Cinematography Mailing List - CML
    advanced

Using Kodak Look Manager System

Message from Georges Leurquin

Does any of my estimated peers have experienced the use of the LMS on a set.....

I am currently on day 25 of 70 using the KLMS.

I think its pretty unreasonable to expect the DP to have time to use the system on set. My primary job is 2nd Unit/Camera Operator. Since we are not shooting everyday the whole day with 2 cameras, I have the time in between 2nd Camera setups, to use the software. It still doesn’t stop the fact that I need at least 2-3 hours everyday after shooting, in the hotel room connected to the Calibrated Monitor making fine adjustments, before burning a CD with all the files for the Telecine Operator to view and compare while transferring the dailies.

We are receiving the dailies on DVD, the director receives his dailies from the Avid, transferred onto a Hard Drive.

For the first couple of days where were also trying to view the dailies from the Off line Avid and soon gave up, due to the terrible quality.

The representee from Kodak has been on set twice, to see in person how I am using the software in a working environment. He has also seen and agrees with the several points I have, why and where I think the system can be improved. Last week we (the whole crew) watched a 1 hour rough cut and I would say 60-75% of the scenes I saw looked fairly similar (at least in the direction) to the KDX files that I had sent with the exposed material.

I am in daily contact with the Telecine Operators and they are very thankful to have a 'Reference', a sort of high tech version of a Polaroid.

The KLMS has definately given the dailies a much 'Better' look, but at the moment the files that I am sending every day are only 'References'.

It will be interesting to see how the final grading with look.

Adrian Cranage
Director of Photography
on location in Leipzig
www.cranage-dop.com


Adrian Cranage wrote:

>The KLMS has definately given the dailies a much 'Better' look, but at >the moment the files that I am sending every day are only 'References'

They always will be unless and until scanning is done at the time of dailies, and those scans used for all post production from that point forward (or unless and until you shoot on something other than film). Today that is not generally the case; dailies are done on a telecine (often HD, but still a telecine) and the scans used for a DI are done after the picture is locked. This yields two completely different interpretations of the film, done on two completely different systems using two completely different sets of parameters - not to mention the likely fact that the images in the KLMS came from something other than the film in the first place (most likely digital stills). Tools such as the KLMS are, to me, much more useful in terms of a predicted print look on a "standard" film finish, and with the understanding that it's all you can accomplish with them.

As far as dailies, I still don't understand why the KLMS is any different or more effective than tweaking frames in Photoshop, since all it produces is an image to be used as a reference. The 3CP program does a bit more than this (it produces an image of what video scopes should look like to implement the colours in that image). Neither system currently produces any information that is of any use anywhere but in a dailies telecine room. And although you state that the system has give the dailies a much better look, how do you compare that to what they would have looked like with a qualified daily colourist looking at Polaroids or Photoshop files - or simply with telephone conversations - when you never did that (or at least I assume you never did)? They look "much better" than what?

Mike Most
VFX Supervisor
IATSE Local 600
Los Angeles


> As far as dailies, I still don't understand why the KLMS is any different >or more effective than tweaking frames in Photoshop

Yes, I never understood this.

If you can, in a telecine or DI finish, get virtually any look you want, tweaking stills in Photoshop et al should be plenty to get a reference image that a competent colourist can mimic. After all, you're not trying to do a final grade, just provide a reference.

I see the need for more specialized tools if you are trying to see what images will look like with a film finish.

One of the best of those specialized tools might be timed film dailies, seen on a projector. (These needn't be timed, they can be printed at a chosen light, obviously). No pixels to get in the way. If these looks are determined in testing, pre-shoot, the colourist can either screen some prints, or look at a few short frames (mounted as slides or scanned) as a reference, just so the dailies don't start looking weird.

Jeff "simplicity is good" Kreines


Jeff Kreines wrote:

> Yes, I never understood this.

