Here in the online realm of the socially distant, there’s a lot of discussion about setting up live performances via the network, and about a jillion questions about gear and techniques and whatnot.  I thought I’d try to at least gather them in an orderly way. I don’t know nearly as much as some (most?) of you do, but I used to make my living organizing information, so this is an attempt at working through stuff in an orderly way. It’ll be greatly improved by your comments and suggestions, both as to the details and as for topics you’d like to see covered.  So please do comment, correct, and bitch and moan.

I did think it would be good to approach this in a way that avoids religious warfare over particular types of gear (there are discussion threads online about audio interfaces that are longer than the King James Bible).  I’ve linked to stuff that does (eg.) detailed comparisons of gear or explores topics in more detail than I can. And again, your comments will help a lot.

Part 1: KISS

Well, actually, kissing’s mostly out what with social distance and all, and so is pretty much any real-world gathering where people listen to music. New York just closed all its concert venues and bars.   But we have networks, and software, and some ingenuity, so virtual concerts via live streaming are well within our reach. And providing we remember to keep it simple, stupid — the KISS in the title — it’s something that pretty much all of us can do, individually and together. 

The simplest thing to do, of course, is to record and mix something and then just put it out there via SoundCloud or YouTube or Bandcamp or DistroKid or whatever means you’re already using to distribute your work.  And maybe that’s an occasion to try new stuff in the studio, or up your game as a mix and mastering engineer, or whatever. Worth keeping in mind.

But this post talks about the second simplest arrangement imaginable: one person playing stuff, and sending the resulting audio to be relayed by a streaming radio station.  There are two reasons to start that way.

First, it’s simpler and easier to understand; it also gives a basis for more complex setups.

Second, you don’t really need anything more.  To put that another way, Everybody seems to be hell-bent on doing video, but there are many situations in which video gives less than it takes and should be avoided like the plague.

Don’t get me wrong. My work and that of many other people is greatly improved by visualists like Steve Mokris or Ken Palmer or Jesse Hawley or Joe Howard, who is the only person who has ever made it look like God was listening attentively to what I was doing on stage. But stop and think about what you’re really getting.  Bandwidth is bad in a lot of small venues, and the lighting is even worse. I can’t tell you how many Facebook Live sets I’ve seen where the video looks like a shaky documentary film of people exploring the sunken wreck of the Titanic or a cave on Pluto. A lot of indistinct blobs moving around. Not a good look. Video eats bandwidth.   And your music doesn’t need to do the stop/start, stop/start, snap-crackle-pop, stop/start that happens in a venue where every kid’s got a phone and is killing the connection capacity of the place while, in the meantime, you administer the bandwidth coup-de-grace by streaming a dim view of the top of your head as you bend over your gear.   So ask yourself: do I really need to be seen? Or do I just want to be heard? (never fear, though, I’ll talk about video stuff later).

So, leaving video out of the picture, so to speak, the broadest possible description of what you will need is:

• Something to gather all the audio you’re creating into a stereo feed.  This is often referred to as a “mixer” ;). It may be fed by things called “microphones” or “synthesizers”. (yeah, OK, I’ll knock it off)
• Something to turn that analog stereo signal into little streams of ones and zeros that pass through USB cables and networks.
• Something to take the ones and zeros and package them up and send them off to an Internet radio station somewhere in an agreed-on streaming format. This doodad is referred to as a “streaming client”, because it interacts with a “streaming server”.  Why? Look here.

A great deal of confusion stems from the fact that those three functional components could be — and are —  packaged in any combination. An iPad or a laptop might contain two of the three (the instruments and the streamer) and not need the third (because the instruments are all digital in the first place).  Some interfaces (the things that turn analog signals into ones and zeros) have multiple inputs for signal and contain software mixers that are internal to the interface and run via external software, or controlled by knobs and switches on the interface, thus acting as mixers.   I believe that there are devices that combine mixers and streaming software, though I have not used them (they seem to be designed for podcasters). So. Many. Choices.

Right now, some of you are asking, “well, WTF do I need all that stuff for when I can just push it all out via Facebook Live or Twitch from my phone and be done with it?”.  I can answer that question for Facebook Live: the audio quality sucks, even if it’s taken from a board feed and not just via the teensy microphone that lives in your phone and is now 400 feet from the stage.  Somebody else will have to speak up for or against the Twitch phone client. Surely it’s just as limited by the phone’s microphone, but there’s a good chance it does better with a direct feed from the board than Facebook Live does, because a bowl of instant mashed potatoes does a better job with an audio feed than Facebook Live does.   I know. I tried it.

