Skip to main content
Topic: (Nitpick) Future WavPack update - display 32-bit float in file properties (Read 368 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

(Nitpick) Future WavPack update - display 32-bit float in file properties

I've noticed that Foobar is unable to report whether a WavPack file contains 32-bit integer or floating point information. I know that WavPack did not truncate the floating point numbers out of the original data, because when I use the WavPack decoder, I get the raw 32-bit floating bit PCM back.

I'm assuming that this is something with the WavPack encoder/decoder and not Foobar? If so, would it be possible for it to have a future release where it tells Foobar whether it's a 32-bit floating point file or not? (Similar to 32-bit wav files?)
-- Brought to you by the guy, alone in the corner, converting all his files to lossless formats...

Re: (Nitpick) Future WavPack update - display 32-bit float in file properties

Reply #1
WavPack does not truncate that information because if it did then the player would not be able to correctly render the audio (which Foobar obviously does).

The WavPack decoder will not convert floating-point PCM to integer in any circumstance, so the only way for Foobar to be rendering the files correctly is that it knows the format.

I did verify that Foobar does show whether WAV files are float or integer (like you say), so it should be pretty easy for it to display the same thing for WavPack files.

Re: (Nitpick) Future WavPack update - display 32-bit float in file properties

Reply #2
So it is a fb2k-specific issue? Maybe move the topic over to the fb2k section then, so @Peter can get yet another good suggestion to put in line for 1.4.1 :-o

(By the way, I don't think I have seen a 32-bit integer PCM file "out in the wild". Those DAEs which work in 32 bits choose floating-point then? Seems like the reasonable thing to do, but what do I know ...)

 

Re: (Nitpick) Future WavPack update - display 32-bit float in file properties

Reply #3
Yes, 32-bit audio is generally floating-point, like in virtually every DAW.

I have seen 32-bit integer files though, but I don't remember where they came from. Some programs do use 32-bit integer internally (like SoX), but I don't know what the original reasoning was.

 
SimplePortal 1.0.0 RC1 © 2008-2018