Feb. 19, 2026
How An HDR Master Let Us See The Invisible Craft Behind 2001: A Space Odyssey
What if the secrets of a 1968 masterpiece were hiding in plain sight—just waiting for modern color tools to reveal them? We dive into 2001: A Space Odyssey with a forensic eye, exploring how a carefully mastered HDR release lets subtle artifacts surface: matte lines stepping in eight-frame rhythms, hand-painted star fields, and reflections that whisper clues about the set. Instead of diminishing Kubrick’s vision, these discoveries deepen our respect for Douglas Trumbull’s team and the analog ingenuity that still holds our gaze at 24 frames per second.
From there, we pivot to the craft of color as a science. Our guest, Paulo Martins of Alchemy Color, breaks down how to treat your camera like a measurement device. He walks us through building rigorous profiles with thousands of color patches under D50 and tungsten, navigating metameric pitfalls from spiky LEDs, and establishing a dependable baseline inside Lightroom and Adobe Camera Raw. Then we get practical with a methodical, 32-bit approach to inverting color negatives—rooted in open-source insights—so the “print-like” result honors the negative without baking in an arbitrary lab look.
If you’ve ever wanted film’s character without surrendering control, this conversation maps a path. We talk chart-driven emulations, creating accurate 3D LUTs, and exporting Cineon log for seamless grading in DaVinci Resolve with print film and halation treatments. Whether you’re reverse-engineering a spaceship window to glimpse a soundstage or building an end-to-end digital film pipeline for stills and motion, the throughline is the same: use precision to serve the illusion. Tune in, get inspired, and see how curiosity, calibration, and careful workflows can bring the texture of cinema to everyday images.
Enjoyed the show? Subscribe, share it with a friend, and leave a quick review on Apple Podcasts or Spotify to help others discover the craft behind color.
Guest Links:
IG - https://www.instagram.com/alchemy_color/
Website - https://alchemycolor.com/
YouTube - https://www.youtube.com/@alchemy_color
YouTube Video on 2001 - https://www.youtube.com/watch?v=TcpDS1r6mQY
Corridor Crew VFX React - https://youtu.be/Dx3Vv6j4tmE?si=EgjUpcG4OuEgop5A&t=1092
PixelTools
Modern Color Grading Tools and Presets for DaVinci Resolve
Like the show? Leave a review!
This episode is brought to you by FSI, DeMystify Color, and PixelTools
Follow Us on Social:
- Instagram @colorandcoffeepodcast
- YouTube @ColorandCoffee
Produced by Bowdacious Media LLC
00:00 - Opening And Guest Intro
01:45 - Discovering 2001’s Hidden Mat Lines
04:45 - HDR Masters And Forensic Color Tools
08:55 - How The Practical Effects Were Layered
13:19 - Ethics Of Revealing Invisible Craft
16:59 - Reverse Engineering The Soundstage
21:14 - Camera Calibration For Still Photographers
26:49 - Building Accurate Profiles And Light
31:44 - Inverting Color Negatives Scientifically
WEBVTT
00:00:00.080 --> 00:00:02.560
I love that film for for a long time already.
00:00:02.560 --> 00:00:07.759
And and there were some comments going around like uh oh, why did you do that?
00:00:07.759 --> 00:00:12.480
Uh this is an insult to the artwork, this is an insult to Kubrick himself and stuff like that.
00:00:12.480 --> 00:00:25.600
And I'm like, no, I actually I revere the men the men in the team, Douglas Turnbull and all of his TV, even more, because being able to see this little piece of history that was always there, and now we can see it because computers.
00:00:26.079 --> 00:00:32.719
Welcome to Color and Coffee, a podcast that's focused on the craft of color and the artists behind it.
00:00:32.719 --> 00:00:43.039
I'm your host, Jason Bowdach, and each episode we'll sit down with some of the most talented artists in the industry and have a casual chat from one artist to another.
00:00:43.039 --> 00:00:49.280
We'll share their stories, their insights, their tips, and maybe even a little gear talk.
00:00:49.280 --> 00:00:55.600
Whether you're a seasoned pro or just getting started, join us for some great color discussion.
00:00:55.600 --> 00:01:00.880
Sit back, relax, you're listening to Color and Coffee.
00:01:00.880 --> 00:01:05.040
Hello and welcome to this episode of Color and Coffee.
00:01:05.040 --> 00:01:11.920
I'm so happy to bring you guys on for another episode as we start to get towards the end of 2025.
00:01:11.920 --> 00:01:17.359
It has been such a great year, and I'm so happy to bring the podcast back for season three.
00:01:17.359 --> 00:01:20.719
I have a really, really awesome guest for you guys.
00:01:20.719 --> 00:01:24.959
His name is Paulo Martins, and he runs Alchemy Color.
00:01:24.959 --> 00:01:30.719
They do custom camera calibrations, they do film emulations, and they do color calibrations.
00:01:30.719 --> 00:01:32.239
Welcome to the show, Paulo.
00:01:32.480 --> 00:01:33.760
Hey, thank you for having me.
00:01:33.760 --> 00:01:35.599
Nice to meet you, and hello everyone.
