Is this normal?

solidstatelogic

Distinguished
Jun 12, 2008
15
0
18,510
1st of all I am a new member so hello and I hope this forum will be helpful to me as much as the old topics on the forum have been. I used them as a source of information before I signed up. Been using them for the last month.

Here are my settings:

FSB QDR: 1333
FSB DDR: 800
Unlinked

Memory timing settings: 4-4-4-12 2T

CPU Multiplier: 9
LDT Freq: 5X

Everything on the overclocker page disabled: Speedstep etc

Vcore Voltage: 1.2v
Memory Voltage: 2.200v
1.2HT Voltage: 1.35v
NB: 1.4v
SB: 1/55v
CPU VTT: 1.4v

The rest on auto.

My GPU temp seems high, it's at 67c when I'm idle which is not to far from 70c which is the max safe temp.

1) Is there a way to lower the temp for my cpu?

2) Do my settings for everything else look sensible? or is there anything you would personaly change?

I did have a few freezes today but when I run Othos it can run for hours. I don't know what to do here. I can run at 3.2Ghz with a CPU temp of 48C but when I overload the cpu it rises to just below 60 and 62c. So I am going to stick to 3.0Ghz seeing as its not a big big difference.

3) Is it normal for orthos to run for hours but still have freeze outs? The last one happened when I had a few IE browser windows open then I went on youtube.

At idle with my current settings my temp is between 42c and 44c with the cpu at full load it adds like 10c to 12c.

Any advice will be much appreciated.
 
GPU temps can take much more than CPU, 67 does seem high, but it could just be a factory OC version, or NVidia sets the fan at 30% as standard lately. 80c is starting to get up there.

What MoBo you have? That always helps!

What is the CPU? That would help too, and the video card, and the power supply...

And sure, why wouldn't your computer be subject to other freezes, though it can orthos for hours? All that does is crank the CPU to 100% and full drain. Thats rather simple, when you do other things, lots of other stuff factor in, did you test the ram already?

--Lupi
 
Sorry about that. I was thinking this was another website where my pc details are already shown under my posts automatically.

Motherboard: Asus Striker Extreme
CPU: Intel E6600 Conroe
Ram: ATA DDR2 1GB 800MHZ x 4
Graphics Card: Asus GeForce EN7600GS (Nvidea)
PSU: Tagan TG-500-U35 Easycon XL (500W)
Cooling: Tuniq Tower 120
Case: Antec P180
 
The Tuniq 120 should really be keeping the temps lower than that, unless you have set the fan speed manually to be slow.

Since you already have Orthos, what I would suggest is doing specific tests, rather then blend. Let Orthos stress the ram overnight one night, then the cpu the next.

As far as the GPU goes, download nTune and you can adjust the fan speed with that!
 
Thanks for the replys. Verillion I think my graphics card is one of the silent ones without a fan. From what I can see it has a silver heatsink on it. It is the AsusEN7600GS/ Nvidia Geforce 7600GS.

Shadowthor I tested my ram again in mem test and it looks like I have a faulty ram slot. Tested 1gb it ran for long, tested 2 that ran for long, tested 3gb that also ran fo long and then tried all 4gb sticks and that failed around 7mins. So I though it could be that specific ram stick but out of curiousity I took out the ram in the 3rd slot and put what I thought was the faulty ram in the 3rd slot and ran memtest again. It ran for aslong as it did with the other rams.

Either the 4th slot is faulty or my system just has trouble running 4GB of ram. The last test I could run would be to shift the 3gb of ram in the system down 1 slot and that way I would really be able to tell if that 4th ram slot was gone. Either way the ram slot is faulty or my system just doesn't want to run on 4gm of ram without a struggle.

I am fine with 3GB and I have the NB set to 1.4v and memory voltage set to 2.100v this seems to be running smoothly so now I am going to work on my Vcore voltage then move on to the rest after that.

My memory timing settings are 4-4-4-12 2T. Is this right for 3GB of ram?
 
get prime and stress the ram it could be your ram
I personally don't like to use Prime95 to test RAM as it's not all ways accurate about RAM (aka. "Blend Test" , from experience I have found out Blend Test runs perfectly fine for 6hrs+, but then the next week BSODs and again run Blend 6+hrs, finally ram Memtest86+ and found out RAM was the issue ). I recommend Memtest86+ for 12hrs+ to root out all RAM related problems.
 

Really? My personal experience has taught me that memtest doesn't test for ram stability, since I have ran it for 12hours+ yet fails in blend test within 12hours. Maybe I need to run memtest longer?
 
There's Superpi, but to my experience passing the 32million test doesn't mean anything. As I said, my ram failed Prime95 blend within 12hours, yet it completed Super Pi without a hitch.