Sign-in / Sign-up
Your question

AMD1.4TB Temperature

Tags:
  • CPUs
Last response: in CPUs
December 24, 2003 1:29:31 AM

Hi,

Lately I've started getting assorted blue screen error messages when I play graphic intensive games. This happens only on hot days generally. My CPU reads high 60's deg C before it crashes. This use to happen when I overclocked the system too much. It now runs un-overclocked. 60 deg C is standard for my system, with no games running.

I have 2 case fans & a decent silverado cooler. I notice sometimes the graphics start to 'break down' and go funny colours just before a crash.

This seems to be a heat releated problem but I was wondering why it has just started now. I have cleaned up any dust build up & checked the fans are all still working.

Could it be a graphics card heat problem?

I know AMD 1.4TB get very hot, is approaching 70deg C too hot? I hate noisy computers so have resistors slowing down the case fans. I'd understand if it had always done this but am a bit puzzled as it has only just started.

Any advice really appreciated.

Thanks,

Ape

Spec:
Gigabyte GA-7DXR (F9 Bios)
AMD 1.4 266 Thunderbird
512 PC2100 DDR RAM
Seagate 60g Barracuda 4
Creative GeForce 2 GTS (4.4.0.3 Nvidia Drivers)
Creative Audigy DE
Plextor 40x12x40 CDR
Toshiba DVD/CD
LS120 Drive
Motorola 4100 Cable modem (V2.1 Drivers)
Network card (disabled)
Win XP Home

More about : amd1 4tb temperature

December 24, 2003 8:50:04 AM

I think it's your graphic card since when a CPU overheats it doesn't usualy give a blue screen but freezes the computer comletely. Also, the funny graphics point to the graphics card. If it has active cooling (a fan) it might have broken down or another could have happened that has affected the cooling of the card.

I know what everyone's been wondering - 2^33634943-1 IS NOT PRIME. Now, wasn't that worth running <A HREF="http://www.mersenne.org" target="_new">Prime95</A> for a month to know?
December 24, 2003 9:21:33 AM

well it sounds like it could be the video card. when the computer on check to see if the video card fan is on.

Last week My Gf3 ti 500 fan stop and got lines on the monitor and some programs like msn messanger would not run. And I did not get the lines untel late. Last two days. So I unplug my video card and plug in my old pci Diamond S3 virge Card Not much power but it save my rear.
So you could be right.

But here the thing about the 60c if you have the old Amd 2000+/2200 or lower and I think intel Not sure 2.2ghz or lower. But amd change on the newer chips to save money. They run on .18 Micron. Which Like my computer 50 to 60c is normal. Intel is a little cooler by five 45c to 55c. Now if you are running the .13 Micron you are running Hot on the new chips they run around 35c to 43.3c .18 chips can hit around 90c but the new .13 can hit up to 80c.
Related resources
December 24, 2003 11:38:03 AM

From my experience,high CPU temperature can also cause the computer to reboot or freeze but blue screen is definitely not caused by CPU. Check your softwares and games. It can cause blue screen too. I think there is a problem with your video card. Do you OC it?If yes, try to lower the core/memory frequency a bit

AMD Athlon XP 2500+@2.15GHz
Abit KD7-RAID
Seagate Barracuda V 80GB
3*256MB Kingston PC2700
December 24, 2003 11:56:48 AM

When my Geforce4MX's fan stopped it didn't do any problems except burn my finger when I touched it, when I put a case fan in front of the card (computer lying down) it wasn't hot anymore, might temporarily solve your problem.

I know what everyone's been wondering - 2^33634943-1 IS NOT PRIME. Now, wasn't that worth running <A HREF="http://www.mersenne.org" target="_new">Prime95</A> for a month to know?
December 24, 2003 3:55:38 PM

Remove and remount your CPU heat sink and fan using some good thermal paste. Be careful if you haven't done this before... I cracked my old T-bird 1.4 with a botched HSF installation!

Your 60 degrees is too hot...

Scout
700 Mflops in SETI!
December 26, 2003 8:03:37 AM

Thanks for all the help.

