bioacoustics-l
[Top] [All Lists]

Re: Timing problems with USB sound cards

To: <>
Subject: Re: Timing problems with USB sound cards
From: Gianni Pavan <>
Date: Fri, 13 May 2005 11:09:15 EDT
We use the Edirol UA5 and UA25 and never had problems by using our own recording software that checks how much time the CPU takes to process each data frame read from the audio device. In any case, it is important to set to max the buffer size in the Edirol settings panel and to avoid too much CPU load.

On a PIII500 MHz laptop we reliably record 2 channels at 96 kHz sampling rate with real-time spectrographic display and two more running programs to read GPS data and distribute NMEA strings on the network. By setting large buffers, that means long latency in the display (in any case less than 1 sec), we can also run a navigation program such as OziExplorer. On faster laptops we can reliably run additional programs with no data losses.

Only very expensive acquisition boards have on-board buffers huge enough to hold data while the CPU is busy for long time. But any sound device has its own buffers and buffering schemes that should be tested in real working conditions to find the most appropriate settings. Data losses can happen with PCI boards too and the most important precaution is to keep the CPU and the disk unit as free as possible while recording (don't forget to suspend email and virus checking!).

Gianni

--------------------------------------------------------------
Gianni Pavan
Email 
Centro Interdisciplinare di Bioacustica e Ricerche Ambientali
Universita' degli Studi di Pavia
Via Taramelli 24, 27100 PAVIA, ITALIA
Tel        +39-0382-987874
Fax        +39-02-700-32921
Web       http://www.unipv.it/cibra

<Prev in Thread] Current Thread [Next in Thread>
Admin

The University of NSW School of Computer and Engineering takes no responsibility for the contents of this archive. It is purely a compilation of material sent by many people to the Bioacoustics-L mailing list. It has not been checked for accuracy nor its content verified in any way. If you wish to get material removed from the archive or have other queries about the archive e-mail Andrew Taylor at this address: andrewt@cse.unsw.EDU.AU