saw141 wrote:So I've been fiddling with this some, and I have to say. I've read a lot on this site, but after reading
https://www.blurbusters.com/motion-test ... it-camera/ some serious kudos are in order. The amount of work that has gone into this research is astounding, and that it's free is beautiful. Thank you so much Mark, and everyone who is involved, I really mean it. I don't mean to be overly sentimental but science is fascinating, and this is science, great work.
Thank you for the compliment!
FYI, the
Easy Pursuit Camera Instructions (end user pursuit camera) are in that thread. The video version rather than the still-photo version.
The old professional-site instruction tell you to use photography (still camera + rail) but for end users using hand waved iPhones should use video recording instead because it's easier. Then just grabbing the clearest video freeze-frames since video is simply like 100-shot burst-shooting.
So that gives you literally 100 chances to find a clear shot in all the shaky-handwaves. That's what makes the hand-wave technique suitable for some end-user troubleshoot purposes. Instead of 1 still photo, video can contain hundreds of still frames, and that's why
video makes rail-less pursuit camera easier for end users.
The overkill number of shots compensates for the lack of rail!
I'm going to post a new BlurBusters HOWTO for end user pursuit camera soon
(rail-less hand waved smartphone via video recording + screenshotting the clearest freezeframe).
saw141 wrote:Here is one of the most clear lines I could achieve with my iPhone 6s, set to 1080p@30fps.
I see each frame capturing 6 refresh cycles. Try a 1/60sec-per-frame camera exposure for video-filming 240Hz blur reduction. To do that, you could just do 60fps video. Or use a third party camera app such as Pro Cam+ or another app that can let you program camera exposure per video frame. But not everyone wants to bother.
Right now, you can just use still photography to capture strobe crosstalk. Just set exposure to one refresh cycle (1/240sec) and the resulting photograph will look almost perfectly WYSIWYG. This is because brief strobing (sub-1ms) actually point-samples the motion (no handwave blur for a strobed refresh cycle) and thus almost eliminates the photographic difference between pursuit camera and still camera.
So as a rule of thumb, to make it easy for users:
Strobe backlight: For WYISIWYG strobe crosstalk photography, it is okay to use a still camera, exposure set to 1 refresh cycle
Non-Strobed: For WYSIWYG motion blur photography, pursuit is better. The current Ghosting track has four tickmark levels, which means it prefers you set camera exposure at four refresh cycles. So you may have to reduce to 1/60sec instead of 1/30sec because of the way I designed the old Ghosting track in the pre-240Hz days. Or just simply use a 60fps video filmrate. The count of refresh cycle exposued should be greater than 2 in all situations, and it's okay to have more than 4, but the Ghosting track was designed to only stack-capture up to 4 refresh cycles accurately.
Maybe we should post a separate thread if we're going to do a massive number of end-user strobe photographs or pursuit camera photographs. I'll probably create a thread once I create an Easy Pursuit Camera FAQ on the Blur Busters website.