Recorded 24Bit files into a 16Bit PT Session File. Help!

[johnjm22]

So I created a new PT session at 16/48, but didn't realize it. I thought I created the the session at 24/48 as usual.

While recording I had PT clocked to my Rosetta which was at 24/48. The vast majority of the tracks I recorded went through the Rosetta and into PT via SPDIF.

It wasn't till later, when I had to start bouncing tracks, that I realized my blunder.

So my question is this; Are my audio files in this session 16 bit or 24 bit? I'm guessing that all the tracks I recorded with the Rosetta are 24 bit, while all the tracks that went through the my 002's converters are at 16 bit. Is this assumption correct?

Also, when I'm bouncing tracks (that are supposidly 24 Bit, but are in a 16 Bit session) should I bounce them at 16/48 or 24/48? I'm guessing that I should bounce them at 16/48 that way the computer won't alter them, causing degredation. But I really don't know.

[johnjm22]

Can anybody help me?

[gsharp]

if your session is 16 bit then your files are 16 bit. the extra 8 bits went bye bye when they hit your hard drive.

I wouldn't stress too hard. If you have more tracking to do on the stuff you could create a new 24 bit session, pull what you've done into it (which will add 8 zeros to make it 24 bit) and do the rest at 24 bit.

bouncing a 16 bit file @ 24 but won't degrade it, it will just add 8 zeros.

Just be glad you didn't have the sample rates mismatched. That's the really fun one.

In the meantime, don't worry we won't tell anyone that there's 8 bits off into the ozone somewhere.

[theblue1]

I'm with gsharp on this one.

If you do the rest of your mixing and processing at 24 bit you won't have lost as much as you're probably afraid you've lost.

While it would have been nice to capture all 24 bits you were putting in, anything on the front end over 20 bits is arguably overkill, anyhow.

Where higher rez really comes to the fore is in mixing and processing of existing signals.


Whether or not to tell the client is a hard one. On one hand, I think it's unlikely your mistake made much if any impact (and that would be highly dependent on the style of music as well as your recording leveles, etc... if you were tracking deathmetal, I wouldn't sweat it. If it was a string quartet with a wide dynamic range... well... that might be different). In general, I favor full disclosure. But such a disclosure might simply plant seeds in the client's imagination. (And imagination is a very powerful thing even in a field where some differences can be measured and quantified -- like some folks who think they can hear the difference between bit-for-bit identical CD burns ("I can tell -- this one was recorded at 4x... I can hear so much more detail than this other one recorded at 24x." You know.)

[gsharp]

If anybody finds out and wants to bitch, you can always use Chris Lord-Alge's response when people whine about him knocking everything down to 16 bit before he mixes it: "16 bits is good enough for this song".

Seriously though, if the difference in quality was make or break surely you would have noticed that it didn't sound right when you were tracking it. Right????

Seriously, finish at 24 bit if you want to/can, but do not lose any (more) sleep.

[theblue1]

Good second point.

And an amusing strategy from Chris Lord-Alge.

[RonB]

Go here, http://digidesign.com/http://digides...igidesign.com/ , select support , next select User Conference then do a search. You'll find your answer.