Re: convert digital sound (Bob Masta )

Subject: Re: convert digital sound
From:    Bob Masta  <masta(at)UMICH.EDU>
Date:    Tue, 25 Nov 2003 09:06:46 -0500

A standard PC sound card has 16-bit audio output that runs through an internal mixer. The mixer has Wave and Master attenuators which would be effectively in series for this use. The dynamic range of the mixer is usually 50 dB or more, but the exact range, number of steps, and step sizes will differ depending upon the model... there is no Windows standard for this, and in fact there is no way under Windows to get this information from the driver or card; you would need to perform a manual calibration. Depending upon your exact application, the extra "volume bits" you could get from this mixer may or may not be useful. The problem has to do with the way Windows operates with sound output devices. In general, your software fills a series of buffers that it passes to the sound card driver via the Windows waveOutWrite function. Windows notifies the program when it has finished with a buffer, so another one can be sent. But the sending of a buffer is not well correlated with sound output of any particular sample; there can be a substantial delay due to Windows and sound card overhead and internal buffering. The mixer, on the other hand, is treated as a separate device. When you send a command to change the settings, it happens "immediately" (for Windows, that's whenever it gets around to it, but usually in a msec or so). So the upshot is that it is very hard to contorl both the mixer and wave outputs together on a sample-by-sample basis. However, it is fine for changing the level of stimuli on different presentations, for example, where there is silence between presentations during which you can change the settings. If you want true 24-bit sound cards, or higher sample rates than 48000 Hz, check out the listings at <>. Hope this helps! Robert Masta dqatech(at) D A Q A R T A Data AcQuisition And Real-Time Analysis Shareware from Interstellar Research

This message came from the mail archive
maintained by:
DAn Ellis <>
Electrical Engineering Dept., Columbia University