00:01:35.920 --> 00:01:39.280
It's truly my pleasure, and I'm so happy you were able to come on the show.
00:01:39.280 --> 00:01:45.840
So, first off, I need to tell people the story about how I discovered you, because it's pretty funny actually.
00:01:45.840 --> 00:01:53.599
For those that don't know about Quarter Crew, they are a VFX YouTube channel and they do these amazing VFX breakdowns.
00:01:53.599 --> 00:02:10.159
And I discovered you via a Quarter Crew video where you were breaking down a Stanley Kubrick 2001 Space Odyssey, where you actually helped them find out that Stanley Kubrick's mat lines were visible on 2001 Space Odyssey.
00:02:10.159 --> 00:02:15.360
They were talking about how Stanley Kubrick, for those who don't know, is obsessive about his VFX.
00:02:15.360 --> 00:02:21.280
He did not like that there were generations on film and he used a really, really in-depth process.
00:02:21.280 --> 00:02:26.400
But you, my friend, were able to discover the mat lines in 2001 of Space Odyssey.
00:02:26.400 --> 00:02:28.479
Can you go into a little bit about how you did that?
00:02:28.719 --> 00:02:30.080
That was a funny side quest.
00:02:30.080 --> 00:02:36.319
I had a Blu-ray rip of the 4K mastering, I think the 2018 mastering of the film.
00:02:36.319 --> 00:02:45.360
And I decided to throw it onto DaVinci Resolve and start contemplating the idea of remastering the film.
00:02:45.360 --> 00:03:03.039
So you naturally you see with digital tools nowadays that there's little things that can be improved, let's say the white point of certain objects in space during the animations of the let's say the moon landing, for example, you have little things that could simply not be achieved with analog means.
00:03:03.039 --> 00:03:04.639
So I decided why not?
00:03:04.639 --> 00:03:08.479
Let's explore how this works in the digital domain.
00:03:08.479 --> 00:03:14.000
And soon I realized that that Ultra HD master was absolutely superb.
00:03:14.000 --> 00:03:17.759
There was so much information in that in that master, in that file.
00:03:17.759 --> 00:03:24.240
Even though it's uh I would say I think like a 50 megabits per second 10-bit file in HDR.
00:03:24.240 --> 00:03:27.680
So it's Rec 2020, 1000 nits, PQ.
00:03:27.680 --> 00:03:28.400
It's fabulous.
00:03:31.039 --> 00:03:34.560
It's scene referred, but it's sort of display referred, scene referred.
00:03:34.560 --> 00:03:37.840
It's not really the final, it's not the original film scan.
00:03:37.840 --> 00:03:42.719
It's intended to be viewed on a display, but still, there is a significant amount of range and data in there.
00:03:42.879 --> 00:03:48.719
Yeah, not only that, and the way that it was transformed and transferred and mastered into into HDR.
00:03:48.719 --> 00:03:50.560
It uh it's it's fair.
00:03:50.560 --> 00:03:58.639
You can still feel the film nature of it, the emulsion, the the little weaving, the shadows are not absolutely stable, the little noise in it.
00:03:58.639 --> 00:04:02.800
Everything was captured in a really, really fair way, let's say it.
00:04:02.800 --> 00:04:15.599
Naturally, I started I decided to do something that I did a while ago with a copy of Dune, the latest Gilles Villeneuve Dune, that I know that that film had a print film pass, even though it was shot on digital.
00:04:15.599 --> 00:04:22.560
So there was a film pass, and while I was watching the film, I realized that there were little specks, little simple little imperfections here and there.
00:04:22.560 --> 00:04:27.519
So I decided to just boost the exposure, and you know, fair enough, that it was there.
00:04:27.519 --> 00:04:31.759
You know you can see that during the film scan there were like little imperfections that got captured.
00:04:31.759 --> 00:04:38.160
So in doing that in 2001 in Space Odyssey, I started seeing the matlines very, very clearly.
00:04:38.160 --> 00:04:49.920
So then I compiled all of the shots that have the matlines visible in them, and decided to make a small, a short video with some text explaining more or less what you should be looking at with the boosted exposure.
00:04:49.920 --> 00:04:58.240
And it got got picked up by Ren on Reddit, and it ended up having my four minutes of fame in a Corridor Crew VFX React video.
00:04:58.480 --> 00:05:00.560
Aw, Reddit, man, it's the best.
00:05:00.560 --> 00:05:06.079
I mean, I can't say enough positive things about Reddit, the things where you learn things that you never thought that you would imagine.
00:05:06.079 --> 00:05:07.199
So congrats on that.
00:05:07.199 --> 00:05:09.680
Corridor crew is a really it's one of my favorite channels.
00:05:09.680 --> 00:05:16.399
I learn things every time I watch them, whether it's about VFX or it's on VFX on movies that I've seen and watched for years.
00:05:16.399 --> 00:05:20.160
So can you talk about some of these for those that haven't seen that video?
00:05:20.160 --> 00:05:22.079
Can you talk about some of these mat lines?
