Manure

[Justin Hayward]

Congrats!

 

Were you pretty happy with the print and projection and all that?

[David Mullen ASC]

I wasn't able to go, but we just finished the D.I. more or less just before the screening so we decided to project digitally from an HD tape -- we didn't have time for a film-out and I'd like a little more time to make tweaks before we commit to a film-out.

[DS Williams]

David you're an inspiration to me and many of us. Thank you!

[Alex Hall]

David,

 

Are you making a lot of changes in the DI for "Manure" or did you achieve most of the look you were going for in camera and with production design?

[David Mullen ASC]

The "brown" look was entirely in-camera through art direction -- even just the converted RAW files pre-correction came up immediately with that look. If you simply corrected the faces to "normal" the rest of the image naturally was brown and muted. The smoke on the sets helped too.

 

The main thing we did in the D.I. was add more diffusion to cause bright areas to glow more, by pulling a luminence key of the brightest areas, defocusing it, and then layering it back over the image. This way we could control the strength of the effect and what was affected in the frame.

 

I had shot with some light Classic Black filters (#1/8 and #1/4), which added a subtle diffusion effect too.

 

Occasionally we had some problem removing unwanted colors in the frame -- for example, we put grass sod down in one outdoor set, spray-painted brown, but simply by animals and people walking around on it, some pale green sort of came through the brown being rubbed off, which was hard to remove in the D.I. -- it almost would have been easier if it had been primary green because then we could at least pull a chroma key from it. I had one wide shot were I had to take the entire lower half of the frame, containing the grass, some chairs, and some people's pant legs, turn it b&w, and the tint it brown just to get rid of the green cast in the grass. But that was only on a couple of shots, the rest of the movie was pretty standard color-correction.

 

The hardest thing was staring at brown images for hours and hours because you found yourself wondering if the brown had too much yellow or too much magenta in it to be "brown", which is not a color on the color-correction knobs.

[Alex Hall]

How much more work needs to be done to the image before you commit to a print? Are there certain things you change about the image that will work better with the print stock?

[David Mullen ASC]

I want to go through the movie and do some selective sharpening, mainly of eyes, just to keep the image from looking too soft -- I know that's a contradiction because I used diffusion afterall, but I feel that the RED image in general needs a slight amount of sharpening because it has a somewhat "smooth" look. But I don't want to overdo it, so I need to see what I'm doing on the big screen of the D.I. theater.

 

There are some shades here and there that still need tweaking. The problem with a monochromatic image is that it is harder to time shot to shot because your eye keeps picking up subtle variations. Plus in shots where a face is in dimmer light, there is a tendency to lose too much of the chroma, so I may need to go back and add a little back into the shadows. I could probably spend months chasing down those variations...

[Alex Hall]

The corrections never end, we just run out of time. Thanks for all the info.

[Thom Stitt]

The hardest thing was staring at brown images for hours and hours because you found yourself wondering if the brown had too much yellow or too much magenta in it to be "brown", which is not a color on the color-correction knobs.

 

Hilarious - I know exactly what you mean. For the most part, color correcting and adjusting becomes kind of an intuitive process once you have some time with it. But muted browns! The earthiest and simplest of colors to understand, suddenly becomes some enigmatic, gloomy thing to properly achieve using primary and secondary colors.

 

And staring at it for so long - I once DPed a short that had a warm, muted palette. I sat down and made mock-up adjustments, but when I revisited it later on, I realized that certain shots (mainly the last ones that I did) just looked YELLOW compared to the rest. My eyes had just adopted some kind of brown-resistance after a while. I can't even imagine doing a feature like that in a sitting. Very appropriate that the experience occurs on a film called Manure.

[John Allen]

Yeah, not too long ago, I was color grading a period scene that me and the director wanted a kind of sepia/cream color for it. After hours of looking at the same thing, I started to put too much yellow in it and when I showed my director he was like, "wha?? They're like really yellow!" So I decided to take a little rest for my eyes. Then when I came back to it, I could totally see what my director was talking about. Since then I've made sure that I take little rests for my eyes. Oh and the scopes help too. But yeah, it's amazing how deceitful your eyes can be when you allow them to adapt.

