Audio engineers: What makes 1960s narrators sound the way they do?
March 5, 2016 3:36 PM Subscribe
In the recent FPP on early computing the voiceover has a beautiful subtle fuzz to it, that isn't apparent in the sound of the people being filmed. The sound is consistently heard in narration of 1960s films - how is it created?
For a long time I've wondered why narrators in 1960s documentary films sound the way they do. I've seen this previous AskMe that covers accent, my question is about the technology.
I guess it's some combination of close-up mic and valve or transistor, but wouldn't know where to start simulating it in something like Ableton.
(I have a minor qualification in Music Technology, so hit me up with your cardioid-electret-frequency-response jargon. I know next to nothing about vintage audio though)
For a long time I've wondered why narrators in 1960s documentary films sound the way they do. I've seen this previous AskMe that covers accent, my question is about the technology.
I guess it's some combination of close-up mic and valve or transistor, but wouldn't know where to start simulating it in something like Ableton.
(I have a minor qualification in Music Technology, so hit me up with your cardioid-electret-frequency-response jargon. I know next to nothing about vintage audio though)
Best answer: I'm not an engineer but I am a broadcasting instructor. Plus I was alive at the time this film was made so I have a bit of historical context. So here goes my best guesses;
Hard to say what kind of mic is being used here. Ribbon mics were popular for some voice work as they could capture high frequencies well, See the RCA 77DX for example. The 77DX had a pretty flat response and was designed for close talking. This would allow the talent to get close to the mic and then be able to relax and not have to project their voice. Giving the performance what was sometimes called the "FM DJ voice" or later the "Barry White" voice. But at the time there were many many microphones to choose from.
Then, as you suggest, there is the fact that the entire audio chain of the day, from Mic pre to playback was "valve" or tube amplification. This would often add some 60hz/50hz hum or harmonics along the way. Adding some EQ at the bottom end to "warm it up" would have the effect of boosting some of that low freq "noise".
Finally based on the date of 1968, the film could be either optical track or mag track. Both would add more noise to the final output - although the mag track could be processed though Dolby A and that would add more apparent "warmth" and reduce high frequency hiss. It could even be double system - but I digress.
posted by Zedcaster at 5:05 PM on March 5, 2016 [3 favorites]
Hard to say what kind of mic is being used here. Ribbon mics were popular for some voice work as they could capture high frequencies well, See the RCA 77DX for example. The 77DX had a pretty flat response and was designed for close talking. This would allow the talent to get close to the mic and then be able to relax and not have to project their voice. Giving the performance what was sometimes called the "FM DJ voice" or later the "Barry White" voice. But at the time there were many many microphones to choose from.
Then, as you suggest, there is the fact that the entire audio chain of the day, from Mic pre to playback was "valve" or tube amplification. This would often add some 60hz/50hz hum or harmonics along the way. Adding some EQ at the bottom end to "warm it up" would have the effect of boosting some of that low freq "noise".
Finally based on the date of 1968, the film could be either optical track or mag track. Both would add more noise to the final output - although the mag track could be processed though Dolby A and that would add more apparent "warmth" and reduce high frequency hiss. It could even be double system - but I digress.
posted by Zedcaster at 5:05 PM on March 5, 2016 [3 favorites]
I like Zedcaster's point about mic technique.
Otherwise, though, I don't know that you could point to anything specific in the signal chain. I'd actually call the frequency range of the voice track fairly limited - almost nothing above 3 or 4K, not much below 150 or 200 Hz. It's missing the "boxiness" around 500 Hz to 1K that you would tend to expect from a "lo-fi" recording, but getting rid of that would be a simple EQ tweak. There's almost certainly some compression in play, too, but exactly where that would be in the signal flow I've no idea.
Whether the subtle fuzz you're hearing is due to the initial recording with tube/valve equipment, or the characteristics of whatever medium it was recorded on, or deterioration over time, is impossible to tell.
