Revolver 2022... it's ON!

[WilliamShears]

Any post of officially released (or known soon to be officially released) material is against forum rules. This is not a new rule. I appreciate the assistance of mods who are taking action against these posts.

[WilliamShears]

No remixes of the SDE are allowed either, at least not until after release. Any posted will be hidden until October 28.

I didn't know you were in charge here now. did someone die?

Here is a reminder of the forum rules. Not sure why so many are acting like they don’t exist.

viewtopic.php?f=12&t=8

[Tex]

OK, did a sync of the early take of "Got to Get You Into My Life" and the final version... It's a hot mess if everything is playing at once so whoever releases a version they are gonna have to make some editorial decisions.

The remix required resampling the final take by a factor of 1.0013. In other words slightly slowing it down.
It appears the drum track is consistent between both version no obvious tape edits in the final version but I did have to do lots of tiny time edits to keep them in tighter sync. They must have ran that 4-track through the wringer back then.

The composite remix is bass & drums (early take), rhythm & lead guitars (early take), backing vocals (early take), lead vocal (final take) and horns (final take). I did have to shift the backing vocals a bit forward in time. This combo produces a very finished sounding version.

As great as the guitar version is it becomes obvious that version just has too many empty areas with nothing to underpin the vocal melody. They couldn't add any more guitar. The answer was to add brass all the way through.

[Lord Reith]

The thing I don't like about the album version of the song is the damned tambourine! It's too loud and corny. It took the place of John's rhythm guitar part which suited the song much better I reckon. The drums are also quite muffled on the album, but the tom-tom fills were definitely dipped intentionally in the mix. You can hear the volume of the drums being decreased at that point, whereas in the outtake that doesn't happen.

The outtake just has a brighter and more pleasing sound overall. Ideally you want the clarity of the outtake in the album version, and less tambourine.

[harrylime]

The remix required resampling the final take by a factor of 1.0013. In other words slightly slowing it down.
It appears the drum track is consistent between both version no obvious tape edits in the final version but I did have to do lots of tiny time edits to keep them in tighter sync. They must have ran that 4-track through the wringer back then.

Can't that be explained by the normal variation in grid frequency? Afaik it normally varies within a 0.30 % margin in either direction (49.85 to 50.15 Hz). Worst case would effectively mean a factor 1.0060 difference, so 1.0013 isn't that strange at all.

[mitchellmichael]

The thing I don't like about the album version of the song is the damned tambourine! It's too loud and corny. It took the place of John's rhythm guitar part which suited the song much better I reckon. The drums are also quite muffled on the album, but the tom-tom fills were definitely dipped intentionally in the mix. You can hear the volume of the drums being decreased at that point, whereas in the outtake that doesn't happen.

The outtake just has a brighter and more pleasing sound overall. Ideally you want the clarity of the outtake in the album version, and less tambourine.

Maybe more cowbell?

[Tex]

The remix required resampling the final take by a factor of 1.0013. In other words slightly slowing it down.

Can't that be explained by the normal variation in grid frequency? Afaik it normally varies within a 0.30 % margin in either direction (49.85 to 50.15 Hz). Worst case would effectively mean a factor 1.0060 difference, so 1.0013 isn't that strange at all.

Yes, the speed difference is no doubt from line frequency fluctuations In UK power. I would suggest almost all the smaller Beatles tape speed variations are from dodgy line frequencies. If there was a big difference between initial recording and later remixing then that would explain a lot. This is why even detecting the line 50hz line hum may not be 100% accurate either. Was it really 50hz that day or 49.8% ?? The more accurate method (I think) would be to spectrally look at the notes of fixed pitch instruments like piano. Back in high school music class in 1989 they even had an analog device with a huge dial with all the key musical frequencies on the dial and it would show in real time what pitch the loudest sound was at.

The finished mix will still need to be speed adjusted to match the early mix. My method is to get a sync point at the beginning of the song and a sync point further into the song (at least the middle). I pan one track to the left and one track to the right while listening on headphones and move the one track back or forth in time to get the common elements in the middle of the stereo picture.

I measure the sample length of those sections that have the exact same content and simply "resample" one to match the other's sample length. In my DAW that gives me a resample "factor". Then I apply that factor to the rest of the track that needs adjusting and bump the pieces back together and back in alignment with the unaltered track.

Resampling is the digital equivalent of speeding up or slowing down a tape and you have tempo and pitch properly adjusted at the same (they are locked together) you don't have to do a manually adjust tempo and pitch as separate processes which seems perilous to me.

[Lord Reith]

The variation in line frequency was tiny. You can't have big variations in line frequency or things on the grid that depend upon it will start to explode. In most countries the variation is required by law to be not more than +/-0.3 Hz In actuality the frequency generally does not vary by more than a few thousandths of a Hz.