[K Borowski]

Even with "rest periods", I have heard that one can only do optimal color-correction work for about four hours a day before the work starts to suffer.

 

One trick at avoiding the eye's natural color balancing process is to periodically look at something neutral, like a grey card or a white sheet of paper under daylight balanced illumination.

[John Allen]

Even with "rest periods", I have heard that one can only do optimal color-correction work for about four hours a day before the work starts to suffer.

 

One trick at avoiding the eye's natural color balancing process is to periodically look at something neutral, like a grey card or a white sheet of paper under daylight balanced illumination.

 

Yeah, it's almost the same as your sense of smell. If you smell a bunch of different smells then after awhile your nose will adjust so that every smell will start to smell the same. That's why at perfume stores you will find little containers filled with coffee beans for customers to sniff in between smelling fragrances. To kind of rejuvenate their sense of smell. Don't ask me how I knew that, I just did. lol :blink:

[Tim Fabrizio]

I love these. David, thank you so much for your insight and lessons. I (we all) truly appreciate. Manure...sounds like a poop show. HA HA HA.

Good luck!

[Chris Burke]

David,

 

I remember you talking about the difficult schedule you had while shooting this picture. Do you think that the intense schedule, had any affect on the end product? I ask because it is always an issue, especially with indie budgets, that time is money and you rarely have enough of both.

Edited February 13, 2009 by Chris Burke

[David Mullen ASC]

David,

 

I remember you talking about the difficult schedule you had while shooting this picture. Do you think that the intense schedule, had any affect on the end product? I ask because it is always an issue, especially with indie budgets, that time is money and you rarely have enough of both.

 

Yes and no -- I mean, all movies are a product, to some extent, of their limitations in time & budget. The end result is what it is. We shot the movie that was written and the actors were able to do their job, so to that extent, we were successful. I also think that we accomplished what we set out to do stylistically. Now maybe some elements didn't achieve the level of scope that the story suggested, but I think if you saw the movie, you'd be amazed at what was accomplished for the budget in terms of "opening up" a small story limited to soundstages.

[Keith Walters]

Even with "rest periods", I have heard that one can only do optimal color-correction work for about four hours a day before the work starts to suffer.

 

One trick at avoiding the eye's natural color balancing process is to periodically look at something neutral, like a grey card or a white sheet of paper under daylight balanced illumination.

For telecine work in TV studios, the usual practice was to have a grade 1 monitor centred in a large panel of translucent white plastic, backlit by colour-corrected fluorescent lamps.

About 20 years ago I used to work under a Chief Engineer who was excessively anal about the surround being exactly 4100K, and I was always crawling in there amongst the dirt, dead moths and cigarette butts, sticking on new bits of red and blue gel!

 

The move to desktop editing on PCs has largely eliminated this practice.

[Chris Burke]

Yes and no -- I mean, all movies are a product, to some extent, of their limitations in time & budget. The end result is what it is. We shot the movie that was written and the actors were able to do their job, so to that extent, we were successful. I also think that we accomplished what we set out to do stylistically. Now maybe some elements didn't achieve the level of scope that the story suggested, but I think if you saw the movie, you'd be amazed at what was accomplished for the budget in terms of "opening up" a small story limited to soundstages.

 

 

I look forward to seeing it as I am a fan of both the Polish brother's work and yours. Congrats on the good review of your work.

[Mathew Rudenberg]

This thread has long gone quiet, but I was checking out Michael Cioni's new company Lightiron for post on a gig and they were color correcting Manure in the pablo - which I recognized right away from the stills Mr Mullen posted.

 

It was absolutely gorgeous at 4k on the big screen. My post sup couldn't believe the backdrops were real instead of greenscreen. I can't wait for it to come out in theatres, amazing work.

[Rich Hibner]

just started reading this thread then looked the movie up. found this on imdb

 

http://www.imdb.com/video/imdb/vi1321861913/

[Alexander Disenhof]

"How about a big unbleached muslin?"

 

Could work, but would probably suck up too much light.

 

Stumbled upon this a little late but...wouldn't the unbleached muslin still suck up less light than a painting overhead, which would block the light entirely?

[aviishek shovakar]

Looks pretty -- no need for Chocolate filters on the lens.

