DVX Geniuses UNITE! 3K video solution. Less than 3K?

Maybe the GH2 40fps frame burst mode will offer some 3K action.

If we can adjust it to 24p and it can burst for 10-15 seconds... could be quite an interesting little feature.
 
For the HDR set up - using a teleprompter mirror looks to be a better idea as the transmittance / reflection is not split 50% 50%.

In this example (below) the light is split 70% transmittance and 30% reflection - so you could expose the darker areas (where you want to get more light) trough the transmitted side and expose for the lighter areas (where you want less of the light) through the reflected angle.


http://cgi.ebay.com/Teleprompter-Mi...ltDomain_0&hash=item3ca72bae36#ht_1744wt_1055


Two way mirror is even more extreme in it's transmittance / reflection difference (1:10) - so would only be good on very very bright days, almost shooting into the sun as you would be getting only 10% of the light from the transmitted side - good for exposing very bright skies (?) and 90% from the reflected angle which would work for the darker exposure.

You would need to do some tests to see what the best ratio of transmittance / reflected light is better for your two over and under exposed exposures - but I expect 50%/50% might not be optimal as the two cameras are getting the same amount of light.

Cheap two way glass:



http://cgi.ebay.com/Two-Way-Mirror-...ultDomain_0&hash=item439c949e04#ht_500wt_1070
 
For the HDR set up - using a teleprompter mirror looks to be a better idea as the transmittance / reflection is not split 50% 50%.

In this example (below) the light is split 70% transmittance and 30% reflection - so you could expose the darker areas (where you want to get more light) trough the transmitted side and expose for the lighter areas (where you want less of the light) through the reflected angle.


http://cgi.ebay.com/Teleprompter-Mir...ht_1744wt_1055


Two way mirror is even more extreme in it's transmittance / reflection difference (1:10) - so would only be good on very very bright days, almost shooting into the sun as you would be getting only 10% of the light from the transmitted side - good for exposing very bright skies (?) and 90% from the reflected angle which would work for the darker exposure.

You would need to do some tests to see what the best ratio of transmittance / reflected light is better for your two over and under exposed exposures - but I expect 50%/50% might not be optimal as the two cameras are getting the same amount of light.

Cheap two way glass:



http://cgi.ebay.com/Two-Way-Mirror-S...#ht_500wt_1070
Click to expand...

You are ABSOLUTELY RIGHT....on both counts. I had actually found the same two-way mirror. Also figured that you'd lose A LOT of light...(too much). I liked the tele prompter solution because it looked good, was already complete and finished...didn't like the price. :( You found a much cheaper one than I did...so yours may actually be feasible? Buying 4 cameras...with 4 lenses is going to be a PAIN. Any added cost from the rig is going to be a growing problem. That said....here's something for EVERYONE TO REMEMBER....make no mistake. If we can do this (still an 'if')....this would be the cheapest 5K RAW cinema option. 35,000usd cheaper than the next competitor. Yes..the workflow would be a pain...but...I'd do it. Heck...a little old school hard work in post...didn't hurt any of the great film makers in the past....so....let's keep chipping away. Thanks Lee.

Here's some stuff I'd like to get done in the next 2 weeks.
1. I'd like to build the multi-splitter rig. I'm pretty busy, but I don't think it's out of reach.
2. I'd like to post/configure the "advertising/donation" site. I already know where I'm going to do this. I think a reasonable target number for this will be 10K-usd. I'll make a video that explains what we're trying to do and what the money will be used for. The bottom line is....we want for other people to be able to PREDICTABLY repeat the process that we create and get 16.5 sec of 5K(ST)24P...and know how much it will cost. This will cost $$...but I think we can get enough interest, based on the potential outcome.

I don't know how many people are actively wathching/reading this....for all I know, it's only Lee and I so far (consistently)...and that's fine. But if anyone else sees what 5K(ST)24P could do for all of us...here's how you can help. (I've already added this in the other threads ...but no-one is using them yet)

Triggers/synching
We need someone to work on the synching. We'll HAVE to get there later (on the rig)...we'll be using triggers I'm sure. So we need some cheap but dependable triggers. (I know there are some reviews on "cheesycam's" site....I was thinking to start there.) If someone wants to do that...it would help. Remember cost is important because we want others to follow our footsteps. So...high quality is important...but cost is too.

Workflow for 5k(st)24P
My thoughts going from camera to timeline...what are the time-lapse people doing? If someone is already doing that...it would be nice to get your input on this part.

Auto Ordering App
We do have one additional step which is ORDERING the images from 4 different cameras. That shouldn't be too tough....but it would be cool to get some software/batch solution to do it for us...you know...drop 4 folders inside one folder and it auto orders them based on number (Low-high), then sequences them. (Any developers in the house?)

DNG Video?
The other thing that I don't think any time-lapse folks are doing (this is a BIG ONE)...is DNG VIDEO. This is Adobe's open source solution for RAW video. They say that it is already being used for scanned film transfers? It would be WICKED if we could keep this stuff in RAW...even once it goes over to video from stills. KEEP IN MIND: we'll have RAW image control when it is in it's STILL form. LIGHTROOM is an app that I LOVE for working on RAW images...version 3 can be used EASILY and EFFECTIVELY when making changes to 100s of images at a time. Remember, we'll only ever have to work 400images with the same image properties at a time (we can't take more than that for now....). But after that...it would be cool to have it in a format THAT WAS STILL 4:4:4. (I think we have that level of control with RAW stills...don't we Lee? 14bit on the Nikon D7000?) Anywho...if we can't...(do DNG RAW)..fine...ProRes/AVID xxx will be fine I'm sure.

We have much more than this to do....but this will keep us busy for sure. :)

Schematics coming later Lee. :) (Sorry for delay...sick kids at home)
 
HHL said:
You are ABSOLUTELY RIGHT....on both counts. I had actually found the same two-way mirror. Also figured that you'd lose A LOT of light...(too much).
Click to expand...


I haven't really given it too much thought, but the cheap two way mirror might not be so bad now I think about it . . .

The claimed ratio is 1:10 (transmitted / reflected).

10% is pretty close to 3 stops difference (12.5%) - which doesn't sound unreasonable as the difference between a bright sky and the subject . . . (guess work ??)


HHL said:
this would be the cheapest 5K RAW cinema option
Click to expand...

P.S. my post (above) is aimed at the HDR idea rather than the 5K idea, for the 5K idea you would, of course, want any beam splitter you use to divide the light equally.
 
I got curious about the possibility of using the Canon 7d or some other camera: (Instead of the Nikon D7000.)

Canon 7d:
7.97 fps
(22 shots)

hmm...1/4 as long on the "burst" mode. Almost fast enough to cut out one camera. (This will be fun over the years....we'll get down to 2 cameras soon enough. One day it will be ONE CAMERA bursting for 200 frames.) Uhh....to save time....I'm aware that there are already cameras that burst faster. (Mostly consumer cams) I'd LOVE to do that except we need synching....if there is one that we can do that with...cool! We also want RAW and interchangable lenses...etc...etc. Anyway. I know that the Nikon D3s is a candidate, but at 5K a pop...:( Not really an option.

OH YA....one other thing. With the Canon 7d...maybe firing at 6fps (instead of it's max)....it's buffer would last longer? How much longer? If someone has a Canon 7d and has the ability to set it to 6fps...and wouldn't mind checking to see HOW LONG it will fire at 6fps.
 
No.

There are commercially available lenses that do just this (Loreo etc).

We are trying to do something different.
Click to expand...

OH...ok...cool.

So...it's funny to think that just a couple of days ago, I was saying how it wouldn't be long before ONE camera would be able to shoot the rates I am hoping to achieve on 4 cameras. GH2...40fps. COOL! We'll see how hard it would be to get it to shoot at 24fps..and how long it can do so. YES...I am aware that it won't likely be RAW....JPEG...and at a reduced size/quality. Still....this will be fun to work with. I look forward to it. :)
 
@HHL

Here you go . . . . buy one of these to play around with, great fun and cheap.


screenshot20100923at010.png



http://cgi.ebay.co.uk/Loreo-3D-stereo-adapter-Lens-cap-/150492924641
 
What about videomode?

What about videomode?

I haven´t read the hole thread, so excuse me if this is coverd... :)

I´m not sure if this will work...

What if you put 3 cameras in vertical posission in a beam-splitter/mirror-rig and shoot in videomode? You´ll have 3 1080 (wide) x 1920 (high) videoframes. if you stich these 3 together you´ll get a total image of 3240(w) x 1920(h) Removing overlapping pixels you´ll be able to end up with a nice 1728x3072 px 3K image.

I know, it will not be RAW but if you scale it to 2K or 1080p you´ll probably get rid of a hole lot of video-noise. :happy:
 
For 3D using simple mirrors is much cheaper and avoids the associated light loss of a beam splitter.
Click to expand...
I just looked at this as I'm planning a dual-GH2 based 3D rig.

If you trace out the widest hFOV you want to use (spreading out from the edges of the lens glass as a reasonable approximation), you'll find that this mirror arrangement only works for ultra-long shots. As soon as you go even a bit wide, you'll need much larger mirrors to cover the FOV, and they quickly get so large that you would get the views much further apart than just using side-by-side bodies.

Btw if you can stay long, then two bodies facing each other, aimed at single 45degree mirrors should be better.

I'll be mainly using a 14mm (28mm equiv.) prime. The mirror sizes you'd need for that are ridiculous. An easy way to verify this is to just mess with a mirror in front of a lens.

So for flexible 3D you're either stuck with side-by-side bodies, or a beamsplitter.
 
g.l said:
Btw if you can stay long, then two bodies facing each other, aimed at single 45degree mirrors should be better.

I'll be mainly using a 14mm (28mm equiv.) prime. The mirror sizes you'd need for that are ridiculous.
Click to expand...

This is what that looks like - the resulting interaxial would be at least 24cm. The bodies are only ~12cm ;).

gh2mirrorrig.gif


EDIT: the extra consideration with this mirror setup is that the reflected view has to clear the body. In fairness, the left body could be brought a little closer as its handle isn't so close to the fov (but not enough to make it useful). Also this is a best-case scenario, as the Panny 14mm prime glass I'm tracing has a really small diameter - for most lenses with larger diams it would get crazy.
 
Last edited:
You must log in or register to reply here.
Back
Top