780M temps in P150SM, P157SM, P170SM, P177SM?

What are your temps in full load for 30 minutes?

Which is either playing a heavy game at constant 99% or running something like Furmark or Afterburner's Kombustor.

Some are seeing around 80C, and some are seeing 90C and above.

Please post your model, the exact temps you see, motherboard revision, purchase date, and any other details that might help. Hopefully we can figure out why there is such a huge range in temps.

Edit, if you can note the room temperature as well, or in unsure, note if it is a hot or cold room.

-----------------------------------

Keep adding temperature information. But note that some have seen improvement with repasting. And for others repasting did not help, but replacing the heat sink with a new one solved the issue. As of this edit two people have solved the issue by ordering a replacement heat sink.

 

P150SM.
90-92 degrees. 93 rarely.
Purchased June 2013.

This is either playing a game or benching.

Unsure on motherboard revision at the minute.

 

I suggest taking air temperature into consideration. This is an international forum and people from different places might have dramatically different weather conditions.

This is not the most scientifically correct way to do it, but it's very easy:
Get your notebook into a air-conditioned room, which is supposed to be dry. Note down the air temperature inside the room. Push your GPU to the limit for some time, note down GPU's temp reading. Calculate the difference between the two values. Vote with the difference but report 3 values in your post.

Remember to put the notebook on a hard surface with nothing blocking the air flow.

 

P150SM, hits 93C after some time in Witcher 3, then throttles slightly and hovers around 90C... (i guess that counts in the 93C or more bracket)

 

when not OCing about 85 °C but when using my 680 Settings (1006/3000) it gets about 90-92 °C using MX-4 as thermal paste and my springs mod.

 

I have P157SM with 3.5 MB Rev.

My temps playing BF3 with everything on ULTRA maxed out around generally in upper 70's sometimes as high as low 80's depending on room temp etc.
If I do some OC they don't really change much.

I see others playing this game posting high temps in other topics. I question if the issue is the P150SM having slightly different vents and the MB maybe have a few components different?? I don't see anyone with a P157SM with temp problems, think all have been the old style 150

FWIW my machine is on a glass table top with the back about 3-4 inches away from a wall. The room is generally 78 degrees F.

 

Hmm weird i just put together another P150SM for a customer and it seems that his 780M isn't nearly as hot as ours. I only cleared the stock Thermal Paste from the GPU Heatsink and put like always just half a pea sized AS5 on the GPU Die and tighened it down like the numbers suggest and here are the results.

This Pic was taken after 1hr Valley for Temp and Stress testing the new P150SM (MB Revision was 3.2), i just used latest nVidia Drivers and Bios 09 and EC 04.

edit: just have noticed that i actually could have used the log option *sigh*, cause CPU-Time isn't the actual time the program is running.

Spoiler

 

Witcher 3 ?? do you mean witcher 2 ??
anyway in my case is around 90c on max load, but on a normal usage it really depends on lots lots looots of things.... you can use a slighlty lower setting and your temp will be lower without losing too much on the graphic...
and another thing, maybe a lots of people will disagree, but you can also use some kind of limiter, I use Bandicam... let's be realistic, do we really need 100 or 200 fps?? or even 60 fps??

I can play with no problem at all with 40 fps, or even 30 fps without any negative effect on the fluidity, and that can reduce your temp by 8-10c

 

I always limit the FPS with Afterburner to 40 with my 6990m, and will limit to 50 when my 780m arrives. BUT, thats not the point, the truth is, there are games out there like Crysis 3, Metro even BF3 with MSAA which does need 99% usage for acceptable FPS, not to mention the ones to come in the future.
Its not about "we dont need", its about knowing that we CAN use our hardware to its limits without worrying about damaging it due to high temps or heavy usage. We want to be able to play heavy games for hours nom-stop and many like to play at 60 FPS or just to see how high FPS goes. So its really a bummer buying a expensive gamming notebook just to be limited by a silly problem like temps, which seems to be commom unfortunately.

 

P177sm bought throu origin pc Australia and repasted using phoyba nanogrease extreme and maxing 92 on stock clocks

 

I've been doing a few tests this afternoon.

Repasted with three different TIMs.

MX-4
MX-2
Gelid GC Extreme.

