Showing results for 
Search instead for 
Did you mean: 

G751JL - OVERHEATING :( help

Level 7
So I noticed my laptop started overheating to 87ºC as of late and then it would throttle. So I sent it to the shop they ran some tests on a different HDD and said the drivers are what is causing the issue. They said I should install asus gpu drivers instead of the geforce website ones. So what should I do?
I'm running W10.

Level 7
what is 87c? the CPU or the GPU?
Laptop: G751JL
Processor: i7-4750HQ
Graphics Card: GTX 965M
Memory: 8 GB Samsung (stock)
Storage #1: Samsung 850 EVO 250gb
Storage #2: 1 TB 7200 rpm (stock)
OS: Windows 10 x64

Level 13
You should call that shop and ask them for proof and a copy of any tests they have to support this claim. You cant just throw another drive in a machine and test it. This is specifically why when you send your machine in for service to ASUS you MUST include the HDD with the OS installed and precisely why the first thing I do when I get a new machine is to clone the original drive then put away the original for recovery and service needs.
If it was drivers why would they not remedy that while they had it and return a fully functioning machine to you? Its not like you have to pay to download ASUS drivers.
My honest opinion, That shop is full of decayed fecal matter.
My guess is you either need a good cleaning or repaste, maybe even new fans.

“Two things are infinite: the universe and human stupidity, I'm not sure about the former” ~ Albert Einstein

My bad. Sorry for the confusion. I was on my phone and in a hurry. What I meant to say is that I sent it in for service to an ASUS Authorized Repair Service Center and I did include my original HDD with the OS installed (but I had my password on--I forgot to take it down). He said he could not log in because he did not know the password and at that point I offered it. However, he proceeded to say how he had already performed stress and diagnostic tests on Windows 8.1, so I presume he used another HDD. He said he thought it was the drivers. He told me to switch back to Windows 8.1 (which I am not going to do as this laptop has all the necessary W10 drivers on the ASUS support page) and that the generic NVIDIA driver is probably causing the GPU to overheat, which is strange I've been running W10 for a couple of months with NVIDIA drivers without any issues. Also, this centre doesn't do any software-related repairs (according to him, even though my W10 is genuine). I did ask him to open it up and clean the fans and any dust that had accumulated--and he did. So now I'll clean install W10 in UEFI mode and install all the latest drivers off the G751JL support page. However, I still can't make up my mind about whether I should install ASUS GPU drivers or NVIDIA drivers (if NVIDIA which is the latest stable build). What do you think I should do?

Level 13
Give me a few minutes and Ill post up what I get on an identical laptop running Nvidia 364.72 drivers after a nice stress.

OK just dogged G751JL.
Win10, upgraded version from Win 8.1
Nvidia 364.72 drivers on 512GB Samsung 850 pro SSD
32GB Gskill RAM at 2133Mhz.

When I say dogged I ran Heaven benchmark 4 times at ultra setting followed by 4 rounds of Firestrike extreme with no breaks in between. The scores were a joke, I knew they would be but the point was to stress it and see how hot it got with completely unreasonably high setting for this machine.

CPUZ HW monitor showed a max CPU core temp of 79C and the Max GPU temp of 62C ambient temp in the room is about 23C.

“Two things are infinite: the universe and human stupidity, I'm not sure about the former” ~ Albert Einstein

Thanks for the help. 🙂 I was getting approx. 90C CPU and up to 87C on the GPU at which point it would start to throttle and that's just in WoW...I used to play Witcher 3 OC for hours on end without ever going above 75C on the GPU. He said the GPU temps didn't go above 80C in his stress tests and that he was getting around 25 FPS (in 3DMARK or something--couldn't hear). So I guess he was also running it on high/extreme. Your advice?

Level 13
My thoughts are its hardware related and I don't get them telling you it was driver related. I just booted mine (second machine I use mostly for travel, primary is a G752VY) and dogged it hard with Win 10 upgraded from 8.1 and the Nvidia drivers they said were to blame. Adding more time would have not made it get any hotter.
Im thinking either the paste between the heatsink and CPU/GPU has dried out, which can and does happen or your fans aren't doing their job. I couldnt get anywhere near 80C on the GPU, it actually took awhile to peak to 62C. At first it was hovering around 58-59C then slowly crept up from there.

25FPS is about what it gets in the standard Firestrike benchmark with standard settings.

“Two things are infinite: the universe and human stupidity, I'm not sure about the former” ~ Albert Einstein

Level 7 even his 80C is too high, not high enough for throttle, nonetheless high. I also remember GPU temps plateauing around 65C, same as you, during the first couple of months. He only cleaned it after the stress tests, so still hoping that did the trick...

Level 13
Download and install CPUZ HW monitor and leave it run while you run 3Dmarks firestrike. I will give you a min and max

CPUZ Download

3dMarks download

“Two things are infinite: the universe and human stupidity, I'm not sure about the former” ~ Albert Einstein

Level 7
jeesus, all test are performed on different area with different air flow, and room temperature. why do you guys compare and confuse each others?
The result will 100% different dude. For GPU test, use furmark, easiest way to tell if the GPU temperature throttling. CPU Test using Realtemp.
No need such 3dmark or anything that need 600mbs or more for test for temperature throttling.

If you want to compare, please provide info :
1. Room temperature
2. Airflow condition ( don't tell if you put your laptop on the bed and run stress test )
3. CPU / GPU process rate, either consistence 99 /100% or 3dmark which not always 99 /100% usage.

Using furmark and realtemp will Consistence burn your cpu and GPU at 100% . Dont use too frequent and too long in period.
If not happy with the resuilt, remove stock thermal paste, use a coollab metal paste or any high end thermal paste. Improve room temperature will help decrease 10'C and more.