Author Topic: How did record makers know what grooves to squeeze together before computers?  (Read 1805 times)

0 Members and 1 Guest are viewing this topic.

Offline BeaminTopic starter

  • Super Contributor
  • ***
  • Posts: 1567
  • Country: us
  • If you think my Boobs are big you should see my ba
Back in the 60's or maybe 70's (I don't know the name of the technology to look it up) record manufactures realized they could fit more music on vinyl, by taking a quiet or high frequency sequence of grooves, then on the next inner grove they could fit it in closer using variable spacing between grooves. Before all groves were fixed/ spaced apart allowing the whole record the margin needed for a loud or low frequency sequence. Depending on the type of music they could add minutes to each side using this technique. Wonder what kind of music or albums were most compressable, soprano operah music?

What I want to know is what kind of tech would it take to do this before there were computers that could just run an algorithim and sqeeze them in. Somehow the device would have to remember or read what the groove next to it was like, which would mean memory and not something that analog does(well, anyways). I bet the inventor was really high one day looking and staring at records grooves when he came up with the idea.
Max characters: 300; characters remaining: 191
Images in your signature must be no greater than 500x25 pixels
 

Offline retrolefty

  • Super Contributor
  • ***
  • Posts: 1648
  • Country: us
  • measurement changes behavior
Grooves? Well I always assumed that a record only has two grooves, one on each side. Perhaps you mean something else  :-//
 

Offline dmills

  • Super Contributor
  • ***
  • Posts: 2093
  • Country: gb
The industry term is 'threads'.

What you did was use a tape machine with two sets of replay electronics spaced a specified distance apart (depended on the make of the lathe), to produce a 'preview' as well as a cutting signal.

Typically the lathe would have a feed motor for the cutting head as well (sometimes, as in the VMS 66) as a voice coil motor that controlled depth of cut.

Depth of cut is basically driven by preview L minus preview R with a multi stage sample and hold driven by a clock from the rotating turntable, and this actually turns out to be the important one because the groove has 45 degree walls so cutting deeper (as you must for high amplitude difference signal) really hurts the spacing.

Something similar is done for the pitch calculation but here you use the preview right and main left in the calculation as the left hand modulation is the inner wall of the groove.  The depth of cut also feeds into this as depth directly influences grove width.

On a VMS 66 this was done by the SV66 assembly (A set of eurocards) with the sample and hold clocked by a photocell reading light and dark quadrants from the underside of the platter.

871560-0



 
The following users thanked this post: SeanB, Zbig, ebastler, newbrain, schmitt trigger, I wanted a rude username, magic

Offline BeaminTopic starter

  • Super Contributor
  • ***
  • Posts: 1567
  • Country: us
  • If you think my Boobs are big you should see my ba
The industry term is 'threads'.

What you did was use a tape machine with two sets of replay electronics spaced a specified distance apart (depended on the make of the lathe), to produce a 'preview' as well as a cutting signal.

Typically the lathe would have a feed motor for the cutting head as well (sometimes, as in the VMS 66) as a voice coil motor that controlled depth of cut.

Depth of cut is basically driven by preview L minus preview R with a multi stage sample and hold driven by a clock from the rotating turntable, and this actually turns out to be the important one because the groove has 45 degree walls so cutting deeper (as you must for high amplitude difference signal) really hurts the spacing.

Something similar is done for the pitch calculation but here you use the preview right and main left in the calculation as the left hand modulation is the inner wall of the groove.  The depth of cut also feeds into this as depth directly influences grove width.

On a VMS 66 this was done by the SV66 assembly (A set of eurocards) with the sample and hold clocked by a photocell reading light and dark quadrants from the underside of the platter.

(Attachment Link)

So there is a lot there to understand: It has a stylus that reads the record, I'm guessing a normally spaced master disk, then that machine cuts the compressed record. How does it know what areas can be compressed and what cant? Does it measure pitch, so a circuit would squeeze more as it got higher? Seems like the electronics in that would have to be pretty high tech unless its just a ratio determined by pitch. That would make a cool tear down.

And it baffles my mind that you or someone here knows this stuff, whats your background with this machine?
Max characters: 300; characters remaining: 191
Images in your signature must be no greater than 500x25 pixels
 

Offline dmills

  • Super Contributor
  • ***
  • Posts: 2093
  • Country: gb
Nope, you are cutting on a blank nitrocellulose coating on an aluminium disk with the audio being sourced (traditionally) from tape.

The audio delay was from the two tape heads and the controls were peak hold and dump with transistor and diode circuitry and capacitors as the memory elements. At 33 1/3rd RPM the disk does one rev every 1.8 seconds, so hold time was only in the order of a second, not too bad with a low leakage cap.

