***The Official Wolfdale 8200/8400/8500 Overclocking Thread ***

[Duke]

Yeah it wasn't stable @ 1.45 (bios) & +0.2v on MCH/Chipset, so I upped it to 1.4675 (or something like that ) and +0.3v on chipset etc and it seems perfect now.

I will drop the vcore back down and try again, as it may have just been the FSB that needed more in the end.

 

[tomanders91]

How much cooler do these run?

If i bought a 8400, along with my gigabyte ds3p p35 board, and 4gb of geil 6400 ultra stuff, what sort of clock should i be aiming at? (with good temps, so not 4ghz lol)

 

[clone]

I had my cpu @ 4.5ghz 1.48v and core temp didnt move from the stock setting at all so they need to do a update pronto.

 

[Demon[KcsF]]

Tried 4Ghz at 1.375v and mine didnt want to know, stepping the volts back up now to find a stable point.

Bry.

 

[Vertigo]

Had this hypothetical conversation with Tom earlier but now it's actually come to pass...

Orthos failed after 8 hours and 22 minutes! Question is, am I stable?

On the one hand, you could argue that, since Orthos failed at all, I'm not stable. On the other hand, 8 hours is commonly accepted as the mark of stability, and indeed is the threshold set for entry into the database. If, therefore, I'd stopped Orthos at 8 hours and 5 minutes, my system would have been "officially" stable and I'd never have known that it was going to fail 17 minutes later I could run Orthos again and, if it makes it to 8 hours, stop it immediately and claim stability, even though it failed before with identical settings.

Frankly, I don't care. I have 4Ghz at very low volts compared to most others here and the system is utterly solid running every app and game that I use, and has managed to Orthos for over 8 hours so, as far as I'm concerned, it's stable enough for me

 

[Tom|Nbk]

Had this hypothetical conversation with Tom earlier but now it's actually come to pass...

Orthos failed after 8 hours and 22 minutes! Question is, am I stable?

On the one hand, you could argue that, since Orthos failed at all, I'm not stable. On the other hand, 8 hours is commonly accepted as the mark of stability, and indeed is the threshold set for entry into the database. If, therefore, I'd stopped Orthos at 8 hours and 5 minutes, my system would have been "officially" stable and I'd never have known that it was going to fail 17 minutes later I could run Orthos again and, if it makes it to 8 hours, stop it immediately and claim stability, even though it failed before with identical settings.

Frankly, I don't care. I have 4Ghz at very low volts compared to most others here and the system is utterly solid running every app and game that I use, and has managed to Orthos for over 8 hours so, as far as I'm concerned, it's stable enough for me

8 hours is widely accepted as a stable PC. By running 8 hours you pretty much guarantee you should never face any stability issues. 12 hours if you really want to be anal about it though

 

[Demon[KcsF]]

Very annoying here that I cant run on a 9x multi

I am pretty sure its a BIOS issue and until ASUS release a new one I am resigned to using the 8x multi.

Anyone disagree with that or know something I dont?

Bry.

 

[siege2]

 

[Demon[KcsF]]

Get it pushed siege

Bry.

 

[siege2]

nice low temps st 3.8ghz

very nice in TA spring "no lag" also COD4

 

[Tom|Nbk]

nice low temps st 3.8ghz

very nice in TA spring "no lag" also COD4

Thats just the socket temp.

Add around 10c on for the actual core temp

 

[siege2]

what do you think off this, Iam worried about the FSb that seams to be red....

 

[Tom|Nbk]

I've just fired up Supreme Commander (CPU Limited game) after someone mentioned it in this thread, bearing in mind I haven't played it since my Athlon 64 days with an X1900XT I was rather surprised with how well it now runs . I did a benchmark for anyone who finds it useful.

This was done with spec in sig, all details in game maxed out, 1680x1050, 16xQ AA

Stats Log Report

Logged frames : 17978
Timestamp : Sun Jan 27 22:20:36 2008

SupComMark (sim) : 9301
SupComMark (render) : 8197
SupComMark (composite) : 17498
(Note: SupComMark scores represent overall system performance. Higher is better.)
FPS ......................................: calls[ 17978] min[ 15.50] max[ 96.63] avg[ 49.436]

 

[w3bbo]

Had this hypothetical conversation with Tom earlier but now it's actually come to pass...

