Cameron 3D Rig

Good evening, it's mail from Japan.

Director James Cameron came to Japan this week. I tried to want to come to the Sony factory in Atsugi to see the camera for 3D which requested development before. As for Mr. Cameron who replied to the interview, 3D of its movie was a thing that it probably changes in use. the purpose of a visit to Japan -- own 3D work it served also as PR of Ghosts of the Abyss -- although that is right – another purpose -- animation of Japan -- natural -- Full CG -- it is said that it is probably in "GAMMU"(It is the sequel of the animation which may be put on the market by the name before called "battle angel" also in the United States.)of 3D work Still, because of copyright or the commercial right, although information did not enter, it seemed to be about interest very in Mr. Cameron. And as for him, technology of Japan is very important for a future movie. It said. It is not only me to have been encouraged by his words now when Japanese economy is ailing. Please watch warmly without forsaking you and Japan! The Japanese movie is also doing its best. About forty HD cinema was made last year. However, I think that there are few movies with the true merit of HD. Effective practical use of a digital signal is not made by E-Cinema for suppressing a budget. Moreover, the Japanese-style division-of-work system also became a neck. A cameraman doing that camera operation for an angle. Lighting has an another part called a lighting engineer, and a video engineer does signal operation in HD.

What a few is separated from the original DoP for is the actual condition. Please allow a long letter. As soon as information enters about "GAMMU", it is made to report, and it obtains.

Shuji Kurata
Cinematographer
http://www.geocities.co.jp/Hollywood-Miyuki/1053/

Stu Maschwitz writes :

>does anyone know if the Cameron 3D rig affects the focal length/focal >plane relationship?

I don't know whether they've stuck with this approach, but the original intention was to interlock the focus and convergence controls so the two fields of view always converged at the focal distance.

As I understand it, with true 3D, within a given scene the lenses remain parallel or at least at a fixed angle of convergence, regardless of the range of distance at which the in-focus action takes place.

By contrast, Cameron's system, by always converging at the focal distance, keeps the in-focus action at a constant perceived distance from the camera. Once the audience gets comfortable converging their eyes at that distance, their attention can be manipulated in somewhat the same way as we normally do it with focus only.

I haven't seen the results, so I don't know how effective it is.

Dan Drasin
Producer/DP
Marin County, CA

Dan Drasin wrote :

> I haven't seen the results, so I don't know how effective it is.

I've shot a lot with the Ikegami LK33 (The actual first Zoom 3D lens) and it uses this approach. It's works very well, but is somewhat limiting in terms of being able to manipulate where the "window" appears to be.

If you have enough stop, and depth of field, then you can move the focus back and forth to change where the convergence is-- yet still have what you need in focus, in focus. So if you want the face to come out of the window slightly, you focus on the eyes, or the ears, if you want the whole face behind the window, you'd focus on the tip of the nose.

We always wanted a version that had adjustable convergence too, but I guess a true Zoom 3D lens set that holds focus and convergence is difficult enough to build. Not been accomplished very many times. Is Cameron's lens for the Imax film? I know Steve Shklair's company has a true zoom for HD work, and then there's the Ike lens for their 1/2" chip SD cameras.

Steven Bradford
Video 3D shooter
http://www.seanet.com/~bradford/
Seattle

Vince Pace of Pace Technologies would be the best person to ask about the 3D rig as his company worked on the 3D rig that was used on Cameron's project. I got a chance to see the rig before they shipped out and it was one hell of a monster.

Vince worked as DP on the job and Mark Goodwin was the gaffer.

Tony Salgado

Steven Bradford writes :

>We always wanted a version that had adjustable convergence too, but I >guess a true zoom 3D lens set that holds focus and convergence is >difficult enough to build.

Cameron's system is actually a two-lens system feeding two HD CCD prism blocks, essentially being two cameras. This composite "camera head" is small enough to be shoulder-held, and feeds the F900's through a cable or two.

When the two HD images are filmed out and projected together, the resulting image is said to be adequate for IMAX. (Granted, Cameron's work with the Titanic wreck is essentially documentary.)

