Film print look

[robbing]

I think you're getting somewhere. This is the first frame I've seen where I got a distinct film vibe. It doesn't look like very good film, and the effect my be exaggerated a bit too much, but I definitely get a film vibe from this. Good work!

Yep, it was extreme as I said.

If someone gonna shot low budget, then his film look gonna look low budget as well, I mean, this is not bad at all, it's just reality.
I have shot motion film stock in low budget, Kodak, Fuji, s8, 16, 35... and looks low budget, looks film because it's film, but not look like a Hollywood blockbuster.
So I think those are points to keep in mind when we try emulate film.

[RyLo]

Yep, it was extreme as I said.

If someone gonna shot low budget, then his film look gonna look low budget as well, I mean, this is not bad at all, it's just reality.
I have shot motion film stock in low budget, Kodak, Fuji, s8, 16, 35... and looks low budget, looks film because it's film, but not look like a Hollywood blockbuster.
So I think those are points to keep in mind when we try emulate film.

I've been thinking. Film usually has more latitude in the highlights than shadows. Digital is the opposite. So would it help with your emulation if the subject was underexposed 3 or 4 stops? You'd obviously need to de-noise, but that might be a more natural starting place.

[robbing]

I've been thinking. Film usually has more latitude in the highlights than shadows. Digital is the opposite. So would it help with your emulation if the subject was underexposed 3 or 4 stops? You'd obviously need to de-noise, but that might be a more natural starting place.

I think 1 stop below the key could work as the "key". -3 or -4 would be too much.

[RyLo]

I think 1 stop below the key could work as the "key". -3 or -4 would be too much.

I see. I was basing those numbers off this (perhaps flawed) chart and my BMPCC:

single-chip-camera-evaluation-over-and-under-exposure-latitude.jpg

Obviously the 4.6K sensor wouldn't need to under-expose as much.

[robbing]

I see. I was basing those numbers off this (perhaps flawed) chart and my BMPCC:

[ATTACH=CONFIG]27193[/ATTACH]

Obviously the 4.6K sensor wouldn't need to under-expose as much.

I know URSA 4.6K have more latitude than the others BM cameras, but I think it's more close to real 13 stops than 15. I mean, dynamic range isn't just detail, it's also how well hold the colors and contrast as well.
I ear some high level colorists saying: Black Magic cameras are super great as long as use it as 10 stops camera!

Maybe it's kinda exaggerated, but those people are working with contrast and color spaces for theaters or TV which have some rules to fit.

Anyway, your concept isn't wrong. With this chart is easy to understand how film works. Easy to see where the key in Kodak is. We can see a lot room in highlights and as the shadows clipping so fast.

So the way to emulate a random ECN-2 negative film color could be:
- huge latitude, specially in high and mids
- great mids luminance and vibrant colors.
- smooth density and roll-off
- subtrative colors underexpose
- smooth edges (it doesn't mean low resolution, maybe on the contrary)
- some HUE adjustments (but not so as people think) yellow is kinda green not orange for instance.
- a few things more....

Some final touches:
-halation: red glow in highlights and red thin halo in clipped highlights. Also make the white clipped highlights a little bit warm (yellow /magenta)
-transparence: blacks are blacks, shadows could be washed or tinted, but black should be blacks.
-grain: I'm still looking for a natural look and non distracting.

Other than that, I' not an expert, just some experience and eye. If I'm wrong I'm open to debate it and to learn more about it.

[RyLo]

I know URSA 4.6K have more latitude than the others BM cameras, but I think it's more close to real 13 stops than 15. I mean, dynamic range isn't just detail, it's also how well hold the colors and contrast as well.
I ear some high level colorists saying: Black Magic cameras are super great as long as use it as 10 stops camera!

Maybe it's kinda exaggerated, but those people are working with contrast and color spaces for theaters or TV which have some rules to fit.

Anyway, your concept isn't wrong. With this chart is easy to understand how film works. Easy to see where the key in Kodak is. We can see a lot room in highlights and as the shadows clipping so fast.

So the way to emulate a random ECN-2 negative film color could be:
- huge latitude, specially in high and mids
- great mids luminance and vibrant colors.
- smooth density and roll-off
- subtrative colors underexpose
- smooth edges (it doesn't mean low resolution, maybe on the contrary)
- some HUE adjustments (but not so as people think) yellow is kinda green not orange for instance.
- a few things more....

Some final touches:
-halation: red glow in highlights and red thin halo in clipped highlights. Also make the white clipped highlights a little bit warm (yellow /magenta)
-transparence: blacks are blacks, shadows could be washed or tinted, but black should be blacks.
-grain: I'm still looking for a natural look and non distracting.

Other than that, I' not an expert, just some experience and eye. If I'm wrong I'm open to debate it and to learn more about it.

I made a slight mistake. This chart is exposing everything at 800. 5219 is a 500 ISO stock and 5213 is a 200 ISO stock. So 5219 is being under exposed 2/3rds of a stop and 5213 is being under exposed 2 stops.

So really:
5219 is -5.7/+8.8
5213 is -5.8/+8.4

That makes a lot more sense. Someday when I get a chance I'm going to test my underexposure theory with my BMPCC.

Also, this might be of some help to you:

https://cmltests.net/Kodak-5219.html

The DPX scans aren't online at the moment, but you might still find the Vimeo video useful.

[robbing]

I made a slight mistake. This chart is exposing everything at 800. 5219 is a 500 ISO stock and 5213 is a 200 ISO stock. So 5219 is being under exposed 2/3rds of a stop and 5213 is being under exposed 2 stops.

So really:
5219 is -5.7/+8.8
5213 is -5.8/+8.4

That makes a lot more sense. Someday when I get a chance I'm going to test my underexposure theory with my BMPCC.

Also, this might be of some help to you:

https://cmltests.net/Kodak-5219.html


The DPX scans aren't online at the moment, but you might still find the Vimeo video useful.

+8.8 stops it's just awesome.

Looking more footage from URSA 4.6K, I think is a great ETTR camera. Even 5 stops can hold details.
An example overexposed +3 stops and looks good.
Above my film print emulation (done over a screen grab) , bellow normal workflow 709 just taking a screen grab from by Shane's grade.
Looks better, worse?... looks smoother I think watercolor vibe. more luminance in mids. Normal 709 looks pretty cool but kinda plastic look.
emulation-film.jpg


And this could be the final look of my film emulation over these +3 stops URSA test: ( I don't know why but here every pic I upload looks more desaturated that the actual frames)
film final look.jpg

[Tatsuo]

Looks great Robbing! Like I said let me know if you want me to do some testing of your LUT/workflow - I have a pocket 4K, the old pocket and an UMP waiting to go!

[LochnessDigital]

(I don't know why but here every pic I upload looks more desaturated that the actual frames)

The built-in attachment function of BMCuser leaves a lot to be desired. For any type of image that needs any kind of critical evaluation, I've been using Flickr. They don't compress images nearly as much.

[stip]

Very nice, robbing.

[mihnea]

Now it gets serious!
I'd pay for your workflow! The lut is not the most interesting part here, but the way you got there.

[robbing]

Thanks everyone.

I'll share it when it finished

[DoryBreaux]

robbing, have you been able to work on these more? They're beautiful!