Fly animation

sushidelic · Post by **sushidelic** » Thu Nov 23, 2017 2:56 am

Pardon the desolate condition of the subject and the lack of diffusion although using a ping pong ball... This is just a feasability study and a test of the UMPLANFL 5x.

https://i.imgur.com/aGgWtKU.mp4

https://i.imgur.com/9a0vKxd.mp4

Best regards,
Michael

rjlittlefield · Post by **rjlittlefield** » Thu Nov 23, 2017 10:05 am

Interesting rotation sequences!

Have you posted a description of your setup somewhere?

--Rik

sushidelic · Post by **sushidelic** » Thu Nov 23, 2017 10:22 am

Hi Rik,
that's all Zerene! 30 intermediates that were then retimed to 144 frames in Nuke.
Best regards,
Michael

rjlittlefield · Post by **rjlittlefield** » Thu Nov 23, 2017 10:59 am

Good grief! So, one stack, repeatedly processed with different X shifts? To be honest, I'm surprised by how clean the results are, over such a wide rotation angle, especially with only a 5X objective at what, NA 0.15? How exactly did you process it?

--Rik

sushidelic · Post by **sushidelic** » Thu Nov 23, 2017 11:26 am

Yes, the UMPLANFL is at 0.15 NA, but this stack consist of roughly 250 images, way enough for a smooth rotation. Quality degrades a bit towards the outer ends, which are -7° to 7°, but it's acceptable. And of course it's a straight PMax stack, no way to get such a result in DMap, there's always a hell of jittering around the edges.

edit: Forgot to mention that the retime also smoothens out the frame to frame brightness differences / jitter. Here are the non retimed versions:

https://imgur.com/Y7PG3En
https://imgur.com/u6bH9v4

rjlittlefield · Post by **rjlittlefield** » Thu Nov 23, 2017 2:55 pm

The normal units for shift in Zerene Stacker are percent (%), not degrees (°). Do you mean percent, or are you working with degrees as explained at https://zerenesystems.com/cms/stacker/d ... wing_angle ?

Either way, I guess it does make sense, now that I run the numbers. NA 0.15 means that the entrance pupil is +-15%, equal to +-8.5 degrees. So yes, the shifts that you're using do still lie within the cone of light that was actually captured in the stack, which is my usual guideline for what's required.

But the result still violates my intuition, in a good way. Nice work!

BTW, are these normal PMax? I have not tested, but it's conceivable that UDR PMax would have less variation in brightness.

--Rik

carlos.uruguay · Post by **carlos.uruguay** » Thu Nov 23, 2017 6:59 pm

SUPER

Jacek · Post by **Jacek** » Fri Nov 24, 2017 3:01 pm

Wow, good job, very nice

Beatsy · Post by **Beatsy** » Sat Nov 25, 2017 1:47 am

Yowzer! Those are really cool. Interesting that the little flares and such are so consistent across the different stacks too - makes them far less objectionable than they can appear in a single frame. Neat.