Film Processing Houses for Film Out

[David Parks]

I've been approached to be involved as a producer in a low budget feature and I'm pushing using my HD-100 with simultaneous Wafian record from analog component outs/HD-SDI and HDV tape for backup.

The director is looking at hire a local DP who is pushingn a different workflow and Baby HD camera solution (I'm not going to say what it is because I'm not looking for a debate on cameras/workflow).

Does anyone know some good film labs that process M2T, QT, or DPX files to 35mm film for a test??

I'm a little familiar with dvfilm but looking for more options.

Thanks David Parks

[Tim Holtermann]

David,

DuArt did the film transfers on the HD100 JVC events: http://www.duart.com/site_main.html

Pactitle: http://www.pactitle.com/services/index.php

[Stephen L. Noe]

Duart is a great resource. There is also I³ (I cubed) here in Chicago. Lab services are available through Cinefilm Laboratory in Atlanta. They price per foot.

[David Parks]

Thanks Tim for the info... I saw (and kept)a WMV film of that film out (shot by a Chicago company I believe)in early Feb and that helped clinch my decision in buying the HD 100. I've already showed that clip to the director and he was impressed. DuArt has been around a very long time.

Thanks again

[David Parks]

Thanks Stephen, I appreciate the info, Did you or Tim attend any of the events where the showed the 35mm prints from that JVC demo?? I'm curious if you have any opinions and recommendations (things to watch out for)when using the film out settings on the HD 100.

[Stephen L. Noe]

Originally Posted by David Parks

Thanks Stephen, I appreciate the info, Did you or Tim attend any of the events where the showed the 35mm prints from that JVC demo?? I'm curious if you have any opinions and recommendations (things to watch out for)when using the film out settings on the HD 100.

I happen to be one of the guys that created a filmout at the Chicago event. I think you should not use the "filmout" preset at all. Instead use any of the supplied scene files at the top of this site in order to get your desired effect in camera that will transfer over to film nicely. Andrew Young is the resident guru on film transfer because his company (DuArt) does the process end to end.

In my experience with I (Cubed) the process was easy as long as I provided the images in the fasion they needed. I cut the 24p video on Liquid which outputs the exact format they need and without any rez/color robbing transcode to another format. In fact I was so impressed with how easy the process was (for me with my "editor" hat on) that I feel confident in the technical aspect of ProHD to film.

More projects seem to be on the horizon (for my group) that require the same film transfer techniques I've become familiar with. I can say that for a budget film maker, it is extremely hard to beat the HD-100 and ProHD as an off the shelf solution to low budget dramatics.

[David Parks]

Interesting. I've looked at the settings from Tim Dashwood and Paulo Ciccone and they're great. But , (speaking as a producer who's not a DP) wouldn't you use the film out setting if you plan on a great deal of post color correction/enhancement or is there something else about that setting that isn't the best for film out? And what does the film out do other than replicate film gammas (as I understand the term)? I'm sure I'll have more questions as we go through this process.

Anyway, the info is great and I need to budget a decent amount for film out tests.

[Stephen L. Noe]

Originally Posted by David Parks

Interesting. I've looked at the settings from Tim Dashwood and Paulo Ciccone and they're great. But , (speaking as a producer who's not a DP) wouldn't you use the film out setting if you plan on a great deal of post color correction/enhancement or is there something else about that setting that isn't the best for film out? And what does the film out do other than replicate film gammas (as I understand the term)? I'm sure I'll have more questions as we go through this process.

Anyway, the info is great and I need to budget a decent amount for film out tests.

My understanding about the 'filmout' setting is that it was put in by request and the idea was to create a linear color curve in order to get the max lattitude for correction via the colorist or post in NLE, however, the filmout settings kills the black and grays and make it un-useable.

Paolo's DSC scene file is the opposite end of the spectrum from the filmout setting, however the "DSC" scene file is not too good for skin tones and oversaturates suttle features when dealing with dramatics (IMO). I think the "DSC" scene file is ideal for TV commercial work or comping the video. Tim's "Warm" scene file is very good for transfer to Vision 2. I'd say it was the closest to what I saw on my timeline. Since then I've derrived the "Panamatch" scene file which I think handles the skin tones very well and there also is the "Any Scenario" scene file which is a decent starting point as well. Any one you choose is better than the 'filmout' setting for literally transferring to film. Concensus is that the "filmout" setting is un-useable.

S.Noe

[Paolo Ciccone]

Hi Stephen.

Agree about the filmout. It puzzled me, I could not figure out why it was included. About my config, I shot several skin tone clips and the results have been very close to real life (IMHO). Kinda puzzled about your comment here.
It is a configuration that is meant to be the basis of post-work, not the end point to achieve a given look. I'm interested about the "oversaturates suttle features when dealing with dramatics (IMO)" comment. Can you give me more details about it?

Thanks!