>If you can, in a telecine or DI finish, get virtually any look you want, >tweaking stills in Photoshop et al should be plenty to get a reference >image that a competent colourist can mimic

I have a demo copy of KLMS on my desk, which was obtained by going to the seminar in NYC. I asked the very question being discussed here. The software can be licensed at various levels and it is expensive.

Part of the value is that built in to the software (as opposed to Photoshop) are LUTS to emulate the various Kodak negative and print stocks, sets of filters commonly used in the MP world, and post processes commonly used. So it is quite easy to build a post production sequence of processes and string them together and get a decent pre viz of what the result could be.

At a certain level the DP can author a look and send the recipe to the post house who can take it and apply to the post process, assuming the post house has also licensed the very pricey software as well.

Kodak is taking a swing at standardizing a digital work flow situation so that one could sit at their computer and design a look through to finished prints and have it all available on one screen.

The way they implement things might actually be worth it if enough people adopt it as a way of working, because it would take the Guess work out a colourist say interpreting your Photoshop still and then trying to apply their take on it to the dailies.

Mark Smith
DP NYC


Usually disclaimers go in the end of message -

1) I'm not receiving any money from,
2) I am an objective observer and ,
3) I am not working for .

In this case the true is only 33.33% - I'm not receiving any money from Gamma and Density yet - cannot find good and free R&D help those days, but I am a non-objective, yet fair observer of all LMS and I am working for Gamma and Density Co. - I am the inventor behind 3CP and a founder of Gamma and Density.

LMS are here. We cannot deny it - we can only discuss details of those LMS - Big K's LMS, 3cP-LMS, etc. - and the "devil is in details", as we all know.

What Adrian Cranage describes is unfortunate. 3cP has none of these problems; not only because we do it in real time, or "our" DP spends on average 15 minutes a day with our system, or we use various scopes, or 3cP is better... No! Simply because 3cP is different.

Last week, 2 more features saw the 3cP advantages: DP Theo van de Sande (ASC)(Cruel Intentions, Beauty shop) and Robero Schaffer (Finding Neverland, Monster's Ball) will be using the system to communicate their vision to Laser Pacific, a Kodak-owned company (sic!) in LA. The next big challenge for Gamma & Density is going to be a few more big, international pictures done outside of the USA. And Theo, who is doing a multi-zillion feature for MGM, is not wasting any extra second of his very valuable DP time with 3cP.

It is because 3cP it is not a software, it is not "a sort of high-tech version of a Polaroid," it is a full-cycle (from-the-set-to-the-screen) Thorough Control System.®

It was created as a concept out of pure necessity a long time ago when I, fresh "from the boat," started filming my first feature in the US, 'Liquid Sky," and got rather quickly frustrated by the level of misunderstanding between myself as DP and the timer of a small non-existent any more "low-budget, affordable" NY lab.

For me, who was used to the then best timers of Mosfilm and Kiev studios, it was inconceivable to see that my timer was not passionate about 5% magenta or 1 point of density, while we would go to blows with Director and Production Designer about shades of wallpaper or hue of fluorescent make-up under black light and 2 points of red were to create a different mood in the mind of our audience, not just a different colour on the screen.

So I had this vision of a different world when where and DP would be , in eyes of some , merely a "supplier of the first exposure", and "Wise and Kind and Understanding Executive Tom" and no less " Kind and Understanding and knowing it all Post-Production Executive Judy" would tell all the world how our picture should be seen and perceived, and then LMS will come to rescue DP, as in good Hollywood film myself and my former 1AC, camera operator and now cinematographer, Sacha Riviere, who is also one of the co-founders of G&D, have been using 3CP in all its incarnations on our own feature and commercial productions and productions of DP friends for a number of projects now.

We "learned to shave on our own beards", so to speak..

3cP does not originate with programmers, chemical engineers, investors or FX specialist - it was thought of, developed and by cinematographers who were "in trenches" every day. 3cP does originate from minds of many and various in style and technique devoted DP .

That is where our difference from all other look management systems comes from.

