Printable View
What is your best rock solid stable overclock (8 hours prime/orthos/etc., give or take) on a RETAIL 45 nm CPU. It is looking like people get crazy high OCs but stable levels are far below. Indicate your CPU, stability test, cooling, vcore, FSB, and mobo.
Especially curious about the new dual cores.
Edit: If you DO post an ES chip anyway, please indicate that fact.
Screenshots always appreciated.
Edit: Don't claim 4.6+GHz stable w/o a screenshot!
Prime 12 hours
QX 9650 @ 4.1GHZ
Water
vcore @ 1.42v
FSB @ 430
Maximus Formula
I guess you probably have to specify which 45nm we're talking about as all of them has different clock speed?
I left out a critical comma. ;)Quote:
Originally Posted by aoch88
List which CPU you are talking about. More interested in the written answers than the poll. The poll is just for a general feel over all types and core counts. Nothing scientific.
From what I see those E8400 are very much a mixed bag. Some does good oc at low voltages and had it stable. For mine, I can boot 4.4Ghz+ easily and do Super PI but it isn't stable. Had a hard time getting it stable even @ 4.2Ghz.
It's all about luck sometimes. I had the same batch number as someone that does high o/c with low volts but mine can't.
e8400 Q748A115
4.0ghz - 8 x 500
1.36vcore
8 hours prime, 30 min occt
Max OC - 4.5ghz with 1.43vcore
I think I could get stable a little higher but don't want to try it for a while until more data comes out on how high voltage effects these chips. For exapmple occt stable at 4.2ghz 1.39vcore but prime failed.
CPU: QX9650 - L740A931T (I still don't know what the "T" means)
Stability: Over 18 hours of folding@home SMP.
Cooling: Air - Noctua NH-U12F + low speed 120mm
vCore: 1.248V
FSB: 400MHz (1600MHz quadpumped)
Mobo: Asus P5K-E
Overclock: 4.0GHz
Remarks: Limited by cooling here.
On my Stepping 4 (B1) e8400es I only hit 3.8-3.9ghz stable for orthos blend. 4.0ghz won't pass orthos no matter what increase in voltage. However, doing 3.8ghz at 1.3vcore.
My retail e8400 arrives next week. Hopefully stable in the low 4's.
Stability: 4.1Ghz
CPU: E8400
Stepping: CO
Week: Q747A374
Code: SLAPL
Vcore: 1.4375v (1.40v load)
Cooling: Air - Thermalright Ultima-90 ES
Motherboard: P35 Asus P5K-D
Bios Version: 603 (705 was aweful)
Operating System: XP
RAM brand: GSkill
RAM rated speed: DDR2 1000
RAM actual speed: 953Mhz
Latency: 5-5-5-15-Auto
anything above 1.4625 and or 4.3Ghz and its not prime, occt, nor orthos stable
Would the people who claim stable at 4.6 GHz and up please provide details? I assume that is one of the new wolfdale cores on phase but confirmation would be nice.
Thanks.
E8400.. Q746A477..
Abit IP-35 Pro BIOS 16.03 Beta
Cooling in Sig..
4.2 Ghz @ 1.34vCore
4.25 Ghz @ 1.36vCore
4.3Ghz @ 1.39vCore
Havent tried for more 24/7 clocks yes, My temp's are VERY high, for Watercooling. Almost 61C Prime Loaded @ 4.3Ghz @ 1.39vCore.
I want to comment on this,Quote:
Originally Posted by Speederlander
Why do you think 8hours Prime is an indication of ROCK stability? There are application which loads Core2 much much higher than Prime, like Intel Linpack 64bit,
but even passing 100+ iterations in Linpack, which is really enormous task for Core2 especially on air, does not guarantee full stability in other application.
As a side note, then i see claims like 4+ Ghz with Q6600 or high 4.5GHz with newer 45nm CPUs, i know this is not really stable for sure.
One might think that as it is pretty much the standard in most credible forums. There are also forums where people consider Super Pi as good enough. I, personally, do not possess enough knowledge to state categorically that 8 hours of Prime proves stability. I do consider it to be an acceptable test as I have read so many posts by folks that have a great deal more knowledge than I that say so.Quote:
Originally Posted by Cronos
What are your credentials?
I do agree with this. I, personally have had a setup that was 8 hours Prime Stable (note I said Prime stable, which cannot be argued with) yet failed 3D MArk 06. Turned out my memory overclock was too high for 3D applications.Quote:
Originally Posted by Cronos
Wow, now this is an emphatic statement. Only people with iron clad reputations and credentials should make statements like this. Please, let us in on your background so that we may judge for ourselves if you are correct, or just have an overinflated opinion of your own self worth.Quote:
Originally Posted by Cronos
I've got Abit IP35 - tell me please what vcore did you set in BIOS.Quote:
Originally Posted by Ad1tya
batch/specs in sig, loadline enabled so bios and CPUZ volts close. Scaling at exactly .05v for every 100mhz over 4. 4.1 10hrs orthos pic in other e8400 thread.
