Welcome to our community

Be a part of something great, join today!

Log in
  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

Interesting information...

Mark Pugh said:
yep - I'd thought about shift- tilts... not very sensible, I think! hehe.
Would having image planes not on an angle matter for anything except really close-focus shots, really?
Click to expand...

If you are close to infinity, then no, but your stereoscopic effect would be zero. When you see those 3D scenes where the subject jumps out of the screen and is right in front of your nose, they are made from close up with large angle of convergence.
 
Pawel Achtel said:
If you are close to infinity, then no, but your stereoscopic effect would be zero. When you see those 3D scenes where the subject jumps out of the screen and is right in front of your nose, they are made from close up with large angle of convergence.
Click to expand...

convergence could be achieved by mirrors in front of the lenses. haha! gotcha.
... camera tilted 90 degrees from subject.
 
Mark Pugh said:
convergence could be achieved by mirrors in front of the lenses. haha! gotcha.
... camera tilted 90 degrees from subject.
Click to expand...

You never give up, Mark, do you :) You're right. You can do almost anything with mirrors.
 
Mark Pugh said:
but... mirrors are a pain in the butt on set, though. damn!
Click to expand...

I think two 2/3" Scarlets would be optimum for most 3D because of increased DOF of small sensor. When shooting 3D you actually want a bit larger DOF to help our eyes lock the two images together.
 
Pawel Achtel said:
When shooting 3D you actually want a bit larger DOF to help our eyes lock the two images together.
Click to expand...

I was hoping you wouldn't spoil the party with that one...
you're right, of course. It's the weirdest thing in 3D for me. I once saw an Imax 3d movie, found it really strange, because the scale of the screen was huge of course, but everything looked miniaturised / macroised due to too - shallow DOF, and maybe lenses slightly too far apart...

BUT... you wouldn't need to use all of the image, Pawel, you could do that parallax correction, then crop to S35, or 2/3 or whatever, so... HAHA!
 
Mark Pugh said:
I was hoping you wouldn't spoil the party with that one...
you're right, of course. It's the weirdest thing in 3D for me. I once saw an Imax 3d movie, found it really strange, because the scale of the screen was huge of course, but everything looked miniaturised / macroised due to too - shallow DOF, and maybe lenses slightly too far apart...

BUT... you wouldn't need to use all of the image, Pawel, you could do that parallax correction, then crop to S35, or 2/3 or whatever, so... HAHA!
Click to expand...


A lot of that comes from using a camera base separation (or inter-ocular distance) that is substantially greater than the natural human inter ocular distance. A wide camera base separation has the effect of being able to record very fine 3d detail but has the effect of making things look like they are a scale model regardless of how large you project it on a screen. We have developed software that allows one to carry out a new type of parallax re-sampling so that it can be viewed/played back with a much smaller base separation (using virtual cameras) but does not compress the fine 3d surface detail, (Nuke/ocular, does not have this technology). Normally conventional re-sampling techniques would flatten the fine 3d detail.

Generally if you shoot with narrow base separation large scenes look very flat and un engaging. However if you shoot or render with a large base separation then greater 3d texture and detail is captured but its painful to view due to a much greater range of parallax. So you can have flat but safe or exciting and rich but painful. The techniques we have developed go long way to solving this issue that you can have rich 3d surface texture but have the whole scene reined in to a range of parallax that is comfortable to view. Basically we have found a way independently control what we call Macro and micro surface parallax independently.
 
jim, i saw an adapter for the 4k projector from sony, it splits the images in 2 2k images that are projected via this adapter. i know, i'm not good explaining this, but yo get the idea? :shifty: like having two lenses connected to the s35 barin and then record two different 2k images from the same sensor? probably for the money the setup you showed is better, dunno.
happy anniversary to all at red. thank you for everything. you rock.
 
Eric/VERAXYZ said:
We have developed software that allows one to carry out a new type of parallax re-sampling so that it can be viewed/played back with a much smaller base separation (using virtual cameras) but does not compress the fine 3d surface detail, (Nuke/ocular, does not have this technology).
Click to expand...

