In Reply to: Re: Another Sample Of An Audio Component That Doesn't Measure Well, But Sounds Great. posted by andy19191 on November 7, 2006 at 13:47:52:
Hi AndyThe oft quoted Nyquist theory says that the sample rate must be at least two times the highest frequency.
Now, if one actually samples a sine wave at 2X F, one finds that depending where in the wave one samples, one gets a totally different result.
For example, imagine what you get as a sample if you sample at the peak amplitude OR at zero crossing.
The later sampling gives Zero signal.
Also, if one has a non constant amplitude (such as in music) then one has a wider bandwidth than the highest “Sine wave†frequency present.
Both issues push to have more samples per period.In the area of test equipment and not audio, these issues are real concerns as “accurate†measurements are the goal and as such, it is normal to have a sample rate much higher than the highest F to be measured.
As some signals are not sine waves, it is desirable to have the rise time of the sampling system to be at least 1/3 to 1/5 the rise time of the sampled signal.
Some stuff to read if your interested.http://www.ni.com/swf/presentation/us/sampling/
http://zone.ni.com/devzone/cda/tut/p/id/2709
Best,
Tom Danley
This post is made possible by the generous support of people like you and our sponsors:
Follow Ups
- Re: Another Sample Of An Audio Component That Doesn't Measure Well, But Sounds Great. - tomservo 07:28:32 11/08/06 (3)
- Re: Another Sample Of An Audio Component That Doesn't Measure Well, But Sounds Great. - andy19191 14:30:44 11/08/06 (1)
- Re: Another Sample Of An Audio Component That Doesn't Measure Well, But Sounds Great. - tomservo 14:23:00 11/09/06 (0)
- The Nyquist criterion was developed for telephony. 'Nuff said. nt - clarkjohnsen 09:30:48 11/08/06 (0)