Another AJA Gen10 Genlock Test

[Phil Bates]

For what it's worth, we duplicated Pedro's fan test. This time with a different frame rate of 29.97. We ran three identical tests in a row, powering everything off in between. All three tests showed accurate genlock (subject to the visual limitations of the "fan" method).

Here are the settings/parameters:

Frame rate: 29.97 fps
Shutter speed: 1000th/sec
Gen10 Jumper switches: 1-2-3 (Pedro used 1-2-7)
Gen10 Outputs 1 and 5
RED MX Camera Build: 30.5.0 (we have not yet upgraded to BETA 30.6.0)
Visual sync method: Clapper slate

Here is an image showing output from both cameras. All 3 tests showed identical accuracy:

http://dl.dropbox.com/u/5981875/Uploads/Fan%20Genlock%20Image.png

We did not test any other frame rates, jumper switches or shutter speeds. I hope this test is helpful, even though the build is 30.5.0.

Phil

PS: Thanks to Annette Gaillard for running this test. :)

[Jason Goodman]

I had initiated these tests when Pedro was working for me as DIT on Julia X just over one year ago. They were conducted at Stereoscope in Burbank with the help of 3D legend John Rupkalvis. At the time, we were testing build 21 which was the first to allow the preview output to be genlocked (necessary for using the Transvideo 3DView monitor). When testing with the fan, Pedro and John got basically the same results you got. Genlock is not a problem with RED ONE.

For ultimate 3D sync accuracy, what you really need to test is shutter phase. To critically test phase, photograph something much faster than the fan like the refresh of LED numbers on a smart slate with narrow shutter angles. As we ultimately concluded from the results, it is close enough that it is rarely noticed.

Since conducting those tests we've shot everything from bikini car washes (http://www.youtube.com/watch?v=M6h7ZieZH9U&NR=1) to drift racing (http://www.youtube.com/watch?v=nHP_guc_i6k) at speeds up to 120fps and nobody has complained.