Q9450 OC / Temps / Settings

[fordf250]

Raising the voltage, especially on the NB is not an option for me.

As you guys might now that on the Asus Striker II Formula, their NB chipsets run quite hot at default voltages.

Mine can range from anywhere from 60 to 70.

Vio|37

I have a 120mm fan blowing on the nb and run 50C at 1.8Vnb on the striker II formula at 475fsbx8 and a q6600, the nb does not get hotter with volts for me, it is at 50C for 1.5Vnb also.

http://good-times.webshots.com/photo...67863468edvpcv

You will also notice in the screen shot 1.8vnb is still not red and is only yellow warning in bios. It also needs +40mV nb gtl to run that speed.

[RunawayPrisoner]

I did it! 487MHz stable in OCCT!



1.344v!

[kickerman65]

Good job man lookin nice!

[RunawayPrisoner]

Thanks. I think this is it. Any higher than this and I guess my board is just going to explode. It was a mere 12MHz jump from 475, and I had to crank my NB volt up by .12v. Not too bad... since the NB chip doesn't get any hotter than 30C thanks to a 12mm fan blowing directly at the Extreme Spirit II cooler than I put in. Still... these results are mind-blowing! Now I'm confident that 4GHz can be achieved... but maybe with even higher NB volt. Probably... 1.80v or so.

[kickerman65]

30c!! thats Downright chilly

[RunawayPrisoner]

Yeah. Try Extreme Spirit II and some insane air blower like a 74CFM fan. Not to mention ambient temp is around 18-20C here... without an air conditioner. And it may get even colder in Winter!

[kickerman65]

Lucky you cause my NB temps hover at about 50c

[RunawayPrisoner]

A 120mm fan blowing directly at the thing. What do you expect? :p

[slim142]

I cant believe people is raising NB voltage up to 1.6 which is dangerous in air IMO.

Unless you have a case like the Armor with a 23cm fan directly on the NB (and thats not gonna help THAT much but anyways)

[RunawayPrisoner]

I have a 120mm on the side door, one 120mm blowing down from the CPU cooler, one 120mm in the back blowing in as well, and one 200mm on top sucking all of the hot air out. That's not to mention the NB chip also actively cools itself (Extreme Spirit II). Antec 900 go!

[kickerman65]

what is the stock NB voltage on a Evga 780i anyway? 1.5?

[kickerman65]

Or is it 1.2?

[RunawayPrisoner]

I have no idea? On my G35 board, it seems to be at 1.25v.

[kickerman65]

yea thats what it is. its like 1.26 i think

[RunawayPrisoner]

You just need to restart your computer and look in to find out.
And GEE! 488MHz... only 30 seconds left, and the motherboard decided to hang on me.

[kickerman65]

so what do you think man? Q9550 or E8500? what do you think will be faster?

[GAR]

i wish 500x8 was stable, im getting my chip tommorow

[RunawayPrisoner]

so what do you think man? Q9550 or E8500? what do you think will be faster?

Either Q9450 or E8400. I have an E8500... and I must say, I don't believe the .5 multiplier helps for anything at all. If you're doing casual 24/7 overclocks, those chips are not going to help much. :P

And 500 x 8 is reachable... just not stable 24/7. You can still do 3.90GHz (or 3.88GHz) with reasonable voltages, though... as tested above.

[kickerman65]

Dang! choices choices

[grimpy]

Got mine this afternoon and put it straight in at 3.4GHz / 1.22v bios - 1.2 cpuz.

Not happy with my idle temps may have to reseat my heatsink,under load it hits 75c:

[RunawayPrisoner]

Try Real Temp instead of Core Temp? I bet your real idle temperatures would be 35/35/37/37. If you are not happy, try another cooler or something.

Mmm... and currently running OCCT at 490MHz. I tried 495MHz previously, but it seems apparent... that I need more vNB. Probably around 1.74v to stabilize 495MHz... and 500MHz would require 1.77v, I guess. 487MHz was barely passing it with 1.67v vNB, so I don't think 488MHz to 493MHz would require anything less than 1.69-171v vNB. Guess this is an indicator... for even MORE fans!

Edit: Couldn't withstand OCCT for 15 minutes with 1.360v vCore... and I REFUSE to actually put in more.

[grimpy]

Didnt think real temp worked in vista64 but just checked and it now does,my idle temps look much better:

[RunawayPrisoner]

Tick the "--" option in the "Idle Temperature Calibration" options. That is intended to be for correct 45nm idle temperatures. It doesn't mess with your load temperatures in any way, anyways.

[Brahmzy]

Dangit, I think I'm gonna have to pick up one of these Q9450's. I've got the right stuff to do it. I've been holding out for an E8500, which would easily go well over 4.0GHz, but if I could get a Q9450 to do 3.6 rock solid, I'd be pretty happy. The price is the same and I still can't even find any E8500's...

The thing is nothing I really do would take advantage of 2 more cores... maybe I should just get an E8400...

[RunawayPrisoner]

E8500s? Hmm... I have one, but another member may be interested in the chip already. I never tried pushing the chip, but I know it might not be worth the effort since it's .5 more than E8400 and... that's it. From what I can see, E8200, E8400, and E8500 all need the same kind of voltage for higher clocks.

As far as quads go, 1.344v for 3.9GHz here. And probably 4.0GHz would require 1.38v, just like the QX9650.