On Mon, Nov 28, 2016 at 19:05 +0100, Gerhard Sittig wrote:
>
> On Mon, Nov 28, 2016 at 14:50 +0100, Matthieu Guillaumin wrote:
> >
> > - The memory is a sequence of recordings that start with a header
> > (containing some flags and a timestamp). This timestamp is obviously a
> > very important piece of information about past recorded measurements,
> > but I haven't found any way to provide them to the sr_session_send()
> > function. Right now I can only print the timestamps on stderr using
> > something like sr_info(), so it's pretty lame, since stderr and stdout
> > need not be synced. Do you have any suggestion on how to handle this
> > before I submit a patch?
>
> As I understand it, sigrok uses a stream of sample data which is
> said to have a specific samplerate. Times are the consequence of
> watching a number of samples pass by at their specified rate. If
> you have a timestamp in your input from the device, you may have
> to provide "padding" samples until the specified time is due.
>
> For example, the Asix Sigma logic analyzer repeats the last
> provided sample in that case (see the sigma_decode_dram_cluster()
> routine).
I was made aware in a private message that the padding approach
which a logic analyzer uses when its input stream is compressed
will not lend itself very well to the situation where a sound
meter is used. So I'd like you to forget I said something. :)
Apparently logic analyzers with their rather constant sample rate
and continuous stream of samples are different from multimeters
or sound meters which optionally apply timestamp to their
recorded measurements which might have been taken at arbitrary
points in time.
virtually yours
Gerhard Sittig
--
If you don't understand or are scared by any of the above
ask your parents or an adult to help you.
------------------------------------------------------------------------------
_______________________________________________
sigrok-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/sigrok-devel
