Revolutionary feature in upcoming camera, what about for Red?

[Wayne Morellini]

Here is the Lytro light field camera links. It is a real revolutionary camera technology that makes present styles of digital cinema cameras obsolete in future. It allows you focus and depth of field in post production instead of on the spot. It also films it in 3d across a range of angles. During playback you can adjust focus spots and regions and direction of view (they don't seem to have figured out that you can warp the direction of spots, regions and planes independent of the whole picture but that is not such a practical feature).

http://www.engadget.com/2011/06/22/l...-lets-you-cho/
http://www.allthingsd.com/20110623/i...lebrate-video/
http://allthingsd.com/20110621/meet-...mera-industry/
http://www.engadget.com/2005/11/04/s...-for-in-focus/

http://www.lytro.com/
http://www.lytro.com/picture_gallery
http://blog.lytro.com/press-release-...field-cameras/

http://www.engadget.com/2010/09/23/a...us-an-image-a/

[Wayne Morellini]

Analysis
Some people may remember the low priced quality motion camera, and new sensor and thin flat lens technology that I talked about previously . It also was based around this sort of filming. Not the same, but different and cheaper, and only a few expected design issues to overcome in quality terms. I can elaborate on the relevant potential of this technology for the film industry.

Since the 90's I have been elaborating on technologies that can reduce production crew to one on a small production (realistically only if you have AI control or are very talented). One of those techniques would be to do a lot of things in post.

With this technology, a professional can point, then pull right through the depth of field and range of focus to check everything is nice, then set an auto recorded suggested focus to save time in post. Check after the shot if you wish. In post you now refine your suggested focus setting, or pick something different.

The real advantage is when you use this with a high latitude sensor (a real single sample variety) with 16+stops and equally high signal to noise ratio. Now you can play with your lighting range in post, with less amount of light needed. Of course this also helps with color balancing.

The problem is that high latitude low niose sensor technology on the cheap side is thin on the ground. One reason I abandoned my own camera ideas was because the high latitude planet82 sensor technology I was going to use as an alternative for, did not pan out. During the digital cinema camera projects at dvinfo I also abandoned alternatives due to competing advancements in technology that might have to be re-examined. I came up with priority techniques for very high latitude and low light filming, recording high definition on standard defintion cameras, 3d, post manufacture bayer or 3 chip filtering, up resolution oversampling, and a swag of compression techniques.

With sufficiently higher resolution you can replace optical zoom. It is interesting some of the gigantic test sensors presently been shown by camera manufactures.

The use of this technology goes so much further, but that is still propriety. For instance, I proposed new designs for regular shaped adjustable aperture mirror lens that could be as wide as the scene. You now can move the camera in post in and out and around. The technology is practical (sensor is another Issue). However, there are limitations to the calculations where the further you push the more potential to run out of data or for visually undesirable results.

The picture dynamics looks slightly familiar, it would be interesting if anybody could work out what sensor technology it is using from the dynamics. I can't remember which, but it reminds me a bit of the pictures from the Sony sr?-17.

How might a camera look in future. I have done many new camera and control format designs over the years, and the one I wanted to use for this camera, was a flat disc up to several centimeters in size millimeters to less than a centimeter thick, all black, held between two fingers or propriety alternative, something even Apple would wet their pants over. It has a few other of those wetting features. Or square it off like a coaster. The sensor and lens are several centimeters and was 3d. This is a consumer style enclosure but with limited pro video use.

Some people may wonder why I don't get overly excited and jump up and down here about cameras like others, this is why, to me it is mostly not particularly revolutionary except for the benefit to the market, but it is more of an evolution on existing technology to me. If it was up to me I would release mid level versions tomorrow for hundreds of dollars and give the market a real big boost. Anyway, lets keep the revolution coming, for the foreseeable future Red is the best game in town.

This light field type of filming technology is a future of cameras until aliens or somebody comes up with ways for freespace lens technology or 3d /holographic volume filming (which I have theorized on).

