Author Topic: 1950s/60s American Video Audio Quality  (Read 7728 times)

0 Members and 1 Guest are viewing this topic.

Offline mc172Topic starter

  • Frequent Contributor
  • **
  • Posts: 487
  • Country: gb
1950s/60s American Video Audio Quality
« on: January 07, 2017, 01:27:09 am »
I see this kind of video crop up on YouTube every now and then, and I quite enjoy watching them as I find them quite humbling and reminiscent of a bygone era, etc. etc. but I can't quite put my finger on the following.

In the following video, as an example:



What is the cause of the distortion from approximately 1:08 to 1:22? I remember hearing similar horrible noises from VHS videos of marching bands that I used to play in during the early 2000s, but attributed it at the time to other people watching wanting to have the volume up to 111 (yes, one hundred and eleven as the TVs at the time invariably went to 100 and 101 wouldn't be loud enough).

The recording in the YouTube video I mentioned above seems quite midrange heavy during the speech, but it's particularly prominent during the orchestral piece between these times (1:08 to 1:22 or so). This seems to follow along with similar recordings from the era and country - a very distinctive, distorted "o" sound, for example, in speech.

I've had a look in Audacity and I can see that it is quite low midrange-rich, 150 to 350 Hz peaks at roughly double the rest, however what I really want to know is, why do these recordings typical of this era sound this way? The signal does not clip like you'd expect of a poorly recorded recording (i.e. recorded from original media into a computer with an ADC going into saturation).

Also, interestingly, without an audio FFT tool such as Audacity, I would have said that the predominant peaks were much higher in frequency, nearer the 1 kHz mark.
 

Offline calexanian

  • Super Contributor
  • ***
  • Posts: 1881
  • Country: us
    • Alex-Tronix
Re: 1950s/60s American Video Audio Quality
« Reply #1 on: January 07, 2017, 03:17:04 am »
I think you are being to hard on an old film strip. This probably had an optical audio track and through use and its eventual conversion to probably tape you are just hearing junk. It ads to its quaint-ness.
Charles Alexanian
Alex-Tronix Control Systems
 

Offline Paul Rose

  • Regular Contributor
  • *
  • Posts: 139
  • Country: us
Re: 1950s/60s American Video Audio Quality
« Reply #2 on: January 07, 2017, 03:25:15 am »
The sound reminds me of the 16mm film we would be shown in elementary school ( late 70s ).

The sound voltage was encoded on the film as brightness, and decoded with a photocell.

I think the upper frequency response was about 6kHz.

There was inevitably some wow/flutter due to the film stretching, etc.  Long sustained low pitch tones made this most noticeable.
 

Offline rrinker

  • Super Contributor
  • ***
  • Posts: 2046
  • Country: us
Re: 1950s/60s American Video Audio Quality
« Reply #3 on: January 07, 2017, 04:43:11 am »
 I once had a toy film projector that used the same optical audio track system - except the light source was one of those prefocused penlight bulbs (the kind with a lens in the end) and an ordinary LDR. Battery powered, and hand cranked - no motor. Audio quality? What audio quality? But it worked, and it wasn't a hugely expensive toy. Only ever had a couple of movies for it - like 5-10 minute cartoons, one was Thundarr the Barbarian, I remember. After I grew tired of it I took it apart and used bits and pieces like the lenses and the audio board for other things.


 

Offline SeanB

  • Super Contributor
  • ***
  • Posts: 16272
  • Country: za
Re: 1950s/60s American Video Audio Quality
« Reply #4 on: January 07, 2017, 10:49:18 am »
Well, this was transcoded from 35mm film to Betacam, probably with a hopping patch converter, and you can see the image drifting in the patch frame as it is quite worn. The audio likely is poor as the original film is worn and distorted, most obvious in the sprocket area, causing it to shift on the sprockets drives a bit, and wander around in the film gate and also in the audio gate, giving the distortion you see, This will be at some multiple of the frame rate and the sprocket pitch You also see film noise, and also quite a lot of tape dropout, but not too much that the dropout compensation has replaced lines much. They are lucky the tape is still playable, probably going through a library and changing format before they lose the ability to do so. Then it was digitised at NTSC rate and stored digitally.
 

Offline CatalinaWOW

  • Super Contributor
  • ***
  • Posts: 5173
  • Country: us
Re: 1950s/60s American Video Audio Quality
« Reply #5 on: January 07, 2017, 02:59:19 pm »
If I remember correctly the frame rate for those films was 24 frames per second.  Lots of chances for fun and games translating to NTSC 30 frames per second and going from there to whatever frame rate is used in the format uploaded to YouTube.
 

Offline Benta

  • Super Contributor
  • ***
  • Posts: 5840
  • Country: de
Re: 1950s/60s American Video Audio Quality
« Reply #6 on: January 07, 2017, 08:32:31 pm »
If I remember correctly the frame rate for those films was 24 frames per second.  Lots of chances for fun and games translating to NTSC 30 frames per second and going from there to whatever frame rate is used in the format uploaded to YouTube.


This was actually done in an easy way: show every 4th frame twice for converting to 30 fps. No complex transcoding or anything was used.

« Last Edit: January 07, 2017, 08:35:53 pm by Benta »
 

Offline CatalinaWOW

  • Super Contributor
  • ***
  • Posts: 5173
  • Country: us
Re: 1950s/60s American Video Audio Quality
« Reply #7 on: January 08, 2017, 12:07:42 am »
If I remember correctly the frame rate for those films was 24 frames per second.  Lots of chances for fun and games translating to NTSC 30 frames per second and going from there to whatever frame rate is used in the format uploaded to YouTube.


