Blue or red pill?

[manisandher]

27 minutes ago, fas42 said:

Okay, now that Mani has confirmed he can hear something in these captures, I'll start looking again for anything can distinguish them.  First step, in comparing 9._A and 10._B, the spectra are perfect copies, visually, in Audacity for matching sections.

 

I did my listening with captures 15 and 16. They're 18dB louder and have the channels the correct way around. You can get them here:

 

https://drive.google.com/open?id=1Y4TiJDFzh6YUNnB4GTnQk1vsqzKvMU9U

 

Mani.

[jabbr]

10 hours ago, sandyk said:

  I accept that you are unable to hear the differences that Mani and Peter have reported hearing too.

 

Alex, what you are claiming to hear is not the same as what Peter and Mani are hearing! You are changing the subject :

 

1) This thread is about mani’s hearing a difference when the SFS parameter is changed in the XXHE software. Now I am not claiming that my own hearing is good enough to distinguish everything @PeterSt can — and maybe he will have to send me a pair of his speakers ??? but the idea that the software parameter might afffect the “jitter signature” of the bits (or some other physical pattern such as rise time, overshoot etc etc etc) is entirely reasonable 

 

2) This thread is not about your own claim that files retain a sonic signature of their path through the internet, or history of compression/decompression etc despite the files being bit-identical. That is your own perma topic not this current topic

 

Claim #1 was demonstrated in an A/B/X test and the mechanism is still being investigated ? do we think @pkane2001 diffmaker is definitive and if so conclusions please from this @PeterSt

[fas42]

13 hours ago, manisandher said:

 

I did my listening with captures 15 and 16. They're 18dB louder and have the channels the correct way around. You can get them here:

 

https://drive.google.com/open?id=1Y4TiJDFzh6YUNnB4GTnQk1vsqzKvMU9U

 

Mani.

 

The data in 9 and 10 is effectively identical - I can swap the channels, and apply gain to match what you heard, so happy to go with those 2.

[fas42]

Just had a more serious listen to clips 9 and 10 - not at the best level for picking things, just in Audacity with clips multitracked, and soloing each, back and forth ... and to my ears I can pick a difference, over the laptop speakers. In particular, the quality of the backing piano - the B version sounds like they used a better instrument, A sounds a touch clangy and lacking in overtones; less "rich". Note, highly unoptimised playback so it doesn't mean that on a top class rig the positions won't be reversed - what matters is if at least some difference can be discerned.

[fas42]

By upsampling and finer aligning I can get an excellent nominal null between 9 and 10 for a short section of the overall clip - but this is telling me that the clock of the ADC is varying too much between the two takes - I'll have to do some very, very slight speed adjustment of one track to improve the correlation in time.

[esldude]

On 4/22/2018 at 3:27 AM, manisandher said:

 

I did my listening with captures 15 and 16. They're 18dB louder and have the channels the correct way around. You can get them here:

 

https://drive.google.com/open?id=1Y4TiJDFzh6YUNnB4GTnQk1vsqzKvMU9U

 

Mani.

For what it is worth, using the last half of those tracks, Diffmaker gives null depth of 91.1 db and 102.7 db for each channel.  I only used the last half because I was working on an older laptop with limited RAM.

[esldude]

I managed to get the entire track to work in Diffmaker by relaxing sample rate correct.  97.2 db left and 96.6 db right channel. 

 

When amplified 75 db you hear smooth noise with the music barely audible deep in the noise.  This with track 15 and 16. 

 

With 13 and 14 null depth was 91.1 and 92 db.  Amplified noise was smooth noise with no trace of the music.  

 

 

[fas42]

Silly bugger me ... I was encouraged by Dennis getting somewhere with Diffmaker, so foolishly gave it a try - which I hadn't done up to now. Well, I was immediately reminded of how it's such an appalling app citizen - it threw my Windows 8.1 laptop into a frenzy of memory and disk thrashing, which effectively froze the machine - it was working, on DiffMaker, but I couldn't see what was going on - impossible to bring up Task Manager. It finally proclaimed that it couldn't digest the tracks, but kept on grinding anyway. Finally, it crashed - but then getting it to really quit somehow caused Firefox browser to crash as well ... ... will I persevere with it, trying to work out how to cajole and tickle it under the chin, to get anything? ... hmmm, I think not ...

