On Mar 26, 2007, at 9:21 AM, David Fleming wrote:

>   I'm working on an audio spectrum analyzer and it sure would be  
> nice to have QT do the FFT math. There may be a way to use QT with  
> live data (from the mic or audio input) by setting up a "movie"  
> audio track in memory, but I can't seem to figure out how to do it.

You can read QuickTime data from a mic/input, but you don't want to  
use a movie audio track for that. Instead, you should use the capture  
APIs in QuickTime, set everything up like you're going to do an audio  
recording, and then there is a flag you can set to make it not  
record. Instead, you catch all the incoming data in a QuickTime  
driven callback. Contact me offlist and I can try to help with a few  
more of the details.

> I've been struggling to code the FFTs in RB and haven't had much  
> success. It's pretty complex stuff. I did find a FFT plugin for RB  
> this weekend. Hopefully that will work.

I haven't tried realtime FFTs with REALbasic before, but I suspect  
it's feasible. There are probably several options here too. I suspect  
that you'll run into problems doing it with pure RB code (the  
realtime part is why). Frank mentioned FMOD. Also, Apple has FFT apis  
available. Or, maybe you could do the coding in pure REALbasic, but  
use Altivec operations to help with the math speed.

Another interesting idea might be to pursue the ability to make a  
REALbasic application act as an AudioUnit host. Then, you have access  
to any AudioUnit plugin.
_______________________________________________
Unsubscribe or switch delivery mode:
<http://www.realsoftware.com/support/listmanager/>

Search the archives:
<http://support.realsoftware.com/listarchives/lists.html>

Reply via email to