Dan Drasin
Producer/DP
Marin County, CA

Dan Drasin wrote :

>Cameron's system is actually a two-lens system feeding two HD CCD >prism blocks, essentially being two cameras.

Yes, that's what we've been doing with the Ikegami camera for 12 years now, the only difference being it has SD chip blocks instead of HD, and more variety in types of recorders that can be used, either on camera or off. Your description sounds very much like the setup Cobalt Entertainment has, http://www.cobaltentertainment.com and so I wonder if that's what was used on Cameron's 3D production. (Warning, that link goes to the usual hype to the skies, "we're the only people who've ever done this or are doing this in the history of entertainment" writing that seems to be an affliction of anyone that gets involved with 3D. But they have done some good stuff, they're not just talk.)

Makes for a heavy camera head, more in the BL range, or some of the early handheld color broadcast ENG cameras. But it is nice to have it all on your shoulder, and able to change shots instantly on the fly. You can even do multicamera sporting event, which we've done, and it sounds like Cobalt has experimented with.

3D seems to benefit greatly from increased resolution, which seems like an obvious statement, but I believe it does more much so than single lens photography does. The whole stereopsis effect improves greatly with resolution. To me, objects become more "round" and less cartoon like, or like 2D cutouts, as resolution increases. SD Video seems to have just enough resolution to be acceptable, By acceptable, I mean, so-so, it still appears very much like cutouts.

Steven Bradford
Seattle

The rig we used was not the setup that Cobalt offers. We checked out the offerings of both Cobalt and Paradise and they fell short of our needs at the time. Added to the challenge is the fact that most 3D individuals are set in their ways, reluctant to take advantage of newer technology especially as it relates to rewriting some of the rules in this medium.

The Cameron/Pace Reality Camera System was designed from the ground up to be used handheld, steadycam, jib arm, dolly, and scuba. This design was done by both Jim Cameron and myself without ANY outside "3D Expert" contribution. In fact, the 3d Experts were certain we would fail in our efforts. We proved that the technology worked taking the system out on the North Atlantic for two months handling all of these areas including 2.5 miles down.

Any person who states that this was done for the last twelve years should show me the other 3D large format shows that used this approach. We blended technology and creativity to make an exciting large format product. The other individuals that continue to use slide rulers to predict the on screen, off screen presence still wish they can present a product like "Ghosts".

Regards,

Vincent Pace
Co developer of the Reality Camera System
Ghosts of the Abyss, Director of Photography

Vincent Pace wrote :

> Any person who states that this was done for the last twelve years

Please sir, I ask you, did you read what I wrote before you posted in anger?

"Yes, that's what we've been doing with the Ikegami camera for 12 years now, the only difference being it has SD chip blocks instead of HD, "

Thanks for the incredibly uninformative 3 paragraph post.

I think I can say truthfully that I learned nothing from it. I know nothing more about what you've done by the last word as the first word.

However, this is par for the course for the 3D biz.

It's good that you got a successful result. But if how you did it is a state secret, why are you bothering to even post here?

Steven Bradford

Vincent,

I have a few questions about the RCS that you seem uniquely disposed to answer. I'd hope others on the list would be interested too. Any help you can offer will be much appreciated!

Can the interocular distance be adjusted? If so, what is the range of adjustment? Is it something that would ever be adjusted on-the-fly? I assume it doesn't (necessarily) change as the convergence changes, but please correct me if I'm mistaken.

As has been discussed, the convergence adjusts to match the focal distance. I would assume that there may be a way to create a static convergence offset on top of this, such that you could place the in-focus object at a depth other than screen depth, and still allow that offset depth to remain constant as the focus follows the object through a compound move. Is this the case? Was this ever done on GHOSTS?

As the convergence adjusts, does one lens move or do both? Another way of asking the same question : Is the triangle formed by the nodal points and the convergence point always isosceles?

In the raw footage I've seen, the right eye record is mirrored left to right. I'm told this is because it is shooting through a beam splitter. How exactly does this work? The photos I've seen of the rig make this look unlikely. I've also noticed a slight exposure difference between the left and right records. Was this an issue on GHOSTS?

