**Official e8400/e8500 Retail OC Thread

[techguy2k5]

That's strange about the memory not liking 445 (890mhz). I have really cheap sticks, OCZ vista upgrade 4gb set that I picked up for $75 and it did fine up to 920 or so with no errors. Can't go much beyond that though even with loosening up the timings considerably.

Have you tried one stick at a time to see if you can pass memtest with one stick or the other?

Both sticks pass at default settings, so I assume it just doesn't like the raised clock or perhaps one of my voltages or timings. Still trying to figure things out though. I've got a help thread here if anyone could share some input.

[unclewebb]

As you can see Coretemp is really messed up with this new one. It's stuck on 62C.

It's not CoreTemp that's messed up. It's only reading the data that your CPU is putting out. If it shows 62C at load and at idle then that means your on chip digital thermal sensors are stuck, like many other users have reported. Blame Intel, not the programmer of CoreTemp.

Send me two register dumps using CPU-Z, one at full load and one at idle and I will confirm this if you like.

[Retrolock]

E8400 Q746A446 is pretty bad, just like emoners already said. 1.3v bios just to be stable for 3.6ghz, 1.296v idle, 1.28v load . @4500mhz I had to pump 1.53v for me to be able do some superpi 1m runs.

I'm still baffled I can't make it stable at 4ghz. I dont wanna go over 1.4v but I guess this batch really blows that much lol.

[PhilDoc]

The 9 second or sub 9 second SuperPI runs with these new processors at insane core voltages is very impressive. I'm more interested in what they can do at a safe 1.40 volts. I realize my abused E8400 is far from golden. That's why I wanted to compare it to other users. With an extra 100 MHz you beat my time by less than 0.050 seconds. Good work.

Sorry, wasn't try to get a great pi time. It was run with stock vista and the memory timings left at auto. Here's one on XP with a little better timings. I'm sure someone who knows all the tricks can shave another couple of tenths off of it. Strange I could run at a little higher fsb in XP. Not to worry, I don't have the golden one either.

[laragirl83]

Just got my new E8400 (Q746A501)
Running prime stable at 4GHz (8*500MHz) @ 1.320V.

[Hughyhunter]

Alright... I couldnt even boot at 4500 in the 1.4-1.465 range. So I decided just to give this a shot... think this will be as high as I can go for awhile until I have more time and money to get some aftermarket coolers for gpu's.

[flesheatinvirus]

Here is my 1meg&32meg Pi at 4500.

[unclewebb]

Thanks PhilDoc for giving me a number to shoot for. 4446 MHz @ 1.36 volts is an excellent chip.
My E8400 needs too much voltage to run at that speed. It's all tapped out here.





flesheatinvirus: How long has your Ballistix been surviving at 2.40 volts? 601 MHz with CL4 timings equals some incredible memory bandwidth.

[ixtapalapaquetl]

Been screwing around a lot today. Finally got my water running, and within 5 minutes of finishing up UPS came with my new memory (SuperTalent 2x2GB Hynix chips for $82 - what a country!). Here you can see a little of each.

[falqon]

Nothing special, Q748A seems to get a bad rap and I don't think it's a terrible chip. It's my only one so I'm not upping the voltage anymore and for whatever reason my Ballistix don't really clock well on this p35-e so I didn't get tight timings or high speed, combine that with Vista and the pi run is so so.

Turned it down to 3.3ghz for daily use, I think it's at like 2.1v, I'm not complaining.

[hersounds]

Q746A500

** Generic rams** ( at the moment )

[Hughyhunter]

Just a few things:

When I set the FSB to 2000 I can get into windows but it BSOD's in the first minute of loading the OS. I am running 64bit Vista right now. I've tried every combination of settings and volts. Anybody having this problem?

I've notice just about everyone here posting CPUZ sets there command rate at 2T... why? What is the advantage of this over 1T? I wonder if it's just that people cant change that in their bios.

[laragirl83]

Aren't u guys worried about running these chips over 1.40V...?

[DStealth]

Nope
1.56-1.58 V here on water over a month

[DStealth]

Here is my 1meg&32meg Pi at 4500.

Very nice memory m8. Optimize
Just to give you an idea how the numbers should look @ those speeds , but with crappier memory Promos 5-5-4-10
1M

32M


keep pushing

Edit:Sorry for double posting ...