If you think about it, if the line frequency had varied by more than a fraction of a percent then overdubbing would have been impossible. Up until mid 1966 (when varispeed recording was introduced), the motor speed of the EMI tape recorders was directly tied to the power line frequency (50Hz). If the frequency wandered all the time, then every time they went to do an overdub the machine would have been running at a different speed.

[harrylime]

a few thousands of a Hz??? fluctuations within 0.15 Hz are normal, peaks and dips up to 0.3 Hz are a common occurrence, see https://gridwatch.co.uk/frequency for recent graphs:


And you say that it 'was' tiny as if it were different before, while this has been the same effect since the introduction of AC power and will remain like this unless we would ever move to DC power. As I said, it's simply because you can never reach a perfect balance between all producers and consumers on the grid. It's like driving a car or sailing a boat, your steering will always oscillate ever so slightly to stay on course.

f you think about it, if the line frequency had varied by more than a fraction of a percent then overdubbing would have been impossible. Up until mid 1966 (when varispeed recording was introduced), the motor speed of the EMI tape recorders was directly tied to the power line frequency (50Hz). If the frequency wandered all the time, then every time they went to do an overdub the machine would have been running at a different speed.

Overdubbing was possible because the artist heard the music in their 'deaf aids' and performed to the perceived rhythm. Tape speed then doesn't matter, even when the tape speed would be different between sessions. This is obvious from the varispeeds that followed using a far lower far higher speed where overdubbing was also possible on (bass and vocals to Rain's slowed down backing track). So naturally a minute difference due to grid frequency also didn't hamper overdubbing, or reduction mixing for that matter.

However what wasn't easily feasible was dependable syncing between two tape machines. This wasn't different for varispeed either, don't forget varispeed still relies on the 50 Hz mains frequency being the circuit's oscillator, it's just changed via a delay circuit. So two semi-tone lowered varispeed recordings made on two different moments will still have the same minute difference as two regular recordings. That's why for example Only A Northern Song was never mixed into mono, as the stereo mix already proved too cumbersome.

And for the same reason, video has always used a clock signal, either on a separate line in-studio to synchronize between every interoperable device (camera, mixing desk, monitor, tape recorder, radio-transmitter etc) or modulated into the radio signal to allow all receivers, even on a different grid, to still perfectly time the representation on-screen.

[Lord Reith]

a few thousands of a Hz??? fluctuations within 0.15 Hz are normal, peaks and dips up to 0.3 Hz are a common occurrence,

Okay but 0.3 Hz is not that much. It would have to vary by at least five times that amount for it to produce any significant change to the motor speed as regards to recording. So if at worst case the EMI machine was running from 50.3Hz, that would not even be a noticeable change of pitch to most people. That's only a tenth of a semitone pitch difference from a machine running from 50hz.

Overdubbing was possible because the artist heard the music in their 'deaf aids' and performed to the perceived rhythm. Tape speed then doesn't matter, even when the tape speed would be different between sessions. This is obvious from the varispeeds that followed using a far lower far higher speed where overdubbing was also possible on (bass and vocals to Rain's slowed down backing track). So naturally a minute difference due to grid frequency also didn't hamper overdubbing, or reduction mixing for that matter.

Well I think we're actually agreeing here. I was saying that a minute tape speed variation (+/- 0.3hz) did not affect the recording. But if it was +/- 2 or 3 Hz like I've seen people saying, that would make overdubbing completely impossible. A mains frequency of 53hz would mean that any track recorded at 50hz would now sound a semitone sharp. You could sing or retune a guitar to match it, but if it were randomly 51.5hz, then there'd be now way to overdub pianos or other instruments that aren't easily retuned, because any track recorded at 50hz would then be a quarter of a tone sharp.

I know that once they started using varispeed (with an artificial mains supply) then they would record overdubs at different speeds, but they still would have had to "tune" the tape speed to the pianos, harmonicas or other instruments of that nature being overdubbed. With a regular mains supply, you can't tune it, so if it were really off then you'd end up with random tape speeds and pitch.

Really all I'm saying is that the mains supply frequency does not affect the tape speed substantially, in response to what Engonoceras was saying. I don't know what accounts for some of the mono mixes being slightly different speed to the stereos, aside from the possibility that every tape machine at EMI ran at a fractionally different speed, so the speed of the resultant mix depended on the particular combination of machines used to record and play back. I've noticed that the outtakes in the Please Please Me session are at a slightly different speed during the tape-to-tape overdubs than on the original takes. Assuming the bootleg is not screwy, that must mean all their machines had slight speed variances. I don't think it is possible to fine tune the speed of a tape recorder that uses an AC motor direct from the mains supply. The accuracy of the speed would be entirely dependent on the pulleys and capstan, so every machine would have a slightly different speed due to mechanical tolerances.