dslreports logo
site
 
    All Forums Hot Topics Gallery
spc

spacer




how-to block ads


Search Topic:
uniqs
657
share rss forum feed


MineCoast
Premium
join:2004-10-06
127.0.0.1

Are these temps right?!

Click for full size


Dissembled

join:2008-01-23
Indianapolis, IN

1 edit

Are they right, or are they good? lol

I run mid 30's idle and mid 50's under heavy gaming load. I have a decent aftermarket cooler on mine. Some processors are slightly different, but yes, that is way too high.



MineCoast
Premium
join:2004-10-06
127.0.0.1

Well, obviously they aren't good lol. I'm idle right now at around 75c but under load well, 110+c. At 100% load even at those temps (if they are correct) the system is rock solid, but it does seem to be lowing the CPU speed. This is a stock intel HSF that is very old and isn't in the best of shape (I have a very difficult time mounting it to the motherboard because the clamps are worn.

I upgraded from a E6300 to this and didn't have any thermal paste so simply reused what was already there.

Either I need to just go replace the entire HSF and new paste or just trying to replace the paste but this HSF is very difficult to get attached to this motherboard cause the plastic clamps are so worn out.



Dissembled

join:2008-01-23
Indianapolis, IN
reply to MineCoast

Have you opened it up and looked at it? The first two easy things are to:

1: Clean
2: Verify your fans are all working.



Dissembled

join:2008-01-23
Indianapolis, IN
reply to MineCoast

I suggest you double check your work on mounting that HSF. I have personally had a HSF that was not fully seated and it will cause this problem.



Dogg
Premium
join:2003-06-11
Belleville, IL
reply to MineCoast

I would first verify the temps from the BIOS. Software solutions rely on drivers and are not always accurate. So verify the temps first, before worrying or taking further action.
--
Google is your Friend


Aranarth

join:2011-11-04
Stanwood, MI
reply to MineCoast

Yep, but do it FAST.



koitsu
Premium,MVM
join:2002-07-16
Mountain View, CA
kudos:23
reply to Dogg

said by Dogg:

I would first verify the temps from the BIOS. Software solutions rely on drivers and are not always accurate. So verify the temps first, before worrying or taking further action.

Absolutely agree about checking the UEFI/BIOS first.

One correction: the above output from Core Temp does not rely on a "driver" -- the temperature readings come directly from the DTSes on the CPU itself, and the interfacing is done via native x86/x64 opcodes on the CPU itself (specifically the MSR).

To the OP: those temperatures are extremely bad, especially for what model CPU you have. The problem is almost certainly not with Core Temp (though I still recommend doing what Dogg See Profile said); I recommend you disassemble the system, remount the CPU + reapply compound + HSF, check the fan, and figure out what's going on. Gut feeling says this is a physical/hardware issue. You're lucky your system hasn't auto-powered-off due to those temperatures.
--
Making life hard for others since 1977.
I speak for myself and not my employer/affiliates of my employer.

n_w95482
Premium
join:2005-08-03
Ukiah, CA
reply to MineCoast

It looks like the CPU is throttling as much as it can from those temperatures (I'd hate to see what it would do at 1.25v). I'd definitely refrain from stressing it at all until you look into it further.

Toss the stock heatsink, it's shit and so is the mounting mechanism. Use something like this instead. Heck, at that price, I might get one or two for spares.

I tried reusing paste on a spare heatsink back when I was running my Athlon XP system. At stock speeds (1.83 GHz), it wouldn't even finish POST before it shut off. I ended up having to drop the CPU to ~900 MHz to run reliably. Once I replaced the thermal paste, it was fine.
--
KI6RIT



MineCoast
Premium
join:2004-10-06
127.0.0.1
reply to MineCoast

I went and bought some thermal paste and cleaned the CPU and heatsink and this is where I am now... It's not great, but much better than before... under load before was 110c at 1.5 ghz and this is 84c at 2.4 ghz. I can live with this for a while until I replace the system in the next year or so.


jchambers28

join:2007-05-12
Alma, AR

I have a 3570K that runs at 85c at full load.



