Interview with Chris A. Church

About three months ago while doing research on time-lapse photography I ran across the website.   Some of the work being done there by Tom Lowe, Jay Burlage aka Milapse and others is amazing. What blew me away is that these are not your typical time-lapse images shot with a camera on a stationary tripod.  These guys have worked out some great techniques for putting the camera in motion.

After doing some more research I found out this technique was pioneered by cinematographer Ron Fricke for has film Baraka.  In a nutshell what Fricke and his team developed was a motion control dolly system that enable them to pan or tilt the camera while shooting time-lapse footage.  The question then was how were the DIY on Timescrapes pulling off building motion control systems without paying thousands of dollars.  

After a little more digging around a found there is a whole Open-Source photographic motion control community whose home is .  The site is run by Chris A. Church aka. Shutterdrone.  Chris is the brains behind the software that’s running some the of the coolest motion control rigs for the DIY crowd. 

1.  So tell me who is Chris A. Church?

As to who I am, well, I’m the kind of guy to say something very humbling about myself, and then follow it up with a statement how annoying it is that people always respond to these sorts of questions with disingenuous inflections of humility.  Most people describe me as a workaholic, but I think I just have a knack for finding fun hobbies and neat projects to work on, and that’s what I spend my time doing.  Generally speaking, I like to split my time between office/shop, and the outdoors – everything else is done en route to one of those two.

2.  How long have you been into photography and when did you develop an interest in time-lapse?

I first got into photography in the mid-nineties, but the bug never really bit me until I got into medium format film and pinhole cameras about ten years ago.  I’ve always enjoyed the technical nature of photography, whether it was building a box camera, or tweaking chemical temperatures to see what effect it had grain sizes in particular films.  At some point, I realized that I was just burning so much money on film to play with it, if I ever wanted to actually get any pictures printed; I’d better switch to digital. It was just over three years ago I decided that time-lapse would be an interesting diversion, and I met up with Tom, Jay, and the other guys from the timescapes forum right as they were all also just starting to get into DIY moco rigs.  As you can imagine, that brought out the technician in me again, and I dove in head-first, forgetting some times that the goal was actually to make time-lapse films…

3.  What are the goals of the OpenMoco Project?

The main goal of the OpenMoco project is pretty simple: to create and share motion control solutions based on open hardware and open software.  Of course, we’re interested in supporting and helping any open-source photography project and helping people solve novel problems without resorting to expensive, closed hardware or software.  Our focus remains largely on the Arduino platform at the moment, but a goal is also to get everything ported over to the ARM platforms out there to craft even more expansive capabilities while still being able to run on hand-held form factors.

4.  What part of the project has been the most challenging to develop software or hardware and why?

One of the difficult aspects of working with hardware hackers is that we all get here by a sense of self-design and a need to express our creativity fully.  In light of this, one of the most difficult challenges is building a system flexible enough, and complicated enough to encourage the brave to not re-invent a new wheel each time, without also hindering those who are just looking to get their problem solved quickly and easily.  Writing software or designing hardware in and of its self isn’t particularly challenging (I’ve spent fifteen years in professional software design and development), but setting strict requirements and boundaries create the hurdles that make it a worthwhile achievement.  “It has to be built for less than x dollars,” or “it has to run using less than 1k of memory…”  That’s where the real interesting and fun challenges lie.

5.  The Openmoco project has two solutions for controlling motors, OpenMoco Slim which is a text-based interface to the your Arduino base timelaspse engine which runs stepper motors.  The other being the MX2 dolly engine which is a combination of Arduino microntroller with an attached Arduino motor shield that runs DC motors. What are the advantages of one over the other?

