We are having an ongoing debate on other forums such as LinkedIN on the current state of 3Dimensional Movie production and presentation.

Here is a document that we believe is the reason that newer 3D films if they feature real world scenario SHOULD be captured with a stereoscopic 3D camera system and not shot in 2D and converted later:

http://www.slideshare.net/clydd/2dto3dconvertedmovies

or can be downloaded from the RealVisionwebsite here: Download

Since that document, one of the very good 3D conversion studios, did take the time to answer. Their reply is here below:

http://www.in-three.com/response2d3dconversionjsk.pdf

In return to their answer, please find our quoted reply below:

Thank you for the rather nice and well written response and article, I will only give my input in points.

1) The Biggest and happiest point that has been made in the Answer document by In-Three is the admission that conversions are a Tool, or toolset for presentation of 3 Dimensional movies. They had previously said that 2D to 3D conversions “can not only be as good as dual-camera capture – but far superior.” Although they did say at that time as well, that its also a good tool.

This was my Main Gripe with everything (about 2D to 3D being superior! and some other conversion houses even said that 3D movies should be shot 2d and converted _ I initially made a mistake and thought that In-Three was the ones saying all movies should be shot 2d. they are not) -

Any how – 2D to 3D has been acknowledged as a toolset only and hopefully will benefit everyone interested.

2) The answer document has very craftily taken Everything that can be (are?) good with stereoscopic 3D camera capture ( true specularity..reflections etc, and managed to make them sound like shortcomings! – how about french flags for lens flares in 2D? or polarizing filters on 2D cameras – should they be any different for 3D camera capture? I can’t allow myself to clarfy these further so I will pass.

3) Most Important – 2D to 3D conversions are creating a “paper world” devoid of the true representation of what is captured stereoscopically – At best I will call conversions a 2.5D movie. not a 3D(imensional) movie – as we take the context of stereoscopic 3D movies from the black and white days of 3D Photography/movies thru to today’s cinematography to mean.

4) Your assumption that I did or did not know whether the ground (pebbled) path was CG or real was ALREADY admitted by me in my document. I’m still guessing that it was a real background and converted by a gradient depth map. thats why the flatness.

If i’m wrong – I stand corrected. with the other obvious choice already pointed by me that it may be an original CG bumpmap shortcut. However…Was I also wrong on the shortcuts taken in the petshop that the rotating postcard stands and the cigarette racks are just rectangular geometry? (notice – it had to dwell long enough for me to register these – i did not pause frames in the cinema).

5) This brings me to the other important point – I’m not suggesting that people go to the cinema to shoot holes in converted 2.5D films! I already mentioned that. My point is that the ” trivialities” that I pointed out.. are what makes perfection. (and this has to be factored in the movie making). Please look beyond the “good conversion” argument into what actually makes a 3 Dimensional film TODAY.. the feeling of Immersion and Psychologial fulfilment of the senses. (we already agree that gimmicks are passe).

When presenting such a ..lets call it.. “Powerful Optical illusion” to the Brain, i.e stereoscopic 3D, there’s much more going on than just eyes watching a nice ‘Z-Depth’ of a scene. The psychological impact of well crafted 3D is what makes people “flinch in reflex” when something comes up close.. and also is what will make the brain feel that something is weird and amiss when peripheral vision registers “Flat PAPER WORLD” backgrounds around the central acting talent in the movie.

For a Director who’s got taste.. I’d assume on the lines of the guys behing L.O.T.R etc.. it would make sense that they are made aware of these, when converting. (of course I would be the first in line to see a conversion of these classics and blockbusters – this I already mentioned, as it’s the only way to enjoy what I would call a “dimensional artists’ handwork”.

To re-emphasize.. I hold objection to NEW movies being pushed to be shot completely in 2D and converted later to 3D.

…contd.

* To put it politely – the “excuse” that gear is un-wieldy etc for live shoots in “proper” 3D, should really be put to rest.

Short of giving a list again, I will quickly say – SIP 2100for all camera /parallax preview and settings in-the-field, Si2K stereo rig (theres even a guy wearing it on his head for crying out loud) how much more portable can you get?  http://www.siliconimaging.com/DigitalCinema/News/PR_04_20_09_3D.html

Other cameras are there too, and for post production and “depth budget” theres a load of solutions out there. If needed I will do a more “mature” document than the lesser one I have been accused of – for a price! to fill in people of what’s available for Stereoscopic Previs, Capture, Production and Postproduction.
…contd

* Which brings me to the last point for now -

Here’s where I’m Stumped – Conversion houses tell you to shoot the Whole movie in 2D, then do the CG compositing pass (if needed).. Then they take this whole 2D film and CREATE a NEW “pseudo 3D world” ONE object at a time (!)…then they shift the parallax according to so-called “eye strain free viewing comfort” – arguably — the biggest USP of conversions —– then they match the real scene lighting with CG lighting… then they render out either BOTH the new left – right eye view or at minimum the right eye view to make the final 2.5D movie?

If this isnt counter-intuitive, in 2.5D and more time consuming..I dont know what is!

Staying in the stereoscopic 3D domain from Shooting to compositing to depth budgeting (yes there are 3D aware matchmoving software whose masks are easier to tweak for right eye once the left eye is done) and yes there is steroscopic color grading one click – and yes there is a more efficient way of applying selective background Blurring WHILE keeping the organic

detailing of the background objects so that “center of attention” as a Story telling techinque can be used for Direction as DOF is used in 2D.

To conclude, I am happy that at least the Document is the seed for more “mature” thinking for Directors to think what’s best for their next 3D epic.

——–