16 44.1

[paul kostabi]

I sent some files as exported .wavs to a PT guy. First he was convinced they weren't there. Once figured out was " bummed they weren't 24bit" All this without even listening. :)

[MMP]

I am always happier to get 24bit files if I am doing any work on them. I am fine listening to 16 bit at the end of processing, though.


Regards,

MM

[Grekim]

Better to have well recorded 16 bit tracks than poorly recorded 24 bit tracks. Still curious why you went 16 bit...

[RBIngraham]

I'm more curious why this guy couldn't just covert the damn things himself. Sound Forge anyone?


PFFFttt.... Blow Tools...

[Dave Labrecque]

I'm more curious why this guy couldn't just covert the damn things himself. Sound Forge anyone?


PFFFttt.... Blow Tools...

In the even that you're not kidding...

I presume he wanted the fidelity increase that potentially comes with the higher bit rate. Something that a simple conversion would not provide. But I'm sure you know all this.

[RBIngraham]

In the even that you're not kidding...

I presume he wanted the fidelity increase that potentially comes with the higher bit rate. Something that a simple conversion would not provide. But I'm sure you know all this.

Yes I was just being my usual charming and sarcastic self. :)

and taking a stab at Slow Tools users while I was at it...

Having said that, we did just get done with a thread about using higher sample or bit rates. And if Paul did what I would expect any good engineer would do (and I have no reason to doubt he wouldn't) and normalize the file, and compensate for good gain structure if it wasn't perfectly recorded in the first place, I would suspect that upping the file to 24 bit wouldn't do all that much, depending on the content of the file. The big differences happen when you manipulate the file or when you're doing the initial recording to have some extra headroom to play with. All of that could easilly be fixed on the PT user's end with a simple conversion to 24 bit of the 16 bit file and then work could move on. Any further editing on his end would be perfomed in 24 bit (and while I don't know PT very well, I would hope it was working with more than 24 bits for it's processing anyway, no matter what the format of the files were, although some do require you to set your software to store higher bit depth audio if you want the processing done at those higher rates). Others allow you to set the rates at which their engine runs seperately from what resolutions are used for file storage.

[Dave Labrecque]

Yes I was just being my usual charming and sarcastic self. :)

and taking a stab at Slow Tools users while I was at it...

Having said that, we did just get done with a thread about using higher sample or bit rates. And if Paul did what I would expect any good engineer would do (and I have no reason to doubt he wouldn't) and normalize the file, and compensate for good gain structure if it wasn't perfectly recorded in the first place, I would suspect that upping the file to 24 bit wouldn't do all that much, depending on the content of the file. The big differences happen when you manipulate the file or when you're doing the initial recording to have some extra headroom to play with. All of that could easilly be fixed on the PT user's end with a simple conversion to 24 bit of the 16 bit file and then work could move on. Any further editing on his end would be perfomed in 24 bit (and while I don't know PT very well, I would hope it was working with more than 24 bits for it's processing anyway, no matter what the format of the files were, although some do require you to set your software to store higher bit depth audio if you want the processing done at those higher rates). Others allow you to set the rates at which their engine runs seperately from what resolutions are used for file storage.

Sounds like you're saying if he doesn't convert the files to 24-bit, the software won't do it's internal processing at that depth. I wouldn't think that's right, and so I would think there'd be no advantage to doing that conversion. As far as optimizing the use of the bit range via normalizing, he could do that in the program with simple gain changes, I'd think.

Am I mad?

[Grekim]

Pretty sure PT and every other DAW will take the file whether it's 16 bit or 24 bit, process it at a higher depth, temporarily storing it this way in RAM as it works its way through the mixer. The processing bit depth should be the same in either case. The only thing affected would be the depth of newly recorded tracks to that session. So a 16 bit session would record new tracks with only 16 bits worth of data, but processing is still 32, 48, or 64 depending on the DAW and plug-ins being used.

[TotalSonic]

"converting" a 16bit file to 24bit only pads the file with 8 additional least significant bits with zeros in them and does nothing to give it greater resolution. You simply get a file with 16bit resolution inside a 24bit container.

While as a delivery format 16bit does indeed work great for the vast majority of playback systems - for a mix format - which will indeed be potentially subject to lots of gain staging and processing - then something which possesses greater detail in its least significant bits and is free of either the distortions that occur from truncating while requantizing - or is free of any additional noise from dithering - to me is indeed preferable to work with. While the end result may be very subtle in any improvement I'd still say an "all other things equal" I think ultimately you're better off giving 24bit files.

So frankly I think the ProTools guy was simply trying to be a conscientious engineer in his request for 24bit files instead of 16bit ones. I would do the same exact thing in his place using SAWStudio (although if they are the only thing available I would certainly go ahead and use 16bit files myself - and once I know these are all there is to work with I don't complain further).

Best regards,
Steve Berson

[Dave Labrecque]

"converting" a 16bit file to 24bit only pads the file with 8 additional least significant bits with zeros in them and does nothing to give it greater resolution. You simply get a file with 16bit resolution inside a 24bit container.

While as a delivery format 16bit does indeed work great for the vast majority of playback systems - for a mix format - which will indeed be potentially subject to lots of gain staging and processing - then something which possesses greater detail in its least significant bits and is free of either the distortions that occur from truncating while requantizing - or is free of any additional noise from dithering - to me is indeed preferable to work with. While the end result may be very subtle in any improvement I'd still say an "all other things equal" I think ultimately you're better off giving 24bit files.

So frankly I think the ProTools guy was simply trying to be a conscientious engineer in his request for 24bit files instead of 16bit ones. I would do the same exact thing in his place using SAWStudio (although if they are the only thing available I would certainly go ahead and use 16bit files myself - and once I know these are all there is to work with I don't complain further).

Best regards,
Steve Berson

Just this week I had a film maker client bring me an OMF with 16 bit files in it. I asked if any of the source files in his Final Cut session were 24-bit. He said he didn't know, but that he'd chosen the 16-bit setting when he created the OMF. I sorted through the raw outtake audio (which was separate from his session) and found that a lot of them were, indeed, 24-bit.

He did some revisions to the film in FCP and this time created two OMF's, because outputting at 24 bits exceeded the maximum allowable size for a single OMF.

The sound is awful throughout. Probably should've just gone with 16-bit. :o :p

[RBIngraham]

Sounds like you're saying if he doesn't convert the files to 24-bit, the software won't do it's internal processing at that depth. I wouldn't think that's right, and so I would think there'd be no advantage to doing that conversion. As far as optimizing the use of the bit range via normalizing, he could do that in the program with simple gain changes, I'd think.

Am I mad?


Yes you are mad... but I wasn't being very clear... :)

What I meant is what Grekim stated. In most DAWs I've seen, if you run the audio engine at 16 for example, it will store files as 16 bit data and I would assume it's doing most of it's processing and mixing in much higher resolutions. However it is possible in some DAWs to run the engine at 24 bits and have it store 16 bit data. Not sure about the other way around, never really tried. So in that case you could potentially be throwing away data.

Also in some editors such as Sound Forge the engine is not really separated from the file. Granted editors like Forge are not really a DAW... but what the recipient was going to do with the files wasn't really specified by the OP either, although one would assume it's being loaded into blow tools....

So for example in Forge, if I have a 16 bit file and do processing to it, I would suspect that some data will be thrown away as soon as I save as the processing would be done at the resolution of whatever the plug-in(s) wish to work at. I think you can also set the temporary files to lock to a certain resolution if you wish. (if my memory serves) Anyway, if I was going to do some work in Forge and I wanted to be able to do my work in steps rather than all in a single work session, the first thing I would do to 16 bit files would be to save them as 24 bit or even 32 bit float. Then resave them to whatever my final format my end delivery was going to be in.

So no, not mad, but it all depends on what the files are, how they were originally recorded, what else needs to be done to them and likely also what kind of material we are talking about as well as some other things I'm not thinking about at the moment.

No point here really... just saying..

[Dave Labrecque]

Yes you are mad... but I wasn't being very clear... :)

What I meant is what Grekim stated. In most DAWs I've seen, if you run the audio engine at 16 for example, it will store files as 16 bit data and I would assume it's doing most of it's processing and mixing in much higher resolutions. However it is possible in some DAWs to run the engine at 24 bits and have it store 16 bit data. Not sure about the other way around, never really tried. So in that case you could potentially be throwing away data.

Also in some editors such as Sound Forge the engine is not really separated from the file. Granted editors like Forge are not really a DAW... but what the recipient was going to do with the files wasn't really specified by the OP either, although one would assume it's being loaded into blow tools....

So for example in Forge, if I have a 16 bit file and do processing to it, I would suspect that some data will be thrown away as soon as I save as the processing would be done at the resolution of whatever the plug-in(s) wish to work at. I think you can also set the temporary files to lock to a certain resolution if you wish. (if my memory serves) Anyway, if I was going to do some work in Forge and I wanted to be able to do my work in steps rather than all in a single work session, the first thing I would do to 16 bit files would be to save them as 24 bit or even 32 bit float. Then resave them to whatever my final format my end delivery was going to be in.

So no, not mad, but it all depends on what the files are, how they were originally recorded, what else needs to be done to them and likely also what kind of material we are talking about as well as some other things I'm not thinking about at the moment.

No point here really... just saying..

OK, well... I'm still seeing no reason for him to convert his 16-bit files to 24-bit. Pain, no gain.

[RBIngraham]

OK, well... I'm still seeing no reason for him to convert his 16-bit files to 24-bit. Pain, no gain.


I'm not sure I would view a few mouse clicks and typing in a new file name as "pain".... :)

But yes if you are thinking solely from working in a SAW environment, there is little to be gained. And likely little lost in sending out 16 bit files either, other than ego points.

In the end why the client of the OP would make a bunch of judgments prior to listening to the work is the real mystery that needs to be solved. :rolleyes:

[Soundguy]

and taking a stab at Slow Tools users while I was at it...



Even decades old versions of Slow Tools automatically convert files to session format on the way in. No outboard conversion required. Slow Toos v10 allows mixed bit depth and sample rates in the same session if you prefer, including 32 bit floating point so you can import from more formats without losing resolution.

So you're stabbing the wrong DAW:)

But as Dave L. said, the guy probably wanted the higher fidelity of files originally recorded at 24 bits. But then there is nothing to be gained by converting 16 bit to 24 bit.

Soundguy

[Dave Labrecque]

I'm not sure I would view a few mouse clicks and typing in a new file name as "pain".... :)


Wow. I think we differ on every point you made. :p With the possible exception of the 'pain' remark. I meant pain as in PITA, which is a bit less serious than actual pain. ;)



But yes if you are thinking solely from working in a SAW environment, there is little to be gained. And likely little lost in sending out 16 bit files either, other than ego points. Actually, I was thinking solely from working in the PT environment, which is what the OP describes as what his client is doing. And, although it can be argued that that little is lost by sending 16 bit files, there's some lost, depending on the quality of the recording. But you know that.


In the end why the client of the OP would make a bunch of judgments prior to listening to the work is the real mystery that needs to be solved. :rolleyes:Well, 16-bit audio can only be "so" good. 24-bit can be better. So, maybe he just has real high standards. Maybe his golden ears are offended by 16-bit files the way mine are by 8-bit files. :)