1.6vcore ok?

Logic Zero

Logic Zero

Logic Zero

Associate
Joined
21 Aug 2006
Posts
648
Hi guys,

Just checking, I am @ 3.6ghz on my E6600, with 1.575 vcore, 62c load, and it lasted 6 hours before failing orthos. I know 1.6v will cure this but am I ok to run it? Temps really dont get any higher than 62-65.

Cheers in advance
 
Core temp :)

Btw, I am using speedfan to see what my CPU voltage looks like, reads as around 1.52-1.55 when I set it at 1.575 and it's under load. Is this accurate, taking into account the vdroop?
 
I've been looking into how programs read C2D/C2Q core temps and what the authors don't tell you is that their programs could easily be wrong by +/- 15c! :confused:

It's because Intel don't provide a guarenteed method of programmatically reading our CPU's TjunctionMax so when it gets it wrong the core temps you are seeing aren't accurate. The only way to find out if it's right or wrong is by running your CPU without a heatsink until it reaches the TjunctionMax temp and starts to throttle.

Reason I'm mentioning this is because many Quad core setups with high volts are believed to be running at acceptable temps (<= 65c) when in actual fact they could easily be 15c higher! :eek:
 
Great post tbh, very good point. However, I have a question, do you think the extra 15c is really doing any harm? Think of all the c2d's out there in systems right now, bought over a year ago, stuck in a closed door cabinet under the desk, running at 60c + idle, they still pull strong tbh. Think of all the abused P4's that home owners buy, and are running hot due to dust/bad TIM etc, they are all still alive and kicking.
 
When I was testing without a heatsink to find out my E2180 M0 stepping CPU's TjunctionMax (CoreTemp 0.95.4 shows 85c) I also tested it with Orthos running (bad idea due to temps shooting up very quickly) and it throttled at 85c and took another ~2 seconds before switching itself off. The shutdown temp was probably not much higher than 100c.

So I guess many on here (myself included) are a little conservative regarding max stress temp but still a decent margin for when the temps increase significantly due to it being a hot day or whatever is probably not a bad idea.
 
When I see 1.6v mentioned the alarm bells go off in my head but I dunno really if it's all that excessive or not. :D

Without using a multimeter you can't know if it's accurate or not. Generally speaking it can be reasonably accurate. In the past I've checked the actual vcore with a multimeter on various motherboards and they were reasonably accurate but I haven't done it with any of my Conroe setups.

My Asus P5W DH appears to vdroop by around 0.05v and I just assume it's correct.
 
Can you have a look see what the motherboard reported CPU temp is when the core temps are at 62c - 65c. You can use Everest or even Speedfan should show it. Also is it a B3 or G0 and is the Tjunction in CoreTemp 0.95.4 85c or 100c?

Screenshot of everything would be nice save you having to write everything down. :)
 
str said:
Can you have a look see what the motherboard reported CPU temp is when the core temps are at 62c - 65c. You can use Everest or even Speedfan should show it. Also is it a B3 or G0 and is the Tjunction in CoreTemp 0.95.4 85c or 100c?

Screenshot of everything would be nice save you having to write everything down. :)
Click to expand...

Tjunc is reported as 85c. Speedfan measures temps exactly the same as my coretemp (poss latest speedfan?), so I cant measure from that, but I'm thinking around 10c - 15c because of the difference that my BIOS reads my temps at idle and my coretemp at idle.
 
id be warey of 1.6v, thats comfort limit on the 90nm athlons... let alone a 65nm chip O.o

then again, as ive said before, if you scrub it down to 2 years lifetime and are still happy with the relative performance in 2 years, ill show you someone who would have been happy with it running at stock when they got it 2 years before ;)
 
1.6v is a hell of a lot if its on air, considering Intel say the TMAX of the processor is 1.55v.

My C2D did 3.6Ghz on 1.425v with a lot of droop (about 1.35v actual) - maybe you would be better off aiming a bit lower? :(

Voltage will kill a CPU much faster than heat will, and it doesn't necessarily track that you will hit the temperature barrier before you knacker the chip. Incidentally I've seen P4s that have been overvolted in their lifetime and they exhibited electromigration effects....

The candle that burns twice as bright, burns half as long... etc. :)

c2d_elec.gif
 
Last edited:
Ive had my CPU in sig at 1.6V Indicated in Bios/Speedfan/CPUz (latest version) for over 6 months now and its had Zero issues, i am WCing by the way and its idling at 23C with an ambiant room temp of 18C.
 
You must log in or register to reply here.
Back
Top Bottom