Week/Batch numbers and highest stable overclocks for Core 2 Duo

[Kantana]

Pls submit your highest stable overclocks with the model processors, the batch numbers, Vcore used, motherboard used, cooling used and highest temps. Trying to consolidate into a table so that we can see if particular batches/weeks do particularly well. Will help those trying to buy a good overclocker.
Let me start with 2 of mine.

Intel E6400
Batch L624A768
Asus P5W DH
3.5G dual prime stable (438FSB)
1.5 Vcore
Corsair Cool water cooling
57 degrees C

Intel E6600
Batch L623A550
Asus P5W DH
3.6G dual prime stable (400FSB)
1.5375 Vcore
Corsair Cool water cooling
61 degrees C

Just changed the Corsair Cool water as the reservoir developed a leak. Didn't have any high powered heatsinks around so went down to the local computer center and picked up a Tuniq Tower 120. Spent 2 hours removing the water cooler and fitting the Tuniq, took a long time!

Anyway started up the system and got some new figures, used the E6600 and left all previous settings alone and noticed that the temps had dropped! Using the huge Tuniq Tower 120 dropped the temps compared to a water cooler, couldn't believe it. The system is still dual priming right now after 90 mins, readings are:

Intel E6600
Batch L623A550
Asus P5W DH
3.6G dual prime stable (400FSB)
1.5375 Vcore
Tuniq Tower 120 on high
55 degrees C

Gives me hope that I can reach for higher overclocks with lower temps and if I reach a thermal wall again I'll just switch to a faster fan in the Tuniq Tower.
Anyone know any better heat sinks than the Tuniq Tower? Thermalright Ultra 120?

[lawrywild]

INtel E6600
Batch L624A600
Asus P5W DH
3.4G dual prime stable
1.575v
Aircooled XP-90C
Around 60C with Asus probe/around 73 with CoreTemp

[Kantana]

Thanks! I see you get the same discrepency in the temps reported in the cores and that of the motherboard. Anyone else encounter the same thing? The difference is ~15 degrees C here on the P5W DH so hitting max temp of 85 Degrees C on the core would be ~ 70 degrees C on the motherboard. I assume this is when the thermal throttleing will start?

[Plaicd]

Are all conroe/allendale from the same plant or some from taiwan,costa rica ect.?

So far all I have looked at in pics here are marked Malay.

Is the plant incoded in the batch numbers?

[lawrywild]

MALAY(SIA) here

[Kantana]

Only seen Core 2 Duos from Malaysia here. Guessing the batch numbers are assigned to 1 plant only.

[Shrekogre]

I believe the chip will be made is USA, only the packaging done in Malaysia

[dissident]

Intel E6400
Batch L624A849
Gigabyte DS3
3.3G prime stable at 1.5 volts (don't want to take it higher)
~50 degrees load on water (lapped IHS)

[theteamaqua]

L625 week 25 is better from what i have seen .. 3 months later ill prolly get anoither one if i can burn my E6400 , i meant literally burn it, so i have a reason to get another one ...

E6400 week 25 can do 3.5 , 3.6GHz ... mine only go as high as 3.4GHz with as much vcore as i can give ...

[davefr]

Are all conroe/allendale from the same plant or some from taiwan,costa rica ect.?

So far all I have looked at in pics here are marked Malay.

Is the plant incoded in the batch numbers?

The assembly plant is meaningless. What's important is the specific Fab plant, Fab lot #, wafer # and die location and where that wafer Fab was running in terms of normal process variance at the time the wafer was produced.

[Plaicd]

The assembly plant is meaningless..

I know with other intel cpu's some plants were considered to put out better overclockers than others.

Maybe cause some fab plants tend to feed certain end assembly plants?

[Eastcoasthandle]

The assembly plant is meaningless. What's important is the specific Fab plant, Fab lot #, wafer # and die location and where that wafer Fab was running in terms of normal process variance at the time the wafer was produced.

And where do you get this info? Is it on the CPU?

[davefr]

I know with other intel cpu's some plants were considered to put out better overclockers than others.

Maybe cause some fab plants tend to feed certain end assembly plants?

Yes and actually some Fab plants feed a batch of sorted material to certain assembly plants at approx. the same time and thus the reason to correlate OC success to an assembly date code. (it's really the Fab material that matters)

The die's specific location on the wafer is even a factor. (ex: I believe the edge die tend to run a tad faster due to slightly lower channel length).