Gordo74
Premium
join:2003-10-28
Monroeville, PA
reply to MineCoast

said by MineCoast:

I went and bought some thermal paste and cleaned the CPU and heatsink and this is where I am now... It's not great, but much better than before... under load before was 110c at 1.5 ghz and this is 84c at 2.4 ghz. I can live with this for a while until I replace the system in the next year or so.

These are much better and well within spec. The original temps in this thread were NOT.

My guess is that the heatsink was not seated properly.


koitsu
Premium,MVM
join:2002-07-16
Mountain View, CA
kudos:23

I'm in agreement with Gordo74 See Profile -- much much better, and the most likely explanation is a badly-mounted HSF (or possibly using too much thermal compound?).

Those temps are still about ~5C too high (idle) and ~14C too high (under load) for my tastes, but that's purely a matter of opinion. I'm used to ~30C idle ~60C load temperatures for the Q9550, for example, but I also use after-market HSFs because I can't stand the noise of the stock fans.

That said, barring a hot spring or summer, you should be okay running with those temps.

And in case it wasn't mentioned earlier -- I strongly recommend going forward you get yourself an after-market HSF that uses a bolt-through kit. I can personally recommend the Noctua NH-U9B SE2 (I have two of them in two separate PCs, absolutely wonderful -- very quiet, very cool, and well-designed).

I've ranted and raved for years about how utterly stupid Intel's "plastic push-pins" are and how they can cause situations exactly like what you're experiencing.

While bolt-through kits make replacing the CPU more cumbersome (you have to pull the entire motherboard out with a bolt-through kit), the advantages easily outweigh that negative.
--
Making life hard for others since 1977.
I speak for myself and not my employer/affiliates of my employer.



MineCoast
Premium
join:2004-10-06
127.0.0.1
reply to MineCoast

Thanks for the replies. I agree the temps are still higher than I'd like, but overall much better. I'm using a stock Intel HSF from a old Core 2 Duo 6300 I believe, so I'm sure this HSF is below specs for the Q6600 but it'll be enough to get me by until I upgrade this system. I still can't believe I've been using the Q6600 for over 5 years now and it still does half-way decent by today's standards. Best CPU I've owned by far.

Either way, the HSF/CPU didn't have hardly any thermal paste on it and what it did have was the stock stuff that intel puts on. I simply pulled the HSF off the old CPU and onto the new CPU without replacing the paste.... I know this is the worst idea ever. I had already done this once with this HSF so I'm sure there was very little paste left and CPU didn't have any at all other than what was left on the HSF from the last two CPU's.

I'm OK for the time being at least. I will say that I HATE these plastic punch through HSF's... pure garbage and because I didn't know how to work it correctly, this one is "broken" and took a bit of work getting it to mount to the motherboard since I messed up the plastic.



Gordo74
Premium
join:2003-10-28
Monroeville, PA
reply to MineCoast

The new thermals are in line with a PROPERLY setup stock heatsink.

You MUST CLEAN the old stuff off with a cleaner like rubbing alcohol from BOTH the CPU AND heatsink. Then apply the new stuff.



MineCoast
Premium
join:2004-10-06
127.0.0.1

said by Gordo74:

The new thermals are in line with a PROPERLY setup stock heatsink.

You MUST CLEAN the old stuff off with a cleaner like rubbing alcohol from BOTH the CPU AND heatsink. Then apply the new stuff.

Yep. That's what I did this time when I reseated it

Aranarth

join:2011-11-04
Stanwood, MI
Reviews:
·Frontier Communi..
reply to MineCoast

I have a q6600 running at 3.2Ghz and under load it hits 55C and typically idles at about 36C.

I'd say your heat sink could really use an upgrade!

I'm using a xigmatech with the back plate (can't remember what number 1236 or something like from 4 years ago).. It is fed by a 120mm case fan blowing right at it and the heatsink fan setup as exhaust.
Heat is then exhausted out the top of the case.



Subaru
1-3-2-4
Premium
join:2001-05-31
Greenwich, CT
kudos:1
reply to n_w95482

said by n_w95482:

I tried reusing paste on a spare heatsink back when I was running my Athlon XP system. At stock speeds (1.83 GHz), it wouldn't even finish POST before it shut off. I ended up having to drop the CPU to ~900 MHz to run reliably. Once I replaced the thermal paste, it was fine.