Thanks again for any help you can provide!

Stu Maschwitz, once drew 3D comic books,

San Fran, CA

Vincent

I would be keen to talk about your system - I have been talking with a director friend of mine for the past couple of years to try and develop a sort of 3D effect and maybe it is something that you have done already or could be interesting in developing.

Cheers

Matthew Woolf
DOP
London

Dan,

F900's were not used on the Pace/Cameron 3D rig.

It was T950's camera which allow removing the block from the head and then via a cable feed the head then another cable feeds a HDW-250 deck.

There was alot more going on with this setup that I can't disclose as I feel only Vince should discuss the specifics regarding his 3D HD invention.

Tony Salgado

Gee....I am feeling a need to rise to Cobalt's defence here, as we may have just been included in a class of "3D individuals who are set in their ways and reluctant to take advantage of newer technology". Although I am sure Vince did not intend a direct slam on us, it may be interpreted that we are of a class of Luddites that will only give up their slide rules when they are pried from our cold dead hands. Heck...I will even confess that I don't know how to use a slide rule, although I take no pride in saying this.

While I have posted on CML a few times, I generally lurk. I have agreed to so many NDA's lately that the safest thing to do legally is to just not post at all. Send personal notes directly instead.

Although I am sure there was no intent to malign, I would like to be very clear about a few things. Cobalt is at the forefront of taking advantage of the new technologies, and "set in our ways" is about the very last descriptive that could ever be applied to either myself or our company.

First, although we lean toward the newer technologies, we have been and are only interested in applying the best, and most appropriate tools to every project we tackle. We owe this to our clients. In direct response to the thread of this conversation, on one recent project we fought very hard with our studio client to get them accept the use of digital cameras, and then the Cameron/Pace camera system. We invested weeks of unpaid time into making this argument. We eventually won, and the show was shot with what if being referred to here as the "Cameron 3D rig".

In terms of embracing new technology, there is very little we have not tried to make work in the field. We have gone as far as to shoot the NFL Pro Bowl in 3D with dual Vipers gen locked together, recording into almost 4 Terabytes of disk space that we brought onto the field. We invested a great deal of time into engineering this show. We have shot other NFL and sporting events with Sony 900 cameras. We've already completed a test shoot with the brand new Sony F950 and the SR Recorders, shooting both 4:4:4 and 4:2:2 into these decks. And using fiber to connect to the decks, which for some of our work is a fantastic thing. We've even recorded the Viper (with the new software upgrades) into the SR decks.

We were invited and have participated in tests with the Olympus 4K camera, and always have our eye on the latest CMOS developments. So please do not include us in a class of any experts that are set in their ways. We produce with proven systems...we test and eventually use all the new systems because they expand our creative ability improve production value, and allow us to continue to expand the language of 3D (and 2D) filmmaking.

We also believe though, that the generation of 3D experts who preceded us do have something very valuable to offer, and that is experience. Even though the tools are now different, they are still two cameras mounted to a computer controlled 3D platform, whether it is a parallel configuration or a beamsplitter configuration. Disregarding the expertise of all of this past experience is like throwing the baby out with the bathwater.

Steve Bradford, I cringe when you make blanket statements like, "However, this is par for the course for the 3D biz". Statements like this just perpetuate the myth that the 3D business is still the bailiwick of laboratory geeks and fit only for carnival side shows.

This is a new "3D biz", and there are new players. Jim Cameron and Vince Pace just made a feature release which was only possible because of the new
tools that they engineered. Additionally, Vince and this equipment have been deeply involved in the about to release Spy Kids 3D film. Disney just announced a 3D film to be directed by Hype Williams. Working with Cobalt, Ideal Entertainment is deeply involved in using the new digital 3D tools and is actively promoting 2 or 3 projects. Watershed Films has projects at the gate. NFL Films is now involved. Christie is making a single DLP projector capable of displaying stereo images. There are so many others.....

