…I’ve got a secret I’ve been hiding under my skin,
my heart is human, my blood is boiling, my brain IBM.
1983 © Styx, Mr. Roboto
So you start with one camera, and all is good. (*)
(*) For a reminder as to why we write about cameras, revisit this rant.
You record for posterity, document activities, accomplishments and accolades, and move on to add style in addition to substance. And this might mean another camera (and another operator), and another camera (and one more operator), and another camera (and yes! yet one more operator), and you reach a point where there is simply not enough people available to act as operators. Especially when said tasks usually involve the occasional pan and tilt, focus pull and zoom operations.
With typical recording scenarios, this is not an issue – recordings can be resumed. As we move towards live production, however, things get a bit problematic as the complexity of getting both a number of operators (not to mention trained operators) within a pretty non-existent budget gets really apparent. And, honestly, having operators beside each camera for the occasional operation seems a bit of a waste in resources for such small scale productions…
…which is why you get a robot. Well, robots.
But before we delve into the robotic apocalypse, a few words about why we need robots – we use up to five DSLR cameras, more often than not with large aperture prime lenses, outputting HDMI to a live production switcher, covering the video side. We then have a minimum of two condenser shotgun microphones and a few lavalier microphones just for kicks into a dynamic microphone preamp, covering the audio side. And then we have lights, and graphics, and live switching adding further complications to what is already a mess.
For more information about how we set up our live production system, you can read through a separate rant.
Suffice it to say, a typical workflow would require a minimum of eight operators. Yet we need to make due with three. Can we do it? Who knows – we will certainly try, and you can read on to see how successful we are, and where we stumble along the way.
Now that we know we need a robot (well, robots), let us see what they need to do when replacing human operators:
- Move the camera along two axes, so pan and tilt. Sounds doable.
- Perform zoom and focus operations on manual lenses. A bit problematic.
- Perform advanced follow focus routines. Still within the realm of possible.
- Provide camera feedback to the live production switcher and implement a tally system. OK, now you are pushing it.
- Perform hardware based live production switching. Fine, I give up.
That said, let us step back a little.
We are not reinventing the wheel. We are well aware of the fact that such systems (ahem, robots) exist, and are probably very, very capable. And even if they don’t, all the components are readily available, and any system combination is readily doable. We are, at the same time, aware that such systems cost a pretty bundle, and as such are out of reach for the nihilist budget of the independent filmmaker.
So it shouldn’t suprise you that we decided to build our own. For a little bit of money, and a little more of elbow grease, it should be a cinch. Right? Right!
Well, somewhat right. But look on the bright side – if we were able to do it, anyone can.
Given the complexity of the project, we are splitting them up to smaller projects, in a somewhat chronological order:
- Pan & Tilt
- Remote Zoom
- Follow Focus
- Live Switching Control
- Remote Tally
Each smaller project represents a category on the blog, and might contain multiple posts (depending on the amount of mistakes we do, of course).
At the end, we either come up with Frankenstein’s monster, or we pat ourselves in the back. Either way, you get to follow our trip and enjoy our mishaps.
Until next time, Kilroy Was Here (*).
(*) Yes, the irony is not lost on me.