Page 1 of 1
1st test on octocopter in "real" conditions

Posted:
Sat Apr 13, 2013 12:40 pm
by Wolf Schiebel
Hi all,
here one of our first test shootings with the BMCC on our new Flightcams Q4 Cinecopter 2.5 K System.
Shot in prores, film log, Samyang 14 mm/3.1 cine prime. This is a quick h.264 playout with FCP, original prores version with Davinci grade will follow next week.
The original clip is really stunning and i believe with this camera and soon the 4K a new aera in aerial filming has just begun.
Would love to hear your opinions.
Best,
Wolf
Re: 1st test on octocopter under "real" conditions

Posted:
Sat Apr 13, 2013 1:14 pm
by Benton Collins
The motion is very smooth, but the picture is too soft IMO. Shows great potential though.
Re: 1st test on octocopter under "real" conditions

Posted:
Sat Apr 13, 2013 1:29 pm
by Wolf Schiebel
Yes, i agree to that. The h.264 codec softens the picture and the samyang is not the sharpest lens also.
We will try with with some older Nikon 16 and 24's soon.
Re: 1st test on octocopter under "real" conditions

Posted:
Sat Apr 13, 2013 2:11 pm
by Felix Steinhardt
I can totally recommend the Nikon 10-24 for wide angle. It´s pretty light, too.
http://www.imagebanana.com/view/ukf8atai/24.jpg(click for 100% view)
Re: 1st test on octocopter under "real" conditions

Posted:
Sat Apr 13, 2013 3:00 pm
by Randy Walters
Really has fantastic potential; my problem is that I'd be totally white-knuckled, with my heart in my throat every time the camera went aloft.
I think the Pocket Camera will do a great job of making that terror a little less overwhelming.
Re: 1st test on octocopter under "real" conditions

Posted:
Sat Apr 13, 2013 3:23 pm
by Christian Schmeer
Looks stunning! Maybe try to sharpen the image a little in Resolve. I have found that BMCC footage sharpens very very nicely in post. How much are these octocopters anyway?

Re: 1st test on octocopter under "real" conditions

Posted:
Sat Apr 13, 2013 8:09 pm
by Wolf Schiebel
Christian Schmeer wrote:Looks stunning! Maybe try to sharpen the image a little in Resolve. I have found that BMCC footage sharpens very very nicely in post. How much are these octocopters anyway?

Hi,
the systems are available in rental only, with menpower. About july we will present a system for sale based on the cinecopter. Thanks, i will try Resolve next week after setting up the new Mac Pro.
Best Wolf
Re: 1st test on octocopter in "real" conditions

Posted:
Sat Apr 13, 2013 8:48 pm
by Jason Hinkle
Those shots tracking towards the house are amazing. With the buzzing heli blades it feels like an ominous scene from a high-budget horror movie. I can imagine many amazing shots with this that would be very difficult if not impossible to get any other way.
I have some simple indoor helicopters so I can appreciate the skill it takes to fly those things. I would be terrified to fly it over a lake!
Re: 1st test on octocopter under "real" conditions

Posted:
Sat Apr 13, 2013 9:04 pm
by Benton Collins
Wolf Schiebel wrote:Yes, i agree to that. The h.264 codec softens the picture and the samyang is not the sharpest lens also.
We will try with with some older Nikon 16 and 24's soon.
I don't think the problem is with the h.264 codec as, unless I'm mistaken, that is currently the best codec for streaming and I have seen some beautifully sharp footage streaming with this format. Plus I'm fairly sure you cannot stream a Prores file.
Re: 1st test on octocopter under "real" conditions

Posted:
Sat Apr 13, 2013 9:06 pm
by Benton Collins
Randy Walters wrote:Really has fantastic potential; my problem is that I'd be totally white-knuckled, with my heart in my throat every time the camera went aloft.
I think the Pocket Camera will do a great job of making that terror a little less overwhelming.
I fully agree that the Pocket Camera is the way to go in aerial rigs like this! I'll bet that the Pocket Camera becomes the #1 camera to fly with.
Re: 1st test on octocopter in "real" conditions

Posted:
Sat Apr 13, 2013 9:09 pm
by Peter J. DeCrescenzo
It's all fun and games until someone gets an eye poked out.
First world problems are fun to watch, though. Cheers.
Re: 1st test on octocopter in "real" conditions

Posted:
Sat Apr 13, 2013 9:39 pm
by ungovernedreason
hahaha I have to say I loved this! defiantly a horror movie look. And good job flying that thing! I am suprised that the rolling shutter was not a bigger issue however...would the pocket cam not have a way more difficult time with the jello?
Re: 1st test on octocopter under "real" conditions