(Steve took a brake in the post...then continued)

Why now....? It is finally possible to make high quality stereoscopic film (and digital) releases on something that resembles a realistic business case. The enabling technologies behind all of this new work are the new digital production tools, which allow 3D rigs to be used that are lightweight, silent, and will shoot for more than 3 minutes per load. For the first time ever, we can shoot on the fly with these rigs, and can shoot on a production schedule close to if not equal to a 2D shoot. We can use standard riggings, and move the camera on cranes, jib arms, and Steadicams.
Because we are using dual HD Cameras, we can also release a 2D version of the film into the 35mm theatrical market from the same material. Imagine the repercussions of this. So, 2D and 3D, from IMAX to 35mm to DVD, we can now shoot once and release to multiple platforms. Not only does this start to resemble a real business case, but is actually better than most business cases for theatrical films as the ancillary 3D releases add a great deal to the potential bottom line with only a minor incremental cost increase.

Only a man of the stature of Jim Cameron is able to get Sony to make a modified camera as a one-off, which is what has enabled their 3D rig, and enabled them to rewrite many of the rules of shooting stereo. I applaud and support their efforts, and the efforts of their collaborators such as Fujinon and Panavision. Thankfully, now that Jim and Vince have broken this new ground, the "rest of us" will be able to get our hands on cameras and lenses that are small enough to allow the same type of parallel shooting for most (but not all) shots. The Cameron/Pace rig is a great advance. There are other great advances right around the corner, although the NDA's mentioned earlier prevent me from saying more.

Our business is producing and enabling 3D content through the best and most appropriate tools available for the job. We work with a number of clients in a variety of ways to help them execute their projects. If any of you were to come to us tomorrow to produce or co-produce a 3D film, odds are the first call I would make would be to see if the Cameron/Pace camera were available. If not, there are others as the shoot must go on.

This honestly was not meant as a commercial. I have learned a great deal from the CML HDTV mailing list over the past two years, and thank you all for your willingness to share your experiences. Thank you as well for your patience in reading this tome of an e-mail. With all of the incredible risks we have taken during the past two years to push a reputable 3D business forward, I was a bit rankled to mistakenly be included into a legion of slide rule users afraid of new technology.

I will now, for the most part, return to my lurking.

Steve Schklair
Cobalt Entertainment

Hello!,

Vincent Pace wrote:

>Any person who states that this was done for the last twelve years should >show me the other 3D large format shows that used this approach.

Two friends and me build a 3D camera in 1998. Although SD it can be upgraded to HD by replacing the CCD minicam bodies used. I know a manufacturer capable and ready for building HD mini cams to the same dimentions as the SD ones used now. However, the price of those makes that we have to wait for a serious project in HD in order to allow for the investment.

Although SD (for now) it is also small and can be used on virtually any support that can hold a 35mm film camera. It was so designed that it also could use many of the same accessories used for 35mm cameras (so we would not have to re-design everything). A description and pictures of the camera can be found on…

http://www.kommer.com/v3d-cam.htm

It was used on several productions, most for 3D video projection shows on 8 to 10 meters wide screens, some actually running in Museums and Local Based Entertainment venues.

It was also used for shooting elements (actors/singers/musicians) to integrate in the 1570 production "Haunted Castle" :

http://www.kommer.com/st-pics5.htm

(and I presented about that on the BKSTS VFX2001 conference.)

>We blended technology and creativity.....

Isn't that what it is all about for all us cinematographers?

>The other individuals that continue to use slide rulers to predict the on >screen, off screen presence still wish they can present a product like >"Ghosts".

I don't understand very well how down Critisizing the way others work will make your work seem better. I doubt if that will really work actually

I have worked on 9 stereoscopic 3D productions till now. I don't use slide rulers. Never have. There may be some rules to be followed for 3D to work well however. Like there are in many crafts. Rules that sometimes can/must be broken as well.