[Stephen L. Noe]

Hi Paolo, I think the DSC scene file is great. For my taste though I like skin to to be much more suttle and not necessarily pop off the screen. I'm more of a fan of the color treatment in movies like "Midnight Cowboy" or "Rocky" which are not overpowering in color.

Either way David, the 'filmout' setting was a good idea but poor execution. I wish JVC would have done an actual filmout to 35mm with the setting. They would have reworked it's parameters immediately.

[Paolo Ciccone]

Originally Posted by Stephen L. Noe

Hi Paolo, I think the DSC scene file is great. For my taste though I like skin to to be much more suttle and not necessarily pop off the screen. I'm more of a fan of the color treatment in movies like "Midnight Cowboy" or "Rocky" which are not overpowering in color.

Thanks, point taken. I'll ponder on it :)

[Steve Mullen]

Originally Posted by Stephen L. Noe

Either way David, the 'filmout' setting was a good idea but poor execution. I wish JVC would have done an actual filmout to 35mm with the setting. They would have reworked it's parameters immediately.

Given the difficulty in getting a company to add something to a product--I've always wondered how it came about. Did someone in the USA have the clout? If so, why is it so bad? Or, do we just not know what the thinking behind it was?

Or, did it come from a video engineer in Japan who guessed at what it should be. And, why hasn't it been altered in the A firmware? In fact, has anyone ever read anything from JVC about it?

Seems like we all should ask the question at NAB because right now it is going to waste.

[Brian Drysdale]

Paolo,

I used your set up as a base line on my recent camera test. Unfortunately, I didn't have a vectorscope etc while I had the camera, but using my cinematography Gamma & Density chart it looked a good match by eye on the monitor.

I also tried a few of Tim's set ups which also looked good.

Reading your comments, your idea is to carry as much colour information into post rather than creating the final look in the camera. For film out I think you really need to run a test with the lab that you're working with to find out the best setup for your production. There are so many variables and workflows that what works with one lab may not work so well with another.

Filmrec on the Varicam looks extremely flat to the eye, you need the LUT to make it look good on video.

[Tim Dashwood]

Originally Posted by Steve Mullen

Given the difficulty in getting a company to add something to a product--I've always wondered how it came about. Did someone in the USA have the clout? If so, why is it so bad? Or, do we just not know what the thinking behind it was?

Or, did it come from a video engineer in Japan who guessed at what it should be. And, why hasn't it been altered in the A firmware? In fact, has anyone ever read anything from JVC about it?

Seems like we all should ask the question at NAB because right now it is going to waste.

I think it came by request from Digital Film Group in Vancouver.
The filmout curve in itself is fairly linear and works as expected if you are familiar with using FILM REC and a proper gamma box on the Varicam.
A linear curve does not extend the latitude in the highlight range, but instead shifts the response down, resulting in apparently dark images (when viewed with traditional gamma monitors.)
The idea is to capture linearly and NOT do a digital intermediate, but instead print to film, let the film stock's native response curve lend itself to the 'film-look' and then use traditional color timing lab processes to finish your film.
However, I've had a conversation or two with Andrew Young at DuArt about this who also found that the HDV codec does not really maintain the video information in the bottom half of the curve the way you would hope, and the results could potentially be disastrous. The information falls to pieces and you get alot of banding and stepping.
I have not tested direct 4:2:2 uncompressed capture, but it seems to me that this would be the only viable way to use the filmout curve successfully.

BTW, we have had this discussion many times before.
http://www.dvinfo.net/conf/showthrea...hlight=filmout
http://www.dvinfo.net/conf/showthrea...hlight=filmout
http://www.dvinfo.net/conf/showthrea...hlight=filmout
http://www.dvinfo.net/conf/showthrea...hlight=filmout


I think it is worth reading this article by Michael Bergeron. http://www.24pdigitalcinema.com/cinegamma.pdf

It refers to the Panasonic Varicam, but the concepts are exactly the same.


FWIW, I'm shooting my feature right now using my wide latitude settings with some slight modifications (only 85% knee.) The plan is to do a D.I. before the filmout, so the more information the better.

[Stephen L. Noe]

Originally Posted by Steve Mullen

Given the difficulty in getting a company to add something to a product--I've always wondered how it came about. Did someone in the USA have the clout? If so, why is it so bad? Or, do we just not know what the thinking behind it was?

Or, did it come from a video engineer in Japan who guessed at what it should be. And, why hasn't it been altered in the A firmware? In fact, has anyone ever read anything from JVC about it?

Seems like we all should ask the question at NAB because right now it is going to waste.

Steve, Tim is exactly right. The filmout setting was put in by a film maker in Vancouver who assisted in making parameters for the original camera. I'm sure the idea was derrived from the Varicam. Tim might be right. Maybe it is useable when capturing component. When I delivered the sequence to I (cubed) we asked them not to color correct the footage and just leave it "as is". This way we could try to figure out what settings in camera would give the best direct to film rendition of the digital equivalent. In that case, as I said then, Tim's "Warm" scene file looked almost identicle to what I saw on my timeline.