What software is this, and when can I get one? Link please?
 
Joseph Ward said:
What software is this, and when can I get one? Link please?
Click to expand...


We are going for launch possibly March/April. My company is called VERAXYZ (sounds like veraxis) and the software is called Objective Reality™. The first product is a fairly simple package designed to extract 3d texture mapped models from stereo sequences. The other thing that it does is that it creates a new type of 3d model called a Coherently Stereo Textured Model or CSTM. In effect the models use stereo textures, correctly mapped to the geometry instead of single mono texture. This means that they have absolutely lifelike models using only the fraction of the number of polygons normaly needed to represent complex geometry. The polygons act as a sort of invisible substrate, so in that sense you can get away with absolute murder, and it still looks visually flawless due to the visual complexity of the stereo textures. One of the strengths of the software is that it stereo immersive throughout and has one of the best 3d and stereo geometry editors in the business. The CSTMS are approximately 400x more efficient than conventional geometries for rendering, and it has a very expedient workflow to make very clean/perfect geometries, I.e. absolute control and you get the polygons and geometry YOU want. The independent parallax re-sampling is in effect an industrial (accidental) by product of the CSTM process. We will also be releasing a surveying grade version of the product so that you can record very large scenes and scenery using static and key frame processes that support the fitting together of many 3d fragments gathered from multiple cameras all at very high precision. We have built and are testing a new generation of medium format high precision (calibrated camera systems) to aid in high precision work. When the FF35s come out we will build stereo metric versions of those for sale, a kind of affordable turn key system. The Objective Reality (Cinematic Lite) is aimed to price for about $7000.00. At the movement we do not have a lot of web presence until we are ready to sell product, as we have a lot of IP in the works and we do not give anything away to potential competitors. The general technology has many other applications besides, particularly for stereo gamming and advanced VR apps. If you are genuinely interested you can e-mail me at ericlange@veraxyz.com for example if you work with stereo footage we could potentialy set up some beta testing. We are based in the U.S. and U.K.

Hope that helps,

Cheers

Eric.
 
Could RED technology be developed into an "eyeball-cam" to benefit the blind (and of course military applications)?

Just imagining...
 
I Guess the thread really started with the 617 and human photo receptors?

What was the motivation or most compelling reason for the idea of the 617?

Jim (if you are listening), what would you most like to do with the 617 (apart from sell it) I.e. what kinds of applications or shoots would you like to do yourself, or how would you like to see people using it. It seems like a very personal project.
 
Jannard said:
There are more pixels in the RED 617 Mysterium Monstro sensor (261,352,000) than there are rods and cones in both your eyes (126,000,000 x 2 eyes = 252,000,000).

Just in case you were wondering... :-)

Jim
Click to expand...

I wonder how Nyquist theorem applies :) by combining the rather regular grid of the Bayer pattern vs. the rather irregular scattered rids and cones? But its amazing numbers anyway...

I once read that I should have a little higher resolution in my eyes, as I am weak and red/green, so have more receptors for B/W which are smaller and result in theoretically higher resolutions. When I was younger I was measured like 150% sharp seeing (100% = average healthy human eye). I guess its less now, but still no glasses :)

Jim, as you are a "glasses" specialist:
What do you think about those correction glasses available for color weak/blind people?
I do have sun glasses that enhance contrast and I have to admit - normal sun glasses are useless for me since I have those...

I guess, given the right optics, the 617 will allow for something what comes close to the term "eagle eye".

I suspect that google will love to make us photos from our homes then...

Cheers,
Axel
 
Steve Freebairn said:
My brain doesn't have to debayer :) Well, I guess it might, but it does it in real time.
Click to expand...

It does - you've got rods and cones, rods sense black and white luminance values (well, not much colour anyway) and the cones pick up bright light and colour information. It's why things look all black and white in the dark, only your rods can see them so you've got no colour. Because your brain's so wonderful, it puts in as much colour as it can, but it's only very pale.
 
You must log in or register to reply here.
Back
Top