I must admit that I feel like I missed the live-stream hype train.  I never really understood why people got into watching other people play video games. I’ve heard people liken it to watching professional sports by referring to the game chart on this site, but honestly I don’t really understand why people get so into that, either.  Despite my inability to understand its appeal, I know that live streaming has produced a significant fan base with incredible numbers.  Twitch’s numbers approach 1 million concurrent viewers daily. With numbers like that it’s not surprising to see more and more indie developers streaming their own game development. And so, despite not truly understanding why someone would want to watch me work, I tried my hand at live streaming my development sessions this week.

---

Disclaimer: I know almost nothing about live streaming or Twitch. I literally made my account this past week. If I’m wrong on anything I say here, please feel free to correct me in the comments below.

---

Before I could try to stream my own content I wanted to see what other developers were putting out there.  I checked a few of the more popular dev streams and I noticed a few trends.

1. Good developers talked to their audience: The most watched devs explained everything they were doing in great detail.  These streams weren’t as much developers working on their games as they were developers explaining how their game works to an attentive audience.
2. Panel information on a channel is important:  Popular devs all seem to have well laid out panels at the bottom of their channel. Information like when the next stream is happening, where people can follow development, and what the game is about is all laid out right there.  Dev’s with an art budget usually spin it up into aesthetically pleasing images. Some even use this space to run contests or allow donations.
3. Audience count tied to developer more than game: I may be wrong in this one, but from watching these streams it seemed to me that the number of people in any stream had more to do with how charismatic the developer was than how interesting the game they were working on was.  I guess this makes sense, though. People can’t play the game yet, they can only really watch the developer, so in this case the developer might be just as important, if not more-so, than the game they are showing off.
4. Use a camera: Developer streams without a camera were always near the bottom when sorted by viewers. I think this plays into the idea I mentioned in the previous bullet point: the developer is as important as the game itself.

So I spun up a twitch channel ( http://www.twitch.tv/jimmothysanchez  for anybody interested in following me) and gave it a go. I streamed both Tuesday and Wednesday night. Some things worked and went well, and some things did not.

 First, the things that did not work well:

1. Music: As a rule, whenever I’m working on something, I need music to keep me concentrated. That’s just how I work. The louder the better and the genre of choice is usually metal.  Unfortunately this doesn’t really work when you are live streaming. I tried turning it down a bit and talking over it but people complained that they couldn’t hear me. I believe I was having some technical difficulties with the stream because even when the music was practically turned off people said they couldn’t hear me.  I plan to look into it, but for this week’s stream I had to work in silence.
2. Distractions: I was a little worried about this one going in.  I wasn’t sure that I’d be able to concentrate on my development if I was constantly checking the chat logs. While this was a bit of a hindrance it wasn’t as bad as I’d expected. Generally people just asked questions about my game or gaming in general, which in turn actually made me feel better about the development work I was doing.
3. Subject Matter: I decided to stream this past week’s development without really thinking if it would be fun to watch. This week I was scheduled to continue work on my procedurally-generated map. This is not fun to watch. It’s painfully boring. Most of the time I was hunting through code looking for minor bugs that were wrecking everything. It’s interesting to me because I’m building it, but it’s a horrible topic for a live stream.
4. Stage fright: This one surprised me. It wasn’t until I got my camera working that I became supremely self aware. It’s weird seeing yourself out of the corner of your eye in your second monitor.  You start picking apart your own facial expressions and being overly critical of everything you do. I had to close that window pretty quickly. On top of that I became really worried every time there was a block of silence or whenever I hit a bug that I was being un-entertaining. It’s a weird feeling that is hard to put into words, and is definitely higher stress than just a regular dev session.

Things that did work pretty well:

1. People watched it:  Despite everything I did wrong, some people tuned in. I kind of expected to have to stream a couple of times before anyone ever joined, but from night number one I got a few viewers. The first time anyone wrote to the chat log, I was so surprised that it took me a second to decide how to best handle the situation, speak aloud or type back. I didn’t get a huge audience, but some people tuned in.
2. New subscribers:  Not only did some people tune in, but a couple of people subscribed to my channel.  I even got some traffic to my blog from my Twitch stream.  It’s nice to know that some people found my first attempt interesting enough to keep an eye on me.
3. Being watched kept me on task:  When you’re live streaming you get presented with the distraction of having to keep an eye on the chat window, but it also removes other distractions, such as the desire to just quickly check Facebook or Reddit. You’ve got people watching you so you can’t procrastinate or get side tracked the way you could otherwise.
4. It was pretty fun:  This was the big one. I wasn’t expecting it to be as fun as it was.  I really liked how interested some people were in my game. I answered a several questions about my game and I got to have discussions about it that I generally don’t get to have. I really enjoyed it.

So in summary, I think my experimental attempt at live streaming went well, all things considered, and I’ll try it again.  With enough time and experience, most of the hiccups should work themselves out.  I’ll still probably go private for more in depth and boring tasks like procedural generation, but I’ll give this won’t be my last live stream.