Welcome to our community

Be a part of something great, join today!

Register Log in
  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

Social Network...

M Most said:
One of the things that Arri got right in its original implementation of RAW conversion on the D21 was the inclusion of multiple color matrices that are user selectable, and that represent different saturation levels, but all targeted to the correct color space. So you had a choice of LogCFilm50, LogCFilm75, or LogCFilm100, for example, that would yield similar results but with different saturation levels, in order to accommodate exactly what you're talking about. To some degree, I think that's what Red is starting to do with things like Redcolor2, and it's a good sign.
Click to expand...

This sounds very attracting..
 
Ian,
It seems that editorial touched almost every frame in some respect in After Effects at least prior to 2K DPX delivery. Do you feel that editorial directly delivering 2K DPX's made your post process restricted at all? were there any points at which you went back to the .R3D file directly in your Pablo?
I'm assuming LightIron was involved in locking/setting up the Adobe DPX pipeline with editorial prior to delivery back to you?

thanks,
Jason
 
Ian Vertovec said:
Correct, and they sent us 2048x1024 dpx. However, I believe they did the de-bayer in REDline.
Click to expand...

So it would be right to understand that the DI wasn't actually done from the RAW R3D files?

Isn't there some advantages to grading from the RAW R3D. Doing all the processing at full 4K internally, (keys for secondaries benefit greatly from this, specially noisy ones...amongst other things). I understand there are no 4K deliverables on your film, so you could process internally at 4K, even if the system downconverts internally the result to whatever format you want should be superior.

Or am I just processing 4x the ammount of data for no real benefit??? (I know thats how we do our movies/spots here)

Thanks!
 
Frank Cueto said:
So it would be right to understand that the DI wasn't actually done from the RAW R3D files?

Isn't there some advantages to grading from the RAW R3D. Doing all the processing at full 4K internally, (keys for secondaries benefit greatly from this, specially noisy ones...amongst other things)..
Click to expand...

That would be true if all of the shots in the picture were directly from a Red camera, as is often the case on a television episode or a small, independent movie. On studio pictures it is rarely the case, as many, many shots go through visual effects. On Social Network that was certainly the case, as Michael has already explained earlier in this thread.
 
Ok, that certainly makes a lot of sense.

Thanks.
 
I do agree that having access to the RAW data gives color the widest range of flexibility. (color temp and tint are quite powerful compared to RGB lift, gamma, gain). However, on a project of this scope there are many things to consider.

That Adobe video shows how David and Angus would spilt the screen and stitch together several performances for timing, reactions etc. What the video doesn't show you is how often they do it and how complex the tracking is on a film with focus this shallow. Tyler did an amazing job there.

There was also VFX to consider. RED has yet to implement an EXR lin to REDlog roundtrip. (or at least a round trip we were comfortable with) So at the time this was coming together we didn't want the several VFX venders to all get R3D's and debayer them 20 different ways. We wanted one house doing all the debayer. Tyler and I worked out the debayer settings we were happy with, then I built a LUT for vfx venders that corrected the REDlog into something that David had approved. There are a lot more VFX shots then you might think.

The other party in this is Reliance Media. While the color was going on, they were doing their custom denoise/sharpen process (used on Zodiac and Button). Of course they need dpx as well.

I could have done all the color, stabilizing, spilt screens, denoise and sharpening in the Pablo, but if I had, I'd still be working on it. Dividing the labor really makes sense on a project like this.

The other thing worth mentioning is the fact that compression and debayering steals processing power. In the same way Charles Lindbergh threw out his seat cushions to make his plane more efficient, we wanted to squeeze every inch out of our performance. And we didn't use rocket because a rocket debayer is softer then a software debayer, it's just true.

So really considering the labor was divided the way it was, it made sense to do a one-debayer-to-rule-them-all mentality and roll with it. Tyler said he would redeliver some shots if I couldn't get there with the ISO / Kelvin selection done in camera, but we never had that problem.

Looking forward, I'd really like to explore linear EXR's with attached metadata to get the best of both worlds. We may be doing this sooner then we think.
 
It's not a bug as much as it's just a different process. The Redrocket does not do any (or very much) post sharpening. However most software based solutions do. If it is ever an issue you can post sharpen in different software and obtain the same results.

For me, ending up at 1080 most of the time, and shooting in 4kHD most of the time the downscaling takes care of any of the subtle differences between software and redrocket process. The time saved with the rocket is worth it to me.
 
Ian Vertovec said:
David and Angus would spilt the screen and stitch together several performances for timing, reactions etc. What the video doesn't show you is how often they do it
Click to expand...

I would love to know which was the most challenging shot this was done to. I wonder if for these you might have reverted to the R3D and cropped in. But paralax really limits camera movement one way or an other, so was it done only to locked down shots?
 
