Symphony I/O: Understanding Latency

What is latency? 

When recording with most computer-based digital audio applications, a delay between the input and output of the recording system often disturbs the timing of the musicians who are performing. This delay, known as latency, means that the musician hears the notes he produces a few milliseconds after having produced them. As anyone who has spoken on a phone call with echo knows, relatively short delays can confuse the timing of any conversation, spoken or musical.   To illustrate the effect of latency, let's think about the typical signal path of a vocal overdub session. A vocalist sings into a microphone, which is routed through a hardware interface to the audio software application for recording. In the software application, the vocalist’s live signal is mixed with the playback of previously recorded tracks, and routed back through the hardware interface to the vocalist’s headphones. Because of the audio application’s latency, the vocalist hears his performance delayed by several milliseconds in his headphones.  

How does Maestro resolve latency?

By routing the hardware input directly to the hardware output and mixing in playback, it’s possible to create a headphone listening signal with a much shorter delay. First, the signal being recorded (in this case, a vocal mic) is split in the hardware interface and routed to both the software applicaton for recording and directly back to the hardware outputs without going through the latency-inducing software; this creates a low latency path from mic to headphones. Next, a stereo mix of playback tracks is routed to the low latency mixer and combined with the hardware input(s). This allows the performer to hear both himself without a confusing delay plus the playback needed for overdubbing. Note that the software application’s mixer is used to set a stereo mix of playback tracks while the low latency mixer is used to set the balance between the stereo playback mix and the hardware inputs.    

Do I need the Maestro Mixer?

The Maestro mixer serves to provide a low latency listening mix while recording. Therefore if you’re using Symphony I/O to listen to iTunes or audio from another program, there’s no need to use the mixer. It’s also possible that the latency of your particular recording system is low enough to be unnoticeable by you or other performers, especially when connecting Symphony I/O via the Symphony 64 Thunderbridge or PCIe card. If you’ve set your audio software’s input/output buffers according to the guidelines below and latency doesn’t bother you or other performers, there’s no need to use the Maestro mixer.    

How do I set my software’s I/O Buffer?

The I/O Buffer setting found in most audio software is one of the most crucial, but often ignored, settings in a Mac-based recording system. When choosing a buffer setting, a compromise between the latency through the application and the amount of computer processor power accessible to the application must be made.

A lower Buffer setting results in lower latency but less available processing power. If the application can’t access enough processor power, processor overruns may occur, resulting in audible clicks and pops or error messages that interrupt playback and recording.

A higher Buffer setting, on the other hand, results in greater amount of accessible processor power (i.e. less chance of overruns) but increases the latency. Determining the best setting requires some trial-and-error in order to find the best compromise. Keep in mind that as tracks and plug-ins are added to a software session, processor requirements increase. Thus, the buffer setting that works during the early stages of a session might result in processor overruns during later stages. The best strategy is to set the buffer to a lower setting during recording and accept certain limitations on plug-in usage, and then raise the buffer during mixing to utilize the computer’s full processor power when latency isn’t an issue. With the processing power of today’s Macs, you may find that adjustment of the Buffer isn’t necessary, and you can leave it at a setting for low latency and still access a sufficient amount of processing power when adding tracks and plug-ins. If you do encounter clicks, pops or software errors, don’t hesitate to experiment with the Buffer setting.

Step by Step instructions to record with the low latency mixer

1. Configure your software application for use with an external low latency mixer. For example , in Logic Pro uncheck the Software Monitoring box (Logic Pro Preferences Audio) so that when a track is in record, audio output is muted from Logic. Thus, only audio through the low latency mixer is heard.

2. Set the software application's mixer output to Out 1-2.

3. In the Maestro Output Routing tab window, assign Mixer 1 to hardware output Line 1/2, Mixer 2 to hardware output Line 3/4.

4. In the Output tab window, assign Headphone 1 to Line 1-2, Headphone 2 to Line 3-4

5. In the Mixer tab window, set the Software Return drop down menu to 1-2 and set both the Software Return and Mixer Master faders to 0.

6. In your software application, play back the session. You should hear playback in headphones connected to Symphony I/O's HP 1 and HP 2 outputs. Next, create a suitable mix of playback tracks. In this example, both Mixer 1 and 2 are in use, but under many circumstances only one mixer is necessary.

7. Connect the signal to be recorded to Symphony I/O's Line In 1 - the signal level will appear in the Mixers' Line 1 meter.

8. Raise the Line 1 fader to create a monitor mix of the input and playback signals.

9. Record!