00:05:22.079 --> 00:05:36.639
Because that to me was what was so interesting was that not only the fact that they were hidden so well, but they almost look like first gen rotoscoping, but they're not rotoscoped on a computer, they're essentially drawn and expanded by humans, essentially.
00:05:36.879 --> 00:05:39.040
I know the basics of the process.
00:05:39.040 --> 00:05:54.959
I didn't ever went to really deep into it, but I know that, for example, when you had, let's say, the sequence where one of the astronauts gets out of the pod and goes down to the spaceship, those two models are in the same frame together, and the background is just stars moving slowly.
00:05:54.959 --> 00:06:04.720
In situations like those, Kubrick most likely shot one of the models first and then shot the other one later on the undeveloped film.
00:06:04.720 --> 00:06:11.279
So it was a first-gen pass, and you just burn data into it, burn information into it.
00:06:11.279 --> 00:06:17.839
And in the case of the mat lines, I don't know exactly how they were achieved, but they are very, very manual.
00:06:17.839 --> 00:06:22.480
You can tell that they happened for a film that was shot on 24 frames per second.
00:06:22.480 --> 00:06:25.279
They happen around eight frames each.
00:06:25.279 --> 00:06:28.160
You see, like the stepping at eight frames.
00:06:28.160 --> 00:06:34.639
You see, for example, an object moving across a screen, and you have eight frames, eight frames, eight frames, eight frames.
00:06:34.639 --> 00:06:38.720
It's it's really fascinating to see that the tactile human aspect of it.
00:06:38.720 --> 00:06:48.720
And my favorite is individual stars getting painted out manually, and you see the brush strokes on the individual stars as a scale model passes in front of them.
00:06:48.720 --> 00:06:50.160
There's a lot there.
00:06:50.160 --> 00:06:56.560
I mean, there's um and it's surprising that all of that could be captured by the HDR master.
00:06:56.560 --> 00:07:07.120
It would most likely be impossible to see on the original 65mm screenings, because I mean, independently of dynamic range, you're looking at something at between 50 to 880 nit.
00:07:07.120 --> 00:07:08.399
It's really dark.
00:07:08.399 --> 00:07:11.439
So, I mean, that's a thing that digital goddess.
00:07:11.439 --> 00:07:18.240
It's the the forensics of it and the fact that you can really reveal parts of cinema history at your fingertips.
00:07:18.240 --> 00:07:19.680
That information was always there.
00:07:19.680 --> 00:07:25.759
And I was lucky to find it and immediately spent the whole afternoon making a video too so I could share it with everyone.
00:07:26.000 --> 00:07:27.920
That's what was so fascinating to me.
00:07:27.920 --> 00:07:38.240
Not that they're so much visible, but to the fact that using the magic of filmmaking, that there are things that we just don't notice.
00:07:38.240 --> 00:07:44.319
And as soon as I started to they were pointed out to me, I did start to notice, oh, there's no stars there.
00:07:44.319 --> 00:07:46.000
Oh, there's no stars there.
00:07:46.000 --> 00:07:50.720
And because our eye is being directed in a specific direction, it just worked.
00:07:50.720 --> 00:07:56.800
It reminds me of tricks that James Cameron did, who I'm sure was inspired by this film, because that's what film is.
00:07:56.800 --> 00:07:58.480
It's magic, it's misdirection.
00:07:58.480 --> 00:08:11.439
And so for me, it was just a peek at film history and how visual effects were done before we had computers, before we had CGI and digital technology, and it still works perfectly.
00:08:11.439 --> 00:08:13.120
That's what I love about it.
00:08:13.120 --> 00:08:14.160
It is so amazing.
00:08:14.399 --> 00:08:24.879
Yeah, like if you talk about suspension of disbelief in cinema and you take it to probably to its most extreme condition, which is believing that you're in outer space.
00:08:24.879 --> 00:08:30.160
And if you think that the film came out in 1968, it's as good as it gets.
00:08:30.160 --> 00:08:38.879
It's and even if you start scraping away and doing all the CSI and DaVinci Resolve with exposure cranked and on six tops up, doesn't matter.
00:08:38.879 --> 00:08:41.519
I love that film for a long time already.
00:08:41.519 --> 00:08:46.799
And there were some comments going around, like, uh, oh, why did you do that?
00:08:46.799 --> 00:08:51.519
Uh, this is an insult to the artwork, this is an insult to Kubrick himself and stuff like that.
00:08:51.519 --> 00:09:04.639
And I'm like, no, I actually I revere the men, the men in the team, Douglas Turnbull and all of his TV, even more, because being able to see this little piece of history that was always there, and now he can see it because computers.
00:09:37.770 --> 00:09:45.050
You know, I originally had that same thought too was man, if he found out about this, he would be livid.
00:09:45.050 --> 00:09:56.010
But it's more of just a curiosity in the same way that I love behind the scenes documentaries, and I love watching how did they make that and behind the VFX.
00:09:56.010 --> 00:09:58.250
It's just an innate curiosity.
00:09:58.250 --> 00:10:02.250
And so for me, it doesn't actually ruin the movie at all for me.
00:10:02.250 --> 00:10:05.530
In fact, it's just I can't believe they did all that.