We at G&D know first-hand what it takes to make a system work for DP on the set and in post. And we know (a big Hollywood secret... t-s-s-s!) how there are still some arrogant producers and executives who will see a look-management system as a "new toy for the boy".

And with our new partner Bob Currier from Synthetic Aperture, author of Colour Finesse known to many of you, we are looking forward to a lot of even better stuff to benefit DP's and colourists alike.

Anyway, the bottom line is not "to look-manage or not to look-manage". We wouldn't want to be like luddits or another IBM chairman who proclaimed that in the whole of the USA the market for personal computers was. 5; the issue, in general, is how to make look-management systems an acceptable tool that would assist DP’s in their struggle for professional survival in this new brave world.

P.S. : And in response to an esteemed, but apparently not too-well-informed CML member Michael Most : 3CP does much more than only "produces any information that is of any use anywhere but in a dailies telecine room" as noticed. It produces extremely accurate results for use during DI colour correction.

Let' s face it - Sony/Columbia Corporation has "favoured" 3cP over all other colour correction systems and trusted Gamma and Density with 20 million worth of Jim Carey face and Tim Robins face, and ABC with Oscar winner Halle Berry face. It didn't happen because we are nice guys - we are! It happened because our system works different than most.

A recent seminar in LA, devoted to the 3cP-integrated DI and VFX workflow on 2 Sony feature films, clearly demonstrated our system's worth.(For those who couldn't attend -- a new seminar is going to be held at Fuji Cine-Club on April 14.)

And as per difference between 3cp and Photoshop - see our Q&A section at : www.gammaanddensity.com

Yuri Neyman,
Director of Photography, Founding Partner
Gamma and Density Co.


mark smith wrote:

>At a certain level the DP can author a look and send the recipe to the >post house who can take it and apply to the post process, assuming >the post house has also licensed the very pricey software as well.

Here's the rub : Nothing in software can emulate a telecine, the setup of that telecine, or a DaVinci or Pogle colour corrector. Nor can it supply any sets of numbers that can make those devices yield a specific result. All it can do is emulate other boxes that Kodak also supplies, but these boxes are, by Kodak's own admission, intended to equip a colourist with a "starting point" for colour correction, after which the colourist does what the colourist does. The bottom line is that the ultimate value of either the Kodak, 3CP, or Photoshop route is that the colourist gets some idea of what you want rather than start with an open book.

I still say that as long as you're shooting film and not scanning at the beginning of the process, all of these things ultimately have limited value. They are basically fancy versions of Polaroids in a more sophisticated wrapper, and, to some degree, high end solutions in search of a high end problem. The loop will only become more closed when either scanning once becomes the norm, or when we're shooting electronically, thus eliminating the intermediate step of telecine and its attendant human intervention.

Mike Most
VFX Supervisor
IATSE Local 600
Los Angeles


>,,,,,might actually be worth it if enough people adopt it as a way of >working, because it would take the Guess work out a colourist say >interpreting your Photoshop still,,,,,

I don't know...

I've had only excellent results when I've supplied prints of digital stills adjusted in Photoshop.

Have people had problems with this sort of workflow?

I agree with Mike Most's appraisal of this sort of software : "...high end solutions in search of a high end problem."

David Perrault, CSC


2 grand a month. For what? To help me to sell more of their film stock.

Dan Bronks
DP
UK


Mark Smith Wrote :

>,,,,,might actually be worth it if enough people adopt it as a way of >working, because it would take the Guess work out a colourist say >interpreting your Photoshop still,,,,,

Guys guys, I am in no way endorsing KLMS method. I'm very sceptical altogether because the DP and the post facility need to have the same software and the licensing is not cheap. Kodak is banking on the idea that production will buy the DP's license for the term of the feature and that post houses will buy the yearly license for 20 large.

I asked pointedly during the seminar so why don't we just use Photoshop? And the response was 2 fold: 1) we built our software using a language and methods DP's are familiar with and 2) the potential for zapping a formula over to post, that they can readily apply. Far as I can see, number 2 it ain't there yet, or even close to being there and Number one is nice but pretty flimsy given the pricing of the KLMS goods.

Mark Smith
DP


Mike Most writes:

>Here's the rub: Nothing in software can emulate a telecine, the setup of >that telecine, or a DaVinci or Pogle colour corrector.

I'm not familiar with the KLMS, etc., but everything Mike says makes sense to me. It has been my experience that a good colourist or a good timer -- and there are a lot of very good colourists and timers out there -- are far better at what they do than I will ever be -- or hope to be -- or even care to be. I don't think I have ever been to a timing session or a transfer in which the timer or the colourist hasn't materially improved the end product.

If I get back dailies that are far from what I expected, it is usually the result of a failure on my part to be clear with my intention to the timer. On a longer term project, timing problems can usually be worked out after a couple of days. I don't know who has the time or the energy after a day of shooting to sit down at a computer and time dailies; I know I don't. What's the point. Especially when the film is going to someone who does exactly that all the time.

The thought of trying to do it on the set gives me heartburn. Commercials are a different matter entirely. However, since the agency art correctors (directors) are going to change everything anyway at least twice, it doesn't really matter there either.

Brian Heller
IA 600 DP


Michael Most wrote:

>I still say that as long as you're shooting film and not scanning at the >beginning of the process, all of these things ultimately have limited >value.

Elegantly put, Mike.

Jeff "low end user in search of, well, something..." Kreines


>"As far as dailies, I still don't understand why the KLMS is any different >or more effective than tweaking frames in Photoshop"

>"Yes, I never understood this."

Important here is A. proper software calibration on the monitor (assuming it's a laptop LCD most likely), using the OS calibrator or a 3rd party utility (freeware SuperCal, etc), and B. choosing your standard working colour space in Photoshop. As much as I dislike sRGB, it's probably your best bet. The "General Purpose Defaults" setting (Colour Settings preference) will work well. For a wider gamut RGB, more filmic, "US Prepress Defaults" C. ICC tag all files. D. Make sure the operator on the other end is set up similarly and reading the tags.

This is assuming emailing the files, not sending prints. Using Photoshop won't emulate specific print stocks or anything, but will communicate intent. DP's send us all kinds of reference files, and we love it and pay close attention, since we might otherwise just get Cineon scans and an EDL.

Tim Sassoon
SFD Vfx & creative post
Santa Monica, CA


> Here's the rub: Nothing in software can emulate a telecine,

My own conclusions on LMS, and fairly forcefully communicated to Kodak, are that if you're doing a "traditional" finish then it's great.

I mean, really really good.

But if you're going for a DI or TK finish then you gain nothing over PS.

I'm looking forward to seeing 3cP at NAB.

Cheers

Geoff Boyle FBKS
Director of Photography
EU Based
www.cinematography.net


>And as per difference between 3cp and Photoshop - see our Q&A >section at : www.gammaanddensity.com

What about using Colour Finesse?

John Babl
Miami USA


Gamma & Density wrote:

>PS: And in response to an esteemed, but apparently not too-well->informed CML member Michael Most :...It produces extremely accurate >results for use during DI colour correction.

Disclaimer: Yuri and I are friends, and when he showed me an early version of 3CP I expressed the same sentiments I stated in my post.

Having said that, I did not mean in that post to demean the efforts of G&D (or Kodak, for that matter) to improve the state of the art. And I did mention that I feel that 3CP, in its current incarnation, does provide very useful information to a telecine colourist via its simulated waveforms as well as the accurate image. I also knew about Yuri's involvement with Bob Currier, and that is a good portend for the future.

At the moment, however (and Yuri, please correct me if I'm wrong here), there is no way to bring information (other than visual information - which, of course, is useful in itself) through the process to a DI suite and "plug in" any numbers that might give you a reasonable starting point. And since neither Kodak nor G&D make full DI systems (these are primarily supplied, at the moment, by DaVinci, Filmlight (Baselight), Discreet (Lustre), Nucoda, Quantel (IQ), Pandora, and Silicon Colour (Final Touch)), it is not likely that such a "closed loop" system is forthcoming based on the current products.

