drewfx1
Had you mentioned in the first place that it was an example of dithering to a 10 bit level then I wouldn't have commented. But you neglected to say that and thus people may have mistakenly believe your example (and any conclusions based on it) were based on 16 bit resolution rather than 10 bit resolution.

Sorry, but to clarify. I have not "dithered to a 10 bit level". I have exported into 16bit with/without dithering, in Sonar, from a clip recorded into 24bit file by the interface with -113 db RMS SNR (RME).
 
But as I have explained in my original post as good as I could (in not my native language), 16bit format is fixed point and so it degrades precision at lower levels. At the same time, not dithering artifact/noise has absolute (-90dB) level, independent from the signal level. So on lower volumes, original signal is degraded (in example case down to 10bit or whatever precision, I have not really checked numbers) while dithering noise stay on the same level as it is for full 16bit. And so, after amplification, it becomes clearly audible.

.

mister happy
For the innocent bystanders, here is the math:

16 - 6 = 10

Hands down the worst understanding of word length I've ever encountered.

John T

mister happy
For the innocent bystanders, here is the math:

16 - 6 = 10

Hands down the worst understanding of word length I've ever encountered.

Actually I think you're the one misunderstanding. 6 bits are indeed not being used, leaving 10 bits of resolution - which is exactly what he said.

Simple question:

It was said earlier in this thread (and another one) that 32 and 64 refer to floating point. Does that mean that if I choose 32 when exporting it's exporting at 32fp?

I can't find anything in the Sonar documentation to support this.

Thanks in advance for any help.

asarghmatey
It was said earlier in this thread (and another one) that 32 and 64 refer to floating point. Does that mean that if I choose 32 when exporting it's exporting at 32fp?

Yes. 32 and 64bit files are floating point. I do not know how to see that in Sonar. But you can check with other programs.

azslow3

asarghmatey
It was said earlier in this thread (and another one) that 32 and 64 refer to floating point. Does that mean that if I choose 32 when exporting it's exporting at 32fp?

Yes. 32 and 64bit files are floating point. I do not know how to see that in Sonar. But you can check with other programs.

Thanks for the response azslow3. I'm confused though; why would it matter what it says in other programs? Are you saying that 32 and 64 are always floating point? In any program? By any manufacturer?

Thanks again

asarghmatey

azslow3

asarghmatey
It was said earlier in this thread (and another one) that 32 and 64 refer to floating point. Does that mean that if I choose 32 when exporting it's exporting at 32fp?

Yes. 32 and 64bit files are floating point. I do not know how to see that in Sonar. But you can check with other programs.

Thanks for the response azslow3. I'm confused though; why would it matter what it says in other programs? Are you saying that 32 and 64 are always floating point? In any program? By any manufacturer?

Thanks again

Sorry. I mean when you select 32 or 64 in the export dialog of Sonar, resulting files are floating point. But I do not know how to check that within Sonar...
32 bit integer WAVs are not forbidden. So I guess some programs can generate them (but not DAWs I have).
 
PS. In spirit of previous post... Unlike dithering in 16bit integer files, 64bit files (not processing!) make no sense. It is theoretically impossible to spot any different between 32bit and 64bit FP files, till you you (export->render) in a loop 16 times. It is practically impossible to hear the different (independent from "tricks" and "cheats" I have described before) till you (export->render) more then 256 times...

Interesting. Do you know of any programs that can analyze a wave file and tell you if it's floating point or not?

https://sourceforge.net/projects/mediainfo/