All the same. When running 3dmark11, it hits 93 degrees max.

Now, I live in the UK and weather has taken a dip today, so little cooler than it has been. Thats hard zero affect on the temps.

I moved next to the window, letting cool air blow onto the laptop, ran 3dmark11 again, it hits 90degrees max.

Think it's just something I'll have to live with.

 

I do see more P150 problems in this thread, but it is way to small of a sample size. Also the P150 maybe more popular.

So maybe if the vents were more open on the P150?

 

Does anyone here have their Clevos purchased from Mythlogic ?

I asked their techs and they said their 170/177SM models average about 77C on 3dmark13. I know they repaste with IC Diamond and also build theirs entirely from scratch instead of through Sager. Could this be the key element explaining the large disparity in temps ?

 

Running a bench isnt really the best way to test temps, anyone trying to do so should run at least 10min of a stress test like furmark, but indeed, sager seems to do a crap paste job. My retailer build themselves too and use AS5, will ask them to do some stress testing to see temps before sending me and hope to see good temps.

 

Agreed, stress tests are probably going to more accurate gauge. That being said, i know for fact even a single run of 3dmark11 had my temps jumping to 85C.

 

I'm in the process of buying a new heatsink from a supplier in the UK, if I can find one who stocks them, or can order them...

Hopefully they'll be pretty cheap. Sure someone said they are £10 roughly.

Do the P157SM models have the same heatsink as the P150SM? If they are different, I might buy one of each?

 

Can answer my own question.

The P157SM and P150SM have the same heatsink.

Managed to source a decently priced heatsink. Should be getting it in the next few days.

 

It appears the only differences are in the model of card, AMD have a single block while the 770 and 780 use a split block.

 

Thanks for that Meaker.

New HS will be with me tomorrow. So will be a good to see if it actually makes a difference!

 

It's something I intend to look at, as well as thermal pad thickness and potential replacement.

 

If the HS doesn't fix my issue, ill be replacing the thermal pads too.

 

I am glad at least one of the ressellers is looking into this.

 

I think I may have fixed it with a new Heatsink.

Think...

Will run some tests and report back.

Edit: Hmmm, temps are creeping back up.

 

Playing FFXIV ARR for 20mins or so, resulted in highest temps of 80degrees.

As you can see, 99% util, full stock clocks. Sitting at 80 degrees, not even with FN+1 fans either.

Looks promising.

Ran two 3dmark11 tests and highest temp was 77degrees.

I don't like Furmark and heard bad things about it, so shall try a few more games/benches etc...

Looks good.

 

So after an hour or so stressing using Unigine Valley. I achieved a max temp of 85degrees.

When I'm home tonight I'll try it in real world scenario, as I'll be on FFXIV ARR for a few hours.

I must admit it does take quite a while to hit 85 degrees too. Whereas with my old HS, it hit 93+ pretty much straight away.

Very impressed with the new Heatsink. It looks identical to my old HS, wonder why I'm getting better results?

 

It could be a variety of reasons, one of the heatpipes may have been faulty on your old HS.

 

Aye, maybe.

Feel bit more comfortable now. Will see how it goes over a period of time. Wait for the Thermal paste to set.

 

Then do we have a lot of faulty heat pipes?

Great news dquancey. This may be worthy of its own thread. So do the two parts look the same? Is the old HS dented or bent differently? Maybe there was a bad batch?

 

I'll like to test it a little more. Stress it a little, see how it goes before I start saying its "fixed".
Going to log onto ffxiv now, so that'll be a good test.

 

Did you order the heatsink directly from your reseller? The stock one comes in aluminum right?

 

I ordered from the Clevo UK supplier directly. Golden Star UK. Cost £42 delivered, so if that's what it costs to put my mind at rest, then so be it.
My original heatsink was copper and the replacement was copper too. Didn't see any differences, they both were identical. I never took pictures, but thinking back I should have done!

 

Well, if the orginal was defective, then the warranty should cover it. Might be worth a shot to see if they will reimburse you, or at least give you a store credit. I would think they would want the bad one back to see what went wrong.

 

Yeah might give it a go. Like I've said previously, I'll give it a couple of weeks before I even say it's properly fixed.

I've got an idea about the bottom plate under the laptop too. Might order one or two of those and mod them to allow more airflow. Thats for another time though.

