Re: [Audacity-devel] Further problem, obvious in Normalize

[Steve t. F. <ste...@gm...>]

On 2 May 2012 01:38, Martyn Shaw <mar...@gm...> wrote:
> Hi
>
> If you have an offset signal, such as created by Steve's Nyquist
> command (sum 0.4 (mult 0.25 (osc 60))) applied to a 30s track
> and then split the resulting signal somewhere, and move the right-hand
> bit further right to make the track about twice as long, Normalize
> (with the patch on Bugzilla 495, and probably without that) will
> calculate the 'space' as having samples of value zero.  So when it
> calculates the 'average' (dc) it will only be half as much as it
> 'should' be, and so not apply the 'correct' offset.
>
> This is not related to bug 495, I just happened across it whilst
> testing on that one, and thought I should record it.
>
> Of course the chances of anyone (using a sensible work-flow) finding
> this problem are slim, but there may be people splitting and shifting
> tracks and then deciding to get rid of the dc.  I think it's a
> WONTFIX, what about you?

The problem goes back to at least 1.3.4, and has probably always done this.
A "sensible" work flow would be to remove DC (if present) before doing
anything else.

I would (personally) prefer it to be a very low P rating.
I don't know how other effects "avoid" white space, but if white space
was "skipped" from the DC calculation it would probably give a speed
improvement as well as doing "the right thing", so I think it would be
better to leave this bug open rather than WONTFIX, but very low
priority.

Steve


>
> TTFN
> Martyn
>
> ------------------------------------------------------------------------------
> Live Security Virtual Conference
> Exclusive live event will cover all the ways today's security and
> threat landscape has changed and how IT managers can respond. Discussions
> will include endpoint security, mobile security and the latest in malware
> threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
> _______________________________________________
> audacity-devel mailing list
> aud...@li...
> https://lists.sourceforge.net/lists/listinfo/audacity-devel