A client’s collaborator heard her voice slightly late when he put reverb on her vocal in Logic Pro, so I showed him how to add the reverb in his RME Fireface’s TotalMix software instead, where it doesn’t add any latency.
He records different singers in his home studio and wanted advice on setting up his monitoring to suit each of them. When he sings himself, he likes to hear only the backing track, with no effects and none of his own voice in his headphones. When his collaborator came round to sing, she wanted to hear reverb on her voice. He added a reverb to her vocal channel in Logic, and they could hear the reverb, but her voice was coming through with a slight delay.
Trying to fix it in Logic
This is a classic latency issue with software monitoring. When you monitor through Logic, the voice goes into the computer, through the plugins on the channel and back out to the headphones, and the buffer size decides how long that round trip takes.
I showed him how to turn software monitoring on and off, and how to change the buffer size in Logic’s audio settings. He had it set to 512 samples. Bringing it down to 64 or 32 reduces the delay on the vocal while the singer is hearing the reverb.
In this song it didn’t work so well. He was using other plugins that introduce latency of their own, and Logic compensates for them, so we couldn’t remove the delay from the software monitoring entirely.
Moving the reverb into TotalMix
So I suggested we approach it in a different way. We put the buffer back up to 512, which gives the plugins plenty of time to process, and left the vocal effects out of Logic altogether. His interface is an RME Fireface, and its TotalMix routing software has a reverb designed to solve exactly this problem.
The TotalMix reverb runs on the DSP inside the interface itself, so the singer hears it with zero latency while Logic stays at a buffer of 512.
His mic preamp sends audio over ADAT into channel 8 on the Fireface, so we went to channel 8 in TotalMix and worked through these settings:
- Channel 8 was set to stereo, so the mic was only coming through on the right-hand side. We took stereo off and set channel 8 to the centre of the stereo field with the pan pot.
- FX send on channel 8 turned up, to send signal to the Fireface’s effects section.
- Reverb turned on, with a Large Room preset that we liked.
- In the headphones section of the control room, at the bottom right of TotalMix, the green bar in the centre of the channel strip raised to add the effects into the headphone send.
- Headphone send level adjusted until he could hear a comfortable balance between the backing track and the live vocal.
Setting it up for each singer
I showed him how to adjust the level of the reverb with the sends and returns on the channel strip, so he can tailor it to the level his collaborator wants at their next session. I also showed him how to change the reverb preset to get a sound that suits the song they’re working on.
Then we stored all of the settings as a mix snapshot in TotalMix, so he can recall them, and he can have a different setup on another snapshot with the reverb turned off.
He came to me wanting to solve this in Logic, but there was a better way of doing it with the hardware he already had. He’d bought an expensive interface with a very helpful feature for this, and my part was showing him exactly how to use it.
Not sure it’s the same problem? Ask a question first