Orthos failed after 8 hours and 22 minutes! Question is, am I stable?

I personally fun my cpu's for 12hrs as a final test of stability. 12 hrs because thats roughly how long it takes to run a full round (take a look at orthos where it says rounds). If it passes a round then I'm happy.

Ask yourself this, if you encounter a BSOD now would you feel confident it isn't due to the cpu? I know it would be playing in the back of my mind even if was most likely windows playing up or a driver issue.

 

[Vertigo]

Ok, as promised, Supreme Commander benchies.

Can't remember the exact graphical settings but, as we're primarily comparing CPU speeds here and the graphics settings remained constant, that shouldn't be too much of an issue.

First up is the E8400 at stock 3Ghz (333x9):

1280x1024:
SupComMark (sim) : 9043
SupComMark (render) : 7528
SupComMark (composite) : 16571
FPS: calls[ 12966] min[ 10.02] max[ 65.31] avg[ 35.289]

1680x1050:
SupComMark (sim) : 9049
SupComMark (render) : 7136
SupComMark (composite) : 16185
FPS: calls[ 10645] min[ 12.76] max[ 53.25] avg[ 29.174]

1920x1200:
SupComMark (sim) : 9019
SupComMark (render) : 6811
SupComMark (composite) : 15830
FPS: calls[ 9494] min[ 12.38] max[ 40.55] avg[ 25.204]

Now the same E8400 at 4Ghz (445x9):

1280x1024:
SupComMark (sim) : 9252
SupComMark (render) : 7944
SupComMark (composite) : 17196
FPS: calls[ 15525] min[ 16.43] max[ 70.65] avg[ 41.118]

1680x1050:
SupComMark (sim) : 9258
SupComMark (render) : 7412
SupComMark (composite) : 16670
FPS: calls[ 11742] min[ 16.34] max[ 55.30] avg[ 32.281]

1920x1200:
SupComMark (sim) : 9227
SupComMark (render) : 7089
SupComMark (composite) : 16316
FPS: calls[ 10395] min[ 14.94] max[ 52.40] avg[ 27.847]

Despite the composite score only increasing around 3-4%, the average FPS increases between 10-16% which isn't bad when you consider the graphics system is identical. The biggest gains are in the minimum frame rates, which demonstrates how CPU bound the game gets at the extreme when there's a lot happening on-screen.

Hope this is of use to someone

 

[Tooks]

what do you think off this, Iam worried about the FSb that seams to be red....

The FSB being read is fine. Intel chips use a quad pumped FS bus, so what you're seeing there is 4 x 399.98 = 1599.93.

Clock it some more!

 

[Topitoff]

what do you think off this, Iam worried about the FSb that seams to be red....

Don't worry about the FSB; your board will do 500 or more.

Try coretemp, it might work and it shows the temps of the cores not the cpu socket.

Try Orthos instead of SP2004 as it does both cores in one window.

Don't like your core voltage, I would have thought that it would do 400 FSB with stock volts ~1.15v

 

[Vertigo]

I personally fun my cpu's for 12hrs as a final test of stability. 12 hrs because thats roughly how long it takes to run a full round (take a look at orthos where it says rounds). If it passes a round then I'm happy.

Ask yourself this, if you encounter a BSOD now would you feel confident it isn't due to the cpu? I know it would be playing in the back of my mind even if was most likely windows playing up or a driver issue.

I hear what you're saying but I won't get that BSOD and you know it

Besides, whilst I appreciate your comments, I believe the rules for entry into the database as "proven stable" stipulate 8 hours, which mine has done and therefore qualifies.

It doesn't matter how long you leave Orthos running for, at some point you have to stop it and you'll never know whether it would have eventually failed. You can run for 12 hours and then stop but it may well have failed 5 minutes later. There's no such thing as 100% stability, only degress thereof and I'm fine with what I've got tbh. If any when I get problems I'll start doing more tests but, until then, I have better things to do that run endless stability tests.

 

[Tom|Nbk]

Vertigo mine and your results vary greatly, mine seems to run a lot better, what were the game settings and what gfx hardware are you on?

 

[Kirth Gersen]

Vertigo/Tom - thanks for the benchies. TBH a 10-16% performance increase against a 33% processor speed increase, is probably as much as one could realistically hope for. I hope these gains translate as well to FSX!