Every year I like to basically unhook my teacher station in the lab and the studio desk. Since our equipment is somewhat patched together, I can adapt the setup to our current needs without “uninstalling” anything. Unlike some other teachers, I would honestly rather have a few cables hanging out and be able to change things than only do this when equipment gets upgraded.
This summer’s goals are to set up the teacher station in the lab as a massive “beat laboratory” MIDI setup (more on this in a later post) and to solve some signal flow issues in the studio.
When we record a band, musicians often complain about a slight delay in the headphones. This started after we expanded our interface to 16 inputs, and hasn’t magically gone away yet.
Delay in the monitor headphones is caused by a too-long signal path that involves the computer. Here’s the history lesson behind this:
In the dark ages, when all was analogue and tape, this wasn’t an issue. The mics went into the mixer, and the mixer split the incoming (post fader) signal between the monitors and the tape machine. Modern digital boards still do this. Since there is no “conversion” stage (the signal remains an electrical signal all the way back to the headphones) the sound is in the ears instantly.
Delay is caused when the signal gets converted from analogue to digital, usually by an interface box, and then again when the computer sends audio out. Most setups with interfaces will default to hearing what’s on the computer (it is basically a sound card, after all).
So how do we fix this? I want my monitoring back!
Here’s the tricky part. Our digital mixer (a pair of Alesis io26 interfaces) is supposed to do hardware monitoring. If I can get this to work, this means it can send the electrical signal back to the headphones without involving the computer, so we can monitor without delay. From a control panel on the computer, I can set whether this is active or not.
The tricky part is actually doing this in practice. A program like Logic (or Pro Tools or Cool Edit ’96 or whatever) will usually monitor the inputs for you, sending the audio back to the interface. This will still result in a delay, as the un-delayed signal is now being mixed with the delayed signal from the computer. You actually have to either mute the incoming tracks in the computer, or use the mix knob on the interface to switch it to monitoring. Confused? There’s more!
This leaves out the fact that lots of musicians and producers like to rely on software-based effects, which can only be heard through (you guessed it) the computer! So, want to hear that auto tune on the vocalist? It will have a bit of delay. Want to hear that sweet Amplitube effect on the raw guitar? Delay. A super good computer can overcome this, especially if you’re using a Lightpipe interface and a tiny buffer size, but when you have 16 simultaneous inputs this starts to not work, and delay (or worse) creeps back in.
The best solution we’ve had to date is when a band can just perform individually and not deal with monitoring issues, but this isn’t ideal. Many bands want to record as they rehearse – all playing together. So that’s a basic overview of one of the pitfalls of recording audio digitally…I’ll update later with our solution to this problem!