Flac vs Wav audio quality

Yes, sorry–I've been fast and loose with the term - in either case, the recommended practice for an ISRC is to utilize the same prefix for each track of a recording, with the last five digits corresponding to the (unique) track.

That said - the same ISRC could be used for tracks released upon two different release formats (e.g., LP and CD) while remaining useful for metadata (identification) purposes, when embedded into the PQ subcode of a CD, for example.

You could still require different descriptive metadata in different market areas; so Simon's right, to be specific you need both.

Um, I'd like to correct myself.  Early on in this thread I stated that I could not hear any difference between FLAC and WAV.  At the time when I did various back-to-back comparisons over a year ago now, I concluded that any perceived difference was my imagination, and ripped my CD collection to FLAC, and have been using said format ever since.  

Today I have experimented with re-ripping some CDs to WAV files, and also converting the FLAC files of high-res albums I have purchased to WAV using dBpoweramp.  Having become completely accustomed to the sound of the FLAC rips, to my surprise the difference between these, and the new WAV files, is quite profound.  Looking back, there were number of things not quite up to scratch in my system when I was doing the back-to-back comparison, using a then newly purchased DAC-V1, and perhaps this is why the difference was less pronounced.  Now though, having gradually conquered getting decent quality from a PC based streamer and the DAC, combined with a year of FLAC based listening of ripped CDs, and the wonderful Qobuz, I can now detect a distinct difference between the two.  In summary the sound from the WAV files, whether re-ripped or converted, is richer, and more dynamic, giving a much more exciting delivery than the FLAC files decoded on the fly.  WAV is better.  Just don't ask me why.

I refer my honorable friend to my previous answer

https://forums.naimaudio.com/to...71#45070656793465871

Noted, now we are agreed.  I might be slow, but I get there in the end.

Double blind test was it?

Nope.

Maybe your FLAC decoding software implementation isn't very good, or your hardware support is defective.  Few would deny that you have weaknesses in other areas of software also.   FLAC should not sound any different to WAV. If your implementation of FLAC sounds worse than WAV then your implementation has failed. It is absurd to condemn a file format because of your inability to decode it correctly.  

If it wasn't double blind, it has no scientific credibility whatever; were anyone else to follow your lead it would just be faith and look where that has taken us this week.  But if you believe it, then store bigger files and enjoy

Took a DVD-A at 96/24 and ripped via DVD-AE to both FLAC and WAV (WAVE). Everything identical and played back through NDS.

The FLAC sounded way off, quite thin and not as full and meaty as the WAV file. It was so far off as to sound like something must be wrong.

So by using the same equipment and ripping the same DVD-A to both formats I found the FLAC worse by a reasonable margin.

Shrugs shoulders????

It has no scientific credibility whether blind or not.  I couldn't care less whether everyone follows my lead, each individual can decide whatever they like, I simply expressed a view.  Apparently that puts me in some kind of hifi extremist cult.  Pathetic!

I did get my long-suffering missus (LSM) to do a double blind test with me and I could tell no difference, unlike the guys above.  But then again I am an aging hippy who has listened to too much Grateful Dead and the Airplane at high levels.

I remember being at a demo once and the salesman had changed a bit of kit and I turned to the LSM and said that sounds better.  The salesman had a smile on his face until the LSM said "No, it ain't better, it's just different" and ya know she was right.  If looks could kill that salesman would just be getting out of chokey about now.

Interpreting testing is difficult, just remember all those people who in the early days of CD, bought a blue magic-marker pen, with a special label (and price).  Not me I hasten to add.

What many people, including myself, would deduce from your experience is that your NDS is no good at playing FLACs.  FLAC and WAV are different ways of representing the same data; they should sound identical - and do on many systems. It follows that if on your NDS FLAC sounds worse than WAV, then it is your NDS that has the problem, not the FLAC format.  Another possibility is that you messed the ripping up. You can check that by using dBpoweramp or foobar to compare the two files. Did you?

I can't understand why anyone would buy a player through which FLACs sounded different to WAVs. To me such a player would be profoundly broken.

@SoundStream

Touchy aren't we?  The post doesn't say anything like that at all.  Can I suggest you read things more carefully before you jump to conclusions?

There is no other way to read it.

BB

Isnt that the point really with our hobby. Who is to say one is better than the other. its only better to the individual who likes it. The person sitting next to you may think its worse.

i can hear a distinct difference between FLAC and WAV, but not FLAC transcoded to WAV. I wonder if others with non Naim streamers can tell.