Having said all that, I think running tracks through "tube emulator" and/or "vintage compressor/limiter" plug-ins would probably not be a bad place to start for your own stuff.
posted by soundguy99 at 7:10 PM on March 5, 2016
Otherwise, though, I don't know that you could point to anything specific in the signal chain. I'd actually call the frequency range of the voice track fairly limited - almost nothing above 3 or 4K, not much below 150 or 200 Hz. It's missing the "boxiness" around 500 Hz to 1K that you would tend to expect from a "lo-fi" recording, but getting rid of that would be a simple EQ tweak. There's almost certainly some compression in play, too, but exactly where that would be in the signal flow I've no idea.
Whether the subtle fuzz you're hearing is due to the initial recording with tube/valve equipment, or the characteristics of whatever medium it was recorded on, or deterioration over time, is impossible to tell.
Having said all that, I think running tracks through "tube emulator" and/or "vintage compressor/limiter" plug-ins would probably not be a bad place to start for your own stuff.
posted by soundguy99 at 7:10 PM on March 5, 2016
Best answer: What everyone else said.
Like most things audio, there’s a lot of things going on, and it all adds up. Just guessing, but the equipment used had a lot of tubes and transformers and compressors softening and distorting the sound. The medium heavily colored the sound. The tastes and practicalities of the day. The intended listening equipment would not have been "hi-fi" in so the EQ was tailored for that; no high frequencies to speak of (and many of them were rolled off due the previously mentioned factors) and no real lows below 200 Hz or so, the low end would be boosted around there. All the noise in the lows and highs could be rolled off without compromising the intelligibility for the audience.
If you’re trying to recreate it; First use a flatter mic if you can, not a modern mic with hyped High’s. Hit some subtle distortion, there are lots to choose from. Softube's Saturation Knob is free. 2 or 3 different really subtle ones each in between the other effects would be more convincing. Compress it well with an old school setting, slow attack and release. Speak clearly and mildly without projecting a lot, really close to the mic. Low pass everything below 4-5 kHz (12dB? you’ll have to figure it out) or even lower and High Pass everything above 200 Hz or so. Maybe put a limiter at the end, not a new brick wall, but a compressor set to a ratio like 20.
PSP’s NobleQ EQ and Old Timer compressor are models of similar equipment that would have been used, and have built in saturation. They’re both excellent. There are many other similar things but they often don’t have the saturation.
That should get you close.
posted by bongo_x at 7:32 PM on March 5, 2016 [1 favorite]
Like most things audio, there’s a lot of things going on, and it all adds up. Just guessing, but the equipment used had a lot of tubes and transformers and compressors softening and distorting the sound. The medium heavily colored the sound. The tastes and practicalities of the day. The intended listening equipment would not have been "hi-fi" in so the EQ was tailored for that; no high frequencies to speak of (and many of them were rolled off due the previously mentioned factors) and no real lows below 200 Hz or so, the low end would be boosted around there. All the noise in the lows and highs could be rolled off without compromising the intelligibility for the audience.
If you’re trying to recreate it; First use a flatter mic if you can, not a modern mic with hyped High’s. Hit some subtle distortion, there are lots to choose from. Softube's Saturation Knob is free. 2 or 3 different really subtle ones each in between the other effects would be more convincing. Compress it well with an old school setting, slow attack and release. Speak clearly and mildly without projecting a lot, really close to the mic. Low pass everything below 4-5 kHz (12dB? you’ll have to figure it out) or even lower and High Pass everything above 200 Hz or so. Maybe put a limiter at the end, not a new brick wall, but a compressor set to a ratio like 20.
PSP’s NobleQ EQ and Old Timer compressor are models of similar equipment that would have been used, and have built in saturation. They’re both excellent. There are many other similar things but they often don’t have the saturation.
That should get you close.
posted by bongo_x at 7:32 PM on March 5, 2016 [1 favorite]
Oh, and add a little hiss back in after the EQ (or a lot before it), or better yet, use a tape sim and add hiss from that.
posted by bongo_x at 7:49 PM on March 5, 2016
posted by bongo_x at 7:49 PM on March 5, 2016
iZotope Vinyl is a free sound degradation plugin that has a "Year" setting, might be worth a shot.
posted by STFUDonnie at 4:26 AM on March 6, 2016
posted by STFUDonnie at 4:26 AM on March 6, 2016
« Older Looking for a good (cheap) laptop for photo and... | Who else went around killing their serial hero? Newer »
This thread is closed to new comments.
posted by pipeski at 4:07 PM on March 5, 2016