E8600 E0 Overclocks

[VTR]

4.3Ghz stable, seeing how far I can go without taking the voltage up any higher. 4.4Ghz required a lot more voltage previously but now I have "burnt it in" for a while and tweaked around with the bios settings I'm wondering if I can get a bit higher....

edit - 4.33Ghz failed within 10mins, looks like 4.3Ghz is going to be the max I can get out of this chip without upping the voltage.

 

[pastymuncher]

4.3Ghz stable, seeing how far I can go without taking the voltage up any higher. 4.4Ghz required a lot more voltage previously but now I have "burnt it in" for a while and tweaked around with the bios settings I'm wondering if I can get a bit higher....

edit - 4.33Ghz failed within 10mins, looks like 4.3Ghz is going to be the max I can get out of this chip without upping the voltage.

Nice clock. Have you done the pencil mod yet? P5E should be good for at least 540mhz.

 

[KillBoY_UK]

Just orderd mine ! see if a can get a good one after my crappy E6600 i got 15months ago , hoping for 4.5/4.6 under my water setup (Swiftech GTX + 120.2 Rad) with maybe shot at 5ghz if only to bench it at that , i gess ill find out at 8:00 in the morning

 

[Dee!]

Your a Joker, you are hands down the biggest moaner and groaner I ever did see on these forums, you joined a few weeks ago did nothing but complain about one thing and another, my case is POS, my CPU is a POS, etc etc depressing stuff to read I can tell u!

Seriously do yourself a favour and assume you are an utter nOOb who knows nothing nada zilch, that way you may be in a better position to learn a thing or two from those that have been in this game a lot longer than you imagined!

You won't last long here Dee! if you don't change your attitude. . .

Personal attacks will not be tolerated.

Post reported.

PS: I'm not a new member at all, I came here 5 years ago, only I lost my original login details. I haven't moaned even once. I expressed frustration because my old case didn't fit many good coolers, that's all. Please stop exaggerating.

EDIT: My exact post:

I hate the Wave Master now. Great build quality, but at least three completely stupid design flaws.

THAT is exactly why you jumped on me for "whining". I can't come to a computer forum and post my opinion on gear I own without being attacked? "Shut up. Be Happy". No.

 

[VTR]

Nice clock. Have you done the pencil mod yet? P5E should be good for at least 540mhz.

Not yet, the weekend if I've got time

What voltages should I need on the NB for what FSB's? What's the highest 24/7 NB voltage for stock cooling?

 

[Bizarre]

I won't be Priming for 8+ hours because I find it a waste of time when IntelBurnTest pushes the CPU harder and in far less time.

In my experiance, no it isn't.

I got better things to do than wait for 8+ hours of Prime to tell me if an OC is stable when I don't fully trust that it's correct. Too may people over at XS have run Prime for longer than 8 hours already, only to find out their CPU fails within 5 minutes of IBT.

I don't trust IBT either, and if you read on you'll find I have a genuine reason not to. And why not leave prime running overnight?

I use OCCT and Super Pi 32M for quick tests while finding a solid OC, then IBT to verify it. Also, I won't be replying to no arguments on this -- too many people outside these forums already proved the doubters wrong, so I don't need anyone here to try and convince me I'm wrong for using IBT. I believe those people can't successfully run it, that's why they doubt. IBT shatters their illusions.
I'll post my results elsewhere, possibly on Friday when I get the new cooler.

Hmm, I think I'm about to shatter your illusions...

It says in the IBT readme file that an 8 min run is the equivalent to 40 hours of prime, correct?

^^ So tell me, why has my E8600 just failed an orthos small fft's run at 2 hours 16 mins when it'll pass 10 tests/15 minutes of IBT without problem?





Honestly, if an 8+ hour prime stable system is falling over after a few mins of IBT then it's more than likely to do with heat issues rather than an unstable oc (as we all know IBT likes to cook the cpu a good 15-20c higher than prime).

One last thing, you'd do well not to believe everything you read on xs.

 

[Dee!]

^^ So tell me, why has my E8600 just failed an orthos small fft's run at 2 hours 16 mins when it'll pass 10 tests/15 minutes of IBT without problem?

One last thing, you'd do well not to believe everything you read on xs.

You'd be better off asking AgentGOD, the guy responsible for IBT. You can find his email address on his website. the most likely cause is it's your setup; RAM, operating system, whatever.

PS: I don't actually visit XS all that much. All the quotes I posted and most of the stuff I've read were from various other forums. XS is not a bad place, I don't understand the silly rivalry here. The fact is, XS has more members and it's one of the most well-established overclockers forums around. Why overlook or brush aside the information of experienced XS members? Sure, I'll ignore the trolls and n00bs, but that goes for any forum.

Anyway, use IBT if you want, or don't. Arguing back and forth about it with you guys is draining, repetitive and negative, and that's bad for your soul.

 

[VTR]

^^ So tell me, why has my E8600 just failed an orthos small fft's run at 2 hours 16 mins when it'll pass 10 tests/15 minutes of IBT without problem?

I'm finding the same, can run 10 tests of IBT but Prime will fail within 30mins-2hrs. Don't rate IBT at all, good for getting temps up though!!

 

[Bizarre]

I don't understand the silly rivalry here. The fact is, XS has more members and it's one of the most well-established overclockers forums around. Why overlook or brush aside the information of experienced XS members? Sure, I'll ignore the trolls and n00bs, but that goes for any forum.

I don't overlook the established members, there's actually a fair few experianced members on here who post on xs as well, I'm not saying it's fully of bs'ers.

Arguing back and forth about it with you guys is draining, repetitive and negative, and that's bad for your soul.

I couldn't agree more.

I'm finding the same, can run 10 tests of IBT but Prime will fail within 30mins-2hrs. Don't rate IBT at all, good for getting temps up though!!

Yep it has its uses, it just won't be replacing prime anytime soon as my no.1 stress tester.

 

[kitfit1]

@BornEvil, you have the wrong mobo listed in the database for my clock. It is the DFI LP UT-P35, not an Asus.

 

[willhub]

I dont know what to believe, loads of people are having this IBT pass like 10 rounds yet fail orthos of prime, they must work differently.

But then again.

Small FFT's on orthos does not stress the ram, IBT does stress ram.

I'm gonna do an 8 hour orthos run now with my E8500 at 4gig 1.32V and if it passes I'll then run IBT, if IBT fails something is wrong, yes orthos is running Small FFT's but it cant be my ram, its at 1062Mhz 5-5-5-15 2.0V, stock is all same but 1066Mhz.

I think if people want to run IBT for testing, then fine but they should do at least a 4 hour.

 

[Rainmaker]

OK I'm a little stuck now. Bear in mind this is my first ever clock, and I'm learning as I go. I got 4.0Ghz stable on stock volts 4.2 however is eluding me even with the CPU volts set to "auto" in the bios:

Speedstep/CE1 etc is all disabled. Memory is set to 2.1v and 4,4,4,15 (its rated speed/voltage) and is at a 1:1 ratio with the CPU FSB. I have the CPU voltage on auto for now, hoping to find a nice clock and then finally manually lower the voltage for the sweet spot. However I'm getting errors even with that setup. It's not like I'm even at 4.5 yet!

Any thoughts, guys?

EDIT: Oh, and I keep getting THIS BSOD when I'm running prime or orthos. Googling suggests it's a Vista x64 bug, but I only seem to get it when overclocked?

 

[kitfit1]

I dont know what to believe, loads of people are having this IBT pass like 10 rounds yet fail orthos of prime, they must work differently.

But then again.

Small FFT's on orthos does not stress the ram, IBT does stress ram.

I'm gonna do an 8 hour orthos run now with my E8500 at 4gig 1.32V and if it passes I'll then run IBT, if IBT fails something is wrong, yes orthos is running Small FFT's but it cant be my ram, its at 1062Mhz 5-5-5-15 2.0V, stock is all same but 1066Mhz.

I think if people want to run IBT for testing, then fine but they should do at least a 4 hour.

Whatever anybody has said in this thread or claimed for it, IBT does not test ram or the memory subsystem. The only proper way to test any clock is to run Prime/Orthos Blend test for 8hrs/24hrs. It dos'nt matter how many years you go back, there have always been people looking for a short cut to a stable clock. I don't know how many times it has to be said, but there are NO shortcuts. Just running small fft's, and then claiming a stable clock is also a myth, just like running IBT is. Getting a good stable clock is a mixture of "Blend", all the "3D Marks", "Gaming", "Encoding". Another myth is "it can't be my ram, it's rated at x and it's only running at y". At a given clock on a given divider, just because ram is under stock, does not mean that it will actually be stable, even if Memtest says it is. In the O/S envirement, it could well be ustable for all sort's of reasons. That's why we run "blend".

 

[willhub]

8 hours Orthos stable is a good sign of stability and its done me in the past.

If IBT does not test ram then why does it use all the ram, it must be stressing it..

 

[kitfit1]

If IBT does not test ram then why does it use all the ram, it must be stressing it..

It's using all the available ram (or whatever you tell it to use) simply to do the calculations. But it is not testing or stressing the ram. The same calculations could just as easily be done using the Page file and you would get the same result. So, as i said, for a test of the whole system Prime/Orthos Blend is the only way.

 

[willhub]

So why when you get an error does the app say it could be your CPU/ram that is unstable, as if the writer of the program is claiming it stresses the ram O_o

And at 3.8Ghz I was unstable in IBT, took the GeIL Ultra overclocked too high out, and it was fine.

 

[VTR]

Rainmaker, you will probably need a bit more Vcore voltage than that. Set it manually in the bios, also, make sure you check what actual voltage you get when you set it in the bios as it will usually be lower due to vdrop (and lower under load due to vdroop). You can check the voltage in CPU-Z.

My chip is the same batch as yours and needs slightly more than 1.3v to get stable at 4.2Ghz.

Set it manually in the bios to whatever setting that will give you a load voltage of around 1.32v and then try running Prime again.

Based on my experience with the same batch you should hit 4.3Ghz with around 1.37v, 4.4Ghz for me then required a lot more voltage (1.44v) - too much for 24/7 use IMO after doing a bit of research.

 

[Rainmaker]

Thanks VTR, I'll give it a go now and report back

 

[Big.Wayne]

Ah this thread is now loads more interesting, some great results getting posted up, very encouraging!

It is rather tempting to source one of these E0 Wolfdales and see how they run with low voltage, like what is the minimum vCore you can get away with for stock speeds and also say 4.0GHz?

I guess I'll have to be patient as it seems everybody is still finding their max clocks . . . seems like 4GHz to 4.5GHz is achievable!

 

[Rainmaker]

I'm currently 4.2Ghz Orthos stable (albeit 20 mins for now, still playing upwards!). I've hit a wall though, at least temporarily. Setting FSB up to 430 (10x multi) should = 4300Mhz, but I'm getting 4295.5 max.

I rebooted, upping the FSB to 432 (only 2Mhz extra but should have pushed the corespeed over 4.3 proper)... still exactly 4295.5Mhz corespeed. I upped the voltage to what Intel say is the max for these chips (1.36) and again booted to Windows - the corespeed hasn't increased, and it's failing Orthos after less than a minute

Maybe time to add a tiny notch onto the northbridge? Is this what I've read being described as "FSB wall"?? Helpz