Hearing Music Before The Music

[The Computer Audiophile]

Hi Guys, since I received the RAAL-requisite SR1a headphones and I've been listening to some Three Blind mice jazz, I've really noticed (more than usual) that I'm hearing music before the music is played. Yes, this sounds crazy but bare with me. 

 

For example, on the track Stella by Starlight from the Terumasa Hino Quintet's Live! album (TBM-17), there is clearly audible trumpet playing a second or two before the actual notes come through at full volume. It's like the recording is in the control room but the sound is coming through the glass before it travels through the soundboard. I know this is a primitive explanation, but I'm at a loss. I've heard something similar on Shelby Lynne's Tears Lies and Alibis. 

 

Can anyone shed light on what this is? Why this happens?

 

@astrotoy do you have any idea? I've heard it's related to recording on tape and think you might know (fingers crossed).

 

Here is the track I'm talking about, but it isn't the TBM version and it's YouTube quality. 

 

 

[Speedskater]

Analog tape recordings sometimes have bleed-thru problems like this.

[The Computer Audiophile]

1 minute ago, Speedskater said:

Analog tape recordings sometimes have bleed-thru problems like this.

So it's actually recored to the tape and it goes through to another "layer" of the tape? One would think this would create an echo effect after the initial playing of the instrument, not before it. But, as I said, I have no clue. 

 

Thanks!

[Kal Rubinson]

50 minutes ago, The Computer Audiophile said:

So it's actually recored to the tape and it goes through to another "layer" of the tape? One would think this would create an echo effect after the initial playing of the instrument, not before it. But, as I said, I have no clue. 

When wound, each recorded portion affects both adjacent layers, the one wound inside it and the one wound outside it. Perhaps more if left tightly wound for very long periods of time.  

[The Computer Audiophile]

4 minutes ago, Kal Rubinson said:

When wound, each recorded portion affects both adjacent layers, the one wound inside it and the one wound outside it. Perhaps more if left tightly wound for very long periods of time.  

Wow. Thanks!

[astrotoy]

Kal and Speedskater are correct.  This is normally called pre-echo and results in having the tape tightly wound in the heads out position, that is ready to play.  Normally one stores tapes in the tails out position after playing - not tightly wound. If you fast wind the tape to the heads out position and store it that way so it is ready to play out of the box, then you can get pre-echo.  If you store the proper way, tails out, after playback, if there is any transfer of signal from one layer to the next it is buried within the sound of the music, not before it.

 

Larry

 

[Ralf11]

1 hour ago, Kal Rubinson said:

When wound, each recorded portion affects both adjacent layers, the one wound inside it and the one wound outside it. Perhaps more if left tightly wound for very long periods of time.  

 

yes, there's a name of this - IIRC it's called print-thru

[gmgraves]

3 hours ago, The Computer Audiophile said:

Hi Guys, since I received the RAAL-requisite SR1a headphones and I've been listening to some Three Blind mice jazz, I've really noticed (more than usual) that I'm hearing music before the music is played. Yes, this sounds crazy but bare with me. 

 

For example, on the track Stella by Starlight from the Terumasa Hino Quintet's Live! album (TBM-17), there is clearly audible trumpet playing a second or two before the actual notes come through at full volume. It's like the recording is in the control room but the sound is coming through the glass before it travels through the soundboard. I know this is a primitive explanation, but I'm at a loss. I've heard something similar on Shelby Lynne's Tears Lies and Alibis. 

 

Can anyone shed light on what this is? Why this happens?

 

@astrotoy do you have any idea? I've heard it's related to recording on tape and think you might know (fingers crossed).

 

Here is the track I'm talking about, but it isn't the TBM version and it's YouTube quality. 

 

 

Sure can. Most of the time it’s caused by a phenomenon called “print through” which occurs when analog tape is not stored correctly. What happens is that the magnetic domains that represent the recorded sound bleed through the tape from one layer to the next when the tape is stored wound on a reel. Professional recording engineers store analog tapes “tails out”, meaning that the tape must be rewound before playback. This causes the print through “ghost” to appear AFTER the actual music note rather than before it. This way, the print through blends with the venue’s natural reverb and becomes part of it and thus usually goes unnoticed. Sometimes, people forget to wind the tape in a tails out storage position and a pre-ghost image of the music appears at the head of a performance, rather than slightly after it.

On the other hand, if the masters of your recording were made digitally, the answers are not so simple and I don’t know if digital tape prints through or not. But my gut tells me that it wouldn’t work like analog print through and thus wouldn’t get transferred to other digital media or decoded by a DAC. 