00:10:05.530 --> 00:10:10.810
And this was the level of artistry that they were able to achieve in the 60s.
00:10:10.810 --> 00:10:13.370
For me, it's just awe-inspiring.
00:10:13.370 --> 00:10:24.730
And it's not that it makes anything we do now easier, we do things just differently now, but it's just a peek at what they were doing and the level of his genius, honestly.
00:10:24.730 --> 00:10:35.530
Like for those that don't know, he had an obsession with not when you were doing analog and opticals back then, it was really difficult to do this without getting another generation on film.
00:10:35.530 --> 00:10:45.930
And Kubrick hated that, and so they were able to avoid that by essentially printing each RGB layer onto a different black and white layer, and he was just meticulous about that.
00:10:45.930 --> 00:10:55.050
And when you look at that combined with his craftsmanship, there was so much time and care placed with these effects for us not to be able to see this until now.
00:10:55.050 --> 00:11:05.290
And so I don't really think it's a bad thing, I just think that it's absolutely incredible that I mean no one saw this until now, other than people that worked on the film.
00:11:05.290 --> 00:11:12.250
And it's also a testament to the technology that we have now and the film scanners and the ability to see into film.
00:11:12.250 --> 00:11:25.850
Like we can talk about how great digital is, but it's just it shows how amazing film is and how awesome celluloid is that we are able to essentially go back in time and get better quality scans of this.
00:11:25.850 --> 00:11:32.730
It's almost like 150, 60 years ago of this amazing film and get more information of how this was done.
00:11:32.730 --> 00:11:34.890
Like it's just awe-inspiring to me.
00:11:34.890 --> 00:11:36.090
That's the best word for it.
00:11:36.330 --> 00:11:46.890
There's another video that I'd recently released, not recently, like two weeks ago or so, where it's another forensic analysis of a sequence in the in 2001 Space Odyssey.
00:11:46.890 --> 00:12:04.890
You have one of the pods rotating and there's a window, and it's funny that the window is deliberately slanted back, so that if you shoot the pod from the front, the window will reflect whatever is dark behind the camera, or better yet, above the camera in the studio.
00:12:04.890 --> 00:12:20.890
But there's a certain rotation, there's two rotations, one of them has better informed data than the other one, where the pod, as it rotates, you see some specks of light, you see a little some mushy information at the bottom, and then you see a streak of light pass really fast from one side to the other.
00:12:20.890 --> 00:12:22.490
And that got me thinking.
00:12:22.490 --> 00:12:24.810
I mean, this is obviously the set.
00:12:24.810 --> 00:12:26.170
This is a sound stage.
00:12:26.170 --> 00:12:27.050
It has to be.
00:12:27.050 --> 00:12:29.930
You see a little speck of light, you see something mushy at the bottom.
00:12:29.930 --> 00:12:31.210
Okay, let's look into it.
00:12:31.210 --> 00:12:40.650
I took that shot, I just rescaled it, made it so that the rotation is corrected, so you just see like an ellipse rotating in front of you.
00:12:40.650 --> 00:12:42.410
It stabilized, it's denoised.
00:12:42.410 --> 00:12:46.490
The 24 frames per second are interpolated to 60.
00:12:46.490 --> 00:12:49.129
Then I decided to export a vertical video.
00:12:49.129 --> 00:12:59.930
So then they got me thinking, okay, there's distortion because of the window, the window is tilted backwards, and there's a little bit of uh, is it concave, let's say, because a streak of light passes by it in a weird way.
00:13:00.170 --> 00:13:00.490
Yeah.
00:13:00.650 --> 00:13:04.570
I ran it through an app that does a slit scan photography.
00:13:04.650 --> 00:13:07.770
So I was able to extract-appropriate for the film.
00:13:08.330 --> 00:13:08.890
Exactly.
00:13:08.890 --> 00:13:20.010
So I decided to extract the single line in the middle of that elliptical window frame as it passes and as it reflects whatever was behind the camera or around the camera.
00:13:20.010 --> 00:13:21.050
And I have an image.
00:13:21.050 --> 00:13:22.410
You can see the soundstage.
00:13:23.290 --> 00:13:23.770
Really?
00:13:24.010 --> 00:13:24.330
Yes.
00:13:24.330 --> 00:13:32.250
You see what it's it's ghostly in the sense that you only have a handful of pixels to work with, even though it's an ultra HD scan.
00:13:32.250 --> 00:13:43.370
But I think you can see you see obviously a light projector, you see a projector hanging in darkness, you see a wall that looks like some sort of sound insulation, and you see a wooden structure with stairs.
00:13:43.370 --> 00:13:48.810
You see two flights of stairs move going up to a place where a light projector would be, and you see the floor.
00:13:48.810 --> 00:13:56.330
I made another video, you can look it up on on YouTube now, and where dissect everything and all the exports from the slit scan technique are all there.
00:13:56.330 --> 00:13:57.930
You can just download the PNGs.
00:13:57.930 --> 00:13:58.250
Yeah.
00:13:58.490 --> 00:13:59.610
That's fascinating.
