If one runs 'sensors' what are reasonable (i.e. reasonably safe) temperatures to see?
On 14 February 2015 at 22:56, Chris Green cl@isbd.net wrote:
If one runs 'sensors' what are reasonable (i.e. reasonably safe) temperatures to see?
Doesn't the output of sensors tell you what the upper thresholds are?
eg. CPU Temperature: +36.0°C (high = +60.0°C, crit = +95.0°C)
So you'd think anything below 'high' is fine.
- srdjan
On 14/02/15 23:05, Srdjan Todorovic wrote:
On 14 February 2015 at 22:56, Chris Green cl@isbd.net wrote:
If one runs 'sensors' what are reasonable (i.e. reasonably safe) temperatures to see?
Doesn't the output of sensors tell you what the upper thresholds are?
eg. CPU Temperature: +36.0°C (high = +60.0°C, crit = +95.0°C)
So you'd think anything below 'high' is fine.
- srdjan
main@lists.alug.org.uk http://www.alug.org.uk/ http://lists.alug.org.uk/mailman/listinfo/main Unsubscribe? See message headers or the web site above!
+1 Sensors should give the information relevant to acceptable temperatures.
As an aside, if you find yourself paranoid about temperatures (I live in permanent fear of my laptop catching fire for example) then you can run a bash script against cron every minute. The feedback can take the form of audible output (as mine does) or by emailing an alert etc...
The script in my cron is below - your mileage may vary (sensors for you might show high AND critical values, for example so you'll need to exclude more results from the regex. Mine only shows a high value of 70, so I use this manually in greps and evaluations. A little modification would sort you out if your system was vastly different from mine.
Apologies to all the real bash coders for the hack-y nature of this.
#!/bin/bash # check_temp
# regex matches numerical values of 2 or more numbers that are followed by # decimal point, then removes known values that will turn up
SensorTemp=$(sensors | grep -oP '[0-9]{2,}(?=.)' | grep -v 70)
if [ "$SensorTemp" -gt 70 ]; then `echo "Danger, danger Will Robinson" | espeak` fi
Cheers
On Sat, Feb 14, 2015 at 11:05:08PM +0000, Srdjan Todorovic wrote:
On 14 February 2015 at 22:56, Chris Green cl@isbd.net wrote:
If one runs 'sensors' what are reasonable (i.e. reasonably safe) temperatures to see?
Doesn't the output of sensors tell you what the upper thresholds are?
eg. CPU Temperature: +36.0°C (high = +60.0°C, crit = +95.0°C)
So you'd think anything below 'high' is fine.
Yes, but surely there are 'normal' running temperatures which are well below the 'high' ones.
On Sun, Feb 15, 2015 at 09:40:35AM +0000, Chris Green wrote:
On Sat, Feb 14, 2015 at 11:05:08PM +0000, Srdjan Todorovic wrote:
On 14 February 2015 at 22:56, Chris Green cl@isbd.net wrote:
If one runs 'sensors' what are reasonable (i.e. reasonably safe) temperatures to see?
Doesn't the output of sensors tell you what the upper thresholds are?
eg. CPU Temperature: +36.0°C (high = +60.0°C, crit = +95.0°C)
So you'd think anything below 'high' is fine.
Yes, but surely there are 'normal' running temperatures which are well below the 'high' ones.
Not to mention that running continuously near the 'high' value will surely not improve the life of the electronics.
On 15/02/15 09:42, Chris Green wrote:
Not to mention that running continuously near the 'high' value will surely not improve the life of the electronics.
Well no it won't improve it, but running way below high most likely won't either.
The spec sheets usually say that running anywhere between the normal and high temp will have no affect on operating life, running between high and critical may shorten life (or give you secondary problems with BGA mounts etc) and running above critical may result in immediate failure.
Also what is high and critical varies depending on the chip, I think it is held within the CPU so sensors/acpi etc can read it.
On Sun, Feb 15, 2015 at 03:23:48PM +0000, Wayne Stallwood wrote:
On 15/02/15 09:42, Chris Green wrote:
Not to mention that running continuously near the 'high' value will surely not improve the life of the electronics.
Well no it won't improve it, but running way below high most likely won't either.
The spec sheets usually say that running anywhere between the normal and high temp will have no affect on operating life, running between high and
Yes, exactly! So what is "the normal"? :-) That's what I'm asking.
Normal is within safe limits. At the moment the desktop I an using has an AMD dual core running at 24c increasing to 28c under load. The GPU is running at 50c. My media server has an Intel Core 2 duo which is bouncing around from 34c to 40c. My laptop cooling fan doesn't kick in until the CPU temperature reaches 50c. To me they are all normal. If the temp goes over 60c continually I would start to get a bit concerned and check to see the cooling fins aren't clogged up with dust/fluff and the fans were working properly.
http://www.buildcomputers.net/cpu-temperature.html saves me having to type the factors that affect temperature and the different temperatures to expect.
On 15/02/15 16:27, Chris Green wrote:
On Sun, Feb 15, 2015 at 03:23:48PM +0000, Wayne Stallwood wrote:
On 15/02/15 09:42, Chris Green wrote:
Not to mention that running continuously near the 'high' value will surely not improve the life of the electronics.
Well no it won't improve it, but running way below high most likely won't either.
The spec sheets usually say that running anywhere between the normal and high temp will have no affect on operating life, running between high and
Yes, exactly! So what is "the normal"? :-) That's what I'm asking.