[flesheatinvirus]

Thanks PhilDoc for giving me a number to shoot for. 4446 MHz @ 1.36 volts is an excellent chip.
My E8400 needs too much voltage to run at that speed. It's all tapped out here.





flesheatinvirus: How long has your Ballistix been surviving at 2.40 volts? 601 MHz with CL4 timings equals some incredible memory bandwidth.

I have been running them for months like that, so far I have been lucky. Sandra Mem gives me over 11,000.

[flesheatinvirus]

Dstealth very nice numbers for the same speed, Promos looks pretty good. Looks like I got a lot more work to do.

[flesheatinvirus]

Just a few things:

When I set the FSB to 2000 I can get into windows but it BSOD's in the first minute of loading the OS. I am running 64bit Vista right now. I've tried every combination of settings and volts. Anybody having this problem?

I've notice just about everyone here posting CPUZ sets there command rate at 2T... why? What is the advantage of this over 1T? I wonder if it's just that people cant change that in their bios.

Its because most of us are running a Intel P35 or similar chipset which runs better at 2T, 1T can run with Nforce based chipsets.

[GFDuke]

Honestly i dont think we can trust the Voltage/Temp readings on the E8400. There are way to many discrepencies thoughout the forums on many sites. But basically who cares if i gets fried? The chip costs less than 250 bucks. Just get another one. Not like the AMD FX60 i got back in 2006 that cost over a grand. POS!! Compared to this. So im just gonna crank the hell out of it.

[sofarfrome]

Honestly i dont think we can trust the Voltage/Temp readings on the E8400. There are way to many discrepencies thoughout the forums on many sites. But basically who cares if i gets fried? The chip costs less than 250 bucks. Just get another one. Not like the AMD FX60 i got back in 2006 that cost over a grand. POS!! Compared to this. So im just gonna crank the hell out of it.

I am with ya dude...I sold my E8400 Q743A that would run 24/7 P95 stable at 4.3GHz with 1.344vcore just so I caould get the Xeon version. That will be here tomorrow and I am going to pour the vcore to it just to see how far it will go. $200 and so what if it fries

[Vinas]

Just got done helping a friend with his e8400. Here is the comparison between his chip and mine. What interests me is that his CPU has a higher "stock vcore" yet uses less vcore to stay stable at 4GHz. Temps are about the same using ultra 120x. I would think it should be the other way around... Maybe the higher FSB needs more vcore for the CPU to be stable? Any thoughts? See below.

My CPU
Week: Q748
Batch: A219
Spec: SLAPL "C0"
Packaged: 1/29/08
CPU = 8x500 (4GHz)
vcore bios: 1.42
vcore CPUz: 1.392
vcore orthos: 1.38
default vcore: 1.13

Friend's CPU
Week: Q745
Batch: A553
Spec: SLAPL "C0"
Packaged: ?
CPU = 9x444 (4GHz)
vcore bios: 1.35
vcore CPUz: 1.33
vcore orthos: 1.32
default vcore: 1.18

[Mickey Padge]

Aren't u guys worried about running these chips over 1.40V...?

Is there a reason to be worried, I mean where are the reported CPU deaths and general OC degradation?

I have not heard a peep about problems with the new 45nm CPU's, aside from the early isolated cases, all from a very, very small section of the OC community.

Anything that has been reported could be chalked up as a handful of defective CPU's, that may have experienced the same problems, even if they were used at stock speeds. OC degradation (if there is any more with 45nm, than any other CPU) could be placed at the door of the fickle nature of OC'ing, and that the settings used, were simply not stable in the first place.....

I have had an older AMD rig go for 6 hours with prime one day, and then not half an hour the next.....

Given that, I would not want to put over 1.45v through these CPU's, unless on high end water and more with phase.

Cheers!

[babalouj]

Aren't u guys worried about running these chips over 1.40V...?

I have been running my Q746 E8400 at 1.53V 4.4ghz since the first day Microcenter got them. No degredation here.

[ante_ante]

Anyone have seen results on a E8500 with Q744A836 stepping ??

Will get one tomorrow ^^

[PhilDoc]

Thanks PhilDoc for giving me a number to shoot for. 4446 MHz @ 1.36 volts is an excellent chip.
My E8400 needs too much voltage to run at that speed. It's all tapped out here.

.

Never really tried to get much out of pi, my thing when I had the benchmark bug was 2001se and that was 4 years ago, lol. Still fun though. I don't think you've tapped her out too much.