Matters of OCing and Heating

crs_cwiz · 12-03-2007, 05:55 PM

Hey everyone here's my config:

Intel C2D E6600
Asus P5n32-e sli (nforce 680i)
4 x Corsair Value Select 512 MB 667mhz ddr2
XFX geforce 8800gts (640mb)
2 x WD 320Gb @ 7200 RPM
Antec Performance TX1088AMG cabinet
Corsair 620W power supply

While running vista home premium/XP, my normal cpu temp is 50-55 and system temperature is 40. GPU temp is 60. Under max load, temp is:
CPU - 60
System: 43-44
GPU-70
Are these temperatures normal or is my system overheating? I have one 80mm inlet fan in front and 120 mm exhaust fan behind. plus my PSU has 120mm exhaust fan.

Also, just now i tried overclocking my gpu using ntune. I have latest drivers (97.92) whenever i click on the button "find Optimal" the screen shows a blue screen. no, not a BSOD, but a blank blue screen, and even the colour tone of the blue is different. I thought it was just detecting the frequencies so i left it for 10 mins but still nothing!! Upon restart the pc works fine. I saw various forums on the net and found that the driver support for 8800 is not very nice and BSODs are common, but i havent had any bsods. How do I OC my GPU? I'm adding extra cooling for my GPU, just to be sure

Thanx

EDIT: By the way, I have arctic silevered my cpu

assasin · 12-03-2007, 06:41 PM

yaeh ur temps r a bit high.if u wanna oc gfx card thru xp,use coolbits.as far ocing proccy is concerned do it thru bios.
attaching a fan may help reducing the temp.

how much did the rig cost u,specially gfx card.

crs_cwiz · 13-03-2007, 03:50 PM

Hey everyone!
I OCed my rig through the bios, not much increase, only 5%, but when i ran the stability test in my nvidia control panel the proccy temperature shot up to 60 C in just a min. Should I overclock my system or let it remain stock?

BTW. my rig came at almost the MRPs, put a few hundred here and there. got my gfx for 33250 (all inclusive)

assasin · 13-03-2007, 05:38 PM

try running ur pc with the cabinet open.

dOm1naTOr · 13-03-2007, 06:45 PM

As of the temps, CPU temps is somewhat OK as system temp is 40c. This may be due to high environment temp these days. Add more system fans. If ur cabby has ny prescotte duct, or any provision of fan on side panel, then get one.

As bout ocing ur GPU, me too gets a blue screen is i select 'find optimal'. So u just have to increase the clocks nd hit 'test setings' nd hit apply nd run some game to see if theres ny unstability of artifacts.
Use softies like nvtemp logger to log GPU temp while playing games and use HOT CPU Tester Pro fo find ur max CPU temps.

darklord · 14-03-2007, 12:02 PM

Looks like you got a bad 6600.Bad as in one with the 'curved' IHS

My 6300 and Xeon 3040 had that problem.My 6600 was fortunately not bad nor is my 4300.

Anyways, you have a Asus p5N32-E SLI mobo.Superb board for OCing.Dangerous in the wrong hands.Also remember,the 680i chipset is a fire breathing monster.The heatpipe cooling just doesnt cut it.You will need much better cooling on it and that board will eat 500 FSB for breakfast.

Also i have seen that Asus has done a bad job of fixing the Heatpipes.When i had removed mine,it wasnt making proper contact with the chipset Core.I cleaned it and applied AS5 and things improved.

Remember, G80 is KNOWN to work hot and the 680i as mentioned above is arguably the hottest of all C2D chipsets out there.
The heat output of these components is in some way transferred to the CPU,resulting in higher CPU temps.

Also,
Here's a hint, use the board in unlinked mode,it owns everything when run in unlinked mode while OC'ing.

Here's a glimpse of what it is capable of,
http://valid.x86-secret.com/show_oc.php?id=163428

crs_cwiz · 15-03-2007, 03:35 PM

Hey, thanks everyone.. looks like i got myself a really "hot" rig. Yeah, I have a provision of fans on my side panel, two infact, so i'm gonna use them. I've got another problem guys. I have 4 chips of Corsair Value Select Ram, 512 MB @ 667 Mhz. I had been having problems ever since I assembled the rig, and found out yesterday that whenever I put in all the 4 modules together, the system becomes unstable. I confirmed it with memtest86 (http://www.memtest86.com/) that when all the 4 moduls are in, the input and output values in ram dont correspond, but these errors occur only within a bracket or around 200 MB. When i remove any one module, system runs fine again. Could my mobo have problems handling all 4 modules together, or is it module defect? I'm really eager to use the modules together... my RAM score in Vista performance bench hasd dropped from 5.4 to 4.5!!!

Thanks..

darklord · 15-03-2007, 04:16 PM

Running 4 DIMMS isnt easy for the MCH.

2 Stick is perfect,anyways you could try 1 stick at a time to find out the bad one.Also flashing to newer BIOS should improve memory compatibility.Also try relaxing the memory timings and use 2T command rate.

Also try unlinked mode and keep memory speed at 533 to test.

Post back ur results.

crs_cwiz · 15-03-2007, 05:35 PM

Hey,

I found a solution to the heating problem, turns out that the CPU fan just wasnt put in properly by my engineer. Next time, Im gonna do all the stuff myself. Now it is running @ 2.7 GHz at just 51 C!!!

As for the RAM, I tried everything, different slots, different modules, lower speed timing, etc. No matter what speed, the modules dont work, but only if all 4 are together. they run perfectly when =< 3. I guess some compatibility issue. Asus hasnt even released new bios versions... guess I'll have to wait...
__________
Hey, could it be possible that 4 modules may require more power to run? I've been reading things like this on many hardware forums. Should I slightly increase the voltage settings??

darklord · 15-03-2007, 05:35 PM

Which BIOS version do you have now ? 2 new BIOS revisions did come out.

Did you try relaxing the Timings for your RAM ? Also did you try running 2T command rate ? Try Unlinked mode [Hint: disable SLI memory setting] That should enable Linked and unlinked mode.
I dont think Vdimm might be the problem,we will look into that later.Try the above mentioned things first.

crs_cwiz · 15-03-2007, 05:38 PM

I have the version 0602. There arent any new bios versions, the 0702 and 0902 on the net are beta versions, I guess, and have already messed up many boards(what i can gather from forums)

The Memory is already running at 2T cmmand rate (whatever that is - could you please explain what it means?), and I also relaxed timings and also speed,but the errors still come

could it be possible that 4 modules may require more power to run? I've been reading things like this on many hardware forums. Should I slightly increase the voltage settings??

assasin · 15-03-2007, 05:51 PM

restore BIOS defaults and run the 4 modules.report ur result.

crs_cwiz · 15-03-2007, 06:04 PM

No reprieve, the problem continues with bios defaults and 4 modules...sigh...

assasin · 15-03-2007, 06:16 PM

if u can try using a diff set of 4 ram modules.also try all 4 of ur modules in another pc.

crs_cwiz · 15-03-2007, 06:29 PM

hmm... that wont be possible, though, i havent got any more modules/computers which support ddr2. i'll try tough...

assasin · 15-03-2007, 09:47 PM

ur friend cud be having one.

darklord · 16-03-2007, 10:30 AM

Quote:

The Memory is already running at 2T cmmand rate (whatever that is - could you please explain what it means?)

Command Rate is the time needed between the chip select signal and when commands can be issued to the RAM module IC. Typically, these are either 1 clock or 2.

Asus P5N32-E SLI Bios Ver 0903

There you go, dload it and update.

Release Notes -
1.Support new CPUs.
2.Fix PreslerXE 3.73G will run at 3.19G after CMOS fails.
3.Use new German translation.
4.Fix PreslerXE & SmithfieldXE CPU fail to boot.
5.Update EZ-Flash module.
6.Fix Vcore error when above 1.80V.
7.Fix CPU Vcore can not be set by user when Overclocking at certain frequency.

Btw are all the 4 Modules purchased together,i mean same batch ?