4.0 orthos stable 1.2875 bios, 1.28 CPUZ
4.1 orthos stable 1.3375 bios, 1.33 CPUZ (turbo, memperf 7, ram 4,4,4,12)
4.2 orthos stable 1.3875 bios, 1.38 CPUZ (stand, memperf 8, ram 5,5,5,15)
4.3 orthos 1+hrs 1.4375 bios, 1.43 CPUZ temps too hot, over 70 load on water, so not doing full run.
My 24/7 will be 4.1 to 4.15. FSB above 465 for 4.2 and 4.3, I have to decrease mem performance to 8 from 7 and loosen mem timings to 5,5,5 from 4,4,4, resulting in 4.3 slower on everest benchmarks, and nearly same spi score as 4.15.
although I'm one of them that isn't impressed by the "suicide shots" of really high clocks that people post superpi shots of, usually prime95 is good enough for testing system stability for me, assuming I stress both the CPU and the RAM separately then both at the same time. I'm hoping to get 4.2 out of this chip.. if I can't, I'll probably just keep my Q6600 as it's a good batch with a low VID that can hit 3.6 stable on 1.35 volts (1.3 in windows, 1.26 with vdroop)
looks likely 4.0-4.1 mostly with a few lucky ones getting 4.2+ from the un-"scientific" poll
Quote:
Originally Posted by Capt Proton
I can't vouch for intel but i can tell you that with AMD 8 hours is not 24/7 stable............In fact when a cpu is close but not quite 100% stable it fails usually within the 10-12 hour mark.......
Take into consideration also that alot of people during the winter toss open a window.......so summer time would be the real deal on whats 24/7 stable.....ambient changes the game alot.
Also taking into consideration that some locks such as PCI rarely work......which can be noted due to your hard drives performance and noise, my old 36g raptor will tell me right away if PCI locks work......NIC's also are prone to be less stable............
100% stable is sort of a myth however most users here can perform most of they're needs wants at they're 24/7 stable settings without failure.......If in fact you placed the hardware at such settings in a mission critical situation such as a server with high traffic........Failure would be imminent..
Now since i made this claim here's my reputation/credentials.....I've been in the amd camp for quite some time since slot cpu's ;) And I've managed some preety insane stuff, I was waterchiiling back in the duron days.....last year I was beta testing reference mobo's for ati...........
All that said, this is XS and we push things to the limit, most of just need it stable to do what we want it to do, game, bench, crunch etc......and 90% are happy with that........
Intel is same, sometimes it craps out from 8 to 12 hours, and one notch hirer solves it. Case in point, just after posting my 10 hr 20 mins orthos stable in official E8xxx thread this am, I left it on expecting to come home for lunch with 14 hrs, and it stopped at 10hrs 52 mins. So now up to 1.34x from 1.3375. Probably wouldnt matter either way...but running again at 1.34.Quote:
Originally Posted by chew*
May i again stress the fact that Prime95 is not nearly as stressing to Core2 as Linpack64 ?
Sure, if you build a computer to run Linpack 24/7. :rolleyes:Quote:
Originally Posted by Cronos
I know I built my machine for folding and playing 3D games, and for that it runs perfectly fine. Prime95, let alone Linpack, is a far more strenuous workload than most normal uses, and as such Prime95 is a perfectly good stability test. I really don't know what you are trying to prove... this is XtremeSystems, not MissionCriticalStableSystems. If we wanted our computers absolutely guaranteed stable we would run them at the manufacturers nominal settings.
QX9650 4.0GHz @ 1.37 vcore
Not 8hrs,but atleast a screen shot.Took the last week off of overclocking this set up.Maybe try again this weekend.
Lets see some screen shots:up: .
http://www.xtremesystems.org/forums/...&postcount=425
I've tried 2 8400's, both are Q748A.
The first one (Q748A242) was very lousy. Couldn't be stable @ 4.0ghz regardless voltage. best stable was 3.8ghz with 1.37v under load.
The second (Q748A141) is a bit better. It's stable @ 4.0ghz but needs a rather high 1.39v under load. BTW i can boot even @ 4.5ghz (spi 1M @ 4.3ghz) but stability is out of the question. :p:
From what i've seen Q748A produces among the worst results.
All these are on air cooling.
E8400 Q740A555T* Note the "T"* So far, nobody has been able to find out what the deal is with the "T" All I can determine is that some XS members from Canada and Australia have this letter at the end of their FPO/Batch#. Any more info would be appreciated:p:
Pack Date: 12/20/07 VID:1.1125
Running 24/7@4005 445x9 Vcore:1.344v Gigabyte P35-DS3R, TRU-X
See sig. for more details.
Longest Orthos run was 8hrs 30 minutes, I then shut it off, did not fail test.
I have had it clocked higher but don't want to exceed 1.4v and am quite satisfied with a 4G stable oc and good temps- hits about 57C running Orthos.