I did a test back in 2000 or so with the AMD T-bird no heatsink and no paste.. I put my finger on the die and turned it on.. damn like 2 seconds it was untouchable :-|
--
It's NOT Ni-kon It's NE-KON!




LG is NOT Lifes Good It's Lucky Goldstar!



Gordo74
Premium
join:2003-10-28
Monroeville, PA

said by Subaru:

said by n_w95482:

I tried reusing paste on a spare heatsink back when I was running my Athlon XP system. At stock speeds (1.83 GHz), it wouldn't even finish POST before it shut off. I ended up having to drop the CPU to ~900 MHz to run reliably. Once I replaced the thermal paste, it was fine.

I did a test back in 2000 or so with the AMD T-bird no heatsink and no paste.. I put my finger on the die and turned it on.. damn like 2 seconds it was untouchable :-|

Pretty dangerous to do in the old Socket A days, as those didn't have thermal protection.

Gem
Premium
join:2005-09-10
kudos:4
Reviews:
·CableOne
reply to Subaru

said by Subaru See Profile
I did a test back in 2000 or so with the AMD T-bird no heatsink and no paste.. I put my finger on the die and turned it on.. damn like 2 seconds it was untouchable :-|

:

I heard AMD chips of that era would literally burst into flames if you left them running without a heatsink long enough.

There may have been the equivalent of an eBay video of that back then.

Intel's chips were also tested in the video I recall. They slowed way down, but did not erupt in flames, iirc.



koitsu
Premium,MVM
join:2002-07-16
Mountain View, CA
kudos:23

said by Gem:

said by Subaru See Profile
I did a test back in 2000 or so with the AMD T-bird no heatsink and no paste.. I put my finger on the die and turned it on.. damn like 2 seconds it was untouchable :-|

:

I heard AMD chips of that era would literally burst into flames if you left them running without a heatsink long enough.

You're talking about stuff like this:

»www.youtube.com/watch?v=XgOmMAasqto

»www.youtube.com/watch?v=ssL1DA_K0sI


2nd video is NSFW (language).

Draw your own conclusions; I pass no judgement.

I have no idea if AMD ever implemented the equivalent of Intel's #PROCHOT signal, but I sure as hell would hope so.
--
Making life hard for others since 1977.
I speak for myself and not my employer/affiliates of my employer.


Subaru
1-3-2-4
Premium
join:2001-05-31
Greenwich, CT
kudos:1
reply to Gordo74

said by Gordo74:

said by Subaru:

said by n_w95482:

I tried reusing paste on a spare heatsink back when I was running my Athlon XP system. At stock speeds (1.83 GHz), it wouldn't even finish POST before it shut off. I ended up having to drop the CPU to ~900 MHz to run reliably. Once I replaced the thermal paste, it was fine.

I did a test back in 2000 or so with the AMD T-bird no heatsink and no paste.. I put my finger on the die and turned it on.. damn like 2 seconds it was untouchable :-|

Pretty dangerous to do in the old Socket A days, as those didn't have thermal protection.

hah yeah, I still have that CPU too sitting inside the clear plastic case it came in, my first AMD cpu.. not too sure if it's useful for anything now really.
--
It's NOT Ni-kon It's NE-KON!




LG is NOT Lifes Good It's Lucky Goldstar!


n_w95482
Premium
join:2005-08-03
Ukiah, CA
reply to koitsu

Hehe, thanks for posting the Tom's video. I was going to do it last night but ended up not getting on the PC .

I don't know for sure about a #PROCHOT equivalent myself, but I have a little story that may help. A few years ago, I helped a friend do a CPU swap in his Socket 939 machine to a dual-core Opteron. We reused the stock heatsink from his previous single-core CPU. I noticed that it was a major PITA to get it back on, but didn't think much about it as it was the first time I'd installed a stock 939 heatsink.

We fired up the machine and proceeded to BS for an hour or so. I then checked the CPU temperature using CoreTemp. It was idling near 70 C and quickly shot up to around 80 C with one core lightly loaded. We turned off the PC and I pulled the heatsink off. The dab of thermal paste I put on it was completely untouched. The PC ran fine the whole time though.

I remounted the heatsink and it went on MUCH easier this time and was definitely making contact with the CPU. Temperatures dropped to normal levels and everything worked perfectly.
--
KI6RIT