[moving discussion back to the list] The higher sample rate would change the way the blocks are serviced by the scheduler. My guess is that the "no input data" time (0.25 sec) gets hit at the lower rate and not at the higher rate. This could change the scheduling order and cause different behavior when the flowgraph stops.
There is a set_blkd_input_timer_value() call that can be set on a per-block basis if you really feel like investigating. Setting the value to something lower (100) or higher (1000) than the default (250) could change the behavior. This is getting into the weeks, so maybe running at a higher sample rate is the easiest thing, if it's reliable enough for you. On Tue, Dec 26, 2023 at 2:23 PM Elmore's <wa4...@comcast.net> wrote: > Jeff, > > I forgot the attachment. here it is. > > Jim > > *From:* Jeff Long > *Sent:* Monday, November 13, 2023 7:12 PM > *To:* discuss-gnuradio@gnu.org > *Subject:* Re: Wavefile Source does not stream its audio to the end of > the file. > > Tried this, and 256 is too small. Something like 4096 seems to work on the > wav file source. And setting the resampler max output buffer to 1024 works. > Together, these may improve the situation. There is no absolute fix, > though. GR terminates the flowgraph when any block reports it is done. > > On Mon, Nov 13, 2023 at 2:13 PM Elmore Family <wa4...@comcast.net> wrote: > >> Jeff, >> >> I guess you saw my reply to your suggestion. >> >> Is there any way around this issue? There must be a way to use a wave >> file source of this length. >> >> Jim >> >> *From:* Jeff Long >> *Sent:* Friday, November 10, 2023 6:04 AM >> *To:* discuss-gnuradio@gnu.org >> *Subject:* Re: Wavefile Source does not stream its audio to the end of >> the file. >> >> This is due to the way GNU Radio decides when a flowgraph is done. Since >> the audio rate is so low, most of the wavfile fits in the block buffer, the >> block sees that there is no more data to load and declares itself done. >> >> There are buffer size controls that might help. On the advanced tab for >> the wavfile source, try setting the MAX output buffer to something like 256 >> and see if that helps. >> >> On Thu, Nov 9, 2023 at 9:24 PM Elmore's <wa4...@comcast.net> wrote: >> >>> I am attempting to send the output of a Wavefile Source to an Audio Sink >>> as a test for inclusion in a larger project. >>> >>> The flowgraph consists only of the Source block ( 12000 samples/sec), a >>> Rational Re-sampler with interpolation of 4 and the Sink (48000 >>> samples/sec). The wavefile is 15 seconds long and plays properly using >>> SMPlayer. >>> >>> When I run the flowgraph there is only a short burst of audio (measured >>> in ms) and the flowgraph terminates. >>> >>> What is wrong here? I expected the flowgraph to run for 15 secs. >>> >>> Jim >>> >>> >>> <http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=emailclient> >>> Virus-free.www.avg.com >>> <http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=emailclient> >>> >>
