[spruce music]

  From small group of participants like 4 people doing informal blind testing, none of them guessed incorrectly from 5 tries. They're elder people and not really familiar with blind testing so we meant to do that in lighted heart manner. But data we found from recording didn''t show any significant result too.
 
There're a few more things audiophiles can perceive difference but there's no measurement to back up their claim. I'm sure if you can try listening to highend equipment in good systems yourself, you'll be able to tell apart too.

So what are the details of this?  Single blind, double blind?
 
What's a good system with high end equipment?  You are doing what such audiophiles do first you impugn the equipment and if that is okay then you impugn our perceptive abilities all while having nothing concrete to back your own claims.  Makes no sense.  No logic.  NO reason, no science. 

 

[WindowsX]

As I told you before, it's informal blind test with 5 tries so it's single blind test and I told you they're listening to Esoteric P-02/D-02 sources.

 

[spruce music]

  As I told you before, it's informal blind test with 5 tries so it's single blind test and I told you they're listening to Esoteric P-02/D-02 sources.

Yes, so who did switching, how did they do it? What was the interval involved to switch?  Single blind isn't great, but better than nothing.  With provocative results why not do more, better, less formally, and more trials?

 

[WindowsX]

 
Yes, so who did switching, how did they do it? What was the interval involved to switch?  Single blind isn't great, but better than nothing.  With provocative results why not do more, better, less formally, and more trials?

 
Yes, so who did switching, how did they do it?
 
I'm the one who do switching. Asking participants to close their eyes and I play music without telling them which device is going to play.
 
What was the interval involved to switch?
 
After playing the song to nearly the end. I paused and asked if it's CD or file.
 
Single blind isn't great, but better than nothing.  With provocative results why not do more, better, less formally, and more trials?
 
There's another trial made by my partner who run highend audiophile in Thailand. The event ended with some customers not being happy with his event so he stopped. It's quite hard to do proper blind test event in Thailand. They aren't opened mind to such topics. Maybe you can try doing ones yourself.
 
Regards,
Keetakawee

 

[gregorio]

  FLAC and WAV really sound different. Many sound engineers prefer wav format over aiff/flac.

 
I'm one of those sound engineers, I much prefer wav to aiif or flac and, so do all the other sound engineers I know. That's absolutely nothing to do with any difference in sound though, because there is ABSOLUTELY NO DIFFERENCE in the sound. It's because the wav container (specifically the bwav variant) contains metadata fields which if employed correctly can turn repetitive, laborious tasks which take many hours (or even days) into tasks which take a couple of mouse clicks and a few minutes.
 
Putting a made up bit of nonsense next to an actual (but unrelated) fact and implying/stating that there is a relation which proves the nonsense is a cheap and not too bright trick to try in the science forum. Such a cheap trick might work very well in an audiophile forum but it's extremely insulting that you assume we're all stupid/ignorant enough to fall for it, here in the science forum of all places!!!
 
G
 
PS, to everyone else: What's going on in the science forum over the last few days or so? Has reddit closed some Trump forums and we're getting the overflow?

 

[asymcon]

  It's because the wav container (specifically the bwav variant) contains metadata fields which if employed correctly can turn repetitive, laborious tasks which take many hours (or even days) into tasks which take a couple of mouse clicks and a few minutes.

FLAC should be able to preserve any foreign metadata with "--keep-foreign-metadata" parameter, even BWF timestamps. Haven't tried it with REX looped material though. Well I got myself a homework. 

 
I use WavPack for project bounces, it can store markers too and comes with 32bit float support which could be helpful for additional volume reserve. Downside is, it's slower than flac-0 for encoding/decoding. 

 

[watchnerd]

   
PS, to everyone else: What's going on in the science forum over the last few days or so? Has reddit closed some Trump forums and we're getting the overflow?

 
Good question.  
 
Usually it's just us number crunches debating amongst ourselves about AES findings and such, not "introduction to audio engineering 101".

 

[asymcon]

   
Good question.  
 
Usually it's just us number crunches debating amongst ourselves about AES findings and such, not "introduction to audio engineering 101".

Consider yourself lucky, where else would you have such fun?
 

   
FLAC and WAV really sound different. Many sound engineers prefer wav format over aiff/flac. Can objectivist explain why flac and wav sounds different with data and evidence?

Sorry to be the bearer of bad news, but when something nulls out bit for bit, there's no disputable difference. 
With music, not all reserved data in WAV file are utilized. Your typical WAV has a constant bitrate of 1411.2kbps - what FLAC or other lossless encoder do is simply convert this unused space/bits which would otherwise be wasted into free space. Figuratively speaking, it removes zero-byte data from the audio stream, much like your 7zip/WinRar does, only more efficiently. The process is completely reversible and lossless, hence the term lossless encoding. 

 
When we get into non-standard formats like Wave64, xac or IEEE754 it's advisable to use lossless encoder, which can handle specifically these formats. WavPack for example can store all PCM data in any channel configuration as well as single precision floating-point and even DSD. FLAC only support up to 24bit linear PCM in specific channel matrix. 

 

[watchnerd]

  Consider yourself lucky, where else would you have such fun?

 
In a prison?

 

[gregorio]

  FLAC should be able to preserve any foreign metadata with "--keep-foreign-metadata" parameter, even BWF timestamps. Haven't tried it with REX looped material though. Well I got myself a homework. 

 
I don't know if FLAC would preserve all the bwav metadata fields. I rely on quite a few of those fields; scene, take and all the various time-stamp fields in particular. However, even if FLAC does preserve them all is not relevant because FLAC still wouldn't be preferred over wav due to it's decoding requirement. While the processing overhead of decoding a FLAC file is very small and well within the capabilities of even old computers, I'm usually dealing with at least 150 channels of audio and on rare occasions, as many as 1,000. That decoding overhead would add up and I need all my available processing power for applying audio effects processing (EQ, compression, de-essing, reverb, NR, etc.) to all those individual channels.
 
G

 

[pinnahertz]

In just a quick look, it appears the available metadata fields in FLAC wouldn't do you any good.  BWAV has metadata fields specific to production, FLAC isn't targeted that way at all, most of what you need isn't there, and a round-trip through BWAV > FLAC > BWAV should effectively wipe all your important data.  As FLAC is targeted at the consumer, and a few pro DAWs support it, not all do, and there's not much motivation to do so.  As you say, there's a need for speed, and an additional process that takes processing resources, but for which the only advantage is to conserve data storage and transmission bandwidth, doesn't seem to lend itself well to Pro applicatons.

 

[watchnerd]