Matrixing for L+R and L-R was done with transformers, and the rest of the computer was transistors and diodes, the nearest thing to a digital bit was the sample and hold state logic (Two transistor bistables) and the PWM generator that controlled the leadscrew rate. The diagrams of the VMS 66 and 74 are out there if you search for em.

Plenty of machines were retrofitted with things much closer to modern computers for control over the years, but you were asking about the 60's tech.

Bear in mind that back then there was HUGE money in this, making records was giving hollywood a run for the consumer disposable income, so complex and slightly fiddly electronics were acceptable.

Once you have the master disk cut the process of making a series run of records is gloriously mechanical, silver then nickel plate the master disk, electroplate up to a suitable thickness, then peel of the resulting metal foil negative, use it to make a positive, then a series of negatives (number depending on expected run rate, these are your stampers, those go into a record press where a ton or so per square inch of pressure (and steam heat) squash the vinyl into the groove, replace steam with cooling water and around we go.

Larry Bodens 'Basic disk mastering' is interesting and the AES have a two volume compendium of collected papers on disk recording.

I once NEARLY took a job in a major studio complex that included disk mastering rooms so I did a little reading.

 
 
The following users thanked this post: SeanB, ebastler, I wanted a rude username

Offline amyk

  • Super Contributor
  • ***
  • Posts: 8492
You do realise computers were around in the 60s and 70s, right...? :-//

A quick Google with the terms "record variable track pitch" gives plenty of material for further reading.
 

Offline dmills

  • Super Contributor
  • ***
  • Posts: 2093
  • Country: gb
Yea, but the computers for this were analog, digital audio of a quality to be useful for music was very, very bleeding edge even in the late 70s. Capacitors and current sources or transconductance stages for integrators or differentiators, transformers for addition and subtraction, caps switched with jfets for memory, all perfectly workable, just annoying to line up.

It was the 80s before digital computers really got a look in in this space, and even then they were part of various bits of sidechain processing not in the main audio path.
 

Online SiliconWizard

  • Super Contributor
  • ***
  • Posts: 15652
  • Country: fr
Oh yeah, so there was actually a life before computers? Damn. ;D
 
The following users thanked this post: eti

Offline GeorgeOfTheJungle

  • Super Contributor
  • ***
  • !
  • Posts: 2699
  • Country: tr
1968:



« Last Edit: November 14, 2019, 05:35:54 pm by GeorgeOfTheJungle »
The further a society drifts from truth, the more it will hate those who speak it.
 

Offline GlennSprigg

  • Super Contributor
  • ***
  • Posts: 1259
  • Country: au
  • Medically retired Tech. Old School / re-learning !
Although slightly off topic.... I remember when using audio cassette tapes to save/load the data
for PLC's. (Programmable Logic Controllers). I finally convinced my old boss that using 'modern'
fancy tape players was fraught with problems!! The cheapest nastiest $10 machine worked better!!

They had crappy 'tinny' sound, but were able to compress the 'data' better, with catching all the
rapidly rising/falling 0's & 1's, and the spikiness of it produced better, faster transfer.   :phew:
Diagonal of 1x1 square = Root-2. Ok.
Diagonal of 1x1x1 cube = Root-3 !!!  Beautiful !!
 

Offline eti

  • Super Contributor
  • ***
  • !
  • Posts: 1801
  • Country: gb
  • MOD: a.k.a Unlokia, glossywhite, iamwhoiam etc
Back in the 60's or maybe 70's (I don't know the name of the technology to look it up) record manufactures realized they could fit more music on vinyl, by taking a quiet or high frequency sequence of grooves, then on the next inner grove they could fit it in closer using variable spacing between grooves. Before all groves were fixed/ spaced apart allowing the whole record the margin needed for a loud or low frequency sequence. Depending on the type of music they could add minutes to each side using this technique. Wonder what kind of music or albums were most compressable, soprano operah music?

What I want to know is what kind of tech would it take to do this before there were computers that could just run an algorithim and sqeeze them in. Somehow the device would have to remember or read what the groove next to it was like, which would mean memory and not something that analog does(well, anyways). I bet the inventor was really high one day looking and staring at records grooves when he came up with the idea.

Lest you forget, MAN invented the computer; not the other way around. Human ingenuity is not constrained by the boundaries of what supposedly "intelligent" boxes of bolts can spit out - that's hugely insulting to mankind, who seems to have managed to design PLENTY of stuff, thousands of years before pooters hobbled onto the scene in their huge glowing racks.
 