In fact, the only manufacturer I can think of that is reasonably close to such a system is Iridas, since they have versions of their software for both production and post production. That's not to say it can't be done, it is to say that the marketplace is somewhat fractured in terms of the equipment used, and in part because of that, general purpose solutions can currently only have minimal impact during the finishing steps.

Mike Most
VFX Supervisor
IATSE Local 600
Los Angeles


Michael Most wrote:

>Having said that, I did not mean in that post to demean the efforts of >G&D (or Kodak, for that matter) to improve the state of the art.

Disclaimer : Michael and I are friends, and when he expressed the same sentiments in his post, I was relieved that nothing had changed and we are still friends...

Michael :

You're right that there are advantages to closely integrated solutions, and we have our own active plans in that area, but it is our intent to remain an "open" system that can work with any DI system.

As a member of the ASC Technical Committee, I'm actively involved in the efforts to make it easier to exchange information between colour correction systems. We certainly plan to create an integrated practical solutions for all parties involved.

But all talk of the "ideal" system aside, what we have now is creating with 3cP very good and much needed for DP and an industry practical results on real projects.

Of course there is room for improvement in working with so many different systems, 3cP including - we are open to all suggestions and proposals - but that's a process that never ends.

All the best,

Yuri Neyman,
Director of Photography, Founding Partner
Gamma and Density Co.
1041 N. Formosa Ave.,Formosa Building, Suite #009
W.Hollywood, USA,90046


Mike (and Yuri), I apologize if this is slightly off-topic, but what would you say is the percentage of Hollywood films that go to DI? Is it close to 50% at this point? And again, forgive my ignorance, but is it no longer customary for the DP to be involved in post production? I can imagine these systems being useful on a very compressed schedule, but is it unreasonable for colour correction to occur in post, with the DP present, rather than half-assing it during production?

Jim Eagan
NY cameraman/editor


Just my two-pence:

>I can imagine these systems being useful on a very compressed >schedule

As I understand it, the DI process is *not* a time-saving process. Yes, it might save you time *on set* because you don't have to spend an hour ND'ing every window in shot because you can bring the windows down in the DI. But the DI offers soooo many options that it can take a long time.

>is it no longer customary for the DP to be involved in post >production?

As I understand it, most DP’s do sit in on the DI... indeed, I think some see the DI as an extension of what they're doing on-set.

Jack Kelly
Dir / Prod / Cam
London


The introducing question about the "workability" of LMS on a set has been replaced by related and more consistent ones : is that kind of system, the Kodak one, the G&C one or others under development, so far technically relevant? Does it help the cinematographer to achieve a visual treatment up to the end ?

As far as the process stays argentic, the parameters are known and are similar everywhere. No twilight zone, or just the amount needed to delightfully spice the job... I guess that's why it is possible to modelize them in order to get, as quoted Geoff Boyle, pretty reliable simulations. But exactly for the same reasons, the need for a simulation is, in most situations, quite limited. The tools the DP already has are efficient and unobtrusive (cf. the report of Adrian Cranage, 33.33% of the report of Yuri being to be confirmed). And so are the calls with the timer for the dailies.

Slightly off-topic. Better if the dailies give a good sense of how the image has to look like, but photography is not all. A 1st AC, for example, is totally unable to check on a monitor an image shot for theatrical projection, digital or not. It's not at all nostalgia, nor in any way resistance towards any technology, it's just that a daily is not anything seen day by day (at best...).

Actually, according to what has been so far said on this forum, it seems to be reasonable for the moment to stay with the pola for personal fast check on the set and, when needed, with digital frames tweaked in PS to send to the timer or the colourist. I'm planning anyway to closely check the KLMS and the 3cP softwares. The could be a good pedagogic support. But that's really another subject.

Cheers.

Georges Leurquin
DP and instructor
Brussels



Copyright © CML. All rights reserved.