Phil Leigh;348711 Wrote: 
> ah...oh... (sound of penny dropping) so it must be  a pretty good
> approximation given what comes out of the DAC!

The math is not so hard to understand.  What it says is that -any-
signal can be decomposed into a sum of sin waves.  In principle, if you
had enough pure-tone signal generators, you could start them off (only
being careful they all have the right phase) and they would play a
perfect rendition of Beethoven's 5th for you.  So anyway, a square wave
can be represented by a sum of sin waves - but the trick is you need an
infinite sum of them going up to infinite frequency.  If you only have
a finite number available, you end up rounding off the edges and/or
introducing Gibbs overshoot.

When you digitize something you band-limit it to frequencies below
Nyquist (otherwise you get lots of nasty aliasing artifacts).  So that
means you only have a finite range of tones available, and therefore
you get plots like that.  A 1kHz square wave can be represented
reasonably well.  Here's a picture:


+-------------------------------------------------------------------+
|Filename: square.jpg                                               |
|Download: http://forums.slimdevices.com/attachment.php?attachmentid=6009|
+-------------------------------------------------------------------+

-- 
opaqueice
------------------------------------------------------------------------
opaqueice's Profile: http://forums.slimdevices.com/member.php?userid=4234
View this thread: http://forums.slimdevices.com/showthread.php?t=53345

_______________________________________________
audiophiles mailing list
[email protected]
http://lists.slimdevices.com/lists/listinfo/audiophiles

Reply via email to