A 414 discovery

My tascam 414 has dbx noise reduction built in. Tape has a significant noise floor - hiss. The louder you record to tape, the better the signal:noise ratio.

Dbx reduces the hiss by compressing the audio on the way in (pre-tape) by a 2:1 ratio to ensure the soft parts of the program are recorded at a healthy level above the noise floor. Then when the tape is played back the audio is expanded (the opposite of compression) by 2:1 to recover the initial dynamic range. The noise floor in the quiet sections drops down in the expansion - however, for a loud signal, the noise is boosted in the expansion, so the noise 'breathes' or 'pumps' with the volume of the audio (one of the main disadvantages dbx has compared to dolby NR).

The cool part is I can turn the NR (noise reduction) on and off whenever I want. I can record to tape with NR on and play back with NR off, which means the audio has been compressed, but isn't expanded on the way out, so I can score some free compression! Alternatively, I can record to tape with the NR off and playback with NR on - I then score expansion - which is a eerie envelope -almost like a gate - where the sounds seem to rush up at you out of silence and then quickly fall off back into the noise floor.

The compressed versions of the sounds are very bright - lots of high frequencies - I suspect the dbx process knocks off some of the tops, so they purposely put a high-frequency EQ boost in the compression stage.

Now to my discovery. There is a third 'sync' mode for the NR. As I was flicking the switch back and forth listening to my City at War recordings sync mode sounded the best. But what is it?

From the tascam manual:

When it is set to the SYNC position, Track 4 is disconnected from the dbx system, so the process does not affect the sync signals going to and from track 4, but tracks 1-3 still go through the dbx encode/decode process. Use the SYNC position for recording and playback of FSKsync or SMPTE time code.This can be used to save slightly out-of-tuneparts, or to create sound effects such as flanging.

I put the vocals on track 4 - by doing this I unwittingly gave myself a choice if I wanted to use vocal compression or not! If I use sync mode, the vocals which were recorded to tape with NR ON (ie compressed) can be played back without the expansion stage, whilst leaving all the other tracks expanded normally by the dbx. Sweet!

I am not the first one to do this, Richard Adams (U2) used a similar technique:

VOCALS: Bono's vocals were largely recorded with an SM58 and compressed with a Summit compressor. Adams: "Instead of using the Summit, what we did on several of the tracks on Achtung Baby was to record his vocals on tape with Dolby SR and play it back without Dolby SR. It tightens up the vocal sound and gives it more brightness and presence. It makes his voice sit really nicely in the mix and easier to balance."

(source: http://www.soundonsound.com/sos/1994_articles/mar94/u2robbieadams.html)

It's a kind of compression you don't find elsewhere - no attack or release times, no threshold, it's just a permanent tightening of dynamic range across the whole signal.