00:13:59.610 --> 00:14:06.410
It's almost like a reverse HDR image, or excuse me, HDRI image that you essentially reverse engineered out of the film.
00:14:06.410 --> 00:14:15.690
Because that's essentially sort of what we're doing now when you have a CGI object, is you try to get an HDRI image to properly light it, and you sort of reverse engineered that.
00:14:15.690 --> 00:14:17.290
And to me, that's fascinating.
00:14:17.290 --> 00:14:29.530
But in its context, we see it and our brain goes, oh, that's just a star or a meteor or something flying past the ship, because we just have the context of where it is and what it feels like.
00:14:29.530 --> 00:14:31.290
And so we don't think twice about it.
00:14:31.290 --> 00:14:32.410
That's what's so cool.
00:14:32.410 --> 00:14:44.650
So, Paulo, I know that you do digital camera calibrations and film emulations, and you actually do a lot of stuff in an area that I hear a lot of people are interested in, myself included, which is working with still photography.
00:14:44.650 --> 00:14:47.690
Can you tell us about your work in that area as well?
00:14:47.690 --> 00:14:55.129
Because a lot of people, including some of our previous guests, have talked about how much they want and how much they do use Resolve for still photography work.
00:14:55.129 --> 00:15:01.050
And you are not only calibrating digital cameras, but you're doing film emulations for Lightroom and still photography.
00:15:01.050 --> 00:15:02.410
So I'd like to hear more about that.
00:15:02.650 --> 00:15:10.410
Well, uh, when I started working professionally as a photographer, there was uh I have a need the need to capture color accurately.
00:15:10.410 --> 00:15:17.290
Either I was doing events or I was doing later on product photography and artwork reproduction, specially.
00:15:17.290 --> 00:15:28.970
You need to have, let's say, a specific artwork with very vibrant colors properly captured, properly processed, and printed on paper as faithfully as possible.
00:15:28.970 --> 00:15:32.410
And so that got me into calibration.
00:15:32.410 --> 00:15:40.730
And so with the the humble, here it is behind me, with the humble color checker passport that I bought back in 2010.
00:15:40.730 --> 00:15:43.129
It tells me that I should replace it every two years.
00:15:43.129 --> 00:15:46.170
There's no mod uh there's no change in the color, by the way.
00:15:46.170 --> 00:15:54.890
I started going deep into what calibration could be and what it how it could improve the way that I work.
00:15:54.890 --> 00:16:12.730
Started off simple with the free software that comes with a color checker, then there's a version for Adobe from Adobe, and then I recently evolved with for uh with the Luma Reaver Profile Designer, which is the the most complete, I think for now, for the time being, is the most complete software that you can use for calibration.
00:16:12.730 --> 00:16:25.770
And so having a handful of professional cameras throughout the years, I always worked, created profiles for them, and I could have faithful reproduction, color reproduction with specific light sources.
00:16:25.770 --> 00:16:30.490
And name those two, the two main light sources will be Daylight and Tungsten.
00:16:30.490 --> 00:16:33.210
So I actually have these two back here.
00:16:33.210 --> 00:16:36.410
When was the last time that you saw a lamp with a filament inside?
00:16:36.730 --> 00:16:38.170
It's been a little while, I'll admit.
00:16:38.170 --> 00:16:40.010
They've all been replaced by LEDs.
00:16:40.250 --> 00:16:43.610
And in Germany, Germany has a lot of fluorescent tubes.
00:16:43.610 --> 00:16:45.850
You see so many fluorescent tubes everywhere.
00:16:45.850 --> 00:16:47.290
You may think, okay, Germany.
00:16:47.290 --> 00:16:47.530
No.
00:16:47.530 --> 00:16:56.410
In some regards, Germany is really, really way behind when it comes to technology at the core level in terms of internet speed and lighting is one of them.
00:16:56.410 --> 00:16:58.490
So it was really easy to buy these.
00:16:58.490 --> 00:17:01.210
And these are actually general electric, I think.
00:17:01.210 --> 00:17:02.570
These are really good light bulbs.
00:17:02.570 --> 00:17:07.129
So this is basically the perfect light bulb, perfect color reproduction, emits a lot of heat.
00:17:07.129 --> 00:17:10.970
You can't you shouldn't use it for artwork in a gallery, for example.
00:17:10.970 --> 00:17:14.329
But I decided to use this to calibrate for tungsten.
00:17:14.329 --> 00:17:16.089
And then I bought this LED.
00:17:16.089 --> 00:17:21.289
This is from a Chinese brand called UG LED, and it just basically measures perfectly.
00:17:21.289 --> 00:17:23.930
It emits a D50 color temperature CCT.
00:17:23.930 --> 00:17:39.849
And so with these two light bulbs, I thought that I could create a series of products that would give photographers the opportunity to have something dependable, a dependable starting point for their edits inside of Lightroom or Photoshop using the Adobe Camera Raw plugin.
00:17:39.849 --> 00:17:44.569
That requires having cameras in studio, illuminated with light bulbs.
00:17:44.569 --> 00:17:48.169
I have more of these, not only two, illuminating a series of color charts.
