Maximize Your Audio Quality Part 1: It Starts in the Box


Audio Mixer

Welcome to a 3 part series that many have been asking for over the last several months. Newcomers to keyboard performance and old hands all have questions and concerns about how to make sure the sounds you are making get to the audience in the best shape possible. Over this series we will look at the different aspects of equipment use and sound design that can positively or negatively impact your signal from computer or synth to the mixer and the audience. This discussion applies to software and hardware based setups. Where there are differences I will point them out.

Step 1: Get your House in Order

If you are working with a soft synth (Ableton, Mainstage, Cantabile, Logic, etc) setup you need to make sure your patches are organized and tested prior to the gig. I recommend removing patches you don’t need and saving into a new file for each specific gig. This keeps the load on your computer as low as it can to be and minimizes the chances of a hardware glitch torpedoing your setup. I personally also keep a set of “generic” bread and butter patches in each set in case something unexpected comes up (things like straight piano, Hammond, pad).

If you are working on hardware it is still important to organize your patches for ease of use and to minimize the search-flail during the gig. I keep the first 5 user patch slots open and save my set patches into those locations. I keep the patch catalog organized by song title in alphabetical order to make other songs easy to find. Like my software setup, I keep a set of generic patches in the second 5 user spaces in case I need them.

Step 2: Find Solid Audio Settings for your DAW

Hardware folks can skip this step. If you are software based you know that the audio engine settings in your DAW can make or break it in terms of stability. Experiment with different latency settings to determine the lowest playable latency that is stable. In my experience most people do not have a problem playing with 128 samples of latency and 256 may even be acceptable. You should also know that the sample rate of your project affects the latency (faster sample rates lower the latency time at the same latency setting). I personally run at 48 khz.

Once you have settings you believe will work it’s time to stress test the system. Load up several of your most complicated patches and switch back and forth between them quickly, playing as much as possible. Do you hear clicks or pops? Are you getting CPU warnings? If things run smoothly you probably have hit on some solid settings. If you are getting audio glitches then you may want to try different engine settings. If changing the audio engine settings doesn’t seem to help you may be running into disk access problems. The only solution there is to find a faster disk or faster disk interface method (SSD or 7200 RPM disk vs 5400; Firewire or USB3 vs. USB2).

Step 2: Volume Balancing

We are going to cover a couple of things here – volume balancing and the concept of peaking. I am going to give a couple of recommendations for volume balancing depending on your particular band situation but first lets go over peaking.

Audio signals are made up of waves and the waves and peaks and troughs just like the waves in the oceans. As you play more sounds the peaks from each sound “add up” and make BIGGER peaks. At some point these peaks may be too big for your audio hardware or software to process and the result is clipping where the top of the waveform is trimmed off. The audible effect of this is a dirty “blip” type sound. Most pro audio software performs a lot of math magic in the box to prevent clipping but clipping can be a real problem with your audio hardware.

So how do we manage our peaks and avoid clipping? As you play through your patches watch the master output meter – if it is hitting the red or even orange often we will want to pull that back a bit (I say orange because realistically you are going to dig in a bit more at the gig). Pay attention to the channel meters too – if you are peaking on one instrument then pull them all back a bit. Your mix will stay the same and you can avoid peaking.

On to volume balancing – this is a subject for both software and hardware users and there are a couple of approaches to volume balancing.

Constant Volume Approach

If you have a professional or advanced volunteer FOH (Front of House – the mixer gal) engineer you would probably do better by balancing all your patches to a constant volume. In this case we are going to rely on the FOH engineer to know that our patch needs to be quieter on song A, louder on song B, and so on. It is, of course, a good idea to ask your engineer if they want you to set up this way. The advantages here are that the volume to the board is constant, and any monitoring will also have a constant volume.

Play all your patches through some sort of speaker and listen for the relative volume. I find that headphones in this case are very misleading and I don’t recommend volume balancing on them. In my experience you need to do this quickly (so play each patch briefly) as your ear will accommodate to the volume pretty fast and it will be harder to tell the difference between patches. If you need to adjust the volume you will want to do that with the channel sliders. Most software will not recall changes to the master volume output as you change patches.

Relative Volume Approach

If your FOH engineer is less experienced it can be helpful to anticipate their needs and balance your patches to the relative volume for the song. This is a bit trickier since the relative volume that is “right” will depend a lot on what other instruments are playing and how loud they are playing but I find it is certainly possible to get things headed in the right direction when I am setting up at home. Just understand that you may need to do some additional tweaking at practice.

In this case we are still going to audition our patches quickly through some sort of speakers but we will try to set up our patch volumes appropriately for the volume of our part in that song. For example, I will usually set my first patch to be a bit louder (since we start with an upbeat song) and my third patch a good bit quieter (since we typically are doing a more reflective song in that spot). The major downsides of this approach are that the volume of the keys in any monitoring and at the board are going to change and you may not enjoy having your keys quieter in the monitor on those songs.

Recap

In this part we talked about organizing your setup for ease of use and minimizing a set to what is needed. We also discussed finding the best settings for your DAW and volume balancing your patches prior to performance. In the next part we will look at best practices for getting your signal to the FOH. In the last part we will talk about working with the FOH engineer to get the most out of your setup. Thanks for reading and stay tuned!

Post 2 in this series is here.

Post 3 is here.

Mixer image courtesy comedy_nose and creative commons.

Advertisements

2 comments

  1. I am running an 1/8 inch to XLR from my Macbook’s audio output to a DI box. Where should I set my laptop’s master volume?

    1. Is it ⅛ inch to one single XLR or a pair? That is not a recommended connection because you may run into phase cancellation between the L and R outputs from the Mac as the receiving equipment tries to phase invert one (XLR is not usually a stereo connector, it is a balanced mono connector). You are better off going ⅛ inch to dual ¼ inch.
      To answer your question directly though, it should be set to 100%.

Leave a Comment

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s