|
Junior Member
Join Date: Jan 2008
Posts: 8
|
System configuration (bought everything new 3 weeks ago):
Q6600 G0 @ 3000 (9*333) VCore 1.28 and stable Case: Antec P182 (1 x Front Intake Fan 120 mm, 2 exhaust fans 120 mm, 1 x 120 mm fan for PSU) MB: ASUS P5K-E WIFI Replaced CPU-Cooler fan with a 1300 RPM (2° down) Temps Idle (all case fans @ ~850 RPM - CPU Fan 1330 RPM): Core 1: 44° Core 2: 44° Core 3: 38° Core 4: 39° Temps load (all case fans @ ~1250 RPM - CPU Fan 1330 RPM): Core 1: 65° Core 2: 62° Core 3: 60° Core 4: 61° Room temp: 24.5° The temperatures where measured with Everest & Core Temp (same temps). Now my questions, why is there a difference in temp of 5° from one core to the other and I thought with the Ninja it should have been possible to bring those temperatures down even when overclocked. To me those temperature seem to high especially under load. Any idea how I could improve the temps or should I just go for another cooler? I used Arctic silver 5 thermal paste (possibly too much). Please advise. Thanks! |
|
Last edited by parkgarage; 01-12-2008 at 03:29 AM.. |
|
|
Member
Join Date: Nov 2007
Posts: 30
|
antec 900+ extra fans, all on medium
q6600 B3 rev. scythe ninja rev b +1600rpm s-flex asus p5ne-sli 8800gt stock ![]() 2.7 oc on silly volts due to it playing up this morning, normally at 40 but i cranked it until i can be bothered to play around with it again. ![]() 65 is about max for the quad i believe. Can't remember the load temps atm but yours look about right for 3.0ghz. just checked my log and 56 was max today, encoded some dvds and played tf2 / cod4 for a few hours. i normally only prime first testing my oc. as for the 5degc temp difference, I have read from many sources 6c is normal for dirrences between any of the cores, dont loose any sleep over it. ![]() my first core is always higher than the others, put it down to windows and most progs running on it. hope this helps. Ry oh and just for proof! antec ftw! ![]() |
|
Last edited by Ryan; 01-13-2008 at 12:14 AM.. |
|
|
Junior Member
Join Date: Jan 2008
Posts: 8
|
Thanks Ryan.
I believe with the Antec P182, which already has a good airflow and the additional intake fan I've put in there these temperatures especially under load are just too high. I don't want to go over 60° under load. Would it make sense to put a second fan to the Ninja: Like in your picture the P182 has also these two exhausts fans on the top left and like you I have the cooler fan directed to the exhaust fan on the back. So, I wonder if there would be any improvement if I have another fan on the bottom of the cooler directed towards the exhaust fan on the top? Unfortunately I only have one set of clips and I'm not sure, if this is possible at all to mount two fans. Of course I know that room temperatures have a big effect on the cooling of the CPU and my 24.5° are quiet high, but I just want to stress that during the hot summer months, the room temperature could raise to somewhat like 27° - 30°. Yesterday I left the windows open for about 10 minutes and the room temperature cooled down to 22° - the CPU on the other hand cooled down within this 10 minutes from 65° to 60°. |
|
Marketing
Join Date: Apr 2007
Location: Hamburg
Posts: 874
|
Hi there.
The temps are read directly from core and 65°C are not the big issue, check the tJunction ![]() There are many ways to improve performance though.
__________________
* Register and enjoy the official Scythe Forum * Compatibility List for Scythe CPU Coolers |
|
Junior Member
Join Date: Jan 2008
Posts: 8
|
Airflow in the case is already good
Thermal Grease = Artic Silver 5 (see: Arctic Silver Incorporated - Arctic Silver 5) is good So it only remains to lap to IHS, which is not a work I'd like to do. Would a second fan on the cooler help, either as mentioned above or a push-and-pull configuration towards the exhaust fan in the back? |
|
Member
Join Date: Nov 2007
Posts: 30
|
I'd be perfectly happy with the temps you have 65 under 100% load is a good temp, as shinigami said tjunction max is miles off.
