• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

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

Room temps about 18C over at Temp gauge on far side of room away from Radiator where as the PC is only 6 feet away from the Radiator.

That was with a 120mm TT Silent Cat Fan fitted to a Noctua 120 Heatsink (not lapped) and new E8500 (not lapped), Case fans were also at 1/2 speed and that was idle, I did not get near 60C on load either, but ideally would run Fans all at 100%, just wanted to stress parts more with Fans slower.

I ran all my Orthos and Memtest and 3DMARK06 For Windows with the above Fan speeds inc GPU Fan at 60% (Default) instead of 100%.

The Mobo Temp is only 30C in Bios where as it reads 40C with these Apps in Windows, you need enter Bios to get real CPU /Mobo / NB /SB Temps, Asus Probe does not read the NB or SB but weirdly enough it can read the Optional Temps (they do not supply Sensors Cables with the Striker II ,but did with the Striker Extreme).
 
Last edited:
helmutcheese said:
A lot of the readings are wrong and high, far higher than the Bios.


no0d01rg9.jpg
Click to expand...

If that Vcore is right then thats a fine cpu you have there. 1.248v for 4 ghz is excellent. One of the guys on XS had an interesting theory about these cpu's. He reckoned his went to pot after putting 1.45v and over through it. After that it needed lots of voltage. Similar with mine. When i first booted up it went to 4Ghz on stock voltages. When i gave it 1.45v while going for 4.5Ghz it's been pants ever since.
 
No, CPU-Z Does not read my VCore correctly on my Striker Extreme or Striker II.

I failed Orthos at with Bios Volts set to 1.30 which was 1.29v in Bios and lower on load in Windows (Vdrop and Vdroop), I then set Load Dampener setting in the Bios on and raised Volts to 1.328v and its now 1.328v with No Vdrop or Vdroop in Bios or on load in Windows.

It nows passes Orthos but I may be able to lower it to somewhere between 1.29v and 1.328v or it may have been another Voltage setting Dropping and Drooping to far as I had forgot to turn on the Dampener. :)

I can pass 3.75GHZ (just the speeds I end up at with current FSB and 1/2 Multi's) with 1.2v and possibily a Tad more but not tried.
 
Last edited:
pastymuncher said:
One of the guys on XS had an interesting theory about these cpu's. He reckoned his went to pot after putting 1.45v and over through it. After that it needed lots of voltage. Similar with mine. When i first booted up it went to 4Ghz on stock voltages. When i gave it 1.45v while going for 4.5Ghz it's been pants ever since.
Click to expand...


Interesting, if that is true and becomes common knowledge, they will be hard to sell on as second hand.
 
Someone over there claimed a E8200 @ 4GHZ with 1.2v.

I found they need no voltage increase to get 1st step of the way to about 3.6GHZ-3.8GHZ MAX, then you need a Tad more for 4GHZ and if you want than 4GHZ+ you really need to start heading towards 1.45v+.

I know every CPU is different though and mileage varies.
 
Last edited:
Got mine at 8 x 450 atm @ 1.240v seems happy enough, i really have no idea what temperature its running at though, 3 different programs say 3 different sets of temperatures, especially everest which has 1 temp for the "cpu" and then 2 more higher temperatures for core 1 and 2 :s

There is some settings in the bios i have no idea what to do with as well ? i ve enabled transaction booster enabled but what is the relax level ??
 
Last edited:
Thought I would share a suspicion of mine that seems to be holding true. The first batch of Wolfdales (inc ES) don't like voltage of more than 1.6v for any period of time (no matter how good the cooling)...Don't go over this or suffer instability issues!
The latest batch seem to have a lower VID. Something to look out for in the future.
 
pastymuncher said:
One of the guys on XS had an interesting theory about these cpu's. He reckoned his went to pot after putting 1.45v and over through it. After that it needed lots of voltage. Similar with mine. When i first booted up it went to 4Ghz on stock voltages. When i gave it 1.45v while going for 4.5Ghz it's been pants ever since.
Click to expand...

This may not be far from the truth! It’s a known fact if you overvolt a cpu too much, you can actually physically destroy the silicon lattice of gates within a processor. These gates are the un-doped silicon that are there to hold the electrons from jumping to an adjacent transistor. With too much vcore and heat the electrons can actually tunnel their way through the un-doped silicon! This will causes irreparable damage to cpu, allowing electrons to easily flow across the junction, causing it to malfunction and resulting in error.

I’m thinking it’s a pretty safe assumption that the silicon thickness of these transistor gates pressent on these new 45nm chips are far thinner than the ones present in the larger 65nm cpus, how else will they be able to fit more transistors in a significantly smaller die size. So I would think the inbuilt resistance to electron quantum tunneling would be less than the larger manufacturing process used on older cpu’s.

So if these chips are possibly more heat and vcore sensitive why are we letting them run significantly higher temperatures and running vcore equal or even higher to that run on old core 2’s??

I wouldn’t be surprised if the silicon degradation, of these cpus usually measured in years with stock vcore and reasonable temps is significantly reduced to weeks or even instant damage due to excessive high temps and high vcore.

This is just me thinking out loud, food for thought maybe??
 
You are spot on Hesky82.

To add; I think with less gate leakage there is less of a requirement or tollerance for excessive voltage. I found with the Q9550 that 1.35v was enough for a good clock, to get a small amount more required a lot more voltage. The most the CPU would take was 1.78v. Any more and the CPU would take a fit.
 
I've now recieved my E8500, but before i go ripping the packaging off and installing it, Is there anything i should check on the packaging to ensure i've go one of the better ones. or is it just a case of waiting till its install. ie the VID, etc
 
Well i have taken off the infinity, and put the stock cooler on. Now my idle temp is 53c with a room temp of 18. load still gets in the 80s tho. and this is with no overclock or voltage

Obviously my infinty isnt doing its job properly or it wasnt seated right, even after 3/4 reseats

am i just extremely unlucky to get a hot chip? my e6600 was running at similar temps too
 
Hesky82 said:
If you feel speedfans results are more satisfactory stick with that, maybe coretemp is working on this issue.
Click to expand...
According to the author of Core Temp the problem lies with the CPU and not his s/w. It looks like some Wolfdale's have either badly calibrated diodes sensors, or they just plain don't work correctly.

As to using the core readings from SpeedFan, this seems to be calculating the core temperature based on a Tjmax of 100c, and not 105. So SpeedFan's core temperature readings are exactly 5c below what they should be.
 
You must log in or register to reply here.
Back
Top Bottom