This was actually done in an easy way: show every 4th frame twice for converting to 30 fps. No complex transcoding or anything was used.

So how is optically coded sound handled in that method?  Bound to create some artifacts.
 

Offline helius

  • Super Contributor
  • ***
  • Posts: 3632
  • Country: us
Re: 1950s/60s American Video Audio Quality
« Reply #8 on: January 08, 2017, 12:54:57 am »
Whether the sound is optical or not makes no difference. It's duped to tape and played back at the appropriate pitch so that the total length is the same as the video. Any kind of sound-on-film can only be played back in the sound gate of a projector, it isn't taken from the telecine.

This was actually done in an easy way: show every 4th frame twice for converting to 30 fps. No complex transcoding or anything was used.
No, not at all. The jerkiness would be completely unwatchable if that method were used.
The method that was actually used is 3:2 pulldown. One film frame is scanned for 3 fields, then the following film frame is scanned for 2 fields. As a result, two film frames require 5 TV fields. This minimizes the jerkiness of motion, but doesn't eliminate it completely. One other odd effect of pulldown is that cyclic motion may appear to be in the wrong direction, as on wheels of a car.
The situation is not so sanguine when pulldown video is digitized to non-interlaced formats. Because each frame pair from the film now occupies an odd number of fields, it is much easier to mangle the asset with a misplaced edit. For this reason, a proper edit suite has a "3:2" pulldown setting for field dominance, and should be able to extract the original frames in 24n.d. fps.
 
The following users thanked this post: tooki

Offline Benta

  • Super Contributor
  • ***
  • Posts: 5840
  • Country: de
Re: 1950s/60s American Video Audio Quality
« Reply #9 on: January 08, 2017, 05:18:41 pm »
Helius, you are right, I used the word "frames", it should have been "fields". The second or fourth field is repeated on alternating fields.
It's a weakness I have, mixing up field and frame, perhaps because they both are with "f"?

Cheers.
 

Offline CatalinaWOW

  • Super Contributor
  • ***
  • Posts: 5173
  • Country: us
Re: 1950s/60s American Video Audio Quality
« Reply #10 on: January 09, 2017, 04:47:52 am »
I will just comment that you are describing good ways to do the conversion.  They may or may not describe what was used.  I have encountered some very poor conversions.  Things like using a camcorder to record the projected image.
 

Online vk6zgo

  • Super Contributor
  • ***
  • Posts: 7563
  • Country: au
Re: 1950s/60s American Video Audio Quality
« Reply #11 on: January 09, 2017, 05:54:36 am »
If I remember correctly the frame rate for those films was 24 frames per second.  Lots of chances for fun and games translating to NTSC 30 frames per second and going from there to whatever frame rate is used in the format uploaded to YouTube.


This was actually done in an easy way: show every 4th frame twice for converting to 30 fps. No complex transcoding or anything was used.

Conceptually easy---practically horrendous.(yep,Thanks helius,I do remember the 3:2 pulldown--we learnt it in passing)
Us folks with 25 frames/sec just sped the projector up .

This didn't work near as well with old WW1 newsreels filmed at 16 fps----The Kaiser's armies didn't look near so menacing when they were prancing along so jauntily! ;D
« Last Edit: January 09, 2017, 05:59:03 am by vk6zgo »
 
The following users thanked this post: tooki

Online vk6zgo

  • Super Contributor
  • ***
  • Posts: 7563
  • Country: au
Re: 1950s/60s American Video Audio Quality
« Reply #12 on: January 09, 2017, 06:12:52 am »
The sound reminds me of the 16mm film we would be shown in elementary school ( late 70s ).

The sound voltage was encoded on the film as brightness, and decoded with a photocell.

I think the upper frequency response was about 6kHz.

There was inevitably some wow/flutter due to the film stretching, etc.  Long sustained low pitch tones made this most noticeable.

This was most certainly not the default frequency response for optical sound.
Theatre released films had far wider frequency response,well before the advent of magnetic sound systems.

Too long ago to remember,but perhaps the frequency response of the theatre projectors was "tailored" to correct any response peaks of the optical system.
Cheaper systems & those used more recently to convert these films to analog tape then digital,may not have had this correction.

From memory,the weird sound on the OP's clip was not unusual on Industrial teaching films,although the ones I watched at Technical school in the 1950s/60s were mainly 1940s issue.
 

Offline mc172Topic starter

  • Frequent Contributor
  • **
  • Posts: 487
  • Country: gb
Re: 1950s/60s American Video Audio Quality
« Reply #13 on: January 09, 2017, 01:21:33 pm »
From memory,the weird sound on the OP's clip was not unusual on Industrial teaching films,although the ones I watched at Technical school in the 1950s/60s were mainly 1940s issue.

That's just it. I've noticed that a lot of these old technical videos sound like it and I'm just not sure how or why.
 

Offline helius

  • Super Contributor
  • ***
  • Posts: 3632
  • Country: us
Re: 1950s/60s American Video Audio Quality
« Reply #14 on: January 09, 2017, 03:40:37 pm »
35mm projection prints had (and have) optical sound with good quality. Initially, of course, talking pictures used synchronized phonorecords such as Vitaphone discs—optical sound came in a little later. 70mm prints have even more area given to sound and can carry 6 channels.

On a small-format film there is less room to put the stripe, and so the sound is much murkier. Ameliorating this issue is the reason that small-format films were given magnetic coating stripes, so higher-fidelity magnetic sound could be used. But if the projector has uneven speed control, you will still hear warbling.

Magnetic sound recording on film was also used to keep synchronization during editing, since the sound record (a magnetic full-coat film) could be cut and spliced in parallel with the image film.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf