• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

E6550 voltage floor Q's

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

Theocnoob

Member
Joined
Dec 1, 2007
Location
Near Toronto Canada
Anybody know the minimum happy voltage for an E6550 at about 3.33?

I'm running 3.15 right now. I'm on standard cooling. I hit 60 under standard max load, 63 under intense gaming load, 65 under max stress (large FFT's). AFAIK, the TJmax on the E6550 (GO, Conroe) is 70 celsius or 72, so at the current voltage, additional clocks are un-doable. Idle is 34.

This is why I'm asking what the lowest voltage I can hit for 3.33 is?

Thanks guys.
 
Run CPUz to know your VID. ( SRY coretemp give you the VID... small error ! )
Monitor temps with something like CoreTemp.
Run OCCT costum test ( Small and Large data set ) for fast stability and a 8hr+ prime BLEND for 24/7 stability.

READ the OC sticky !!!
 
Last edited:
We cant tell you that. Every chip is different. That is the point of stability testing and such to find a stable system at the minimum amount of volts it takes.

Also, SMALL FFT should get you the highest temps in P95 with your chip. I find it quite odd that your temps in gaming are higher then your temps in P95. Must be the large FFT.

keep that chip under 70C for 24/7 operation.
 
Run CPUz to know your VID.
Monitor temps with something like CoreTemp.
Run OCCT costum test ( Small and Large data set ) for fast stability and a 8hr+ prime BLEND for 24/7 stability.

READ the OC sticky !!!

I got my VID from CPUz. The bios is set to 1.31 for the CPU but I Vdroop down to 1.27 all the time.
coretemp's what I use.
it's prime stable in current config.
 
We cant tell you that. Every chip is different. That is the point of stability testing and such to find a stable system at the minimum amount of volts it takes.

Also, SMALL FFT should get you the highest temps in P95 with your chip. I find it quite odd that your temps in gaming are higher then your temps in P95. Must be the large FFT.

keep that chip under 70C for 24/7 operation.

No P95 runs hottest. Sorry- may have poorly worded OP. The maximum temperature the CPU ever hits in coretemp is 65 with current settings.

I'll try going a little lower on voltages.

Should I be worried about the motherboard getting hot? It isn't cooled very well and it's not like it has great passive cooling onboard to begin with- it's a P5Q-E. Not exactly high end.

We cant tell you that. Every chip is different.


Sorry to get off topic-- I just don't understand that. Aren't chips made on big discs some 100+ at a time then laser cut apart? If so shouldn't each chip have, at very least, identical characteristics to atleast the rest of the chips on it's particular disc or whatever it's called? Or are the differences due to the inherently imperfect process of mounting the actual CPU into it's 'package'?
 
No need to double post... EDIT! :)

I dont think you understand. Within P95 there are 3 tests, Small/Large FFT and Blend. Small FFT for your chip should get things the hottest. In your post above you BROADENED the scope to the application.

I wouldnt worry about the mobo, no. Maybe put a 40mm fan on teh NB is all I would do if its HOT to the touch.

Yes, chips are made from the same piece of silicon. No they dont have identical characteristics as your as asking it. VID's are different across the wafer and so are the voltages needed to hit X.xxGhz clocks. I may be able to get 3Ghz at 1.3v and you may be able to do less. It will absolutely vary and "we cant tell you that. Every chip is different". Welcome to overclocking, thats the tweaking part of it.
 
Some people theorize that chips near the center of the wafer tend to be 'better'. If you read up or go to school to learn how lithography works you will see just how inaccurate it is. Those Intel guys in white suits are really pushing the limits and the features they are drawing are not perfect lines and spaces. They look pretty rugged when you zoom in enough.

Chips definitely have different characteristics, though. I test drove several Q6600 G0's last year, and they were all very different.

CPU-Z doesn't display VID. It displays vcore...there is a difference. RealTemp will show you the min-VID and max-VID if you hit the settings button. Min-VID and max-VID are just what the chip has been programmed to send the mobo so it knows how much voltage this particular chip needs to be stable in speedstep and at stock speed, respectively. In the Q6600 days there were several threads where VIDs were compared and it did seem that the low VID Q6600 G0's did tend to be the better OCers overall. People sought after certain batches and chips w/ 1.2-1.225v VIDs looking for the chip that could hit 4GHz.

Every since 45nm hit the market I haven't seen any evidence one way or the other that VID matters anymore. But people still tend to idolize low VID chips for some reason. Batch seems to be the best indicator overall.

Keep lowering the voltage to see when it conks out. You won't be risking anything except a corrupted OS, but that's an issue anytime you OC or UC.

Look at my sig at my file server. I OC'd the FSB to 333 from 200, but then I lowered the multi to 6 to have it run at 2GHz...file serving duties don't require much! Anyway, I was able to drop the voltage down to 0.856v and it remained stable in p95-small-fft. I've also got all the other mobo voltages set to the lowest value. Even w/ the stock cooler temps barely rise at all from idle to full P95 small-fft load.
 
Well, apparently my CPU doesn't want anything below 1.27. Started getting dropped frames and wacky artifacts. Put it back up and everything is perfect..

Is that 5200 you have the 45nm version?

.8V that's pretty cool.
 
Yep, 45nm all the way! I would have gone 65nm to save money, but decided that going w/ the smallest process CPU (45nm) and mobo (P45 = 65nm...all others are 90nm) would consume the least amount of power (save more money over time). Since the file server is on 24/7 I thought that would be best. Plus, I was also concerned w/ heat b/c I have my PC, the wife's PC, and the file-server in the same small office. It can get warm in here.


What speed at 1.27v? If it's 3.33GHz that's not bad at all.
 
Back