Re: [PATCH 1/2] iio: add INT_TIME (integration time) channel info attribute

[Jonathan Cameron <jic23@kernel.org>]

On 08/18/13 22:30, Peter Meerwald wrote:
> Hello,
> 
>>> integration time is in seconds; it controls the measurement
>>> time and influences the gain of a sensor
>>>
>>> used by adjd_s311, tsl4531
>>> the following drivers have similar controls:
>>> * tsl2563 (integration time is controlled via CALIBSCALE among other things)
>>> * tsl2583 (has integration_time device_attr, but driver doesn't use channels yet)
>>> * tsl2x7x (has integration_time attr)
> 
>> I know this has come up before (and I'm not actually expressing oposition
>> to this going into info_mask) but could you sumarise the arguments for why
>> this property cannot be covered by calibscale or scale?
>> I'd like to have a little stronger argument for the patch description in Git.
> 
> sounds good; I'd add:
> 
>> There are two typical ways that scaling is implemented in a device:
>> 1) input amplifier
>> 2) reference to the ADC is changed.
>> These both result in the accuracy of the ADC varying (by applying it's
>> sampling over ta more relevant range).
>>
>> Integration time is a way of dealing with noise inherent in the analog
>> sensor itself.  In this case a mixture of Photon noise and device specific
>> noise.  Photon noise is dealt with by either improving the efficiency of
>> the sensor, (more photons actually captured) which is not easily varied dynamically,
>> or by integrating the measurement over a longer time period.  Note that this
>> can also be thought of as an averaging of a number of individual samples
>> and is infact sometimes implemented this way.
> 
> altering integration time implies that the duration of a 
> measurement changes and the user of the device may be interested
>  
>> Hence it makes sense to distinguish between integration time and simple
>> scale. In some devices both types of control are present and whilst they
>> will have similar effects on the amplitude of the reading, their effect
>> on the noise on the measurements will differ considerably.
> 
> integration time may or may not have an impact on the range of values
> measured
> 
> both characteristics cannot be expressed with scale/calibscale in a 
> satisfying way
> 
> let's see if further comments come up, and I'll repost
Not much come out of the wood work so something like what you propose will do nicely!

I would also like a short description in the documentation file so we don't have
to keep refering people back to the original patch.

Jon, you still out there?  Not heard anything from you for a while and your input
on this sort of thing is always valued!

Jonathan
> 
> please note that there is a fix for the adjd_s311 driver posted which you 
> may want to fast-feed to Greg
> 
> thanks, regards, p.
>