Oh, and if you want to take a direct feed from the board, you’re going to need an interface to take the board’s analog signal into the phone, so….  Some modern digital boards will send USB audio directly from the board, but I’m not clear on whether any of them do that in a way that a phone or tablet can deal with.  Commenters, please chime in on this. Pretty sure my Allen and Heath doesn’t.

Now, where were we?  Oh, yeah — about to talk about the two components that some will be less familiar with:  interfaces and software to do the streaming part. As to interfaces, people tend to have strong opinions.  Here are mine:

• Lots of people love Zoom’s products and I hope they’ll tell you about them in the comments.  I’ve never owned any. The one time I had to troubleshoot one I could not for the life of me figure out a way to monitor the signal with headphones using the device itself.  Not sure what the issue was. Again, lots of people love these.
• My pick for “least likely to bewilder” is the Focusrite 2i2. It just does the right thing and is indestructible.  It does not have a software mixer, which is reason to rejoice. You stick two XLR or TR cables in the front and you get two channels of digitized audio via the USB connector in the back.  It’s the SM-58 of interfaces, though I have never hammered a nail with one. It’ll take a stereo signal and give you USB audio and that’s all, no other nonsense. They’re pretty cheap, and frequently found for sale by former aspirants to the thrones of James Taylor or Suzanne Vega who have discovered that not a lot of other people care about the deeper details of their angst.
• Lots of people like MOTU interfaces because they’re very high quality and their latency is very low — a big plus if you want to play collaboratively with others over the network.  I have owned and used one for over two years and I still do not understand how its software mixer is operated. That may be inattentiveness on my part. It may also be stunningly poor mixer-interface design and documentation.  But it *is* a good audio interface.
• Not treated here: PreSonus, Behringer, etc. etc.  I will say that I own a Behringer UCA-202. They cost 30 bucks and they work great if all you need to do is USB-ify some analog signal that you can deliver to it via RCA cables.  
• Equally not treated here: more sophisticated interfaces like the iConnectivity products that also handle MIDI routing. I’ve got a few and I love all of them, but they’re a little out of scope here.
• Also not to be overlooked: most manufacturers of these, or anything else, put their manuals online now.  VERY handy when you’re trying to figure out what cable you need to feed device X from device Y, what kinds of inputs and outputs things have, and so on. NEEM techies have started to gather all the relevant manuals for the gear used at NEEMFest in one place, and it’s a godsend for planning.

Useful comparisons of inexpensive audio interfaces are here and here and ten million other places you can Google.  SoundOnSound gathers its reviews here. 

So now you’ve got a thing that makes noise delivered as audio signal, and a thing that takes that analog audio signal and converts it to a digital signal. Next, you need to  send all those digits out in one of the special formats expected by a streaming radio station. “Format” is a pretty all-inclusive term here — it’s really a system in its own right that describes how the connection is made, which audio formats are supported, what metadata you can send along about the song you’re playing, and so on and so forth.  There are two main general-purpose systems used for internet broadcast — Shoutcast and Icecast. Most of the software I know of supports both of those. Other companies — Twitch, Facebook, and YouTube — use proprietary formats that are only supported by clients that they write and distribute. Those are generally designed for phones or tablets or browsers.

Streamers that work with one distribution service and one distribution service only are in some ways easier because they already know what they’re going to connect to and how, and you’ve probably already set up an account with Twitch (gear/software summary here) or YouTube (goofy instructions here)  or Facebook (terse help here) or whoever so they know who you are and have gathered much of the necessary information from you when you got the account.   

More general systems like Icecast and Shoutcast, which work with any number of streaming radio stations, are going to take a little more time to set up.  It’s not hard — it just requires you to provide a bunch of detailed information that you should gather in advance.  I’m not going to go into particulars here, but to get an idea you might look at this page from electro-music.com.  In fact you might look at it anyway, because it is a pretty good rundown on what you need to do to relay through any streaming radio station. And down at the bottom there’s a list of streaming clients you can use.

About those streaming clients…. I’ve used BUTT successfully, LadioCast for the Mac, and I’m now playing around with izicast for the iPad, written by the author of BUTT.  There are many others — a good summation (though probably out of date)is here (most Icecast clients support Shoutcast too).  Again, not talking here about proprietary stuff like Facebook Live or YouTube or Twitch.  I do have a dream of one-piece-of-gear radio performance using an iPad, but I’m hung up right now on the fact that izicast likes Audiobus 3 and I am an AUM kinda guy, and attempts to hook the two up are getting me a lot of snap, crackle, and pop.

One final note:  all of this is only as good as your connection to the network.  I would STRONGLY recommend using a hardwired Ethernet connection to your cable modem or router.  To get really geeky for one second, I suspect that mesh setups that have independent backhaul channels might be almost as good, but I’ll never use them because JamKazam won’t let me.  More about that in the next episode.

All for now.  Next up: musicians in multiple locations.