Neither I nor the magazine article said that decompressing the FLAC file results in any loss of data, so in that sense there should be no harm in creating a WAV file from the FLAC file, but there wouldn't be anything to gain either. The magazine's contention apparently was that the process of decompressing at the same time as the D/A conversion introduces a slight degradation of the audio playback (not the data itself). I tend not to believe the perceptibility of such small details, and I'm not suggesting that anyone else needs to. For those with any further curiosity, the article is in the December 2014 issue of The Absolute Sound.

Love your explanation Craig
 

Lots of good points about FLAC, and Craig's explanation is spot on. The input and output of a FLAC encode-decode process are identical.
 
Just one query: people are saying FLAC renders faster than WAV?
 
I've not checked, but I can't even see how this could be possible. Indeed, if it were correct, that would suggest a bug in rendering to WAV.
 
I'm sure this must be an illusion. Rendering to WAV means writing the data. Rendering to FLAC means writing the data *and* compressing it.

There's a general computer science Hard Problem about compression, which is that compression is slow, while de-compression can be really, really fast. This is why when you upload a file to youtube or soundcloud, it takes hella time to process, but then playback later is more or less instantaneous.
 

How much space do you save? I remember saving a bunch of space back in the days of furthurnet.org. Now, it seems to not be such a big difference. Is Windows compressing our wav files? Did the wav standard improve?

If anyone cares,I export wav.

I just ripped a song from a CD and saved the WAV file. I used Sound Forge to convert to FLAC and loaded the FLAC and saved back to WAV.
The two WAV files are not identical.
I summed them with one inverted and there was a very tiny error component as shown below. After about 45 seconds of playback the error dropped below the SPAN threshold.
 

It would be interesting to see if the FLAC and WAV files null.

The FLAC compared the same way as the WAV. Error below -100db and then after 48 seconds the error is essentially 0.
 
I don't know what happens magically at 48 seconds, but the error signal drops out very quickly.

John T
people are saying FLAC renders faster than WAV?
I've not checked, but I can't even see how this could be possible.

Sure, it is possible.  If the FLAC is substantially smaller, it might require less disk I/O or less network transmission, at the expense of additional CPU processing on the FLAC.  Probably not worth fretting over.  I am in the camp that says stick with WAV until you are ready to distribute.

eikelbijter

That's absolute nonsense! Sorry, but LOSSLESS means no loss!
 
What is the world coming to.....

But lossles does not mean identical. Whether you hear the difference or not, that's another topic,
but the differences could (?) cumulate in some way depending on the path of refinement.