What is this?

[matthewacbroad]

SPL = sound pressure level (i.e. "volume")

Yeh, that's what I thought - that you could actually hear it when listening to the recording. Just double-checking because it's the kind of thing that most adults (and probably many children) couldn't hear. No doubt the engineers who recorded, mixed and mastered the track couldn't hear it, otherwise it wouldn't be there in the first place.

In that case I'd try my suggestion in post #17.

thanks.
I notice it more in quiter spots and start-ending when its guitar only (no drums or vocals). What makes it so noticable for me its off tempo and short sound. It feels like a sensation like a tickling of my inner ear if that makes sense at all...

Ill read more about audacity from that link provided, thanks for the help.

 

[matthewacbroad]

Messing around with in audacity I applied EQ as high pass filter @ 20Khz and pressed play:

- Default track volume (w/ max dac volume) = short tiny burst of high pitch, if a sample that got cut off
- Amplification without clipping (+23db) (w/ max dac volume) = Hear long high pitch sweeps and makes my heart beat faster

However I refuse to believe that I'm hearing this "V" sweep my only guess is harmonics or saturation into 17-18khz range where my hearing is more detailed?

 

[AnalogSteph]

At this point I think you should give the intermod test files from this page a shot. Either you have some darn good hearing or a resampling problem somewhere. Last time I heard the test tones loud and clear it set me on the trail of some sound driver issues that apparently caused downsampling of clearly less than ideal quality. My speaker setup is making them somewhat audible at higher levels, with headphones I really have to be pushing the levels to hear much of anything.

 

[Blumlein 88]

What I see in the fft looks like resampling artifacts. Maybe from processing if only this track. I too suggest a notch filter at 21 khz. I say choose a Q of 6. There's a Notch filter option under Effects.

 

[matthewacbroad]

At this point I think you should give the intermod test files from this page a shot. Either you have some darn good hearing or a resampling problem somewhere. Last time I heard the test tones loud and clear it set me on the trail of some sound driver issues that apparently caused downsampling of clearly less than ideal quality. My speaker setup is making them somewhat audible at higher levels, with headphones I really have to be pushing the levels to hear much of anything.

Fiio X3 v1 @ 120/120 Volume (+6db gain @ OS-menu)

30kHz tone + 33kHz tone (24 bit / 96kHz): Slight High Pitch Tone
26kHz - 48kHz warbling tones (24 bit / 96kHz): Quiet Warbling
Song clip shifted up by 24kHz (24 bit / 96kHz WAV): Hear some type of shaker

Warbling came out to be the highest volume of the 3 to my ears. How does one fix audible intermodulation on a DAC?


EDIT: added low-pass @ 24000hz in JRiver & after repeating these test (same volume) I cannot hear it anymore. Would it be recommended I select 48Khz 24-bit instead of 96khz to avoid any distortion from DAC options?

"Not encoding such a wide frequency range to begin with. You can't and won't have ultrasonic intermodulation distortion in the audible band if there's no ultrasonic content."


EDIT #2: 24-bit 48Khz in sound options = no noises coming from test files

 

[Guermantes]

What I see in the fft looks like resampling artifacts. Maybe from processing if only this track. I too suggest a notch filter at 21 khz. I say choose a Q of 6. There's a Notch filter option under Effects.

It does look like aliasing but I notice that it seems to generally occur when the vocals are in the mix as it's not in the intro and not in the guitar break at 40 secs. I wonder if it is something on the multitrack -- maybe some phaser effect with ultrasonic artefacts that has been applied to the vocals and bounced to tape. Then on mixdown the engineer has punched the vocal tracks in and out which is why it's intermittent.

Curious

 

[matthewacbroad]

It does look like aliasing but I notice that it seems to generally occur when the vocals are in the mix as it's not in the intro and not in the guitar break at 40 secs. I wonder if it is something on the multitrack -- maybe some phaser effect with ultrasonic artefacts that has been applied to the vocals and bounced to tape. Then on mixdown the engineer has punched the vocal tracks in and out which is why it's intermittent.