 

If you're averaging 80-85C in that laptop that's fantastic and really eases my concerns.

 

I've limited time as I've got two small children, so can't exactly bench/stress it all day.
But...I'm hoping to do a mini review with my findings in a new thread.

 

I have yet to order my laptop. I wonder if I should order an extra HS with it?

 

I'd get them to stress it before sending. Make sure it doesn't reach 90degrees under full load.

 

Which test might they have on hand that I can ask for? (I need to check to see if the Haswell USB issue has been fixed in the shipping units yet.)

 

I'd just ask them to stress the machine. Using unigine valley for 30mins+ is a good start. See what temps hit.

Just played FFXIV for another few hours. Not a notch over 83 degrees.

 

Crisis 3 during the initial training in the simulation section is great for loading the GPU.

 

Doesn't the Nvidia GPU Boost 2.0 let you set a target temperature? Is it exceeding the target temp?

 

It kicks in at 92/93.

 

Hmm sorry, new guy here~ i'm thinking of getting a P150SM (So far my region only got this and P170SM). After so many days of research i just throw away everything i've seen on the other reviews and all focus Temperature@780M. I prefer P150SM over 170 i guess is because of stylish which 17" look a bit box to me. Read from Page 1 to Page 5, so can we conclude that if we put weather aside, the temperature of 780M full load for over 1 hours is holding below 88C?

 

Looking at the votes, 4 out of 15 are seeing 91C and above. And 6 out of 15 are seeing 90C or above. This is a small sample size, and very uncontrolled. But at least some are seeing high or max temps. It throttles at around 92-93C.

 

I must admit, I've been gaming on my 780m for the past week and never seen temps over 85 degrees.

The new Heatsink must of fixed the temps issue. Not sure if it was the heatsink or maybe even the thermal pads?

I won't be opening my P150SM again anytime soon, so can't compare to show you guys.

All I can say, is that if you are having problems with heat. Get your reseller to send you out a new heatsink, one that has not been pulled from another machine, an actual spare part. Then give it a go. I'm pretty sure that if I was to repaste again, I'd have better results, even change the stock thermal pads might work too.

 

Considering the extreme differences I would say heatpipe.

 

I have done as dquancey suggested. New HS shipped today, I installed it, and first tests suggest its 10 degrees Celsius better under load than the old one (i tried repasting that one mulitple times). With the old HS i had temps of 90-92 C and some throttling too. I can actually feel REALLY hot air coming out of the back of the notebook now as it should. Thank you dquancey for these findings. Really great
Idle temp is also about 4-5 degrees C lower (when GPU isnt switched off ofc.)

 

Shaz2bot, had you tried repasting before? If the paste was fine, then it sounds like there might be a bad batch of heat sinks out there. Not good.

 

Great stuff. Glad that I took the plunge and had a new HS shipped. Temps are far better and I now have the perfect laptop. No more worrying about temps.
Glad that its worked out for someone else too!
Must of been a bad batch of heatsinks somewhere along the line.

 

Two loops of Unigine Heaven 4.0, max temp = 84 C.
ajnindlo, I tried repasting the old HS before I got the new one maybe 5 times. I tried AS5, and Prolimatech PK-3, so im sure the paste-job was fine.
One more thing I noticed: The new HS had a strange shape, the GPU core spreader seemed to be bent off to one side, I was a little shocked when I saw it, since my old one looked MUCH better in terms of being able to sit FLAT on the GPU core. I carefully bent the new HS into a flatter position, and just installed it, surely it wasnt perfect yet.... But once I tried some benchmarks everything was fine, temperature much much better then the old one.
It is definately a new revision or refresh of heatsink. Theres a small difference that I could tell. The aluminium part of the GPU memory HS is extended to the copper fins of the HS. Theres a metal piece that connects the aluminium and copper heatpipes where they just enter the copper heatfins (looking a the thing from top between copper heatfins and HS screw number 5.
This is not present on the old version of the HS. This piece certainly doesnt look to be for thermal conduction, it looks more like a structural connection piece, maybe something that makes sure the thing survives manufacturing or whatever
Since the new HS is in the computer I am typing this on, I cant show a picture at the moment, but I can take one over the weekend if anyone wants me to, let me know