My comment is not meant as criticism, but just as an observation:
First, it's not only one file, but two files that have the channels swapped: In file 1 and 2 the cymbal in the very beginning is on the left, and the guitar straight afterwards is on the right, and in file 3 and 4 it's reversed.
I know this will sound a bit harsh, but if you're making a post about the readibly audible differences in cables, then I think you should at least be able to attach the cables correctly. Spending two minutes listening to the recorded files before posting them would easily solve this. Again, not to sound harsh, but it makes you seem very sloppy, rushed and careless when you're trying to present yourself as thorough and attentive to the even most minute details - meaning if you claim that cables make an audible difference but you connect them wrong, people will get the impression that either you're not as trained and attentive a listener as you claim, or that the differences you hear are simply the swapped channels, which will open you up to ridicule.
Again, this is not meant as criticism, but just observations and suggestions on how to make a better test that will satisfy even the harshest critics.
As for volume level, it's true that the files are not matched. First of all, when matching volume levels it should be RMS values, not peak levels that are matched. It should, however, be noted that in some cases volume levels need to matched by ear. This is especially the case if there's a noted difference in frequency response/EQ, especially if this difference is in the harshness region where our ears are the most sensitive.
Anyway, the volume level on these files don't match. Again, not criticism, just an observation.
Here are the numbers (note both the differences in peak values but also in RMS values):
foobar2000 1.3.7 / Dynamic Range Meter 1.1.1
log date: 2020-05-10 09:45:32
--------------------------------------------------------------------------------
Analyzed: ? / ?
--------------------------------------------------------------------------------
DR Peak RMS Duration Track
--------------------------------------------------------------------------------
DR11 0.00 dB -13.96 dB 1:06 ?-052020Comeaway1
DR12 -4.20 dB -19.64 dB 1:04 ?-052020ComeAway2
DR12 -0.10 dB -16.87 dB 1:02 ?-052020Comeaway3
DR10 -0.10 dB -17.86 dB 1:10 ?-052020ComeAway4
--------------------------------------------------------------------------------
Number of tracks: 4
Official DR value: DR11
Samplerate: 96000 Hz
Channels: 2
Bits per sample: 24
Bitrate: 4608 kbps
Codec: PCM
================================================================================
All the files were slightly different lengths, but especially File 4 was quite a lot longer, so I converted the files to wave to edit them to the same length, and when I then did the same reading of those edited wave files, this is the reading I got (again, note both peak and RMS differences):
foobar2000 1.3.7 / Dynamic Range Meter 1.1.1
log date: 2020-05-10 10:29:40
--------------------------------------------------------------------------------
Analyzed: ? / ?
--------------------------------------------------------------------------------
DR Peak RMS Duration Track
--------------------------------------------------------------------------------
DR11 0.00 dB -13.83 dB 1:02 ?-052020Comeaway1
DR12 -4.20 dB -19.45 dB 1:02 ?-052020ComeAway2
DR12 -0.10 dB -16.87 dB 1:02 ?-052020Comeaway3
DR12 -3.18 dB -18.78 dB 1:02 ?-052020ComeAway4
--------------------------------------------------------------------------------
Number of tracks: 4
Official DR value: DR12
Samplerate: 96000 Hz
Channels: 2
Bits per sample: 24
Bitrate: 4608 kbps
Codec: PCM
================================================================================
In file 4 there was a loud peak in the section that was only included at the end of this particular file, which explains the big difference in peak value between the reading of the aiff file 4 and the wave file 4.
As for cables making an audible difference, the difference cables can make are changes in frequency response and volume.
If all you did was change the cables, then volume levels could have occurred, but since it seemed like you tried to match the peak levels, this explanation goes out the window.
As for frequency response, I used Voxengo's CurveEQ to compare the files, and there are very light frequency response differences between the four. It's impossible for me to say if anybody could actually hear those differences, but except for file 3 they are all between 0.1 to 0.3 dB, so it's unlikely that this would be audible. But we could of course match the volume levels, swap the channels on two of the files and then do a blind test and see if anybody would pass, and that would be proof that there's an audible difference. I would be happy to help set this test up, and I would also be happy to participate.
I should, however, say that file 3 did seem to differ by 0.5 dB between around 2 and 12 kHz, so that particular file would be most likely to have an audible difference in a level-matched and channel-matched test.
Then I would also guess that file 3 is either the $5700 cable or the $15 cable. Since file 3 is the only cheap cable and the one that differs the most, I would imagine that file 3 is the $15 cable.
Just for the record: I find that it's perfectly okay to pay for a change in frequency response if you like the effect, as that's also partly what we're paying for when we change amplifiers, phono cartridges, etc., but I would find paying for a volume level difference to be unnecessary.
www.analogplanet.com