[fas42]

3 hours ago, Kal Rubinson said:

When wound, each recorded portion affects both adjacent layers, the one wound inside it and the one wound outside it. Perhaps more if left tightly wound for very long periods of time.  

 

With the latest comments, how does this post stack up?  There will always be a layer of tape on the inside, and a layer outside. any point in the tape - what makes the print through occur in only one direction, if that indeed is what happens?

[Kal Rubinson]

1 hour ago, fas42 said:

what makes the print through occur in only one direction, if that indeed is what happens?

It doesn't although it may not be the same in both directions.

[John Dyson]

3 hours ago, gmgraves said:

Sure can. Most of the time it’s caused by a phenomenon called “print through” which occurs when analog tape is not stored correctly. What happens is that the magnetic domains that represent the recorded sound bleed through the tape from one layer to the next when the tape is stored wound on a reel. Professional recording engineers store analog tapes “tails out”, meaning that the tape must be rewound before playback. This causes the print through “ghost” to appear AFTER the actual music note rather than before it. This way, the print through blends with the venue’s natural reverb and becomes part of it and thus usually goes unnoticed. Sometimes, people forget to wind the tape in a tails out storage position and a pre-ghost image of the music appears at the head of a performance, rather than slightly after it.

On the other hand, if the masters of your recording were made digitally, the answers are not so simple and I don’t know if digital tape prints through or not. But my gut tells me that it wouldn’t work like analog print through and thus wouldn’t get transferred to other digital media or decoded by a DAC. 

Digital print-through doesn't happen for a few reasons -- one is that digital tape is more difficult to record/erase so changing the bits is pretty difficult (also the encoding isn't amenable to print thorugh.)   The other reason is that since the signal is digital, a mutual interfernece, if it really did happen, would be error corrected.

The encoding methods just dont add/subtract with each other in the same way as analog.

On analog tape, the signal only has to be modified a little bit to be noticeable, but digital signals are usually more "sticky" and need more gross, large scale changes to be noticeable.

 

John

 

[astrotoy]

Print through occurs primarily (overwhelmingly) on one side of the tape.  Tape has two sides.  One side is coated with iron oxide - that is the side which faces the inside of the reel and faces the heads of the tape recorder and can record music by lining up the iron oxide particles with the record head of the tape recorder. The other side is usually coated with a protective non magnetic coating (so called coated tapes which have been the norm for professional tapes for decades).  When tape is tightly wound and stored the inside oxide side of the tape can pick up the magnetic field of the tape that lies below it.  Since there is no oxide on the other side of the tape, the coated side, that side acts as a protective buffer against picking up the magnetic field from the tape that lies above it and there is little or no effect.  That is why if you play the wrong side of the tape, the signal is very weak. 

 

When the head side (where the tape starts) is at the outside of the reel, the tape has no signal recorded on it, that is, it is quiet for the first few seconds of the tape playback. If the tape has been tightly wound and stored so that the tape layers are in very close contact, the tape may have picked up the signal of the music lying in the layer below it, so that when played back there is a "pre-echo" a faint playing of the music a couple of seconds before the music actually starts.  A professional reel is about 10 inches in diameter (circumference is about 30 inches - 3.14 x 10).  The speed of the tape is 15ips - so the layer above comes across the tapehead about 2 seconds before the layer below - hence the pre-echo occuring about 2 seconds before the music starts.  Since the print through signal is at a much lower level than the music, one can only hear it when there is otherwise silence.  In an album with several songs and a few second gap before each song, the pre-echo can be heard in the silence - with a slightly shorter delay the further into the reel one goes. 

 

If the tape is properly left tails out after playing and not wound tightly, then there is both less print through since the layers are not so tightly packed and the echo is a post-echo, buried in the music, with the same delay, but with a much lower volume.  It is possible that you might hear a post echo at the very end of the tape, but then the music is usually fading out and the signal is getting weaker and any post echo is difficult to hear. 

 

So in the case of Chris' files, my presumption is that the tapes used to make the files had some pre-echo from improper storage.  The pre-echo was captured by the digital transfer and not edited out, resulting in the sounds Chris hears.  BTW, if you have four track stereo tapes, sometimes one can hear a faint signal of the sound of the adjacent tracks (for example tracks 2 and 4 when you are playing tracks 1 and 3). They are much fainter and running backwards to the tracks you are playing.  They typically are noticeable when the passage you are playing is very quiet and the opposing tracks are very loud (much like hearing your neighbor's children practicing the drums or trumpet in the adjoining apartment.)

