E8500 overclock result..so far !!

[hansje]

Thanks

Hans

[hansje]

Back again

Thinking of buying a 4870, maybe crossfire later...not sure, but it isn't possible with my 680i SLI board

Which board will be the best choice with my E8500..at 4.0 stable now !!
Will a better mainboard give an extra higher overclock-result too ??

What to do ??

Hans

[hansje]

My memory is set at 889 with 4-4-4-12-1T...before with my E6600 I could do 1167 in my bios, I think it was 2:3 sometihng like that.

Is it possible to get my memory at 1167 again with 5-5-5-15-2T...and this E8500, looks like there is something left in my meory, running 1:1 now.

HELP please

Hans

[hansje]

My memory is set at 889 with 4-4-4-12-1T...before with my E6600 I could do 1167 in my bios, I think it was 2:3 sometihng like that.

Is it possible to get my memory at 1167 again with 5-5-5-15-2T...and this E8500, looks like there is something left in my memory, running 1:1 now.

HELP please

Hans

[ctal]

Hi Hansje -- I wouldn't bother trying. First, I think you are doing well to have DDR2-667 running at 889. Second, you would be slowing down your latencies by 25% (5 instead of 4), while increasing clock speed by only a little more than that (1167/889 = an increase of 31%). As indicated in various threads in the last few months, including one named "memory doesn't matter," there will be negligible difference in system performance, for recent generation memories, between faster clock speed/longer latency and slower clock speed/shorter latency.

That said, if you really want to try it, raise the memory voltage to near the max it is rated for, and change the memory:fsb ratio in the bios to increase the memory speed. Keep in mind that that may make your system unstartable, and you might have to clear cmos to get the system functional again.

Regards,
-- Al

[hansje]

Hi

Nice info my friend...will keep it the way it is.

Just bought my new 260GTX TOP from Asus.

3Dmark06 >> 17601 ...so I can't complain

Hans

[hansje]

I know we talked about it before, but forgot, sorry

BIOS gives CPU 53C IDLE
Core Temp gives 56C
Speedfan 46C
Real temp 46C

Which is right

Hans

[Sandog]

I would trust Speedfan and Realtemp. The CPU runs very hot in BIOS, this is normal.

[ctal]

Originally Posted by Sandog

I would trust Speedfan and Realtemp. The CPU runs very hot in BIOS, this is normal.

Yes, I agree. For an even more accurate reading, perform the calibration procedure recommended by Realtemp:

Real Temp - Documentation

Regards,
-- Al

[hansje]

Thank you guys what will I do without you

Hans

[hansje]

Is enabling SpeedStep and C1E in the bios enough ??

I don't like to lower everything again Vcore and so on ??

Forgot, ORTHOS LOAD...RealTemp 67C after 10 min. normally not to hot ??
Room is 27C

Hans

[ctal]

As I recall, after a great many experiments you had to raise Vcore significantly to get to 4.1 gHz or whatever it was, which is the main reason you are seeing the 67 deg reading. I wouldn't touch anything, assuming the system has been working reliably, or you'll wind up investing a lot more time with nothing to be gained. Chances are the 67 deg would read lower if you were to perform the calibration procedure recommended by Realtemp, which is pretty simple to do. That would be my one recommendation.

Regards,
-- Al

[hansje]

Thanks for helping me out...again

On another forum somewhere a guy could do 4.0 with a E8400 and only 1.28 VCore, can this be true ??
He said, that 1.41 was way to much for a WOLFSDALE, I don't think so

What about a reseat of the hsf with that MX-2 I bought ?? using ARCTIC SILVER 5 now.
Just to be sure ??

Hans

[ctal]

Originally Posted by hansje

Thanks for helping me out...again

On another forum somewhere a guy could do 4.0 with a E8400 and only 1.28 VCore, can this be true ??
He said, that 1.41 was way to much for a WOLFSDALE, I don't think so

What about a reseat of the hsf with that MX-2 I bought ?? using ARCTIC SILVER 5 now.
Just to be sure ??

Hans

The Intel datasheet for the E8xxx's lists an absolute maximum rating for Vcore (beyond which permanent damage or severe degradation of reliability can occur) of 1.45V. There is also a spec for the maximum voltage at which functionality is guaranteed, and beyond which there might be some impact on long-term reliability. That spec is defined in a very complex manner, and can vary from chip to chip, and can range (for the entire E8xxx family) from 0.85 to 1.3625V. I would expect E8500's to be at or near the top of that range.

Keep in mind also that engineers will allow some margin in defining specification limits.

So I would say that 1.41V is a bit aggressive (I run my E8400 at 1.35V, and it has been rock solid at 3.8gHz), but it is not unreasonable. Also, keep in mind that results will vary from chip to chip, and mobo to mobo, even when the model numbers are the same. So just because someone gets 4gHz at 1.28V with an E8400 doesn't mean that the next person will. Also, people will often make claims like this without specifying how thoroughly they have tested for stability, which makes the claims meaningless.

It's your call as to whether or not you want to take the time to re-do the cpu seating and thermal paste. Just make sure you follow the guidelines for applying it, for a dual core cpu, that are provided at the Arctic Silver site:

http://www.arcticsilver.com/pdf/appi..._dual_wcap.pdf

Regards,
-- Al

[hansje]

Thanks !!

Have some new results...running Vcore 1.32 now !!
3Damrk 05 and 06 and my games are running stable so far

I have done a reseat with MX-2.

IDLE 44C and LOAD with 3Dmark 56C..not bad.

I disabled >> Execute Disabled BIT in my BIOS...it was always ENABLED, something to do with the lower Vcore ??

Hans