then what is your "Yes" meaning? that he's right about "the added resolution doesn't mean higher quality"?
or wrong?

sean

For what?
Playback? Not really.
Recording with no editing or processing at all? Not really.
Recording with editing and processing? Yes, absolutely use 24 bit.

You´re not completely right. If you process and save a 16 bit file or process a it off-line in your daw it will be saved again in 16 bit format (yes after being processed in 32 or 64 bit float) and errors will slip in due to quantisation errors, which will increase if you then process again. 24 bit is much better in that respect (32 bit float even more so). The better noise specs of 24 bit are only somewhat useful. It is not easy to find a source (mic, instrument or even a room) with better then -96db self noise. Since the OP is asking this question it is unlikely he/she will have extremely sophisticated/ super low noise gear.

Agreed up to a theoretical point. Maximum converter dynamic range is about 110db in reality though (not 144 bit yet unfortunately) but the dynamic range of real world sources is even less. So yes you could record 14db lower with 24 bit vs 16 bit to get the same noise level, which is nice when you record sources with a very wide dynamic range, which need levelling afterwards.
But, I always record 24 bit. The only reason to use 16bit for me would be; If file size is a real issue or if I need fast access (without converting) on low end media players that only support 16 bit wav files (lots of those around).

Manolo, I think that in real conversation we would agree very soon, on a forum it´s always a little slow.
This was all about advice to the original poster. Yes 24 bit has lower noise floor and distortion, but this has only audible impact if the music contains useful information below -96db (and almost none does and even if it does you would have to play it very very loud). I don´t think we should discus this any further as we might just as well be boring the OP to death.....LOL.

That´s what is thought, but not true. The loss of resolution going from 24 to 16 bit causes noise and distortion at -96db. The main reason to work at 24 bit is that you don´t have to be paranoia about levels. You can do a mix (that needs mastering later) at -20db and still have 120db of dynamic range in 24 bit. If you would do that with 16 bit, you would have only 76db (not to little either) but add some compression later which adds 20db of gain to a low level passage and suddenly the dynamic rang at that passage is only 56 db. Now that distortion might be audible!

Hope this helps!

I said we would agree!
Still remains if you do a simple recording at decent level, no post processing, not for DAW use, 16 bit is fine. 16 bit sounds the same as 24 bit. Only after processing you will reap the benefits of 24 bit.

A: Recording levels

Since 24-bit allows to record at lower gain-levels to begin with you don't have to:

- use compression or extra talent with high dynamic sources (vocals!) *before* going digital

- watch hard for clipping during recording or verify the recorded material afterwards (time and concentration saver!)

- drive the analog-gain stages closer to their distortion point by staying close to 0 dB

B: Virtual instruments/effects

If a virtual instrument/effect introduces digital artifacts like aliasing (bad filter design) and you have to add gain-staging (compression, level, whatever) afterwards it's better to have those artifact in the lowest possible range.

With a 16-bit signal-path the lowest level any digital artifact can be is -96 dB, with 24-bit it's -144 dB = 48 dB more room to hide artifacts in. And since we are in the digital realm these are real values and not something limited by analog reality (up to the final DA conversion).