Some alternatives With the announcement of this research years ago, I talked about one of my proposals to covert normal pictures into 3d, by examining the spread of de-focus blur. It is less precise because of the blur overlap. The other was mapping the 3d image from 2d footage, which I had techniques for and also Intel sub-sequentially was showing off a technique.

[Evin Grant]

This technology may indeed have a place in the future of professional filming but a few very important questions will have to be answered first...

1. Will the resolution (both spatial and temporal) be capable of cinematic results without adverse artifacts like aliasing, skew and moire?
2. Will the cinematographer still have the ability to use the lenses that he/she desires to achieve a specific look and will those lenses retain their character?
3. Will the Bokeh or out of focus area retain the pleasant look of the original lens chosen?
4. Will there be an ability to set a primary focus point on set or will the editor have to focus each and every shot before he/she knows what's on the slate?
5. Will the cameramen and director really want to give up the very important tool of focus selection to the editor? And if producers force the issue will the role of the DP be even further dimished?

[Wayne Morellini]

Too late, and hope to go to bed ;) . But I think ill give it a go.

I like your choice of blue in your Sig.

1) As far as aliasing, skew and moire, I did say there were performance issues that I wanted to check on my own technique. I do imagine this is at least as good or better. I know what you mean, the microlens array could introduce problems. I was wondering about the color softness I could see, I mean I wonder if the lens system has introduced some washout that has been subtracted leading to satuation, though pleasant satuation. A good design could even get around some of these issues in the sensor.

2 & 3) I wonder about this. During the diy digital cinema camera projects we found that we got problems (see 1) with microlens when using fast apertures, and light bounces off even a flat sensor surface in very fast enough apertures. However, the lens now serves to gather light and the look is created electronically. I don't think the color is going to stay the same through the array without post processing to compensate.

4) With the nature of the technique, everything will be on the slate and recorded, all focus points canbe reset in post. As I illustrated, you would quickly check dof and focus ranges electronically that everything looks fine, and use what you think works, and refine it in post. What is recorded is the focus settings you pulled, and a combination of the image from different angles across the angle of the lens in focus. They use this image to remake focus. See the science link. The closer the width of the lens area to human vision used the better the image for 3d and use for 3d should be.

5) Yes to an extent, but it saves a lot of time if somebody pulls reasonable or good settings during shooting. You are going to have a talented editor to trust them to do that alone. I imagine the director will be more involved or somebody from the camera side. Eventually the camera skill required will be less, and the pay, and post positions will be more lucrative, dp might remain unchanged. I am of course including post latitude lighting and colorisation adjustment ability with this. It is something for refinement, or for low to very low end indie productions, where the editor might be doing multiple roles and wanting to cut costs. That is more the angle of interest for me.

I think we might be able to get to that 80% for those low end story teller productions, now if only they could act and write ; )

[Steve Marshall]

This has been discussed quite a lot here and here FYI

[Wayne Morellini]

Evin, here is a post that sums it up nicely.

http://www.reduser.net/forum/showthr...l=1#post781128

Somethings I forgot to mention: I identified this procedure as close enough for a perfect nausea/problem free 3d overlap for processing, focusing and convergence and the rest, which I what I was aiming for. You can get increase resolution by designing the lens for pixel overlapping to get cinema resolution. Purists might not like that but it can offer very acceptable results when done right. My own scheme offers greater pure resolution separately. With super resolution sensors it does not matter so much. You can use larger sensor formats easily, even medium or large format which allow very high resolutions and makes lens design easier. This maybe the separating issue for the digital cinema camera industry, 6 cm and above.

Noticed the haziness, also the focus on the Oscar in their picture gallery almost looks stepped. Forgot to mention the tilt lens possibilities.

Thanks Steve.

Should have been talked about 6 years ago. I had not mentioned it, maybe to Jim, because it competed with my scheme and the less competition the better.

I think I will continue on here as it is about using it forced in the right forum, covers more indepth look into more than focus, and my own scheme.

I forgot some things will have to re-edit hopefully.