Test tracks to see if Offset is audible

[Empirical Audio]

Offset is the leading nulls and trailing nulls that enclose the music data in the total data record of any track at any sample-rate. It seems logical that this should not matter for sound quality, only the data words themselves.

Maybe not so logical....

So I have made four test files available here that all have identical music data fields, but different offsets, so you can see if you can hear any difference in your system. It is interesting to determine if there is any DAC or re-clocking sensitivity here, so please list your DAC and if you use a re-clocker or if the DAC has a re-clocker inside being used. Also please list the playback software you are using.

The files include two 16/44.1 files and two 24/96 files of the same track. It is a piano track. I will not tell you yet which tracks are untouched and which have more nulls inserted into them.

Download the four tracks here:

https://www.dropbox.com/sh/g9jz9lwgvozepic/AADx1d8YLCr5YCPQl23fUkFDa?dl=0

I would like anyone interested to listen to these tracks and report back on what they hear, describing things like dryness, ringyness, attack, decay, warmth, depth or shallowness of soundstage, clarity, wooliness etc. with each of the four tracks and rate them 1-4 with 1 being the most live sounding and 4 the least.

PS: I suspect that this might be one of the reasons that one playback software sounds different from another, or differences between one release and the next.

Thanks for your help,
Steve N.
Empirical Audio

 

[cat6man]

could you please explain in some technical detail what is meant by 'offset' and 'leading nulls and trailing nulls'

 

[Empirical Audio]

There are essentially zeros, before and after any track, lots of them, sometimes thousands. Not music, just header and trailer. Changing these should not change the sound quality. We will see..

 

[cat6man]

so you added some digital silence (all zeros) at the beginning and end of the music?

 

[Empirical Audio]

so you added some digital silence (all zeros) at the beginning and end of the music?

No, it is not silence. These nulls are already there as part of the formatting. The modified tracks just have 200 more nulls added. This is the kind of thing that poor rippers get wrong or maybe CODECs. Maybe even on-the-fly decoding of some formats.

 

[cat6man]

i am not familiar with your terminology. what is the difference between a 'null' and a 'digital zero' and 'digital silence'?
i see 200 samples of zero added at the beginning of 2 files. how is this not 200 samples of silence?

even more interesting, what is it that 'poor rippers or codecs or on-the-fly decoding gets wrong?

please provide some technical details and support for what you are saying.

edit: am i being trolled here? i see no one else is responding and it is too early for April 1

 

[Empirical Audio]

Offsets vary from one track to another. You will not "hear" any silence in one track versus another one.

It is best not to examine the files to see the differences before listening to them. This creates a bias.

I am trying to determine what is wrong, DAC, CODEC, rippers etc..

 

[NorthStar]

edit: am i being trolled here? i see no one else is responding and it is too early for April 1

No you are not. There is not a more human and respectable man than Steve.
He's offering a test into the greater deaths of music exploration; nothing is more noble in the audiophile cause.

 

[cat6man]

@NorthStar
thanks for the reply.

@Steve
i find your test very very interesting and i do hear differences in files that logically should not!
something is going on here clearly, and whether or not it is system dependent is of huge interest.

what i don't understand is how/if the files differ other than the insertion of 200 samples of zero amplitude at either the beginning or end of each track. is there anything else in the 'packing' of the data that differs? i'm not sure i even understand what is meant by 'packing'.
could you please explain if there is more going on than just adding 200 samples of zero amplitude?

i have long sought (with zero success) to understand why FLAC and WAV files can sound so completely different.
perhaps this is a clue. if one file has 46734 samples of zero before the music begins and the other has 46934 samples of zero, how can that possibly impact the sound............yet it seemingly does

 

[Empirical Audio]

@cat6man
There is nothing else going on. The files have been studied on other forums. There is only 200 sample nulls added to two of the four files.