Posted:
Sun Apr 14, 2013 11:08 am
by Wolf Schiebel
Benton wrote:
I don't think the problem is with the h.264 codec as, unless I'm mistaken, that is currently the best codec for streaming and I have seen some beautifully sharp footage streaming with this format. Plus I'm fairly sure you cannot stream a Prores file.
Yes, that's right, but uploading a prores version to vimeo results in a better quality after the internal vimeo compression, as far as i know. We will try that after setting up our Davinci next week.
I personally don't believe in the pocket cam for (big screen) aerial video, since the rolling shutter might be a much bigger issue and with the small sensor post-stabilizing might be far more difficult. But we will try that also, for sure

.
Best Wolf
Re: 1st test on octocopter under "real" conditions

Posted:
Sun Apr 14, 2013 1:59 pm
by Benton Collins
Wolf Schiebel wrote:Benton wrote:
I don't think the problem is with the h.264 codec as, unless I'm mistaken, that is currently the best codec for streaming and I have seen some beautifully sharp footage streaming with this format. Plus I'm fairly sure you cannot stream a Prores file.
Yes, that's right, but uploading a prores version to vimeo results in a better quality after the internal vimeo compression, as far as i know. We will try that after setting up our Davinci next week.
I personally don't believe in the pocket cam for (big screen) aerial video, since the rolling shutter might be a much bigger issue and with the small sensor post-stabilizing might be far more difficult. But we will try that also, for sure

.
Best Wolf
So I can upload a Prores file directly to Vimeo? Can it be 4444 and 422? Thanks!
Re: 1st test on octocopter in "real" conditions

Posted:
Sun Apr 14, 2013 2:56 pm
by Frank Glencairn
The 14mm is pretty soft wide open - especially at 2.8 on a 3.1 lens

(T-stop, I know, just mocking)
Stop down to 8ish and make sure infinity is really infinity, and you are golden.
Frank
Re: 1st test on octocopter in "real" conditions

Posted:
Sun Apr 14, 2013 4:19 pm
by Wolf Schiebel
@Benton: As far as i know both is possible, since converted and compressed by Vimeo internally.
Hi Frank, yes, still learning the cam. We had stop at 16 but i didn't check infinity properly... First goal was to get the system in air and stable, without jello and rs, and that worked out quite well. Now comes finetuning....and learning to grade. Next time we will shoot in raw also. What shutter angle would you recommend, specially for the wide shots and good post-stabilization?
Thanks, Wolf
Re: 1st test on octocopter in "real" conditions

Posted:
Sun Apr 14, 2013 4:31 pm
by Frank Glencairn
Stay with 180deg if possible.
Faster shutter my introduce temporal aliasing, that is not fixable in post.
With a wide lens, you should not need much stabilization anyway.
f16 may be already too much on this glass, and things getting soft.
Around f8 is the sweet spot - compensate exposure with a ND if needed.
Re: 1st test on octocopter in "real" conditions

Posted:
Sun Apr 14, 2013 5:32 pm
by Wolf Schiebel
Ok, thanks will try next time. But that explains the soft image, since we were on 90 deg. and 16.
Best Wolf
Re: 1st test on octocopter in "real" conditions

Posted:
Sun Apr 14, 2013 6:12 pm
by Benton Collins
Frank Glencairn wrote:Stay with 180deg if possible.
Faster shutter my introduce temporal aliasing, that is not fixable in post.
With a wide lens, you should not need much stabilization anyway.
f16 may be already too much on this glass, and things getting soft.
Around f8 is the sweet spot - compensate exposure with a ND if needed.
I agree with Frank on this. My rule of thumb is to stay away from the last two stops on ANY lens. A stop somewhere in the middle of the range is usually the best and that is usually f8. You can push that to f11 at the most, but only if the lens goes to f22. While it's true that as you stop down the apparent sharpness in the depth of field increases in the foreground and past your actual focus point, but as the aperture goes past the mid point and the foreground and background increase in apparent sharpness, the actual critical focus point begins to soften. The softening of the critical focus point continues to get softer as the lens is further stopped down. So, instead of achieving an "everything in sharp focus" we have nothing that is in true sharp focus. Yes, the close foreground can appear more defined, but at a big cost to the rest of the image. f8 is great!
Re: 1st test on octocopter in "real" conditions

Posted:
Sun Apr 14, 2013 7:24 pm
by Wolf Schiebel
Ok, thanks guys!
Still have to learn a lot.... That's why we always work in teams of three (pilot,Dop and tech. supporter) normally.
Re: 1st test on octocopter in "real" conditions

Posted:
Tue Apr 16, 2013 7:54 pm
by Wolf Schiebel
Ok,
here my first Resolve tryout:
For comparison the old/ungraded file:
Next time i will shoot in raw, and with the right exposure..

Cheers,
Wolf