 

Any professionals or other knowledgeable folks please correct anything in my post.

 

Larry

[The Computer Audiophile]

3 minutes ago, astrotoy said:

Print through occurs primarily (overwhelmingly) on one side of the tape.  Tape has two sides.  One side is coated with iron oxide - that is the side which faces the inside of the reel and faces the heads of the tape recorder and can record music by lining up the iron oxide particles with the record head of the tape recorder. The other side is usually coated with a protective non magnetic coating (so called coated tapes which have been the norm for professional tapes for decades).  When tape is tightly wound and stored the inside oxide side of the tape can pick up the magnetic field of the tape that lies below it.  Since there is no oxide on the other side of the tape, the coated side, that side acts as a protective buffer against picking up the magnetic field from the tape that lies above it and there is little or no effect.  That is why if you play the wrong side of the tape, the signal is very weak. 

 

When the head side (where the tape starts) is at the outside of the reel, the tape has no signal recorded on it, that is, it is quiet for the first few seconds of the tape playback. If the tape has been tightly wound and stored so that the tape layers are in very close contact, the tape may have picked up the signal of the music lying in the layer below it, so that when played back there is a "pre-echo" a faint playing of the music a couple of seconds before the music actually starts.  A professional reel is about 10 inches in diameter (circumference is about 30 inches - 3.14 x 10).  The speed of the tape is 15ips - so the layer above comes across the tapehead about 2 seconds before the layer below - hence the pre-echo occuring about 2 seconds before the music starts.  Since the print through signal is at a much lower level than the music, one can only hear it when there is otherwise silence.  In an album with several songs and a few second gap before each song, the pre-echo can be heard in the silence - with a slightly shorter delay the further into the reel one goes. 

 

If the tape is properly left tails out after playing and not wound tightly, then there is both less print through since the layers are not so tightly packed and the echo is a post-echo, buried in the music, with the same delay, but with a much lower volume.  It is possible that you might hear a post echo at the very end of the tape, but then the music is usually fading out and the signal is getting weaker and any post echo is difficult to hear. 

 

So in the case of Chris' files, my presumption is that the tapes used to make the files had some pre-echo from improper storage.  The pre-echo was captured by the digital transfer and not edited out, resulting in the sounds Chris hears.  BTW, if you have four track stereo tapes, sometimes one can hear a faint signal of the sound of the adjacent tracks (for example tracks 2 and 4 when you are playing tracks 1 and 3). They are much fainter and running backwards to the tracks you are playing.  They typically are noticeable when the passage your\ are playing is very quiet and the opposing tracks are very loud (much like hearing your neighbor's children practicing the drums or trumpet in the adjoining apartment.)

 

Any professionals or other knowledgeable folks please correct anything in my post.

 

Larry

Wow, thank you so much Larry. 

[fas42]

11 minutes ago, astrotoy said:

When tape is tightly wound and stored the inside oxide side of the tape can pick up the magnetic field of the tape that lies below it.  Since there is no oxide on the other side of the tape, the coated side, that side acts as a protective buffer against picking up the magnetic field from the tape that lies above it and there is little or no effect. 

 

Forgive me for being obtuse, but I still can't register why the "protective buffer" is one-way in its action - is it a special material in its behaviour in this regard?

[Le Concombre Masqué]

2 hours ago, John Dyson said:

Digital print-through doesn't happen for a few reasons -- one is that digital tape is more difficult to record/erase so changing the bits is pretty difficult (also the encoding isn't amenable to print thorugh.)   The other reason is that since the signal is digital, a mutual interfernece, if it really did happen, would be error corrected.

The encoding methods just dont add/subtract with each other in the same way as analog.

On analog tape, the signal only has to be modified a little bit to be noticeable, but digital signals are usually more "sticky" and need more gross, large scale changes to be noticeable.

 

John

 

how about the ghost orchestral music in Gould's 1981 digital recording of the Goldberg ; where does it come from ?

[Norton]

An interesting and informative thread.

 

I’ve noticed the same phenomenon in the lead in to certain (analogue) LP tracks.  Up till now I thought it was maybe a pressing fault or possibly anti-skate being out, but from the above I presume it was most likely a fault on the master tape?

