Re: audio-video asynchrony with a USB soundcard (Sylvain CLEMENT )


Subject: Re: audio-video asynchrony with a USB soundcard
From:    Sylvain CLEMENT  <sylvain.clement@xxxxxxxx>
Date:    Fri, 5 Dec 2008 16:45:39 +0100
List-Archive:<http://lists.mcgill.ca/scripts/wa.exe?LIST=AUDITORY>

Dear Massimo, you didn't tell what kind of audio driver you are using. I assume that all of your tests are done on a windows platform (correct me if I'm wrong). One important parameter is the kind of audio driver you use. To increase accuracy of latency, you should use the ASIO Driver on which you'll be able to set the buffers size which determine the audio latency : a buffer size of 128 samples with a sampling rate of 44.1kHz leads to a 2.9 ms latency. The results should be consistent with asio driver, the only uncertainty being the time between you display & soundplay function calls. The standard windows driver (WME) is quite unaccurate in terms of latency. Hope this helps Sylvain Clément Le 5 déc. 08 à 14:03, Massimo Grassi a écrit : > Dear list members, > > I measured yesterday the audio-video asynchrony with an USB sound > card (M-AUDIO Fast Track Pro). > > In practice, I was running a visual display and, in correspondence of > one particular video-target-frame, I was switching on a 1-cycle sine > wave (1000 Hz). I checked whether the video-frame and the audio > signal were synchronous. > > In general, the video preceded always the audio (although, if any, I > had to expect the opposite result because in the program the > soundplay command preceded the play the video-target-frame command). > The average video-anticipation was of 6.3-ms (SD=3.0, maximal video > anticipation 12-ms, min=2-ms). I forgot: the video was running at > 100-Hz refresh rate. > > To summarise: > 1) audio and video signals are often asynchronous; > 2) the asynchrony (unfortunately) does not have a fixed value but > changes within a certain range. > > It is not as bad as I was expecting but with a PCI sound card (a > very cheap one) and identical set-up I observed an average > asynchrony of 2-ms (SD<4-ms). > > All the best, > m -------------------------------------- Sylvain CLEMENT JE 2497 "Neuropsychologie et Cognition Auditive" Université de Lille 3,UFR de Psychologie BP 60149 59653 Villeneuve d’Ascq Cedex France Phone: +33 3 20 41 64 42 Fax: +33 3 20 41 63 24 Email: sylvain.clement@xxxxxxxx www: http://nca.recherche.univ-lille3.fr


This message came from the mail archive
http://www.auditory.org/postings/2008/
maintained by:
DAn Ellis <dpwe@ee.columbia.edu>
Electrical Engineering Dept., Columbia University