Trying to get it to burn in.

Had this CPU for about a week now. Got it from NewEgg, when they had the $190 free shipping on. But due to a new PSU breaking my perfect record of no DOAs I set it aside until I got a new back-up PSU. Anyway, without further ado ... e8400 Q809A161 at the tail end of a OCCT stability check.

http://img102.imageshack.us/img102/5...6x9occthq7.jpg

First impression is that it is slightly better better than my Q746A510 with slightly less voltage for 4ghz and more room for FSB getting weird at 550 instead of 538. Am I imagining things or has Intel adjusted temp sensor? I'm testing on a mild Swiftech H20-220 Compact kit for my bench set-up but temps seem suspiciously low.

So bumping it up just 200mhz is pretty much another .1v to the vcore which seems to be about where most people have decided to stop fearing for the long term health of their 45nm CPUs.

http://img404.imageshack.us/img404/4...8x9occtec8.jpg

I'll probably start goofing around with the memory set-up now.

What is your Step man? It is not bad. I had a Q805A465, demanding lost of vcore, and I did my learnings on those 45nm fragility with it... now R.I.P.
I stood too long at 1.72vcore idle and POUF! no more cpu :shocked:

Now I got a favor from the shop where I bought it: they replaced it, and gave me the opportunity to get another stepping :p: I really had to keep quiet when the person asked me if a Q807 was ok for me !!!!!!!
Man, I almost jumped over the bureau to catch all cpu's they had in store :D

Ok, wPrime1024, 3DM03, and Occt priority cpu 1h stable at the program. This can from now be tweaked ;)
http://pix.nofrag.com/c/1/5/f7c4468b...58c17e73tt.jpg I opened Setfsb machinally :p: didn't used it I booted with that fsb, to test a bit with lowest possible tensions.
http://pix.nofrag.com/1/4/9/40287a3b...faabba42tt.jpg
http://pix.nofrag.com/3/e/b/d83ead0f...0412664att.jpg The droop is too much, no mod at all on my X38T-DQ6 (rev 1,1 and bios f7b beta).
I think I reached the limit for voltages...this cpu is not a king, but no complain at all, as I should test it with another mobo. That one for example http://www.jmax-hardware.com/forum/i...ic,2213.0.html

Regards, Toy

Quote:

---

Originally Posted by gundamit

So bumping it up just 200mhz is pretty much another .1v to the vcore which seems to be about where most people have decided to stop fearing for the long term health of their 45nm CPUs.

I'll probably start goofing around with the memory set-up now.

---

Are those 200mhz worth the 0.1 ? If I was you I would run ya rig 9 x 450 or even 8 x 500... at 1.3volts, seems more assuring to me...

and @Toyten : great CPU, tu as bien mérité mon brave...

Quote:

---

Originally Posted by Plester

cpu .57x, nb .61x or .61x .67x or who the hell knows...

---

Wait, what's the difference between .57x and .67x for CPU ? which is more aggressive ? What's the default ? What does it change actually ?

Quote:

---

Originally Posted by gmat

Wait, what's the difference between .57x and .67x for CPU ? which is more aggressive ? What's the default ? What does it change actually ?

---

The higher values should be more aggressive.

As long as your FSB is 500mhz or lower Just keep it at cpu .57x, and NB .61x.,
and and keep FSB Term (vtt) @ 1.10v (actually it's 1.20v real when set to 1.10v in bios).

And FYI, set your PLL to 1.6 in bios, because 1.5, and auto in bios sets it to 1.8v real, not good!

Does the VTT makes a difference for these 45nm's?

ı guess ı've found the most stable setting with this batch q802a570

Quote:

---

Originally Posted by aoch88

Does the VTT makes a difference for these 45nm's?

---

Try not to put more than 1.3v for everyday use, 1.5 in short benchs

Edit: pushing this value too much can simply kill your 45nm... as for Vcore and Vpll (vpcie on Gigabyte)

Quote:

---

Originally Posted by ToyTen

Try not to put more than 1.3v for everyday use, 1.5 in short benchs

Edit: pushing this value too much can simply kill your 45nm... as for Vcore and Vpll (vpcie on Gigabyte)

---

Thanks but in my case, it made no difference by increasing the VTT. I don't see the real purpose of it on 45nm C2D's but someone please correct me if I'm wrong :)

Quote:

---

Originally Posted by aoch88

Thanks but in my case, it made no difference by increasing the VTT. I don't see the real purpose of it on 45nm C2D's but someone please correct me if I'm wrong :)

---

For example, if I want to boot at 534 fsb, i need 1.4 vfsb otherswise, nothing.
In every day use I keep 500 fsb with 1.3v. Maybe the mobo would demand that.

How far have the E8400's gone on stock vcore? Here's mine running @ 3.6 on 1.080, but past 3.7 it requires more. I'm sure 4.2 is at least within reach and maybe more. But all my volts on the board are minimum and I can physically feel the heat of the case on low. I'd actually like to see how low I could go on volts and keep 3Ghz, heh.

http://img.photobucket.com/albums/v2...e8400_3600.jpg

Quote:

---

Originally Posted by ToyTen

For example, if I want to boot at 534 fsb, i need 1.4 vfsb otherswise, nothing.
In every day use I keep 500 fsb with 1.3v. Maybe the mobo would demand that.

---

That seems about right. The VTT voltage normally needs to be bumped up to make high FSB stable,
however some are able to do it with less.

well im pissed. the promised 807a i was told i would recieve was a lie.. i got a Q805a500 from what i have read these 805's suck.. anyone have any luck with them or should i send this back to get a different one?? pack date is 4/15/08

not gonna open it to check vid in case i decide to send it back.. what should i do??

Zfactor I have an OEM 805A246, does 3.6ghz at 1.08, 3.8 at 1.17 and 4Ghz at 1.25Vcore OCCT 1H stable... pretty happy with it compared to my other E8400 I had (4Ghz at 1.39Vcore)... pretty a hothead though temps ( with Intel Extreme cooler) at 4ghz going up to 70°C on the cores...

Now testing FSB and it's priming away at 7 x 500 now at 1.1volt... if it can do 4ghz at 8 X 500FSB at 1.3volts I'm pretty pleased...

is ti a e8400 or 8500?? also im on air though.. not going to water this one.. i was hoping for a bit over 4 with this.. i know someone with a 807a at 4.3 24/7 without a single issue.. i doubt ill be able to do that seeing your temps at 4

seems you are one of the few with this batch that are having luck though most i have read around here and other forums hate their 805's.. ?????

just ticks me off since they said they had a 807a held for me and they would send that.. arghhhhh i always get crap for chips..

giorgos th.: what GTL's and vtt you have on 580 fsb?