The core TimeLapse Engine, which forms the basis of the reference design, is to enable any sort of motion control, for up to four axes, using stepper motors.  This is the Swiss Army knife of moco tools – it can do just about everything, but if you just got one thrown in your hand, it’s going to take some time to figure out where that right tool is for the job.  In its current form, it’s really designed to entice the software developers and the seasoned makers to focus just on what they have to to come up with the next great motion control tool.  It has accuracy, scripting capabilities, keyframing, the ability to react to the environment and its own activities. Slim is just a scripting interface into it, meant to demonstrate what the API it provides can do, to experiment with capabilities, and to support debugging of software development around it.  Until we finish the GUI for the TimeLapse engine, its complexity keeps it a bit out of reach of the novice.

The MX2 DollyEngine was designed to be an absolute lowest-cost solution with minimal entry requirements.  Whereas the TimeLapse Engine requires four additional motor drivers, a whole lot of soldering, a separate device for user interaction, the MX2 packages it all into a single board (plus arduino).  The choice for DC motors here is just to get the job done easily, and at a low-cost.  For about 80% of what people want to do in time-lapse, DC motors are perfectly fine for the task, and the theory behind their operation (not to mention the wiring) is simple for a newcomer to get quickly.  Of course, the MX2 has several broken-out I/O ports, so one could easily use it to drive external stepper drivers, making it a great learner platform for the first time experimenter.

6.  On October 1st you started selling a version of your dolly. Do you anticipate any conflicts with selling this product and the open-source philosophy of the project?  Bottom-line will you still be out there helping others to develop Openmoco systems?

Absolutely, the Dynamic Perception project was not to distract from the OpenMoCo project, but instead to enhance it.  I kept running into two problems: the first was that there were a lot of guys with a nice garage workshop who could put all of the mechanical stuff together, but no electronics experience who kept asking if they could buy this circuit or that from me, so that they’d have a stable platform to start their learning experience on.  The second was that, like in any project, you can pick money or time, and this is an expensive hobby!  There are a lot of things we wanted to achieve, but funds were always limited – we could sell systems and put that money back into R&D to create the next generation of open systems to share with everyone.  Our hope is that the success of DP can be turned into an accelerator for all things in the OpenMoCo project.  The continued success of DP is intricately tied to OpenMoCo:  I am a firm believer in open-source software and open hardware, and truly believe that when one eliminates price and the bottom 20% of the features (those that 80% of the shooters rarely use), there isn’t much left to differentiate players in the same field.  Open-ness and helping the community grow – even if it means that others can capitalize on our work – is a key tenet of the business. We’re even in talks with a few people who are building competitive systems on the OM reference design, and helping to enable them rather than hinder them.  We hope to be seen as a knowledge base more so than just another company selling moco gear.

7.  After reading so much about stepper motors on the site I was a little surprise to find you went with a DC motor on the Dynamic Perception Time-lapse Dolly Stage Zero, are you planning on coming out with a stepper base version of the dolly?

Absolutely, we’re working right now on a four-axis stepper driver board that will be both low-cost and powerful, and a GUI that can be run on numerous hand-held platforms.  The Stage Zero is aptly named, as it is a building block – we will be providing lots of upgrades for it, stepper upgrade, 2nd and 3rd axes, and some fairly neat configurations that I don’t think anyone’s seen before.  These upgrades will open up some shooting techniques that have only been imagined at this point, and put them within reach of the hobbyist.  Obviously, the whole repeatable move aspect is going to be there, but we think the envelope needs to be pushed even further.

8. Okay thanks Chris, how that you have the Dynamic Perception Time-lapse Dolly Stage Zero up and running what are your future plans?

At the moment  were starting to work together with the Apertus(open-source cinema) project to merge both the open motion control and the open video camera together into one integrated system.

I would like to thank Chris for taking the time to answers the questions here and putting in so much time in development of the OpenMoCo software.

2 Responses to “Interview with Chris A. Church”
  1. Tom says:

    love this rig. is there any rental house in NYC with this?

Leave a Reply

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

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

Google+ photo

You are commenting using your Google+ 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 )


Connecting to %s

%d bloggers like this: