Bit depth conversion for effects?

Seanmorse79

Seanmorse79

New member
Okay ...as I understand it, the main disadvantage (albeit, not the only one) to recording at 16 bit vs. 24 is that you haven't got enough resolution to use effects as cleanly. So here's a question:

Would there be any advantage to:

1. Recording in 16 bit
2. Converting to 24 bit
3. Mix and master, printing all of your effects in 24 bit (pain in the neck, I know)
4. Dither the final mix back down to 16 bit.

..or would this just be a big ol' mess?

I was thinking of it in terms of graphics... If you take a low-res photo, double the size (amount of pixels), then do your shadows, smoothing, etc.. before resizing it back down, you get much better results.
 
Um, that's what's already happening. Your audio software will most likely use a 32bit float internal resolution.

Slackmaster 2000
 
Yes, but you want to "digitize" the original source with the largest word-size feasible (ie, 24-bit) to maximize the resolution of the digital signal.

If you start with 16-bit, sure the DSP processes internally at 32-bit, but doesn't add anything to the bitsize of the original signal. Starting with 24-bit gives the DSP more to work with.

It's kinda like scanning a low-res graphic and trying to blow it up -- it will become all fuzzy. If you start with a hi-res graphic and blow it up, it will be that much clearer.

Bruce
 
Right, I didn't mean to imply that starting at 16 and 24 would yeild the same end result! I just meant that there's no reason to convert a bunch of 16bit tracks to 24bit, unless of course you're working with other 24bit files, but even then it's just for convention.

Slackmaster 2000
 
Well, it would be even MORE pointless if you recorded in 24bit, because then you'd be converting from 24bit to 24bit. :)

Yes, don't do it.

Slackmaster 2000
 
Where can I find out more info on bit depth, 24 bit, 16 bit, etc... I'm trying to grasp these concepts as I read. I mean I understand the words, but I get lost on the meaning of the phrases. Is there a beginner course somewhere...? I figure since I'm trying to build a new DAW, I want to try and understand why I'm using the hardware and software I think I would like to use...

Thanks,
Vice
 
Bottom line, its best to record in 24 bit and keep it 24 bit until you have to mix down / master for CD-ROM, at which point you HAVE to convert to 16 / 44.1
 
Very informative article, Gidge, thanks again for the link. I think I understood the concepts (not fully of course), but I get the idea. It will probably be awhile before I do any mastering. Nevertheless, I have questions, I'll ask two for now.

If my hardware and software all say 24bit/96kHz, then, I'm good, right?

How does applying 32bit effects do to the word length? Does it automatically truncate itself to 24bit? Or am I confusing things?


Vice
 
You must log in or register to reply here.
Back
Top