if this only happens while you are playing games, it could be due to heat.
There is a setting in the Bios to set the "Minimum Temperature" which should be set to 65 degrees by default.
Try to change this value to 55 degrees and see how this works while playing.
All this does is to start the cooling cycle 10 degrees earlier.
The issue applies to any game or watching movies for extended time. It does have symptoms of overheating but..why? This thing was designed to attach to the back of a monitor, gaming, and 4k resolution, you would think heat would not be an issue. Also I have read numerous threads on here saying that a replacement fixed the issue or that the Samsung 950 Pro could be the issue. At this point I wish I could return it for a refund but I am a few days past that deadline, so I need to find out if I need to RMA or figure out some better cooling solution.
Get the Intel XTU to check the temperatures:
The XTU does have built-in stress test, so simply run the CPU or GPU test and keep watching the temperatures. This way you can check if the problem is related to the temperature.
Just so you know...
The KY units have a number of temperature sensors. There are the many Digital Temperature Sensors (DTS) contained within the processor (exposed as a single "package" temperature for fan speed control purposes), there is the DTS in the Platform Controller Hub (PCH, the "chipset") and there are two thermal sensors on the motherboard itself (one under each of the M.2 devices). By default, the temperature of the processor and the temperature of the thermal sensor under the primary M.2 device are used for fan speed control purposes.
If you would like to monitor these temperature, the best application to use is AIDA64 Extreme (available here: Download AIDA64). If "free" is an absolute criteria (AIDA64 Extreme costs US$40), remember that you get what you pay for. You can try Speedfan (available here: Download SpeedFan), HWiNFO32/64 (available here: Download HWiNDO32/64), HWMonitor (available here: Download HWMONITOR) or even the Open Hardware Monitor (available here: Download Open Hardware Monitor). As I said, you get what you pay for. Most of these free tools, while they understand how to access the various ICs used for environmental monitoring and fan speed control, do not understand how the various sensors are actually used on the board. This results in non-existent/unused sensors being displayed (a typical problem with temperature sensors), incorrect sensor readings being displayed (a typical problem with voltage sensors), etc.
Hope this helps,
P.S. BTW, AIDA64 Extreme has a try-before-buy option, so you can try it for 30 days - and take advantage of its stress test capability to see whether thermals are really the cause of your problems.
The NUC is well ventilated in a open area, and the room is cool. The fan is working, I did check the fan power connector. When I ran the stress test from the XTU, the CPU was around 92c, I know that is extremely hot but it did not crash, I ran a game afterwords and the CPU hovered at 90c and crashed. Thanks everyone for the suggestions, my new ram sticks will be here today ill see if that helps. I purchased Kingston Hyperx HX421S13IBK2/16.
Are you thinking maybe its the game that causes the crash? maybe its possible to find something else that runs the unit hot for a while. I am not sure but perhaps some instruction from WoW that isn't present on the Iris brand video chip could be causing such a crash..
have you tried running WoW on the lowest settings? maybe it is some shader.
**edit nvm, read one of your previous replies - seems gpu related - maybe if you have a portable deskfan or some way of cooling the unit you could see if it runs longer if you are watching movies or playing games that normally cause the crash.
Ive seen some indepth reviews where they show thermal images and, well, there is pretty much no active cooling on the ram or ssd - although I don't understand how the air vacuum is supposed to work in this case design - but i am pretty sure it was designed for it. One thing that I found interesting is that although the cpu can run at 90-100c, the ssd's are rated to function only at like 65 or 70c, i think the ram too.
Thank you very much to all the peers participating viewing this thread, your comments are suggestions are very useful in order to fix this problem.
Please let us know the results of testing the new RAM, that might be the reason why the NUC crash.
The memory for the NUC is the: DDR4-2133+ 1.2V, 1.35V SO-DIMM, as you can confirm on the following link:
Also, a BIOS update might be useful for this situation, the latest version is 0041.
Instructions to do a BIOS update using the F7 method:
And this is the .BIO file you need in order to do the update:
Also, there are some recommended settings for this game, you can see them here:
For testing purposes, please test the NUC with just one memory stick and try to reset the BIOS to defaults.
If you have the option to test the NUC with a different SSD, that will be helpful too, to rule out a possible problem with the SSD.
Besides that please run the Intel® processor diagnostics tool:
And also, since the NUC works with Intel® Iris™ Pro Graphics 580 controller, the following is the latest graphics driver for it, which is 4501:
Any questions, please let me know.
I just wanted to check if the information posted previously was useful for you and the problem was resolved or if it still persists?
Any questions, please let me know.
We will be more than glad to assist you in trying to fix this problem.
For us to be able to start an investigation on this subject, please submit your inquiry on a new thread, the reason for that is that every scenario is different, even if the same product is being used, so, for us to further assist you and also in order for us to help you with the specific problem, please submit a new thread.
Let me apologize for any inconvenience.