@Dan43 was the FLAC uncompressed or compressed at usual level5 ?

Graeme

Big Bill originally posted:

"Interpreting testing is difficult, just remember all those people who in the early days of CD, bought a blue magic-marker pen, with a special label (and price).  Not me I hasten to add."

Nor me, Bill. I wasn't so gullible - the marker pen I bought was green, not blue! That made all the difference!.  

Actually, it didn't cost so much, and I don't think I have admitted buying one to anyone up until now. But now the world knows my secret.

And no - I didn't detect any difference in SQ.

Hi SongStream -

By "decoded on the fly", I take you to mean streamed as FLAC and decoded by your Naim device (sorry I don't recall which streamer you are using). Can your UPNP server transcode on the fly? If so, I'd be interested to hear your thoughts on any sonic differences between a transcoded FLAC file streamed as WAV and a native WAV rip coming from the same server, all other elements in the chain being equal. 

I do prefer the prsentation on my SuperUniti when transcoding to (24 bit) WAV from my FLAC rips, but the difference isn't as profound as you report. I've been a bit lazy, and have not converted a FLAC file to WAV to do the experiment myself (and I'm loath to give up the metadata, or learn how to do that with WAV files...). You seem to be in a position to do this from a files perspective, and if running eg MinimServer, switching on transcoding in your UPNP server is straightforward. 

From the the explanations offered as to why one file format might sound different than another, given both contain identical data but local computational effort differs and might impact audio for second-order reasons, I'm curious to know if others find a difference between WAV streams from different original files. 

Thanks for for whatever you're willing to share!

Regards alan

Nice to hear this Graeme, this feels sensible to me...and I happen to have just asked SongStream to do this same experiment as a follow up to his file format explorations. Interesting stuff!

Regards alan

@Soundstream

Don't be silly; read things carefully.

Believe what you wish. But setting out your opinion as evidence is not scientific; it is just your belief or faith.  Nothing wrong with that, as long as everyone else sees it as what it is and nothing more

PS If you are going to name call, then look up the word "pathetic"; you are using it wrongly at the moment

Have fun

Well I started this thread; and my tests were single blind. My wife had no idea anything had changed in the system and she asked me why the HiFi sounded different. I played her the different versions. She told me she could tell significant differences and what they were.

Obviously a test subject to observer bias but hey ho. Nobody suggested people could only post opinions here subject to a rigorous p value proof.

To all who doubt the effect I say a) try it yourself and b) note Paul Stephenson's comments in this thread. He does not dip into threads often, and something tells me he may know what he is talking about.

I'd also re-iterate suggestions it is about playback rather than the file per-se as the differences through my HDX were far more marked than through the NDS. This may explain why some are convinced, and others less so.

Bruce

A different way of looking at this.

Most players do not have the ability to play WAV properly. The most that they can do is reduce WAV to the level of FLAC. The manufacturers of these players do not realise that their players are compromised because they believe that the file format doesn't matter and that because the data is the same then so must the replay.

I don't know why people get so passionate. This matter comes up reasonably regularly. I believe all one is hearing when comparing FLAC and WAV replay is minute traces of crosstalk from the decoding circuitry. Cross talk is a feature of system design, it almost always exists in systems.

Thereforevif you can hear a difference, transcode before the data reaches your system. These days most Upnp servers support transcoding. I

I prefer WAV to FLAC on Naim, so I transcode on the fly. It works perfectly, and it's effortless. If I didn't want to do this, it's simply a single mouse button click to change. If you hear no preference, don't click that mouse button. Surely it no big deal?

Simon

I think that it is being suggested that a file stored as wav and played back without the need of transcoding sounds "better" than a flac file which has been transcoded to wav prior to reaching the streamer (assuming this is what SongStream means by "decoded on the fly" since the streamer will play flac files natively).

For my own part I when I originally bought my ND5 streamer I tried listening to both flac and "pure" wav files and found that the latter sounded a little louder which I think influenced my decision to store files in flac and decode via my upnp server which to my ears sounds fine.

+1 Simon - reminds me a little of the fine tuning debates back in the good old flat earth days; stylus profile's, bearing oil, platter mats etc.  The saving grace in those days was we we had no 'tinternweb so keyboard warriors had to write letters & that kinda damped the enthusiasm.