00:17:48.169 --> 00:18:08.649
I have a lot of color charts here next to me, and then creating a product that tells you that gives you the confidence that any variant that could be present in the scene will just be a consequence of, you know, like a stray light, something with a white point that's slightly different, an LED that's not quite as good as your reference color.
00:18:08.649 --> 00:18:11.210
You have that ease of mind of a good starting point.
00:18:11.210 --> 00:18:13.129
And so that's how calibration started.
00:18:13.129 --> 00:18:17.289
And uh the I started off with Adobe Camera Raw, the plugin.
00:18:17.289 --> 00:18:19.210
The implementation is pretty good.
00:18:19.210 --> 00:18:24.569
I would even say that it's better than capture one, and naturally also started doing that with video.
00:18:24.569 --> 00:18:33.689
DaVinci Resolve has a little tool, color checker, not a color checker, color chart tool in the color panel that allows you to match colors.
00:18:33.689 --> 00:18:34.809
It's not great.
00:18:34.809 --> 00:18:39.849
It feels like a feature that was kind of forgotten, but every so often it works.
00:18:39.849 --> 00:18:42.250
But it is an approach, it's a possibility.
00:18:42.250 --> 00:18:42.730
Yeah.
00:18:43.049 --> 00:18:48.889
So what are you doing to essentially for some of your like color negatives that you're doing?
00:18:48.889 --> 00:18:54.889
Because I know that you're taking a lot of film emulations and essentially creating them for still photography.
00:18:54.889 --> 00:18:58.169
So can you talk through us your process a little bit about that?
00:18:58.169 --> 00:19:04.329
Because I know you have to negate both the digital camera and then you have to emulate the film, which is super interesting.
00:19:04.649 --> 00:19:12.089
The way this all started was when I have the let's say the interest in buying a film roll.
00:19:12.089 --> 00:19:13.609
It's photograph on analog.
00:19:13.609 --> 00:19:20.250
I bought many years ago, I bought a EOS 50E for like 25 euros.
00:19:20.250 --> 00:19:22.409
That's like 30 something dollars.
00:19:22.409 --> 00:19:23.769
So cheap that camera.
00:19:23.769 --> 00:19:26.730
I bought it so cheap, and it's really good and it's really good condition.
00:19:26.730 --> 00:19:27.849
Let's do some film.
00:19:27.849 --> 00:19:28.970
Uh it's been a while.
00:19:28.970 --> 00:19:35.210
I've I've shot film back in the day, I've developed film manually, I've printed color and black and white.
00:19:35.210 --> 00:19:45.369
And so that got me thinking, what if I can bring all of the knowledge that I that I developed in camera calibration into film negative, color film negative in this case.
00:19:45.369 --> 00:19:55.129
So I started doing research, I started reading about it, and then my main concern was is there a scientifically accurate way of reproducing the color negative?
00:19:55.129 --> 00:20:01.049
Because if you look at a color negative, you don't see an image, it's not an image that is ready to be seen.
00:20:01.049 --> 00:20:03.849
It has to be either scanned or printed.
00:20:03.849 --> 00:20:09.450
So with that in mind, you will always be capturing what the paper does.
00:20:09.450 --> 00:20:14.250
So it's kind of like with cinema film with the print film or the scanner.
00:20:14.250 --> 00:20:22.490
So is there a way to take a step back and interpret the color negative in the in its truest form?
00:20:22.490 --> 00:20:30.970
So what I did was take my digital camera, my workhorse, the Sony a7 IV, the camera that does it all, including overheating.
00:20:30.970 --> 00:20:36.169
And so I decided to calibrate that camera to the nth dimension.
00:20:36.169 --> 00:20:43.289
So I decided to create a very special profile for that camera with I think 2000 and something color patches.
00:20:43.289 --> 00:20:49.129
It's basically the combination of all of my color charts illuminated with a D50 LED.
00:20:49.129 --> 00:20:56.009
So after have I created that profile, and that allowed me to turn my digital camera into a measurement device.
00:20:56.009 --> 00:21:03.929
So I was no longer aiming for something kind of generic that kind of looks good, but it's there's not a lot of tension between colors.
00:21:03.929 --> 00:21:05.129
It kind of just works.
00:21:05.129 --> 00:21:08.490
No, this was like super high quality, super rigorous.
00:21:08.490 --> 00:21:16.169
It works in the real world, but it's not like completely agreeable sometimes with real world colors, but it's measures really, really well.
00:21:16.169 --> 00:21:34.329
So I decided to, okay, I'm gonna have my digital camera photograph the negative, illuminate it again with this light bulb, a full spectrum light bulb, and rely on the capability of that profile to do all of the color separation that I read that I need in post-production.
00:21:34.329 --> 00:21:38.009
Something that a dedicated film scanner would do by other means.
00:21:38.009 --> 00:21:53.289
And in this case, we're talking about the fact that the film, the color color negative, compresses the red, green, and blue channels, let's call them channels, in these narrow band, very narrow band areas of the visible spectrum.
00:21:53.289 --> 00:21:55.289
So it's really compressed.
