24 bit WAV

[pfairfax]

I've been recording at 24bit and exporting as Other as Audacity doesn't have a 24 bit wav option.
This has worked fine up to this week. Now when I openthem in Audacity it brings back the info that they are 32 bit float. Is this a problem that I can deal with?

Peter

[steve]

This begs the question "Why do you want to export as 24 bit WAV"?
If you're after top quality, use 32 bit WAV - files are bigger, but the format is a lot more "standard" than 24 bit WAV.
If you're thinking about DVD's, then by the time it's been encoded I doubt that you would be able to tell the difference between 16 bit and 24 bit (assuming that you can hear the difference even if the sound is uncompressed). I would stick with 16 bit or 32 bit, depending on what you are wanting to do with the files.

[pfairfax]

Hi,

I've been trying out Click repair which doesn't support 32 bit float, but does seem to do the automatic click removal rather well. Audacity is rather easier to do the final repair of difficult damage as the spectrum view is really good.I was trying to go for maximum quality while I'm doing the processing, but I'm beginning to wonder if the hassle is worth it!! I probably won't notice the difference once burnt to CD!

It's frustrating though why something that works one time doesn't another.

Peter

[waxcylinder]

Hi,

I've been trying out Click repair which doesn't support 32 bit float ....... I probably won't notice the difference once burnt to CD!

I doubt too that you will notice the difference - I use ClickRepair too (and agree with you on how good it is at it's job). I capture LPs and export 16-bit PCM stereo 44.1kHz to ClickRepair -reimport the repaired WAV into Audacity and do the final editing. The export to CD at the same 16-bit PCM stereo 44.1kHz. This seems to produce excellent results (even to my older ears - albeit on good equipment). In fact older LPs proper tidied up can often sound better than their modern "remastered" counterparts which have too much compression applied by the engineers in pursuit of the loudness war.

WC

[Harry_O]

This begs the question "Why do you want to export as 24 bit WAV"?
If you're after top quality, use 32 bit WAV - files are bigger, but the format is a lot more "standard" than 24 bit WAV.
If you're thinking about DVD's, then by the time it's been encoded I doubt that you would be able to tell the difference between 16 bit and 24 bit (assuming that you can hear the difference even if the sound is uncompressed). I would stick with 16 bit or 32 bit, depending on what you are wanting to do with the files.

Hi,

I have to say I´m a bit after the same feature as the waxcylinder. The reason is,
that I´m using the 24 bit/48 KHz in my main sequencer (Logic 5.5.1).
When I´m editing the Audacity-recorderd files in my Logic, it would be nice if the files
had the same bit debth (24 bit).

But now I´d like to ask the expert opinion: if I open the 32 bit file in the Logic, do I have
to convert the file firts for the 24 bit? (my Audiointerface doesn´t support higher than 24 bit).

My second question: what does the "float" mean in this connection (32 bit float).

My third question: how do find using 48 KHz in sample rate in stead of the 44,1, do you
find that I´m winning something in terms of the sound quality by using this rate.

Harry

[kozikowski]

<<<My third question: how do find using 48 KHz in sample rate in stead of the 44,1, do you
find that I´m winning something in terms of the sound quality by using this rate.>>>

I do all my production at 48000 for a couple of reasons. 44100 is a compromise delivery format designed in the thirteenth century to get audio CDs out the door for the first time. The audio golden ears complained bitterly that it wasn't high enough and you cold hear problems way up at the high audio frequencies. True, but only if you had golden ears. Most people really liked that they could get 75 minutes (originally) of music on the disk and that was the tradeoff.

48000/16 is the video editing format. I slosh back and forth.

Koz

[steve]

I´m using the 24 bit/48 KHz in my main sequencer (Logic 5.5.1).
When I´m editing the Audacity-recorderd files in my Logic, it would be nice if the files
had the same bit debth (24 bit).

Yes I can see that.

if I open the 32 bit file in the Logic, do I have
to convert the file firts for the 24 bit? (my Audiointerface doesn´t support higher than 24 bit).

I think Logic should be able to import 32 bit (float) format directly - try it and see.

My second question: what does the "float" mean in this connection (32 bit float).

23-bit mantissa + sign bit + 8-bit exponent.
In English - Real world recordings at 32 bit would be burying the bottom 8 bits in background noise, which is pretty pointless, so, since computers use binary and there's little point in recording at greater than 24 bits, the extra bits from a 4 byte number are used for floating point calculations. This makes calculations far more accurate (and quicker) than fixed integer format, so rounding off errors due to the calculations involved in editing are negligible.

My third question: how do find using 48 KHz in sample rate in stead of the 44,1, do you
find that I´m winning something in terms of the sound quality by using this rate.

Take two identical microphone, and two perfect digital records and make one recording at 44.1/16 and another at 28/32 and I doubt that you could tell the difference - even if you can tell the difference, you would not be saying "this one sounds so much clearer/sharper/smoother/nicer than that one" you would just be saying "that's version A and that's version B", however, the differences can be compounded by doing multiple edits and processing, which if using 44.1/16 can become far more noticeable. My sound card runs internally at 48 kHz, and Audacity runs internally at 32 bit float, so If I am doing a project that I want at the highest quality, I will use 48 kHz 32 bit. If I'm just sticking something onto a CD so that I can listen to it in the car, I use 44.1 kHz/16 bit.