In Reply to: Re: Please retract your misteaching above posted by Dan Banquer on June 16, 2006 at 13:12:51:
"but it still needs filtering in the D/A process in order to work as intended."
Can you explain that please.If you took a sine wave of the amplitude of an LSB and quantized it with dither, you'd see the sine wave represented as the LSB randomly triggered, with higher density near the peaks, and lower density near the zero points. If you played this digitized signal back *without* filtering in the D/A process, the output would be that randomly-triggered LSB. Now if did apply filtering, convolving the randomly-triggered LSB with a "sinc" function (the most-common of the D/A filter algorithms), and calculated the results into words of greater length than the raw data (a 24-bit DAC), the result will be an approximation of the sine wave that is close to the pre-digitized waveform. The extra bits in the digital filter's output enable the signal to be approximated below the media's LSB.
This post is made possible by the generous support of people like you and our sponsors:
Follow Ups
- Re: Please retract your misteaching above - Todd Krieger 13:29:58 06/16/06 (5)
- Totally wrong, as usual - real_jj 13:52:41 06/16/06 (1)
- Hint: - real_jj 13:56:16 06/16/06 (0)
- Re: Please retract your misteaching above - Ted Smith 13:52:32 06/16/06 (2)
- Re: Please retract your misteaching above - Todd Krieger 14:02:55 06/16/06 (1)
- Wrong again... - real_jj 14:13:48 06/16/06 (0)