00:21:55.289 --> 00:22:04.490
So the way that color printing paper and film scanners work is that the backlight itself aligns with those peaks.
00:22:04.490 --> 00:22:14.329
So that works if you're using a digital scanner, a dedicated film digital scanner where the sensor is calibrated and tuned for that narrow band backlight.
00:22:14.329 --> 00:22:19.769
A backlight that aligns with those peaks is, for example, an iPhone screen.
00:22:19.769 --> 00:22:22.329
It aligns almost exactly with those peaks.
00:22:22.329 --> 00:22:32.329
So I decided to set that aside because if I use a narrow band light source to illuminate my negative with a digital camera that expects to see.
00:22:32.329 --> 00:22:38.409
Continuous transitions between colors, the camera would incur into something that is really nasty.
00:22:38.409 --> 00:22:51.369
That I mean most of the audience will know, which is metameric failure, stands for metameric failure, which is, for example, if this LED that's illuminating my face was bad quality, I would appear green to my blackmagic camera.
00:22:51.369 --> 00:22:54.089
So it's tense, it's a tense process.
00:22:54.089 --> 00:23:05.289
So I decided to rely totally on the value of having a good profile and a really good light source to illuminate my color negative.
00:00:00.080 --> 00:00:02.560
00:00:02.560 --> 00:00:07.759
00:00:07.759 --> 00:00:12.480
00:00:12.480 --> 00:00:25.600
00:00:26.079 --> 00:00:32.719
00:00:32.719 --> 00:00:43.039
00:00:43.039 --> 00:00:49.280
00:00:49.280 --> 00:00:55.600
00:00:55.600 --> 00:01:00.880
00:01:00.880 --> 00:01:05.040
00:01:05.040 --> 00:01:11.920
00:01:11.920 --> 00:01:17.359
00:01:17.359 --> 00:01:20.719
00:01:20.719 --> 00:01:24.959
00:01:24.959 --> 00:01:30.719
00:01:30.719 --> 00:01:32.239
00:01:32.480 --> 00:01:33.760
00:01:33.760 --> 00:01:35.599
00:01:35.920 --> 00:01:39.280
00:01:39.280 --> 00:01:45.840
00:01:45.840 --> 00:01:53.599
00:01:53.599 --> 00:02:10.159
00:02:10.159 --> 00:02:15.360
00:02:15.360 --> 00:02:21.280
00:02:21.280 --> 00:02:26.400
00:02:26.400 --> 00:02:28.479
00:02:28.719 --> 00:02:30.080
00:02:30.080 --> 00:02:36.319
00:02:36.319 --> 00:02:45.360
00:02:45.360 --> 00:03:03.039
00:03:03.039 --> 00:03:04.639
00:03:04.639 --> 00:03:08.479
00:03:08.479 --> 00:03:14.000
00:03:14.000 --> 00:03:17.759
00:03:17.759 --> 00:03:24.240
00:03:24.240 --> 00:03:27.680
00:03:27.680 --> 00:03:28.400
00:03:31.039 --> 00:03:34.560
00:03:34.560 --> 00:03:37.840
00:03:37.840 --> 00:03:42.719
00:03:42.879 --> 00:03:48.719
00:03:48.719 --> 00:03:50.560
00:03:50.560 --> 00:03:58.639
00:03:58.639 --> 00:04:02.800
00:04:02.800 --> 00:04:15.599
00:04:15.599 --> 00:04:22.560
00:04:22.560 --> 00:04:27.519
00:04:27.519 --> 00:04:31.759
00:04:31.759 --> 00:04:38.160
00:04:38.160 --> 00:04:49.920
00:04:49.920 --> 00:04:58.240
00:04:58.480 --> 00:05:00.560
00:05:00.560 --> 00:05:06.079
00:05:06.079 --> 00:05:07.199
00:05:07.199 --> 00:05:09.680
00:05:09.680 --> 00:05:16.399
00:05:16.399 --> 00:05:20.160
00:05:20.160 --> 00:05:22.079
00:05:22.079 --> 00:05:36.639
00:05:36.879 --> 00:05:39.040
00:05:39.040 --> 00:05:54.959
00:05:54.959 --> 00:06:04.720
00:06:04.720 --> 00:06:11.279
00:06:11.279 --> 00:06:17.839
00:06:17.839 --> 00:06:22.480
00:06:22.480 --> 00:06:25.279
00:06:25.279 --> 00:06:28.160
00:06:28.160 --> 00:06:34.639
00:06:34.639 --> 00:06:38.720
00:06:38.720 --> 00:06:48.720
00:06:48.720 --> 00:06:50.160
00:06:50.160 --> 00:06:56.560
00:06:56.560 --> 00:07:07.120