Brian Smith said:
Saw it... great movie, but the picture didn't look quite right. Maybe it was the digital projector, but it seemed kind of soft, video-ish, and, at-times, weirdly colored. Almost the whole film seemed either underexposed, or improperly color corrected (gray blacks and stuff). Wondering what the deal is with this. On a positive note, it was extremely clean (noise free) in all situations.
Click to expand...

I'm with you on this. I just saw it at the Arclight in Hollywood. At times it looked like there was an extremely fine scrim in front of the screen, things just looked slightly hazy and soft. I don't know if this was caused by diffusion in front of the lens, or some process in post.

I also thought the colors looked odd and kind of electronic. And I didn't see the rationale for it.

People may not have liked the look of "Public Enemies," I certainly did like the look. I understood it; to make a period film look contemporary, Spinotti and Mann intentionally went for a modern almost video look. I really had no idea idea what look Fincher was going for except kind of under exposed and maybe kind of maroon/crimson (in homage to Harvard?).

I was going to say I never thought 4K could look that soft, but I read earlier that all the .R3D's were rendered out at 2K (for understandable reasons).

I do want to thank Michael and others involved with the film for posting here and congratulate them on their work. Making any movie is like going to war. But they made a blockbuster with cutting edge technology, so hats off to them.

That said, I'm not going to lie about my opinion of what I saw, FWIW.


P.S. I do want to say that the compositing of the Winklevi was simply amazing.
 
Peter Moretti said:
I'm with you on this. I just saw it at the Arclight in Hollywood. At times it looked like there was an extremely fine scrim in front of the screen, things just looked slightly hazy and soft. I don't know if this was caused by diffusion in front of the lens, or some process in post.

I also thought the colors looked odd and kind of electronic. And I didn't see the rationale for it.

People may not have liked the look of "Public Enemies," I certainly did like the look. I understood it; to make a period film look contemporary, Spinotti and Mann intentionally went for a modern almost video look. I really had no idea idea what look Fincher was going for except kind of under exposed and maybe kind of maroon/crimson (in homage to Harvard?).

I was going to say I never thought 4K could look that soft, but I read earlier that all the .R3D's were rendered out at 2K (for understandable reasons).

I do want to thank Michael and others involved with the film for posting here and congratulate them on their work. Making any movie is like going to war. But they made a blockbuster with cutting edge technology, so hats off to them.

That said, I'm not going to lie about my opinion of what I saw, FWIW.


P.S. I do want to say that the compositing of the Winklevi was simply amazing.
Click to expand...

Don't go see it in the arclight dome. The lens on the projector flares incredibly and the curved screen is painful to watch what it chops off on the composition.

The print I saw at the DGA looked beautiful and organic.
 
I saw TSN at the Arclight Sherman Oaks. I thought it looked great for a film print- I'd love to see it digitally projected, just to see the difference. If anybody knows where its playing digitally, I'll give David Fincher & Co. another $14 to see it again. It has the best night shots that I can remember seeing in a while. To finally see the ambient light of the city above all the buildings was very exciting to say the least. I could see inside the buildings on campus and see desks and people and books and papers on the desks-- just really impressive. I can't remember seeing a film print that had such discreet detail of the insides of buildings in a night exterior crane shot like that.

I also had a strange realization while watching it. Here we have a story of a wildly successful tech start-up and the people behind it--- shot with cameras from a wildly successful tech start-up. While Facebook has certainly bloomed and made for transformative social experiences, I couldn't help but wonder if the imaging technology that shot the movie won't, in the end, be more transformative socially or culturally. It certainly has the potential to be so. I mean we use Facebook to communicate, but its not an authoring platform, its a social context. Cameras have the power to forge ideas-- sometimes intimate, sometimes globally influential-- with huge ramifications.

It was a very cool little moment that will stick with me, kinda like when I was learning to walk at the same moment men were walking on the moon. Its cool to step back a bit and appreciate how much everything has changed in your lifetime. I loved the movie.
 
Saw TSN at Munich's "Cinema" (by far the best local option for OVs) on a Barco 2K. I'm still in awe. To me it looked beautiful, stunningly unique, rich and deep. TSN's look is still, a week later, burnt-in, can't quite put my finger on it. That opening scene, isn't that a rather unusual grade? And yet from there on you're taken into a very special space and it works great. Sorry for the drooling but it's such a great piece of work. Thanks to all involved.
 
Jeff Coatney said:
I saw If anybody knows where its playing digitally, I'll give David Fincher & Co. another $14 to see it again.
Click to expand...

The Rave Cinemas at Howard Hughes (formerly the Bridge Cinemas) off the 405 in Culver City/Westchester (I've never really been sure which....) are now all digital and one of the better places in town to see just about anything.
 
Will we see a movie by Fincher titled "The Camera Company" within the next 10 years?
 
You must log in or register to reply here.
Back
Top