Re: Iterating through audio data
Re: Iterating through audio data
- Subject: Re: Iterating through audio data
- From: Paul Davis <email@hidden>
- Date: Thu, 26 May 2011 19:36:14 -0400
On Thu, May 26, 2011 at 7:04 PM, Brian Willoughby <email@hidden> wrote:
[ ... ]
of course, the slightly contrary view is that presenting time-domain
representations of audio data doesn't tend to convey a lot of useful
information except for min/max values, and that editing based on
visual assessments of a time-domain representation is sloppy, lazy and
generally a bad idea.
this isn't a reason to do a bad job with waveform drawing, but it can
be useful counter to spending too much time on something that actually
isn't really that useful to the people who will be viewing it, even if
they mistakenly believe that it is.
the hardest problems with waveform drawing, as far as i can tell, come
from scaling up to the challenges faced when drawing representations
of *huge* amounts of data on-screen rapidly and smoothly. when the
user zooms out to show 12 hours of 100 tracks of audio (something some
of our users do on a regular basis), they are showing a representation
of a collossal amount of information. the drawing scheme has to work
as well for that case as it does when zoomed in to show 1 pixel per
sample value. this isn't computer science, per se (although
bresenham's algorithm is a useful thing to be aware of), but it can be
hard to implement correctly and efficiently.
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Coreaudio-api mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden