It's time for Saw to rule the world (ProTools 9) !!

[Tom Roberts]

Bright observation.
Did you ever engineer at Quad Recording studios?

Sorry, not familiar with that name.

Thomas

[Dave Labrecque]

By the way if someone finally comes up with a sample accurate solution for this problem please let me know - I have a lot of free wheeling multi track tapes that I'd like to sync up...

Sounds like Cary's post about the guys using the bias frequency recorded on the tape is the ticket, no?

[Tom Roberts]

Analog tape synchronization has never been, and never will be "sample accurate". It's not possible. There are no digital samples involved. But it has been accurate enough for music or sound to picture when done correctly. Back in the day, the test was to play back both machines locked together and pull up faders of identical sounds. If there was phase cancellation, or more accurately, a "very phased out swishy sound", you were right on the money.

I don't know how many microseconds or milliseconds that was, but certainly close enough, and as good as it could get. Probably less than some of the latency we deal with today with digital audio.

It's the same story today with syncing analog and digital. Not possible to be sample accurate. Only digital and digital sync can be sample accurate, and should be if done correctly.

Thomas

[Dave Labrecque]

I've been in touch with a friend and former colleague who was a young tape-op at Abbey Road in the late 60s. He has read this thread with interest, and passed on the following information:

"Cary has it about right. I can recall the method because it was me doing it, night after night. A 50Hz hum was recorded on Tk 4 of the first machine. This was routed through a Vortexion 200 watt tube amplifier and used to drive the capstan of the slave machine. A further four tracks could then be recorded on the slave. During the first pass only, the count in from the master was also transferred. The trick was to judge the preroll required by the slave on each start up. It was necessary to chinagraph the tape on one of the pre-rollers as a guide but it wasn't exact. You got better with practice. After the first track was recorded it got more difficult. To play that back and be able to proceed with further overdubs the machines had to start up in reasonable synch.....I remember one had cans with the two machines and the studio feed. As soon as you got to the count in you knew: about 3 times out of 10 I'd say it was OK, but otherwise it was "s**t or f**k" from the studio! It could get quite stressful if you got several misses in a row. Stuck in the corner of No.2 control room with 2 J37s and the Vortexion amp with no air conditioning and those sodding jossticks was not hugely pleasant. I'm sure you could have cooked breakfast on the deck plates by the end of the night."

Always good to hear this kind of stuff first-hand

Yikes. Three times out of ten. That sounds like a lot of Beatle cussing! Not a fun time to be a tape op.

Great to hear the inside scoop. Thanks, Mark.

[Dave Labrecque]

Analog tape synchronization has never been, and never will be "sample accurate". It's not possible. There are no digital samples involved. But it has been accurate enough for music or sound to picture when done correctly. Back in the day, the test was to play back both machines locked together and pull up faders of identical sounds. If there was phase cancellation, or more accurately, a "very phased out swishy sound", you were right on the money.

I don't know how many microseconds or milliseconds that was, but certainly close enough, and as good as it could get. Probably less than some of the latency we deal with today with digital audio.

It's the same story today with syncing analog and digital. Not possible to be sample accurate. Only digital and digital sync can be sample accurate, and should be if done correctly.

Thomas

Right. No samples in analogue. So, by definition...

[Cary B. Cornett]

Sounds like Cary's post about the guys using the bias frequency recorded on the tape is the ticket, no?

No. Not for attempting to synchronize different tapes with each other, anyway. There are a few reasons, but I won't bore you with the details (good excuse for me being too lazy to trot them all out right now) .

[Tim Miskimon]

Analog tape synchronization has never been, and never will be "sample accurate". Only digital and digital sync can be sample accurate, and should be if done correctly.

Thomas

I certainly know that but once those analog tapes are transfered to the digital domain getting those samples to accurately line up is the goal otherwise nothing will sound right that's what I was referring to.

[Tim Miskimon]

I've been in touch with a friend and former colleague who was a young tape-op at Abbey Road in the late 60s. He has read this thread with interest, and passed on the following information:

"Cary has it about right. I can recall the method because it was me doing it, night after night. A 50Hz hum was recorded on Tk 4 of the first machine. This was routed through a Vortexion 200 watt tube amplifier and used to drive the capstan of the slave machine. A further four tracks could then be recorded on the slave. During the first pass only, the count in from the master was also transferred. The trick was to judge the preroll required by the slave on each start up. It was necessary to chinagraph the tape on one of the pre-rollers as a guide but it wasn't exact. You got better with practice. After the first track was recorded it got more difficult. To play that back and be able to proceed with further overdubs the machines had to start up in reasonable synch.....I remember one had cans with the two machines and the studio feed. As soon as you got to the count in you knew: about 3 times out of 10 I'd say it was OK, but otherwise it was "s**t or f**k" from the studio! It could get quite stressful if you got several misses in a row. Stuck in the corner of No.2 control room with 2 J37s and the Vortexion amp with no air conditioning and those sodding jossticks was not hugely pleasant. I'm sure you could have cooked breakfast on the deck plates by the end of the night."

Always good to hear this kind of stuff first-hand

Your friend makes it clear that what I'm saying is in fact true.
Even using the 50 cycle tone to sync machines together it was still hit & miss.
The sync between those tapes were not exact and the drift can be heard on that Yellow Submarine CD.
That's been my point since my first post several days ago.
If you fiddle around with it for days you can get it close enough that most people might not even hear it.
When I first listened to Eleanor Rigby I felt uneasy about the timing. Something was wrong and after several playbacks I noticed that Paul's voice was not locked with the track the way it is on the Revolver album.
I compared it over & over and realized there was a timing issue.
Don't take my word for it - listen to the version on Revolver & than the 1999 Yellow Submarine version.
I also hear timing problems on All You Need Is Love & It's All Too Much.

[Dave Labrecque]

No. Not for attempting to synchronize different tapes with each other, anyway. There are a few reasons, but I won't bore you with the details (good excuse for me being too lazy to trot them all out right now) .

Yeah, I guess they could get it close, but nothing that would be a meaningful lock relative to each other. Over time. In the rain. (sorry... a little Hemingway snuck in there for no apparent reason)

[Dave Labrecque]

Your friend makes it clear that what I'm saying is in fact true.
Even using the 50 cycle tone to sync machines together it was still hit & miss.
The sync between those tapes were not exact and the drift can be heard on that Yellow Submarine CD.
That's been my point since my first post several days ago.

That's not where the issue comes in, Tim. In fact the method they used produced the sync you hear in the original record mix. It's the transfer later, when that original sync was gone, that causes the loss of sync.