04-18-2016 01:38 PM
I have this laptop for more than 4 months now and I cannot play on it. I am using it on two external monitors. One on HDMI and one on a USB3 display link (also hdmi). (Perfectly works when im in work using 3D Studio Max, Photoshop, Premiere pro, etc. but useless with games)
I am able to play with any games for like 5-10 minutes and then the Nvidia driver randomly crashes with various error messages:
Images of various error messages
I have tried to solve it for many many hours but I am out of ideas. Here is what I did:
- Download and install newest and some older Nvidia drivers (incluiding and excluding Nvidia Experience)
- Upgrade Bios to 35WW
- Set everything to High performance NVIDIA processor in Nvidia control panel.
- Set always Nvidia in physx
- Tried to download and install the newest (including betas) intel GPU drivers and some older.
- Set everything to performance in Intel® HD Graphics Control Panel
- Not using any external monitors (I have crashes that way as well)
I don't know where to go now... It seems that it is a driver problem, or anything like that but I don't think that it would be a hardware problem because when it is runnning it is smooth and good performance. The temperatures are totally ok, max 70-75 celsius degrees. I cannot see throttling at all.
Is anyone out there with similar problem?
The games I tried to run:
- Star Wars: Battlefront 3
- Battlefield 4
- World of Tanks
- Alien Isolation
I would like (very very want to!) to play on this machine but currently it makes me angry with these 5-10 minutes crashes.
Please help if someone has an idea or solution.
04-18-2016 01:53 PM
Could you try running Furmark, and let me know if it crashes?
04-18-2016 02:30 PM
Thank you for your answer! I did it.
First I have started the stress test on 1920x1080 and waiting but no crashing. I decided to run prime95 simultainously to stress it harder and yes it has been crashed in very short time. I've received the following errors:
Nvidia crashes and graphs
If I run only furmark it is not crashing. If I run both CPU and GPU stress it is starting to crashing nearly immediately.
With these high loads the CPU starting to throttle but I cannot see that the Nvidia has started to.
04-18-2016 03:35 PM
Ok, some further tests and a thing maybe...
So I made a lot of stress to both the cpu and gpu and I tried to monitor it with multiple graphs. I can say that every time (10+ tries) the gpu reached the 80 celsius degree and the nvidia driver has been crashed.
Is this GPU has an 80° limit and then disconnects? Is could be why the driver crashes?
04-19-2016 03:34 AM - edited 04-19-2016 03:37 AM
I've tried to get some information from other sources and found an app called Nvidia Inspector. Here as I understand the max GPU temperature for throttling is 92°.
So I don't understand how is it possible that my laptop crashes it's Nvidia driver always exactly at 80°. It is burned into else where.
Anyone has this problem besides me? I cannot beleive that only I am who faces it. If yes it could be a hardware problem...
04-19-2016 06:10 AM
If you are hitting 92 degrees, then there's no issue with the GPU itself. The driver crashing may actually be a result of bad memory (there's still some shared RAM services which will crash the driver if you run out of RAM).
Could you try running a memtest86 test to see if your RAM is healthy?
04-19-2016 06:51 AM
I am unable to reach 92° the driver stops working when the GPU reaches 80°.
I am going to test the memory now, but I don't think it would be the problem. It is clearly temperature related.
04-19-2016 12:54 PM
Just wanted to cover the memory. Also, have you run the Hardware Scan in the Lenovo Companion App? It may help identify if there is a hardware issue.
04-20-2016 05:55 AM - edited 04-20-2016 06:32 AM
I did memtest for a night long running but it returned no errors.
I did all the quick tests and then a comprehensive in Lenovo Companion (Solution Center) but everything is alright.
I have removed the back cover and cleaned the fans and radiators with high pressure air (Cleaned the dust filter as well).
I have tried to lift the laptop upper from the table with a cooler pad, I could save some time but 80° reached again.
I have tried to turn off turbo boost on CPU to save heat, but not helping. A bit later but the GPU reaches the 80° and driver crashes as before.
My room temperature is arround 21-23° so I am out of ideas. I don't want to sit in colder environment because of the laptop.
For this model what is the usual working temperature for the GPU when gaming?
I don't think that the cooling paste / contact is wrong between the GPU and the heat pipes because the temperature goes sligthly slow to arround 80° (It takes arround 5-10-20 minutes in game. Or 5 minutes when benchmarking with Furmark and Prime95).
Because of the CPU's higher working temperature which is arround 90° (On benchmark) and based on they are on one heating system, I think that is impossible to stay lower with GPU temperature because the CPU heats it (Or am I wrong?).
What could be the source of the problem?
04-20-2016 01:37 PM
I'm currently about 36,000ft in the air, so setting up the Y700 is difficult right now, to say the least.
I'll try running both apps tonight simultaneously to see if I can kill the driver.