Curious

If what I read up "Tom Scholz" is a engineer and played almost all instruments in the band. He wanted to record the album in his own equipment built basement studio. I assume he would of known of this artifact or something he missed hand building equipment?

(skip to 5:54)

 

[Blumlein 88]

If what I read up "Tom Scholz" is a engineer and played almost all instruments in the band. He wanted to record the album in his own equipment built basement studio. I assume he would of known of this artifact or something he missed hand building equipment?

(skip to 5:54)

Maybe maybe not. He used a DAW and likely plugins. Those don't always work cleanly. He may not have heard it. So it may not have come to his attention.

 

[matthewacbroad]

Maybe maybe not. He used a DAW and likely plugins. Those don't always work cleanly. He may not have heard it. So it may not have come to his attention.

Sorry, maybe I missed something or not understanding but the song was recorded in 1976 so DAW wasn't around? I assume it was most analog

 

[Blumlein 88]

Sorry, maybe I missed something or not understanding but the song was recorded in 1976 so DAW wasn't around? I assume it was most analog

My mistake. But it had been thru some digitizing which had something weird going on digitally.

 

[Guermantes]

We don't really have any details about the provenance of the track under analysis. What particular hi-res release is it? Has it been "remastered" or even "remixed" for release? Or is it a flat transfer from a master tape?

 

[matthewacbroad]

We don't really have any details about the provenance of the track under analysis. What particular hi-res release is it? Has it been "remastered" or even "remixed" for release? Or is it a flat transfer from a master tape?

can't speak for the hi-res file but I did buy this album from a thrift shop long time ago and YES it's there even with 44.1Khz

EAC (Exact Audio Copy) -> Uncompressed Wav (16-bit)

 

[Guermantes]

can't speak for the hi-res file but I did buy this album from a thrift shop long time ago and YES it's there even with 44.1Khz

Great! Now all we need is a rip of the original vinyl release to confirm if it was on the analogue master

 

[Guermantes]

I found another CD copy of this album with catalogue number CDEPC 81611. Same artefact present.

I also found this at the 54 second mark which looks like evidence of an edit on the master or even the multitrack:

 

[Blumlein 88]

Somehow, between posting on a phone etc. I lost track of what we were looking at as in it being Boston. I've that album and it shows some of these artifacts on tracks #2, #4 and #7. Interesting to say the least.

 

[Blumlein 88]

https://www.guitarworld.com/features/interview-tom-scholz-recalls-making-boston-and-don-t-look-back

Okay, so if you have the Scholz remaster, he made those from the original two track master tape according to the above. If those are in his remaster, then I take it they were on the LP. I see them in the Epic very early CD I have of it.

 

[Guermantes]

@Blumlein 88 Yes, you're right. In track #7 they are all over the place with little correlation to anything. I suspect some homemade processor of Scholz's is the culprit.

 

[Guermantes]

https://www.guitarworld.com/features/interview-tom-scholz-recalls-making-boston-and-don-t-look-back

Okay, so if you have the Scholz remaster, he made those from the original two track master tape according to the above. If those are in his remaster, then I take it they were on the LP. I see them in the Epic very early CD I have of it.

I suspect the 24-96 version may be from the remaster . . .

 

[matthewacbroad]

@Blumlein 88 Yes, you're right. In track #7 they are all over the place with little correlation to anything. I suspect some homemade processor of Scholz's is the culprit.

holy cow track 7 is weird it's more detailed. This part is noticeable for me.

IF you can't hear that high just lower the sample frequency to 48Khz or "rate -1.0x" which pushes these artifacts into 10-15Khz range.

 

[Blumlein 88]

Okay, earlier you filtered out everything below 20 khz. You can also change speed in audacity. I can hear it if I slow it to half speed. Even better at 1/3 speed. @matthewacbroad that you can hear this is insanely good hearing. Maybe we have a hiresolution test listener for the forum.

I've got old ears so I couldn't hear it without slowing it down. Slowed down it almost sounds like tape catching and then letting go like a high speed scrape flutter. But scrape flutter normally isn't found at this frequency.