I got confused with max tcase which is put down as intel as 62.2c for B3 rev. My bad. see below for explanation, the tjunction is the thing measured in core temp so you are 5deg below the SAFE level according to below. Maybe oc some more? ![]() whole article here for you: Core 2 Quad and Duo Temperature Guide - Overclockers Forums Regards Scale 2: Quad Q6x00: Tcase Max 71c, G0 Stepping, Tjunction Max 100c, Vcore Default 1.372, TDP 95w, Delta 10c -Tcase/Tjunction- --70--/--80--80--80--80-- Hot --65--/--75--75--75--75-- Warm --60--/--70--70--70--70-- Safe --25--/--35--35--35--35-- Cool Scale 4: Quad Q6600: Tcase Max 62c, B3 Stepping, Tjunction Max 100c, Vcore Default 1.372, TDP 105w, Delta 10c -Tcase/Tjunction- --60--/--70--70--70--70-- Hot --55--/--65--65--65--65-- Warm --50--/--60--60--60--60-- Safe --25--/--40--40--40--40-- Cool |
|
Junior Member
Join Date: Jan 2008
Posts: 8
|
Thanks for all the patience with me.
I changed some settings in BIOS and got a prime stable (30 Minutes) system with a Vcore of 1.22 and OC 9*333. Temps are now: Idle (Case fans @ 850 RPM / CPU cooler @ 1300 RPM) Core 1 38° Core 2 38° Core 3 34° Core 4 35° Load (Prime) (Case fans @ 1200 RPM / CPU cooler @ 1300 RPM) Core 1 59° Core 2 58° Core 3 54° Core 4 55° If I want to OC to 9*355 the temps under load goes quickly up to 70° with a Vcore of 1.29 (stable). I know I have safe temperature now with 9*333 OC, but with this Vcore of 1.22 to me the temperatures should be even lower. Am I wrong? |
|
Member
Join Date: Nov 2007
Posts: 30
|
Quote:
I'd kill for these temps and that G0 chip. at 9x300 I have to use a vcore of 1.3750 just to get stable which is 1.3825 in my bios. And then I idle at 37-43. I have a hotter B3 chip and it takes alotta volts to get stable. If you want to swap then be my guest! Stop moaning about good temps, you're making me want to cry. 70c is still 30 away from the chip throttling back. Excuse me while I go cry some more.. |
|
Member
Join Date: Jul 2007
Location: Moscow, Russia
Posts: 74
|
parkgarage
1. Different tasks generate load on different cores (and even on different sections of each core), so uneven temp reading distribution is ok. 2. You need a cure for your thermophobia. The critical (that can cause physical damage) temp for almost _any_ semiconductor device is about 120C, any given figure below that can be reasonably considered BS. This also applies to throttling settings on Intel CPUs: if throttling can de disabled via BIOS - feel free to do it, chances of frying the proc are _very_ slim even if the CPU fan has been jammed. |
|
Marketing
Join Date: Apr 2007
Location: Hamburg
Posts: 874
|
Hi everyone.
Well, I have to agree on what Ryan and EndoSteel says here. Please don't panic becasue of that temps. You got a lucky G0, I have B1, not even B3. So with that B1 I can not even overclock, even lapped and with Mugen it does not give you any OC, temps are around 85°C voll load. So I should be the one who cries ![]() SOme people mistake the CoreTemps with the old-school temps from Mobo-Sensor. So back then the temps 30°C were kind of good and 60°C is already like hell. It takes time to make people udnerstand, that CoreTemps are read different and have nothing todo with the temps of the socket 939 and socket A times.
__________________
* Register and enjoy the official Scythe Forum * Compatibility List for Scythe CPU Coolers |