[astrotoy]

One side of a tape is coated with iron oxide.  That means that side is going to record whatever signal is fed to it.  The other side normally has a protective coating that won't record the signal above it.  There can be some leakage through the coating.  However, the key to pre-echo - what Chris is hearing, is that this is at the beginning of the tape where there is no signal in the tape for the first several winds of the tape.  So there is no signal above this blank tape to record.  However, there is a signal starting at some point below the blank tape. In the case of a 15ips 2 track tape on a 10.5 inch reel (the vast majority of master tapes used for these recordings) it is about 2 seconds before the music starts (one layer of tape) which can record the pre-echo.  This is most easily heard if the tape has been stored heads out after doing a fast rewind and stored that way for a long time.  This gives the time for the tape below with the signal to print through to the layer above - particularly when the beginning of the music in the tape is loud.  There may be a pre-echo in tapes with very quiet beginnings, but that is so quiet, it cannot easily be heard. 

 

Larry

[John Dyson]

3 hours ago, Le Concombre Masqué said:

how about the ghost orchestral music in Gould's 1981 digital recording of the Goldberg ; where does it come from ?

I would have to see information on the entire signal chain to figure out that one.   All I know is that digital tape, by virtue of the fact that the material is recorded in an encoded form that such interference, if big enough to interfere, would not manifest as echo, but instead would be a full-on error, possibly would be corrected up to a point.  After correction failure, then it would be a glitch -- and would seldom happen.

 

It is possible that such a digital recording might have been stored for some time on analog, or ther could have been other things going on in the signal path.   Early on, stuff wasn't always 100% digital.   Also, however unlikely, an LP is more likely to create echo than any reasonable digital tape scheme itself.

When commenting on this question, my answer was a very narrow, specific answer ONLY about digital tape.  The signal paths can be more complex than just the digital tape.

 

John

 

[fas42]

Not having seen a response yet that clarified the value of the tails out storage, I did some further digging, and found this very informative bulletin, http://www.aes.org/aeshc/docs/3mtape/printthrough.pdf.

 

In particular, this explained the reason clearly,

 

Quote

Normally, the outer layers of tape from the printing signal will receive more print-through than the inner layers, since the signal must reach the top of the oxide coating to be noticeable. To reach the top surface of an outer layer of tape, that signal's magnetic field needs only to pass through the base material. To reach the top surface of an inner layer of tape, it must pass through both the base material and the oxide. So by storing tape tails out, the strongest print-through will be postprint

 

 

[gmgraves]

12 hours ago, John Dyson said:

Digital print-through doesn't happen for a few reasons -- one is that digital tape is more difficult to record/erase so changing the bits is pretty difficult (also the encoding isn't amenable to print thorugh.)   The other reason is that since the signal is digital, a mutual interfernece, if it really did happen, would be error corrected.

The encoding methods just dont add/subtract with each other in the same way as analog.

On analog tape, the signal only has to be modified a little bit to be noticeable, but digital signals are usually more "sticky" and need more gross, large scale changes to be noticeable.

 

John

 

That’s pretty much what I figured. I have limited experience recording on digital tape. In the early Eighties, I recorded using a modified Sony digital encoder to a Beta VCR, but I went to DAT as soon as possible and then to HDD and Solid State memory with a Korg MR-2000 and a Korg MR1. I’ve never heard any digital print through, and figured that it wouldn’t be a problem. Thanks for the info John.

[astrotoy]

8 hours ago, fas42 said:

Not having seen a response yet that clarified the value of the tails out storage, I did some further digging, and found this very informative bulletin, http://www.aes.org/aeshc/docs/3mtape/printthrough.pdf.

 

In particular, this explained the reason clearly,

 

8 hours ago, fas42 said:

 

Thanks for the article.  Explains things very well.  I didn't know that rewinding before playing (necessary when the tapes are stored tails out) dissipates much of the print through, even though it is buried in the music.  I just started playing my 1/2" safety master of Kind of Blue.  No pre echo that i could hear.  

 

Larry

[Skip Pack]

I've heard it on lots of LPs starting back it th '70s. While I think it's only on a minority of the recordings, it wasn't a tiny minority.

The pre-echo is probably more audible simpliny because it's has no other 'musical' sound at the same time. The post echo would be masked in almost every instance.

[gmgraves]

1 hour ago, Skip Pack said:

I've heard it on lots of LPs starting back it th '70s. While I think it's only on a minority of the recordings, it wasn't a tiny minority.

The pre-echo is probably more audible simpliny because it's has no other 'musical' sound at the same time. The post echo would be masked in almost every instance.

I think that pre-echo on LPs has a different source. The first couple of grooves in a record “cut” are not modulated - in other words, silent. But when the music starts, the groove modulation can actually distort the adjacent un-modulated grooves. The distortion disappears from the subsequent grooves because the cutting stylus “erases” as it cuts the next groove.