Offline eti

  • Super Contributor
  • ***
  • !
  • Posts: 1801
  • Country: gb
  • MOD: a.k.a Unlokia, glossywhite, iamwhoiam etc
Oh yeah, so there was actually a life before computers? Damn. ;D

I am often incredulous to the point of despair, at how many people appear to think that human history was arbitrarily "on hold" for thousands of years, until, as if by some fluke of fate, the computer was "magically" invented, and rolled along pre-loaded with solutions for all our problems, we having somehow known this would come to pass...  :-DD
 

Offline BeaminTopic starter

  • Super Contributor
  • ***
  • Posts: 1567
  • Country: us
  • If you think my Boobs are big you should see my ba
Back in the 60's or maybe 70's (I don't know the name of the technology to look it up) record manufactures realized they could fit more music on vinyl, by taking a quiet or high frequency sequence of grooves, then on the next inner grove they could fit it in closer using variable spacing between grooves. Before all groves were fixed/ spaced apart allowing the whole record the margin needed for a loud or low frequency sequence. Depending on the type of music they could add minutes to each side using this technique. Wonder what kind of music or albums were most compressable, soprano operah music?

What I want to know is what kind of tech would it take to do this before there were computers that could just run an algorithim and sqeeze them in. Somehow the device would have to remember or read what the groove next to it was like, which would mean memory and not something that analog does(well, anyways). I bet the inventor was really high one day looking and staring at records grooves when he came up with the idea.

Lest you forget, MAN invented the computer; not the other way around. Human ingenuity is not constrained by the boundaries of what supposedly "intelligent" boxes of bolts can spit out - that's hugely insulting to mankind, who seems to have managed to design PLENTY of stuff, thousands of years before pooters hobbled onto the scene in their huge glowing racks.

OK one thing I still haven't figured out is how do I get a stamp into my computer to send an "electronic mail?" it must be like the dollar bill slot on a soda machine but email came out before bill and therefore stamp readers. I must owe so much unpaid postage or have a huge pile of electronic mail with return to sender stamped on it.

Yes there was life before computers but until then 99% of your day was spent looking for food or hunting game then almost over night we had the transistor and a supermarket. Hell when my grand mother was a little kid "they didn't even have electricity" and I'm not even that old. Now I have friends who's grandmothers "didn't even have computers" when they were kids.



So how many extra minutes could you potentially add to a side of a 33 1/3?
Max characters: 300; characters remaining: 191
Images in your signature must be no greater than 500x25 pixels
 

Offline dmills

  • Super Contributor
  • ***
  • Posts: 2093
  • Country: gb
It is highly content dependent, and you do much better with classical then modern highly compressed pop for fairly obvious reasons.
Generally you get somewhere around 25% or so, but it varies hugely.

Interestingly the automatic depth of cut control is actually the important one, because the cut is a 45 degree groove wall, so going deeper (as you must for very wide material) increases the width of the groove and means you must lower the pitch of the spiral. This turns out to be at least as important as responding to the side to side modulation for efficient use of space.

Obviously varying the depth of cut introduces a subsonic 'rumble' to the L-R component which must be filtered out on replay...
 
 

Offline BeaminTopic starter

  • Super Contributor
  • ***
  • Posts: 1567
  • Country: us
  • If you think my Boobs are big you should see my ba
It is highly content dependent, and you do much better with classical then modern highly compressed pop for fairly obvious reasons.
Generally you get somewhere around 25% or so, but it varies hugely.

Interestingly the automatic depth of cut control is actually the important one, because the cut is a 45 degree groove wall, so going deeper (as you must for very wide material) increases the width of the groove and means you must lower the pitch of the spiral. This turns out to be at least as important as responding to the side to side modulation for efficient use of space.

Obviously varying the depth of cut introduces a subsonic 'rumble' to the L-R component which must be filtered out on replay...

I think I have noticed that when having a big subwoofer close the table your turn tables sit on. It seems to go away when you press down on the turntable compressing those little shock absorbers they sit on. The best way to fix this is to put your turntables on a scrap piece of carpet, for a cheap turntable you can use egg crate foam.
Max characters: 300; characters remaining: 191
Images in your signature must be no greater than 500x25 pixels
 

Offline dmills

  • Super Contributor
  • ***
  • Posts: 2093
  • Country: gb
That's just acoustic feedback.

The patent cure in the loud sort of clubs is turntable on a paving slab, paving slab on halves of squash balls. The soft rubber hemispheres provide the spring and damping and the concrete slab provides the mass.

 
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf