cpu temp too hot??

Status
This thread has been Locked and is not open to further replies. Please start a New Thread if you're having a similar issue. View our Welcome Guide to learn how to use this site.

Chudders

Thread Starter
Joined
Jun 23, 2005
Messages
8
hi!
I want to know what the right temp for running a 2.8 ghz p4 should be?
I have 512 kingston RAM, g-force 5200 fx. I do alot of video editing so i use the ol cpu alot for that!
At the mo it seems to be running at 38 c idle and upto 50 -55 c when in use! It's prone to resetting its self as well i supect thats down to overheating but i don't know.
I want to buy a new case so can fit more fans etc but what is the ideal temp for it to run at the overall system temp is about 40c!
Many thanks
ian
 
Joined
Sep 1, 2004
Messages
1,756
Those temperatures are entirely normal for a 2.8GHz P4. As for the resets, I would suspect a weak powersupply, you should be using a good 400W at least.
 

Chudders

Thread Starter
Joined
Jun 23, 2005
Messages
8
Thats what i have been told! But the cutout temp in the bios is 60 c i'm just wondering cos its very close!
Not that just that but with the very hot weather and all its not helping! Ne other ideas what it could be?
 
Joined
Mar 18, 2005
Messages
3,390
G'day Chudders, I usually stick the temp setting up to 65, that gives a bit of leeway, if you have a hot day you very easily run out of degrees otherwise.
If you obtain "everest home" free program of the net it runs within windows and is very helpful for watching what is happpening. Use the sensors icon in it to display temps and volts. Watch that the 5 volt rail does not droop below 4.95 volts that will give potential problems from my experience. (odd shutdowns and hung boots)
Cheers, qldit.
 
Joined
Mar 18, 2005
Messages
3,390
G'day Chudders, I usually stick the temp setting up to 65, that gives a bit of leeway, if you have a hot day you very easily run out of degrees otherwise.
If you obtain "everest home" free program of the net it runs within windows and is very helpful for watching what is happpening. Use the sensors icon in it to display temps and volts. Watch that the 5 volt rail does not droop below 4.95 volts that will give potential problems from my experience. (odd shutdowns and hung boots)
Cheers, qldit.
 
Joined
Sep 1, 2004
Messages
1,756
60C is too low, set it to 70C. The P4 automatically clocks itself down if it overheats. They are designed to run in the low 60s under heavy load.

You can't kill a P4 processor by overheating, even if you remove the heatsink while it's running.
 
Joined
Mar 18, 2005
Messages
3,390
G'day Arcadion, can't you sleep! I am about to turn in here, just after midnight!
Yes I concur with your suggestion.
Cheers, qldit.
 

Triple6

Rob
Retired Moderator
Joined
Dec 26, 2002
Messages
52,933
Arcadion said:
You can't kill a P4 processor by overheating, even if you remove the heatsink while it's running.
I have a P4 1.8Ghz that would say otherwise.
 
Joined
Sep 1, 2004
Messages
1,756
Triple6 said:
I have a P4 1.8Ghz that would say otherwise.
I'm just remembering that video Tom's Hardware Guide made where they took the heatsinks off a P3, a P4, an AthlonXP and an AthlonMP while they were running to see what happened, the P3 froze, the P4 kept going but slowed down, the Athlons went up in smoke (literally). But if your P4 really did die from overheating then I must take back what I said.
 
Joined
Apr 30, 2005
Messages
2,113
Those temps are normal. Random reboots typical causes: Weak power supply(as already suggested by Arcadion but no response) or bad memory.
 

Chudders

Thread Starter
Joined
Jun 23, 2005
Messages
8
Yea thats what i thought, i only have a 300 watts power suply which is abit crappy, might bump it upto a 400 or maybe 500! what ya think?
 

Chudders

Thread Starter
Joined
Jun 23, 2005
Messages
8
well i bumped it up2 a 400 watt psu and got a new case with 3 fans in! lol i didnt realise there was one in the top as well ! All is going well so far it seems to be running alot smother and hasnt cut out yet (touch wood).
 

Triple6

Rob
Retired Moderator
Joined
Dec 26, 2002
Messages
52,933
Arcadion said:
I'm just remembering that video Tom's Hardware Guide made where they took the heatsinks off a P3, a P4, an AthlonXP and an AthlonMP while they were running to see what happened, the P3 froze, the P4 kept going but slowed down, the Athlons went up in smoke (literally). But if your P4 really did die from overheating then I must take back what I said.
Yeah, I saw that video myself. I can't be 100% sure it died from heat but the CPU bracket was broken and the heatsink was barely contacting the CPU. I've actually replaced a lot of broken Socket 478 brackets; anyone else see lots of broken brackets? Several of the ones with the broken brackets suffered from random lock-ups and reboots; on one of them the temp hit almost 80 degrees. I can't say why my experiences contradict what seems to be widely accepted.

Faulty caps I see at least once a week on a variety of motherboards.
 
Joined
Mar 18, 2005
Messages
3,390
G'day gentlemen, well done chudders, as a matter of interest how do your voltages and temps look now.
Yes Triple6, I have seen a couple of broken bases but as far as I could determine it was through brute force being used to clip the "latch". Or unclip it, either way.
I think logic tells us that doped silicon is only good for so many transitions af thermal cycling before one or two transistors amongst the milllions fails, with avalanche effect.
It is a bit like playing chicken on a freeway, you might get away with it a few times buy inevitably you will pay.
I do feel that the quantity rule applies, lots of cycles at low temperature or fewer cycles at higher temperatures.
I really can't believe these things actually still work after some of the temperatures I have seen reached.
With the old germanium stuff there was no "second" prize, the first overheat normally was terminal.
In most metals there is a condition we used to call thermal corrosion, it was referring to heat cycling on metal strength, in view of the complexities of a processor and it's incredibly complex construction I think it is miraculous that they work as long as they actually do, so in view of this concept, longevity is relative to cycling variations. (as apart from electrical transient abuse)
I was interested in chudder's 5 volt rail voltage using "everest" sensors, I would expect something like 5.1 or something like that.
Cheers gentlemen.
 
Status
This thread has been Locked and is not open to further replies. Please start a New Thread if you're having a similar issue. View our Welcome Guide to learn how to use this site.

Users Who Are Viewing This Thread (Users: 0, Guests: 1)

As Seen On
As Seen On...

Welcome to Tech Support Guy!

Are you looking for the solution to your computer problem? Join our site today to ask your question. This site is completely free -- paid for by advertisers and donations.

If you're not already familiar with forums, watch our Welcome Guide to get started.

Join over 807,865 other people just like you!

Latest posts

Staff online

Members online

Top