 

Good luck with using the RED One; i"m sure you'll make it sing. And if you need any hign speed footage I know the perfect companion camera...

hi I am shooting a Feature on Red and i have a few explosions along with a glass house exploding so i do require a high speed option, i was going to go photo-sonics but i am not keen to mix mediums, nor is my producer (but for different reasons )i know the phantom does high speed, unless u are suggestion another camera , would appreciate the help, also I know that Red loves daylight balanced sources, but if tungsten sources are used is there noise ?

 

thanks

avii

[David Mullen ASC]

hi I am shooting a Feature on Red and i have a few explosions along with a glass house exploding so i do require a high speed option, i was going to go photo-sonics but i am not keen to mix mediums, nor is my producer (but for different reasons )i know the phantom does high speed, unless u are suggestion another camera , would appreciate the help, also I know that Red loves daylight balanced sources, but if tungsten sources are used is there noise ?

 

thanks

avii

 

Hopefully you can use a Red One with the new M-X sensor, it doesn't really have a problem with 3200K lighting even if it prefers daylight, like all sensors actually do.

 

You can shoot tungsten on the old sensor, just be conservative in your exposure and ASA rating.

 

You can shoot up to 60 fps in 3K I believe, but if you have a lot of footage to shoot that is well-over that speed, I'd get a Phantom HD camera rather than shoot 2K on the Red One... the Phantom should intercut fine with Red One footage, maybe even better than 2K Red One footage intercuts with 4K Red One footage. On the other hand, I haven't tested 2K on the new M-X sensor, perhaps the improved OLPF has made 2K look more acceptable.

[John Sprung]

On the other hand, I haven't tested 2K on the new M-X sensor, perhaps the improved OLPF has made 2K look more acceptable.

 

You could crop and blow up 2K's worth out of your existing tests and see if it looks any better. My guess is it won't, because 2K Bayer just isn't enough samples. It's the Nyquist limit....

 

 

 

 

 

-- J.S.

[David Mullen ASC]

You could crop and blow up 2K's worth out of your existing tests and see if it looks any better. My guess is it won't, because 2K Bayer just isn't enough samples. It's the Nyquist limit....

 

But technically, isn't the Phantom HD only a 2K Bayer camera? But the OLPF was optimized for the whole sensor at 2K, as opposed to an OLPF optimized for a 4K Bayer sensor which is then cropped to 2K.

 

That's why it would be good to test the differences, though in the past I would have said that the main difference was the noise level of the Red frame cropped to 2K, compared to the Phantom HD 2K frame, which was cleaner. But now that the new M-X sensor has a lower noise floor, maybe the difference is more in terms of resolution, which in theory should be similar.

[John Sprung]

But technically, isn't the Phantom HD only a 2K Bayer camera? But the OLPF was optimized for the whole sensor at 2K, as opposed to an OLPF optimized for a 4K Bayer sensor which is then cropped to 2K.

 

I dug through their web site, and it doesn't actually say "Bayer" anywhere that I could find. But it sure sounds like it could be. Their chip is 2048 x 2048 photosites, 25.5 x 25.5 mm.

 

The sensor and OLPF are fairly close together and never move relative to each other -- at least in all the cases I'm aware of. So, when you window or crop the sensor, you're equally windowing/cropping the OLPF. Both Phantom and Red are doing the same kind of thing when they window down the image size to accommodate faster frame rates.

 

The OLPF acts very locally within the image -- it's just there to knock down the maximum resolution to below the Nyquist limit for the sensor, to keep it from aliasing and giving you a moire pattern. It depends not on how many photosites there are, but rather on how big they are. Detail finer than twice the size of a photosite (or more precisely, twice the pitch from one photosite to the next), in theory, needs to be filtered out. In practice, they cheat on that a little bit. Especially on Bayer chips, where you have a Nyquist limit for red and blue that's only half what it is for green. Luminance comes mainly from green, and resolution mainly from luminance, so we seem to be OK with aliasing in the red and blue channels. That's why Bryce Bayer's idea was a good one. You can shoot a zone plate

 

http://en.wikipedia.org/wiki/File:Zonenplatte_Cosinus.png

 

that runs through the Nyquist limits and see wicked red/blue aliasing, but there aren't many things in the real world that look like that.

 

 

 

 

-- J.S.