On 2001-08-07, Devon wrote:
>Mine is the same. I have:
>set temp1_over 55
>set temp1_hyst 50
Similar chipset. :)
Mine is: VIA Technologies VT 82C686 Apollo ACPI
(Driver via686a, i2c-viapro)
>Output is:
>SYS Temp: +45.8�C (limit = +50�C, hysteresis = +55�C)
>
>In this case, despite the labels, the alarm will go off at 55, and turn
>off at 50.
Yes, because it is swapped. :) See your quote at the top. I'm
confident it is to be interpreted like this:
Alarm will start the first time the temperature goes above 55 and
will stop only if and only if the temp goes below 50. Between that
the temp may oscillate arbitrarily without toggling on/off the
alarm.
That would make sense to me with what I have learned about the
meaning of "hysteresis" (first in magnetism, I think).
>Although I'd still argue that if it were changed, people would find
>fault with the following scenario:
>
>If the values are swapped, and limit is 50, hysteresis is 55....
>Once the alarm is triggered, you could have output like:
>
>SYS Temp: +52.8�C (limit = +55�C, hysteresis = +50�C) ALARM
Here I disagree. The history of the temperature level matters, not
just the current level. Your example is to be interpreted like this:
Temperature has been OVER the limit recently, but has not fallen
below the hysteresis threshold yet. Alright! Keep up the alarm
until we are safe and below 50�C again.
_______________________________________________
Seawolf-list mailing list
[EMAIL PROTECTED]
https://listman.redhat.com/mailman/listinfo/seawolf-list
