Frozen Tracks/Final Mix

I'm not familiar with Ozone but cake needs to be set to render at 44.1/16 bits.  Ozone should dither only once, when you do the downsample.

AT
I'm not familiar with Ozone but cake needs to be set to render at 44.1/16 bits.  Ozone should dither only once, when you do the downsample.

AT, not to sound argumentative, because I know you know your stuff, but Sonar shouldn't be set to 16/44 if he is taking a mixed wave back into Ozone to master. In my mastering business every day, I preserve all audio in its highest format before I CSR and dither. He's doing the same as me minus me using Ozone because I hate it. LOL!
 
Export the mix out of Sonar the way the project was created (24/96) re-import the stereo mix wave back into Sonar/DAW of choice for mastering while using Ozone, then dither and CSR as the final stage to 16/44. Why would you want to dither and CSR before you finish the mastering procedure?
 
Maybe I'm not understanding something here?

BrFrBob
What I'm wondering about is; when I export the mix at 24/96, are those 32 bit frozen tracks rendered down to 24 bit without dithering?  I read somewhere you're not supposed to step down the bit rate without dither.  I'd like to export the mix without dither so I only have to apply it once on the finished project.  Should I unfreeze the tracks first so they stay in 24 bit format?  Do frozen tracks with plug-ins lose something the way I'm doing it?
 
Thanks

No, they are not because they are already 24bits. Don't confuse the file format with audio bit depth. They are two different things.
 
Regards,
Dan
 

You only ever dither a project ONCE and that's right at the end when you convert to a lower bit depth from a higher one.
 
If your end product isn't bit reduced then there is zero need for dither.

AT
Danny,
 
cool.  As I said, I don't know ozone at all, except for what people here say.  I use Sound Forge for stereo editing/mastering, and simply open the 24 bit file in it and do everything I need within it, including dithering.  I thought Ozone was just a plug-in to be used inside of SONAR.  If you set the render/dither inside of Ozone I didn't realize that and can see how my advice would be confusing.
 
dither on, you crazy stars. ;-)
 
@
 

It's all good! I figured you read it wrong or just wasn't sure and I knew you probably didn't dither and CSR before you used your final limiter. But yeah, it was a little cornfusing to me which is why I asked. LOL! :) Glad you saw it that way too...haha! I'd never pick on ya man...too much respect. ;)
 
To be honest, the other thing was, (nothing to do with you) I didn't know how to answer the op's question because I don't use 32 bit file sizes. That whole debate will go on forever..lol...but I figure, if my sound card is 24 bit, I'm keeping it there. I learned years ago that when you bounce something digital, it stays as it was because it's digital...even at 16/44. Someone came up with some sort of math error somewhere that I sure can't hear, thus, I have never felt it was needed to tax my system and use higher bit rates/floating point.
 
I've never been able to hear a difference, I've never been able to notice any artifacts, and I've never had an issue staying at 24 bit without using 32 bit render or float. When I've tried it on projects to compare which were done from the ground up to compare, I shook my head and just felt "um, ok...if you math guys say so...but I sure don't hear anything blatant that makes me say "oh yeah, I'm so using 32 bit float from here on out!" Maybe because I have good interfaces?
 
But it does bring forth an interesting question made by the OP that I'm not sure anyone has answered blatantly....unless I've missed it or misinterpreted the replies? If so, I apologize and admit I know nothing about this sort of thing simply because it just didn't make enough of a difference in my world to care or waste any time on. That's not ignorance, mind you, that's "if I can't hear a major difference after 3-5 plays...there is no difference that I need to worry about." and my mentality on just about everything in life that doesn't make a huge impact or blatantly obvious difference. :) Bob Katz has preached this to me several times. I still think it's a bunch of hooey in my world. At least I'm being honest. ;)
 
So if the op has recorded at 24/96 and has it all mixed, and has frozen tracks that were frozen and rendered to 32 bit, and those rendered 32 bit tracks are then exported along with the rest of the project at 24/96, what happens to those 32 bit files? They get reduced to 24 bit upon export without dithering, right? If that's the case, wouldn't there be a need for dithering on those tracks since we changed a bit rate?
 
However, (and this is where I think the op is having problems because I gotta tell ya, so am I lol) if he uses dithering in Sonar before exporting the project because of the 32 bit frozen/rendered tracks, what happens to the tracks that are already 24 bit during the dithering process that are remaining at 24 bit?
 
Shouldn't we dither any files that are 32 bit when going down to 24 bit since we are "going down" in bit rate? Or, is it because it's 32 bit float and not an actual bit rate, there is no need to dither? It can definitely be a little confusing and you guys know me...I got no problems learning from you and admitting I'm clueless when I simply....am clueless. LOL!
 

bapu
If (Mr. Golden Ears) Danny cannot hear the difference, I'll not worry my pea brain/ears over it.

LOL you're too good to me Ed. :) I've gotten bashed in other places about it. Stuff like "huh, you don't upsample to 32/96 and above?! You must not be a very good mastering guy."
 
I just respond..."no, I suck....but I'm booked with jobs up to November of 2017....but what do I know?"

BrFrBob
"Don't confuse the file format with audio bit depth."
Now THERE's a piece of information I didn't have before.  Danny, you nailed the intent of my question!  Using Dan's answer I did some surfing, now that I know what to look for.  Let's see if I got this:  the CALCULATIONS done on that 24 bit pre-frozen track are done to a 32 bit accuracy to allow for the effects the plug ins will have on the original wav file.  The resulting "24 bit" frozen track is therefore more accurate and lopping off the last 8 bits from the calculations would make no difference?
 
BrFrBob

Well, you aren't really lopping anything off. If your project is 24bit, and you export at 24bit, nothing has been changed and dither is not needed. The file format is 32bit, but the audio is 24bit. While rendering the frozen track, the full 32bits were used, but it is 24bit when it hits the hard disk. 
 
Regards,
Dan