I’m having problems too, if it’s any consolation.
I agree, an extract from someone that isn’t having problems would be interesting.
Alright, I have to revive this thread as I’m still not able to play 60fps UHD/HEVC content without overheating. This time I, again, experienced temperatures above the shutdown threshold of 116C (as I did earlier in the thread). This was seen both the CLI and the GUI:
This worries me a lot as I’m afraid of breaking the device.
Please post some new logs
Your memory usage seems high: was the system at rest there?
Sam
I had just finished scraping a bunch, that might be it.
Here are the logs when this happened. After finishing scraping, I tried playing the 400mbps and 200mbps Jellyfish video and after that Gemini Man. After I realized the temperatures were going through the roof, I stopped it and it started cooling down and after a minute or two, it shutdown with the red light on (but much later than it should have).
Logs (not debug logs as I hadn’t enabled that at the time I experienced the high temps): https://paste.osmc.tv/sopoxuxika
Are you disabling the thermal limiter or adjusting the trip points ? Otherwise there is no way it would reach 130C without shutting down…
I haven’t touched anything, I don’t even know what/where they are.
your cli screenshot shows the thermal limiter settings in /sys
What is the temperature at rest after 10 minutes on the home screen?
Don’t leave it on System Info as that will generate unnecessary heat.
All I did, was cat /sys/class/thermal/thermal_zone0/temp to get the temperatures.
Has been idling on home screen for 15 minutes now and it’s at 75C.
can you run the following command and paste the output ?
for i in /sys/class/thermal/thermal_zone0/*; do echo "$i: $(cat $i 2>/dev/null)"; done
probably easier to do that using SSH instead of taking a photo of the TV. 
osmc@osmc:~$ for i in /sys/class/thermal/thermal_zone0/*; do echo “$i: $(cat $i 2>/dev/null)”; done
/sys/class/thermal/thermal_zone0/cdev0:
/sys/class/thermal/thermal_zone0/cdev0_trip_point: 1
/sys/class/thermal/thermal_zone0/cdev0_weight: 1024
/sys/class/thermal/thermal_zone0/cdev1:
/sys/class/thermal/thermal_zone0/cdev1_trip_point: 1
/sys/class/thermal/thermal_zone0/cdev1_weight: 1024
/sys/class/thermal/thermal_zone0/integral_cutoff: 0
/sys/class/thermal/thermal_zone0/k_d: 0
/sys/class/thermal/thermal_zone0/k_i: 10
/sys/class/thermal/thermal_zone0/k_po: 220
/sys/class/thermal/thermal_zone0/k_pu: 440
/sys/class/thermal/thermal_zone0/mode: enabled
/sys/class/thermal/thermal_zone0/offset: 0
/sys/class/thermal/thermal_zone0/policy: power_allocator
/sys/class/thermal/thermal_zone0/power:
/sys/class/thermal/thermal_zone0/slope: 1
/sys/class/thermal/thermal_zone0/subsystem:
/sys/class/thermal/thermal_zone0/sustainable_power: 2150
/sys/class/thermal/thermal_zone0/temp: 85000
/sys/class/thermal/thermal_zone0/trip_point_0_hyst: 1000
/sys/class/thermal/thermal_zone0/trip_point_0_temp: 95000
/sys/class/thermal/thermal_zone0/trip_point_0_type: passive
/sys/class/thermal/thermal_zone0/trip_point_1_hyst: 1000
/sys/class/thermal/thermal_zone0/trip_point_1_temp: 105000
/sys/class/thermal/thermal_zone0/trip_point_1_type: passive
/sys/class/thermal/thermal_zone0/trip_point_2_hyst: 5000
/sys/class/thermal/thermal_zone0/trip_point_2_temp: 114000
/sys/class/thermal/thermal_zone0/trip_point_2_type: hot
/sys/class/thermal/thermal_zone0/trip_point_3_hyst: 1000
/sys/class/thermal/thermal_zone0/trip_point_3_temp: 118000
/sys/class/thermal/thermal_zone0/trip_point_3_type: critical
/sys/class/thermal/thermal_zone0/type: soc_thermal
/sys/class/thermal/thermal_zone0/uevent:
Everything looks fine there. CPU speed should start to throttle at around 105C and maintain the temperature between 105C-110C even under the heaviest load, assuming that the device has sufficient ventilation. If it doesn’t it would get even hotter and at 118C it should shut down automatically.
I have to wonder if the temperature reported in Kodi is misleading. Try leaving the following running in an SSH session while you try to reproduce excessive temperature to monitor the temperature in realtime:
watch cat /sys/class/thermal/thermal_zone0/temp
It should shut down before it gets to 120C. (And shouldn’t get there in the first place if there is enough ventilation)
I can’t reproduce the issue now with the temperatures going above the 118C. But it has happened to me twice now - today and earlier as indicated by previous posts in this thread. Just thought I’d make you guys aware that I wasn’t making things up the last time…
However, you say that temperatures shouldn’t go near 120C if there’s sufficient ventilation. That is exactly my issue; temperatures go this high when playing 60fps UHD/HEVC content. I just recreated a crash by setting the skin to the default OSMC skin, leave it idle in the main screen for about 15 minutes (idle temperature was 75C) and then play a 60fps UHD/HEVC file until crash. This is with the Vero box sitting like this: Image
Logs: https://paste.osmc.tv/iyonisipol
Temperatures: http://paste.osmc.tv/fowixizequ.css
Ok, so I updated to the 4.9 kernel which definitely helped the issue. Playing the same 60fps UHD/HEVC video now yields temperatures that are consistently between 109C and 113C. So that’s good!
However, I’m getting a shit load of skipped frames to the extent that it’s clearly visible upon playback. I guess it’s the high temperature that’s causing those.
I have checked your logs.
Do you see Wireless under My OSMC -> Network?
Sam
I see it. It says ‘Status: Disabled’.
Are you able to enable it and see WiFi networks?
Yes, no issues with that.
What kind of CPU usage should I expect during playback? kodi.bin sits between 70% and 250% when checking via top.