I did a bit of research and read that a 1.4TG can go up to 95degC. Whether they meant that after that it is dead or starts to go wrong, I don't know.

I actually tried changing my graphics drivers from the Nvidia ones I was using to the Creative ones (4.4.0.4)that were released quite a few months ago. I figured as my card is getting a bit old now, the newer drivers written for all GeForce cards may not be the best to use. I had been using the newer drivers for quite a while without issues though.

So far so good.

I replaced my graphics card fan a few months ago after it started whining. It's still going ok. Thanks again for all your help. Sorry to bother you all for what seems to just be a driver problem.

Ape
December 26, 2003 9:47:12 AM

Right, the 95 degrees is in the processor core, but they measure the heat under the CPU socket on these older Athlons so it's going to register cooler than the die internally. I know my old T-bird 1.4 became unstable in the high 50's. 'Course, the accuracy of the measurement can be quite a bit off too... I just thought your 60 degrees, combined with instability indicated it may be too hot. Glad you got it fixed!

Scout
700 Mflops in SETI!
December 29, 2003 7:47:33 AM

Well, it's not a driver problem, at least the change in drivers didn't help.

It's 30degC ambient today in my lounge. Gwum, the Gigabyte motherboard monitoring software, shows my CPU temp at 65degC at idle.

I just got another blue screen error, while playing Generals Zero Hour. No notice was given in the way of display problems. The blue screen error mentioned 'Memory_management' as the reason.

This is the third error since changing drivers. I wonder if my graphics card cannot handle the heat, not my CPU. Maybe the fan I used to replace the dead one on my graphics card is running too slowly or something.

I've been in Australia for 2 years now. Previously I was in the UK. There, I never had a problem. Now I am in Australia, maybe my system hasn't met these temperatures before. It did manage last year though.

I'd be interested to know if there is any software that can read the temp on my CPU and store it so after a crash I can see how hot it was on crashing. A modification to the Gwum software maybe.

Anyway, from your advice so far, it is a graphics card problem, not a CPU problem. Maybe I'll just have to upgrade. It's not the money, it's the Mrs!!

Any other advice welcome.

Ape
December 29, 2003 9:58:18 AM

Wusy, you are answering two of my questions at the moment, thanks for that.

It was indeed mentioned it must be my graphics card overheating. I suppose I can take the cover off, yes, I'll try that.

Ape
December 29, 2003 10:46:51 AM

you could be having memory issues too under heat, you said one screen mentioned a memory issue? run memtest and see if you have errors.....

Fat, Drunk, and Stupid is no way to go through life Son.
December 29, 2003 6:57:43 PM

Wusy has a good idea... open the case and put a house fan next to the PC blowing in... That'll tell you for sure if it's a cooling problem...

Scout
700 Mflops in SETI!
December 30, 2003 2:42:11 PM

Quote:
if there is any software that can read the temp on my CPU and store it

Motherboard Monitor 5. Just google it, there's a free version. You can set it to log to file at certain intervals, so you can see more or less how hot it was before crashing.

---
<font color=red>Those of you who think you know everything are annoying to those of us who do.</font color=red> :wink:
December 31, 2003 10:03:54 AM

I tried Memory Test 86 Ver 3, after 10 hours I got fedup with waiting, it was still testing.

It did come up with some errors but I've no idea what they mean. I do have a screen shot but couldn't find a way of posting it here. I'll give you one of the error messages, they are all very similar, just with different addresses:

Test: 6
Pass: 20
Failing Address:0000688dcac- 96.8MB
Good: 00000000
Bad: ef000000
Err-Bits: ef000000
Count:1

After many of these messages, I did get it to output the following, maybe this helps:

badram=0x0206995c,0xfffffffc,0x020698cc,0x028698cc,0x02069b8c,0xfffffffc,0x00069c5c,0x00069c5c,0x02069d1c,0xfffffffc,0x0406980c,0x0486980c,0x0008d80c,0x0008480c

I'd really welcome any comments as this means nothing to me but a problem with the RAM. Should the test really take hours & hours to complete?

Thanks,

Ape