Best regards,

Kommer Kleijn.
VFX Cinematographer
Brussels, Belgium, Europe
http://www.kommer.com

Steve,

Well said however took me a while to take it all in. You are one of the few that remains "interpretation independent" and I applaud that effort. However, I do personally feel (and it is just a personal comment) that the market is reluctant to embrace the advantages of new technology to change, delete, modify, or confirm many of the so called "3D shooting rules".

The design of the Cameron/Pace rig was to develop a tool that would give every cinematographer on this list the opportunity to contribute creatively to 3D production. Steve, you and I both agree that the industry needs better 3D content and some of that content is hindered by lack of suitable technology. The rig is now developed and we are working hard to build more even though they are very costly at this time. I am not here to bang the drum about technology. Like you I want to move forward with discussions that center around the creative possibilities that now exist using this equipment.

For me, that is were the excitement lies.....

Regards,

Vince Pace

Kommer,

>Two friends and me build a 3D camera in 1998. Although SD it can be >upgraded to HD by replacing the CCD minicam bodies used.

Capable and ready is admirable. We have already built six systems and
continue to build more to supply demand.

>We blended technology and creativity...Isn't that what it is all about for >all us cinematographers?

Agreed.

>I don't understand very well how down Critisizing the way others work >will make your work seem better. I doubt if that will really work actually

Wasn't trying to down criticize others. However, being in the 3D arena, you are very well aware how stubborn individuals can be about experimenting for better creative results.Our goals were set very high. Sony, Evertz, Panavision, and Fujinon all participated in the effort to build special components to meet these goals. Motion controllers were designed, lens control interfaces were integrated, and a High Def 3D system was built.

Everyone involved deserves credit for the results. Call that up criticizing if you will.

Rules that sometimes can/must...

>be broken as well.

Well said....

Regards,

Vince Pace

Speaking of James Cameron, has anyone read the article on Cameron on the CineAlta website?

http://www.sonybiz.net/cinealta^/templates/std_page.jsp&OID=105536

Putting aside the 3D aspects that people have been discussing on the list, here's a quote that's sure to inspire debate :

"Asked whether HD was good enough to replace 35mm, Cameron is unequivocal; "I think we've soundly and quietly passed that mark, and it's so far in our rear-view mirror that it's not really an issue of whether it's the equivalent to 35mm. It's an issue of 'Is it the equivalent to 65 original negative, or beyond that?' because that's the kind of data that you're getting." "

George Hupka
Director/DP, Downstream Pictures
Saskatoon, Canada

George Hupka wrote :

>whether it's the equivalent to 35mm. It's an issue of 'Is it the equivalent >to 65 original negative, or beyond that?' because that's the kind of data >that you're getting." "

As much as I am fervent proponent of HD as an acquisition medium.

After seeing "Ghost of the Abyss" and "Corals of the Pacific" within hours of each other I have just one word to say: no way!!!

Daniel Villeneuve, c.s.c.
Directeur-Photo/Director of Photography
Montréal, Canada
http://pages.infinit.net/davil

Vince Pace wrote :

>I want to move forward with discussions that center around the creative >possibilities that now exist using this equipment. For me, that is were the >excitement lies...

And as much as I love new technology, that is where the excitement lies for me as well. I'm on board Vince...we have shows to shoot. You already have a few of the dates...more coming when we lock these down.

Like you, I am evangelist for the stereoscopic medium. Properly done, the images are extremely powerful. Not only visually, but they enable a filmmaker to add an entirely new branch of language to their existing vocabulary. I know the market will quickly embrace this new technology, and these new rules, when the market has had the opportunity to see more great content released into the commercial exhibition venues. The next films that are made also have to be great films. They need to be almost as compelling in 2D as they are in 3D.

Ghosts was a great start, and has earned a spot in the history of cinema. Now lets get keep the momentum going with action, music, sports, musicals.....

BTW…not sure I understand the meaning of "interpretation independent".... but it sounds like a good thing. Thank you.

And thank you all who wrote directly to my e-mail address. Your comments and support were more than encouraging, and no...I promised both Thomson and Sony that it was not a shoot-out. So I will not post the "results" as if it were a race.

Best to all,
Steve Schklair