Camera Decisions

#0004

What a headache I’ve given myself!

I was reading this forum about filming 3D with some more advance cameras such as the Canon 7D DSLR.

Forum Link - http://www.dvinfo.net/forum/3d-stereoscopic-production-delivery/474855-stereoscopic-two-7ds.html

Now granted this forum is a few years out of date now but some of the points made have given me quite a cause for thought.

First of all, an interesting term that I hadn’t come across before, retinal rivalry. It means that while watching a 3D film your eyes are being presented with 2 slightly out of sync images. Apparently this problem can occur if your cameras are recording as little as 20 milliseconds, or more, out of sync from each other. This adds a whole other level of complexity to filming in 3D which I hadn’t even conceived.

Alright, I knew that there was defiantly more to 3D than I understood but this forum has come as a bit of a blow in the face.

However there might be a silver lining in this bombshell.

Now the following conclusions are only my best guess with the knowledge that I have, but it seems that these problems only occur when using more complex cameras such as DSLR’s. For the past year I’ve been using 2 GoPro’s to film my 3D, these are a lot less sophisticated that a DSLR. The article also said that watching very short clips, such as the ones I have on YouTube, wouldn’t reveal these errors, yet watching longer length films could cause server headaches and nausea.

Call me mad if you like, but it seems that the only way I can tell whether my footage has fallen foul of these sync errors and may cause retinal rivalry, is to edit last year’s footage into a one hour long film and see what happens. The results of this will be posted here when available.

What this does mean for certain is that my development of my Mk5 3D rig is now on hold until I understand this mindboggling challenge enough to overcome it.

On the plus side though, this also means that my tests for my Mk2 performance capture setup can go ahead and seeing as the test script for this is nearly finished, the time to turn years of developing theories into something practical may come sooner than I think.