The view from 2005: This was the first article I ever wrote for Recording Magazine (Home and Studio Recording then). Boy, have I learned a lot in the meantime! At the same time, the techniques described here have become the basis for a standard and widely used signal processing trick in the digital realm: the spreader!

Anyway, virtually all of what I said still holds up, and I use spreaders as a matter of routine today (in Pro Tools, it is a cinch), pretty much exactly as I did back in 1992. Enjoy!

Just so you know, I expanded this article into a section in my book, Total Recording.

The early days of delays

Back when I got into this crazy business (in the late 60’s), delays were a wonderful but not very subtle effect. They were generated by utilizing the travel time between the record and playback heads of tape recorders to create the delay. Varispeed was not a standard item then (in fact, we were still just trying to get the decks to run on time - forget about varying their speed!) and so the available delay times were simply the times that occurred at 7.5 and 15 IPS (about 130 and 65 milliseconds respectively). Sending the delayed version of a sound to the opposite channel was awesomely, totally creative, even if it was obvious and audible to a fault.

However, when I tried sending the delayed signal to the reverb and then to the opposite channel, that was beyond totally awesome. That was, well, mega-awesome. As I worked with this effect, which has now become a cliché and the basis for an entire family of signal-processing tricks, I began to think about the acoustics and psychoacoustics which make these effects work. Since then, little by little, I’ve worked up my own model of an approach to the use of early delays which is based on the acoustics of rooms.

The general use of early delays didn’t become possible until digital delay lines became commercially available in the middle ‘70s. The tape delays we used before were too long, and were blatantly obvious when we used them. 65 milliseconds is a delay that we perceive as an echo. The dividing line between early, inaudible delays and audible echoes seems to be around 40 milliseconds, depending on the material. When I began to play around with delays, I was struck by the fact that really short delays aren’t consciously heard (which is what the Haas Effect is all about). However, the presence of these short, inaudible delays in normal everyday acoustical listening makes use of a very powerful set of features in our auditory perception system. Because of this, the control of these delays is a basic and useful tool for the producer and recording engineer making recordings . What makes the use of these time-based effects even more powerful is that their impact is subliminal or pre-conscious, so that we are not aware of just how much impact their presence is having on us.