A Few Thoughts On System Design

I've done a lot of system design in my career. In college, I was tasked with coming up with some pretty elaborate systems for pulling together large multi-image festivals. Afterward, I designed temporary systems for very elaborate corporate sales meetings. When I started my company, I did all my own edit suite design and install, and built more than one flypack. In fact, I really enjoy doing system design. There's a certain sense of satisfaction that comes from figuring out the most elegant way to solve a problem and seeing it through to fruition. When I think back on how I've approached each design, a few things stand out.



Ask a Lot of Questions

The first step in the design process is not design at all. Rather, we need to ask a lot of questions. What are we trying to accomplish? Who will be running the system? What is the budget and time frame? What are our current assets? Any physical constraints? Often these questions beget more questions. For example, in designing a live video system the first question may be, "What are we trying to accomplish?" The answer might initially be, "IMAG and maybe internet streaming." At that point, we need to know a lot about the room, light levels, projectors, current (if any) infrastructure, desired level of quality and the like.

The answers to those questions start to shape how you approach the design. In the above example, if the room is 100 feet deep, you had better allocate a big chunk of the budget to lensing. If there is not much existing light, HD is going to be an issue unless you upgrade the lighting grid. There are a lot of factors that come into play, and design is all about compromise. The trick is to not compromise on the important items.

Think a Few Steps Ahead

Too often systems are designed to simply handle the task at hand. Lighting systems are a great example. Let's say you currently have 48 lighting fixtures. You may think putting in 48 channels of dimming is sufficient. However, the day will likely come that you will need additional fixtures. If you planned well, you can plug them right in (because your dimming channels can handle more than one fixture) or you can slide some more dimmer trays in the rack. If you didn't plan well, you have a mess on your hands. Fixing it will involve bringing in electricians, adding racks, running new lines, and significant cost. If you really didn't plan well, you may have to replace everything.

Always look down the road 3-5 years (more and more, I'm trying to look 5-7). Think about what you may want to do in the future, and the best way to accommodate that. Sometimes, designing in a few extra pulls of wire, a couple extra conduits and buying a slightly larger controller than you think you'll need will save you thousands of dollars and a lot of headaches in the future.

Don't Be Afraid To Blow Up The Plan

I'm actually notorious for this. After we spent 3 weeks getting our new lighting plan in place, I blew the whole thing up. I started thinking about losing dimmers altogether and going all LED. I re-configured everything. Then I started getting pricing. It was all over budget. So I blew it up again. This time, when we came back to the original plan, some of what we learned in the re-design altered the original. We now have a really good plan in place that is more cost-effective than it was originally, yet gives us more capability and more future expansion.

You see, it's easy to design what you know. In our case, we brought in an outside designer to help. He's good, and he's used to doing things a certain way. When we blew it up and said, "What if we did it this way?" it changed the approach. That led to, "Well if you do that, you could also do this, this and this. And come to think of it, that's better."

I could go on for quite a while longer on this topic, as it's one that's close to my heart. Mainly because I see so much bad design out there. Later on this month, Dave Stagl, Jason Cole and I will be hosting another webinar on this very topic. We're excited to have special guest, Bob Nahrstadt of Clark ProMedia joining as well. This is good because when I say I've done a lot of design, it needs to be put in perspective; Bob's probably forgotten more designs than I've done. It should be an interesting talk.

Join us on TokBox for the webinar. We're moving it to Friday night in hopes of accommodating more people who have Thursday night rehearsals. Save this date: January 29th at 10 PM EST, 7 PM PST. Watch our Twitter feeds and this blog for a link. And we'll post the audio here and on iTunes if you can't make it.