Q6600 on the way - advice

[j.col]

I remember some saying the tj max was 90, others stated 100. Did intel ever release a statement/article stating which was correct?

this could be why C64 pulled me over the temps.
i have the tjmax in realtemp set at 90.
core temp has it by default at 85 but i add 5 in adjust offsets
so if for example realtemp is saying 71.C. if the tjmax should be 100, then my temp would be 81.C
if that is the case, that is too hot for my liking.

 

[WingZero30]

this could be why C64 pulled me over the temps.
i have the tjmax in realtemp set at 90.
core temp has it by default at 85 but i add 5 in adjust offsets
so if for example realtemp is saying 71.C. if the tjmax should be 100, then my temp would be 81.C
if that is the case, that is too hot for my liking.

Afaik Q6600 G0 TJ max is 100C. (this is what it shows in my coretemp)

Q6600 B3 TJ max is 90C

Plus the B3 runs hotter than G0 generally and C64 is right withe regards to ibt vs prime temp difference.

Download the latest version of coretemp. Realtemp shows it as 90C in your case so it is correct.

Your temp of 71C should be correct aswell.

 

[chipps101]

whats the tj max meant to be on my q9550 c1 ? core temp has it as 100 and i've always thought this chip runs hotter than my q6600 ever did.

just ran ibt on maximum and also with the amount of available ram on custom setting and the results are fairly close and the temps maxed out pretty much the same

 

[C64]

81 isn't too hot I ran IBT loads under my tuniq and the q6600 G0 hit 95c.

I am hitting 81 under a themralright archon with 1.46 volts.

 

[WingZero30]

whats the tj max meant to be on my q9550 c1 ? core temp has it as 100 and i've always thought this chip runs hotter than my q6600 ever did.

just ran ibt on maximum and also with the amount of available ram on custom setting and the results are fairly close and the temps maxed out pretty much the same

Q9550 runs cooler than Q6600 as the former is a 45nm cpu so your temps will be lower. You can see taht you are getting higher GFlops values in custom stress level than on maximum stress level.

 

[chipps101]

well my q9550 @ 3.4 with 1.24 vcore runs a lot hotter than my q6600 @ 3.3 with 1.4525 vcore ever did so either the zalman flex fitted with a viper running full pelt is garbage compared to the tuniq 120 or something is amiss

 

[C64]

set threads to 4 and select just under free memory not available free.

 

[WingZero30]

set threads to 4 and select just under free memory not available free.

Well spotted

Run it for atleast 15 passes. that way we can get broader view of your temps.

Also close core temp and then start again. So we can get fresh load values

 

[chipps101]

lol i just ran it for 5 there, went into resource monitor and i had just over 2700mb free so set it to 2700



also downloaded the latest real temp and still the same scores and temps

 

[j.col]

right. i ran ibt and now my temps are higher.
when i first set my oc, after 20 passes of ibt my max temp was 72.c (i have no proof )
but tonight after just 3 passes its at 74.C and i got one of the temps displayed in orange in core temp. so i stopped the test
this is obviously too hot now, so i will turn down oc to 3.2 and see if that helps with temps
edit new pic

 

[WingZero30]

At those GFlops values my [email protected] reaches 70c on the hottest core and your [email protected] seem to reach 65c so I think it is good enough as Q9550 is cooler than Q6600. Plus my Q6600 has a low VID of 1.2625v; though it doesn't matter much in terms of overclocking but for comparison purposes I would say you are getting good temps.

The temps may well vary with :

1) different thermal compound
2) application method
3) pressure applied
4) heatsink itself
5) case size
6) case cooling (no of case fans and their cfm rate)

So many factors can affect the performance .

 

[WingZero30]

right. i ran ibt and now my temps are higher.
when i first set my oc, after 20 passes of ibt my max temp was 72.c (i have no proof )
but tonight after just 3 passes its at 74.C and i got one of the temps displayed in orange in core temp. so i stopped the test
this is obviously too hot now, so i will turn down oc to 3.2 and see if that helps with temps
edit new pic

I think it shows yellow may be because you are reaching close to 85c tj max in coretemp. I still don't understand why realtemp and coretemp are showing different tj max value. In this case I will trust the realtemp.

You have set the threads to 4 so this should load the cpu more. Did you choose 'free' memory for you custome stress level?

 

[chipps101]

i wish i had some screens of my q6600 still but i got rid of it ages ago and about the only screen i have of it is from the fritz thread -



as you can see that was at 3.5 with 1.4525 vcore and the tuniq's fan at 1540rpm which i normally kept it at unless running prime or ibt, but with the fan at 2000 rpm it wouldn't go over 60c in ibt or prime at 3.3.

it wasn't even a low vid, it was 1.325 if i remember rightly.

 

[j.col]

You have set the threads to 4 so this should load the cpu more. Did you choose 'free' memory for you custome stress level?

i have 2550 free, so ran it at 2500

 

[WingZero30]

i have 2550 free, so ran it at 2500

Ah thats good.

I just noticed that your FSB is 413MHz as you are using multiplier of 8.

[email protected] :
367MHz(FSB) x 9(multi) = 3.3Ghz
413MHz(FSB) x 8(multi) = 3.3GHz

FSB of 413MHz will make your cpu faster than using 367MHz for same speed of 3.3Ghz. However it will also put more stress on your northbridge and temps will be higher than if you were using 376Mhz in theory.

Run it at 367Mhz x 9 using same stress test values and see if you get same temps or lower.

 

[j.col]

i couldnt get it stable with a x9 multi
i might try and reduce my vid a notch

oh! just realised we have hijacked this thread
sorry op

 

[chipps101]

At those GFlops values my [email protected] reaches 70c on the hottest core and your [email protected] seem to reach 65c so I think it is good enough as Q9550 is cooler than Q6600. Plus my Q6600 has a low VID of 1.2625v; though it doesn't matter much in terms of overclocking but for comparison purposes I would say you are getting good temps.

The temps may well vary with :

1) different thermal compound
2) application method
3) pressure applied
4) heatsink itself
5) case size
6) case cooling (no of case fans and their cfm rate)

So many factors can affect the performance .

both were as5 with the spread method and as much pressure as i could put on them

q6600


q9550


as you can see the q9550 has a much roomier and neater environment

 

[setter]

WingZero, by free do you mean simply input the amount of ram, (or slightly below it) indicated in the system resource monitor.

 

[chipps101]

that is what i understood him to mean and have done

 

[WingZero30]

Was your Q6600 passively cooled? That heatsink looks massive on the Q6600. What was it again?