In the days when we recorded music on tape, manufacturers had a simple automatic solution to the problem of latency. Why isn’t there a similar standard for recording into computers?
Remember Simul-Sync? The picture at the top of this story is of IK Multimedia’s Tascam Tape Collection, which has software models of the famous reel-to-reel machines.
But on the actual machines, overdubbing would have been all but impossible without Simul-Sync.
That’s because the tape passed the record head before it got to the playback head, so you’d get a delay if you tried to record parts on top of existing ones. (And when that effect is exploited creatively, it’s called tape delay.)
So Simul-Sync simply switched to monitoring off the record head when the machine went into record. It didn’t sound quite as good as the playback head, but so what – it was fine for the purpose.
Today, instead of tape delay we have computer delay when we record through audio interfaces into our DAWs.
The latency, as it’s called, is much shorter than tape delay. Depending on your audio buffer setting, the latency in and out might be around 15 milliseconds.
Now, sound travels at just over 1 foot per millisecond. So considering that this is like playing along with speakers 15 feet away – picture a musician playing 15 feet away from an amp – that may not sound significant.
But it’s more than enough to throw off most musicians’ timing, and that 15mS figure may even be optimistic. So many if not most audio interfaces compensate for this with a direct monitoring feature – the input is routed directly to the output for monitoring, either digitally or it’s mixed in the analog domain. In other words, it bypasses the path through the computer.
Great, except it’s not automatic. At the very least you have to push some buttons on the audio interface’s panel, or more commonly you have to go in and switch to a direct monitoring configuration in software.
We have simul and sync, but no Simul-Sync.
And that’s the point: this should be automatic. Put a DAW track in record, and the audio interface should switch the input routed to it to direct monitoring.
30-odd years in, why is there no universal standard for this?
There should be!