Did you rate the 4 files from best sounding to worst? This is very important to determine if you like the unchanged or the modified file. Also whether you are using USB, what DAC you are using and what software app for playback. All of this info will be helpful to identifying the smoking gun.

I have some feedback already from other forums and it's very consistent. There are of course those people that cannot hear any difference in the files. I expect that some systems are not resolving enough to detect it.

I also want to find the smoking gun with file formats. With my system, I can hear differences in every one of them, compressed or not, lossy or not.

I also want to understand why different playback software sounds different, and even different releases of the same playback software.

I want to understand why it is so hard to make USB sound stellar.

Once we understand these things I think it could eliminate a lot of variability in sound quality and maybe improve everyones sound quality.

 

[cat6man]

very repeatable for me, i prefer 1 over 3 (44k rate) and prefer 2 over 4 (96k rate).
i find it hard to compare 44k and 96k samples since my converter makes an audible click when changing rates.

digital music path in system:

WAV files on QNAP nas ==> hqplayer embedded (no processing at all) on ubuntu NUC ==> NAA on cubox in totaldac ==> totaldac re-clocker ==> totalDAC d1-direct

note: NAA on cubox ==> reclocker is generic usb cable, not Vincent's gigafilter. i sold my ultraRendu and am waiting for an opticalRendu and the sound of a plain vanilla cubox without the gigafilter is a major step down in sound from the ultraRendu.......i'll revisit test next month when my system is hopefully back up to snuff, but the differences i hear are very repeatable. compared to files 3 and 4, files 1 and 2 have more detail, separation and space, less clutter/murkiness and extended high frequencies.

edit: BLESS YOU for trying to get to the bottom of this. as an engineer, i am more than a little frustrated by things that can't possibly impact the sound IMPACTING the sound.

 

[Empirical Audio]

??
The tracks are:
frame1, frame1a, frame2, frame2a

What do you mean by 1,2,3,4?

Is that reclocker on the USB cable?

 

[cat6man]

1=frame1
2=frame1a
3=frame2
4=frame2a

1 and 3 are 44k
2 and 4 are 96k

in my test, 1 >> 3 and 2 >> 4

 

[Empirical Audio]

1=frame1
2=frame1a
3=frame2
4=frame2a

1 and 3 are 44k
2 and 4 are 96k

in my test, 1 >> 3 and 2 >> 4

Perfect, thank you very much.

 

[cat6man]

reclocker is usb in, aes/ebu out to dac

 

[Rocoa]

Hello,

I've listened the four tracks right now (eyes closed) some times and the difference is clearly audible.
The system is: Jeff Rowland Aeris DAC, Corus preamp (+PSU), Model 625 S2 power amp., Avalon speakers, Cardas Clear and Clear Beyond cables.
As source an MacBook Air. Tellurium Q Black USB cable (very analog like and time coherent cable). Between Aeris and USB cable an iFi iPurifier 3.
Audirvana Plus.

Well, the most live sound track to me is, no doubt, track 4. I can listen much better the decays, and the piano's wood. More ambience and harmonics, more body and depth, better image, perhaps less attack than previous one, but a more relaxed and natural sound to my taste.
The worst is the first track 1. The sound is more artificial, mechanic, edgy (less like a grand piano, like a clavecin as overstatement).

So, my preferences: 4, 3, 2, 1 (2a, 2, 1a, 1).

 

[cat6man]

@SteveN

have you been able to draw any conclusion regarding this experiment and why it impacts the sound?

 

[Empirical Audio]

@SteveN

have you been able to draw any conclusion regarding this experiment and why it impacts the sound?

I can only speculate. The evidence does not point to any particular smoking gun. It happens with all playback software, with both NOS and Sigma-Delta DAC's.

 

[DonH50]

I have noticed some processors handle digital "0" differently, leading to "interesting" muting of sounds, and there may be (in general are) differences in how long (how many samples) the filters take to reach steady-state... No idea if that is what is being heard.