[PeterSt]

2 hours ago, esldude said:

 

When amplified 75 db you hear smooth noise with the music barely audible deep in the noise.  This with track 15 and 16. 

 

I have been working on it as well last Sunday and saw that the Motu too caries some noise. Say that this noise is of similar exhibit as the Tascam but way way lower. At first (in the lower level tracks Mani put up) I did not even recognize it as noise.

Dennis, do you think it is possible that this noise alone can make the Diff express music (the way you heard it) ?

I tried to see a different noise pattern in both tracks but although I thought I could see differences in one over the other, I am far from sure it is dedicated to one of the tracks only. So, sure, the differences are there regarding this noise (which is logical in itself) but can it dig up the music as such ? What is your experience ?

[esldude]

11 minutes ago, PeterSt said:

 

I have been working on it as well last Sunday and saw that the Motu too caries some noise. Say that this noise is of similar exhibit as the Tascam but way way lower. At first (in the lower level tracks Mani put up) I did not even recognize it as noise.

Dennis, do you think it is possible that this noise alone can make the Diff express music (the way you heard it) ?

I tried to see a different noise pattern in both tracks but although I thought I could see differences in one over the other, I am far from sure it is dedicated to one of the tracks only. So, sure, the differences are there regarding this noise (which is logical in itself) but can it dig up the music as such ? What is your experience ?

Yes you'll get this with music in the noise pretty often.  I think the noise corrupts level matching.  Let us say it uses the peak in the track to set level.  Well the peak in track A is signal plus a little higher noise while the peak in track B is the same only this time a little less noise.  It makes those the same and the noise prevents perfect level matching.  In this case (and most of the time with Diffmaker) the residual music does NOT have a 6db/octave upward slope.  That upward slope is a sure sign of residual timing mismatch.  So with Diffmaker timing is matched very well typically and level may be the tiniest bit off which leaves this music residual mixed in with noise. 

 

I'd think this is quite often the dither the ADC does.  That is intended to be random anyway, and is higher at the ultrasonic frequencies.  

[PeterSt]

Thank you Dennis.

[fas42]

Very good to see Dennis and Peter working on this as well! Something of interest is that I earlier manually got the tracks to reach a deep null, for about 5 seconds in towards the middle, and I just spent a little bit listening to that section of the null, in comparison to the B track. The slightest thing that comes through is, guess what, the sibilance of the singer - now, I may find that disappears as I improve the correlation for longer through the track - to be continued ...

[STC]

This is so much of a trouble to enhance sound fidelity. More of engineering curiosity. It will be a point of interest though for audiophile gathering. 

[fas42]

A lot more than that, IMO - the holy grail is to be able to identify the actual variations in the output of real life equipment, that can be subtle, but very important for people who take listening to music
"seriously" - this has been an ongoing tussle, and it's worth getting one's hands dirty somewhat, trying to get a handle on things.

[STC]

I don’t think anyone here can positively identify which one of the parameter really improves the SQ. 

[Andyman]

4 hours ago, STC said:

I don’t think anyone here can positively identify which one of the parameter really improves the SQ. 

I suspect at least one (double) Dutch man can...

[pkane2001]

12 hours ago, STC said:

I don’t think anyone here can positively identify which one of the parameter really improves the SQ. 

 

At least in theory, this can be done with the tests conducted in this thread.

 

Since we have two analog captures with audible differences (A and B), and bit-perfect digital captures, it should be possible to use diffmaker (if only it would work well!) to subtract digital capture from both analog captures. Whichever delta has the lowest amount of noise/harmonic distortion would be the one that improves the SQ. Since the only difference between A and B was a different SFS setting, we'd know which SFS setting produces the best SQ. In theory...

 

[esldude]

3 minutes ago, pkane2001 said:

 

At least in theory, this can be done with the tests conducted in this thread.

 

Since we have two analog captures with audible differences (A and B), and bit-perfect digital captures, it should be possible to use diffmaker (if only it would work well!) to subtract digital capture from both analog captures. Whichever delta has the lowest amount of noise/harmonic distortion would be the one that improves the SQ. Since the only difference between A and B was a different SFS setting, we'd know which SFS setting produces the best SQ. In theory...

 

I'll try that when I can.  I'll predict they'll be similar in null depth around -40 dB.

[esldude]

Just did a quick run with digital vs analog.  I only have 176 rate digital captures.  I resampled those to 192 khz.  

 

Digital vs analog gave the following results. 

 

(INV)parameters: -30.44msec, 5.809dB (L),  5.855dB (R). rate adj=19.023 ppm.Corr Depth: 24.6 dB (L), 22.3 dB (R)

(INV)parameters: -30.66msec, 5.814dB (L),  5.860dB (R). rate adj=19.0289 ppm.Corr Depth: 24.5 dB (L), 22.2 dB (R)

[pkane2001]

2 minutes ago, esldude said:

Just did a quick run with digital vs analog.  I only have 176 rate digital captures.  I resampled those to 192 khz.  

 

Digital vs analog gave the following results. 

 

(INV)parameters: -30.44msec, 5.809dB (L),  5.855dB (R). rate adj=19.023 ppm.Corr Depth: 24.6 dB (L), 22.3 dB (R)

(INV)parameters: -30.66msec, 5.814dB (L),  5.860dB (R). rate adj=19.0289 ppm.Corr Depth: 24.5 dB (L), 22.2 dB (R)

 

Close. Would be interesting to see the spectrum plot for both A-digital and B-digital to see if there's any obvious difference..

 

[esldude]

13 minutes ago, pkane2001 said:

 

Close. Would be interesting to see the spectrum plot for both A-digital and B-digital to see if there's any obvious difference..

 

In this case neither level nor timing was well matched.  So it sounds tinny and has the original spectrum mostly with a 6 db/octave tilt added to it. So this one needs some work. I'll look at it again later when I have time. 

[manisandher]

I've now set things up so that I can make digital and analogue captures simultaneously, and perhaps sync the clocks of the two capturing devices as well. I'm working away from home for the rest of this week, but if anyone feels it'd be useful, I'll do this on my return.

 

Mani.

[pkane2001]

46 minutes ago, manisandher said:

I've now set things up so that I can make digital and analogue captures simultaneously, and perhaps sync the clocks of the two capturing devices as well. I'm working away from home for the rest of this week, but if anyone feels it'd be useful, I'll do this on my return.

 

Mani.

 

Certainly would be another useful datapoint, Mani.

 

I’m still hoping to do some comparisons, but have had little luck with diffmaker and my own rewrite is coming very slowly.

[austinpop]

On 2/9/2018 at 3:38 PM, manisandher said:

My key take-away from this experience is that the 'red pill' is way more bitter than I could ever have imagined, for those who are entrenched in their beliefs. No amount of evidence contrary to their beliefs seems to help.

 

 

Thanks for your summary in the first post, and for your patience and forbearance during the exercise.

 

Your conclusion is unfortunate, but certainly not surprising.

[fas42]

4 hours ago, pkane2001 said:

 Whichever delta has the lowest amount of noise/harmonic distortion would be the one that improves the SQ. Since the only difference between A and B was a different SFS setting, we'd know which SFS setting produces the best SQ. In theory...

 

 

I strongly suspect i's going to be very much less clearcut than that - the key artifacts may be buried in noise, and hence dismissed as being irrelevant; but it's already very well understood that the human hearing system is remarkably adept at extracting information within noise. This is usually expressed in a positive context, but that doesn't stop it also being true in a negative sense - bury a repeating, very irritating sound glitch in random noise, and people's hearing will register its presence.