However the biggest factor is downbinning. If a CPU manufacturer is yielding fast material but most of the demand is for medium to slow then die that have been tested to yield to fast speeds can be downbinned to fill that market.

However I have a feeling that Intel isn't doing any Conroe downbinning at this point!! However they probably will after the product matures.

[fhpchris]

Yes and actually some Fab plants feed a batch of sorted material to certain assembly plants at approx. the same time and thus the reason to correlate OC success to an assembly date code. (it's really the Fab material that matters)

The die's specific location on the wafer is even a factor. (ex: I believe the edge die tend to run a tad faster due to slightly lower channel length).

However the biggest factor is downbinning. If a CPU manufacturer is yielding fast material but most of the demand is for medium to slow then die that have been tested to yield to fast speeds can be downbinned to fill that market.

However I have a feeling that Intel isn't doing any Conroe downbinning at this point!! However they probably will after the product matures.

I think they do have alot of problems getting good 4mb cache yields, and that is why 6600/6700 stuff is nowhere, or really slow, where alot of good 6400 stuff seems to exist.

IMHO even if you can buy a 6600 now, it will probably be crappy...

[theteamaqua]

from what i have seen and many PMs... in many forums .... week 25 which is batch #L625XXXXX is much better OCer compare to week 24

[davefr]

I think they do have alot of problems getting good 4mb cache yields, and that is why 6600/6700 stuff is nowhere, or really slow, where alot of good 6400 stuff seems to exist.

IMHO even if you can buy a 6600 now, it will probably be crappy...

It's much more likely that Michael, Steve, and Mark have first dibs on the E6600 and E6700's!!!

[irev210]

I think they do have alot of problems getting good 4mb cache yields, and that is why 6600/6700 stuff is nowhere, or really slow, where alot of good 6400 stuff seems to exist.

IMHO even if you can buy a 6600 now, it will probably be crappy...

i think that intel is more interested in trying to get the product into the retail market... while woodcrest is busy sucking all the 4mb cache cores.


Your best bet is to judge major OEM's like DELL and HP. When they have mass shippable quantities of core 2 duo's, so will we.


Right now it is looking like 3 more weeks before much better avalibility.



I really wish I knew the yields at intel, but they have been on 65nm in full production for over 8 months now. they have been making fully functional ES YONAH's back in may/june of 05.


I think they have more of a supply chain/capacity problem. You cant flip the switch and say bye to all those other processors that they are stuck supporting, making, and supplying.


Wish they could though

[gundersausage]

Anyone know what week 26 is like?

[Evilsizer]

E6400
7/23/2006
L624A869

no oc yet still working on installing windows.

[theteamaqua]

E6400
7/23/2006
L624A869

no oc yet still working on installing windows.

try go for 3.6GHz ... i bet it will stop after 425fsb ...

[nicepun]

BATCH:



Well my friends, here what I have so far. It is true, the ES chips kick butty because this bastard likes voltage as you can in AiBooster. I'm not sure if this is stable so I'll be doing some more testing to find my sweet spot. I don't see 4Ghz happening, sorry.



I'll do some more testing tomorrow, so far I think 3.56Ghz @ 1.6vcore in bios is my highest stable OC. I was aiming for 3.5/3.6 so and/or 14secs 1m SuperPI and so far 3.56 does 14.6 and I still have stuff running in the background. For 24/7 use I'll leave it at 3.4Ghz@1.5vcore in BIOS.

[ceevee]

X6800 "Xtreme" Edition

3.6Ghz rock solid dual-prime stable stock cooler

1.50V BIOS

55C full load air cooling

32C resting temp

right now 277 FSB, x13 CPU multiplier, but this is subject to change

I still have a lot of tweaking to do, and will add water cooling. I am positive this thing will get to 3.8Ghz

3.8Ghz will be great!!

BATCH # L623A465






--edit--

and post #100 !!! It took me six months! At this rate I will have as many posts as Charlie in about 48 years!!

[ArcTan]

E6400 L624A774
Gigabyte DS3
2 x 512MB 5400UL

so far I've done 400x8 dual 32M fine but prime will fail immediately
aircooled with XP90C and 18db 92mm fan

edit: also got E6300 Step 5 QPHC
480x7 small fft's overnight stable stock vcore though limited by board

[Pete]

E6700 Es Stepping 4 B0 QLLO

4.2ghz phase!
4.6Ghz LN2
3.7ghz air stock cooler

[nicepun]

Unless you're on Phase or LN2, I believe the average OC for these E6600 Retail (OEM also) will be 3.4Ghz~3.6Ghz!!