00:07:07.120 --> 00:07:08.399
00:07:08.399 --> 00:07:11.439
00:07:11.439 --> 00:07:18.240
00:07:18.240 --> 00:07:19.680
00:07:19.680 --> 00:07:25.759
00:07:26.000 --> 00:07:27.920
00:07:27.920 --> 00:07:38.240
00:07:38.240 --> 00:07:44.319
00:07:44.319 --> 00:07:46.000
00:07:46.000 --> 00:07:50.720
00:07:50.720 --> 00:07:56.800
00:07:56.800 --> 00:07:58.480
00:07:58.480 --> 00:08:11.439
00:08:11.439 --> 00:08:13.120
00:08:13.120 --> 00:08:14.160
00:08:14.399 --> 00:08:24.879
00:08:24.879 --> 00:08:30.160
00:08:30.160 --> 00:08:38.879
00:08:38.879 --> 00:08:41.519
00:08:41.519 --> 00:08:46.799
00:08:46.799 --> 00:08:51.519
00:08:51.519 --> 00:09:04.639
00:09:37.770 --> 00:09:45.050
00:09:45.050 --> 00:09:56.010
00:09:56.010 --> 00:09:58.250
00:09:58.250 --> 00:10:02.250
00:10:02.250 --> 00:10:05.530
00:10:05.530 --> 00:10:10.810
00:10:10.810 --> 00:10:13.370
00:10:13.370 --> 00:10:24.730
00:10:24.730 --> 00:10:35.530
00:10:35.530 --> 00:10:45.930
00:10:45.930 --> 00:10:55.050
00:10:55.050 --> 00:11:05.290
00:11:05.290 --> 00:11:12.250
00:11:12.250 --> 00:11:25.850
00:11:25.850 --> 00:11:32.730
00:11:32.730 --> 00:11:34.890
00:11:34.890 --> 00:11:36.090
00:11:36.330 --> 00:11:46.890
00:11:46.890 --> 00:12:04.890
00:12:04.890 --> 00:12:20.890
00:12:20.890 --> 00:12:22.490
00:12:22.490 --> 00:12:24.810
00:12:24.810 --> 00:12:26.170
00:12:26.170 --> 00:12:27.050
00:12:27.050 --> 00:12:29.930
00:12:29.930 --> 00:12:31.210
00:12:31.210 --> 00:12:40.650
00:12:40.650 --> 00:12:42.410
00:12:42.410 --> 00:12:46.490
00:12:46.490 --> 00:12:49.129
00:12:49.129 --> 00:12:59.930
00:13:00.170 --> 00:13:00.490
00:13:00.650 --> 00:13:04.570
00:13:04.650 --> 00:13:07.770
00:13:08.330 --> 00:13:08.890
00:13:08.890 --> 00:13:20.010
00:13:20.010 --> 00:13:21.050
00:13:21.050 --> 00:13:22.410
00:13:23.290 --> 00:13:23.770
00:13:24.010 --> 00:13:24.330
00:13:24.330 --> 00:13:32.250
00:13:32.250 --> 00:13:43.370
00:13:43.370 --> 00:13:48.810
00:13:48.810 --> 00:13:56.330
00:13:56.330 --> 00:13:57.930
00:13:57.930 --> 00:13:58.250
00:13:58.490 --> 00:13:59.610
00:13:59.610 --> 00:14:06.410
00:14:06.410 --> 00:14:15.690
00:14:15.690 --> 00:14:17.290
00:14:17.290 --> 00:14:29.530
00:14:29.530 --> 00:14:31.290
00:14:31.290 --> 00:14:32.410
00:14:32.410 --> 00:14:44.650
00:14:44.650 --> 00:14:47.690
00:14:47.690 --> 00:14:55.129
00:14:55.129 --> 00:15:01.050
00:15:01.050 --> 00:15:02.410
00:15:02.650 --> 00:15:10.410
00:15:10.410 --> 00:15:17.290
00:15:17.290 --> 00:15:28.970
00:15:28.970 --> 00:15:32.410
00:15:32.410 --> 00:15:40.730
00:15:40.730 --> 00:15:43.129
00:15:43.129 --> 00:15:46.170
00:15:46.170 --> 00:15:54.890
00:15:54.890 --> 00:16:12.730
00:16:12.730 --> 00:16:25.770
00:16:25.770 --> 00:16:30.490
00:16:30.490 --> 00:16:33.210
00:16:33.210 --> 00:16:36.410
00:16:36.730 --> 00:16:38.170
00:16:38.170 --> 00:16:40.010
00:16:40.250 --> 00:16:43.610
00:16:43.610 --> 00:16:45.850
00:16:45.850 --> 00:16:47.290
00:16:47.290 --> 00:16:47.530
00:16:47.530 --> 00:16:56.410
00:16:56.410 --> 00:16:58.490
00:16:58.490 --> 00:17:01.210
00:17:01.210 --> 00:17:02.570
00:17:02.570 --> 00:17:07.129
00:17:07.129 --> 00:17:10.970
00:17:10.970 --> 00:17:14.329
00:17:14.329 --> 00:17:16.089
00:17:16.089 --> 00:17:21.289
00:17:21.289 --> 00:17:23.930
00:17:23.930 --> 00:17:39.849
00:17:39.849 --> 00:17:44.569
00:17:44.569 --> 00:17:48.169
00:17:48.169 --> 00:18:08.649
00:18:08.649 --> 00:18:11.210
00:18:11.210 --> 00:18:13.129
00:18:13.129 --> 00:18:17.289
00:18:17.289 --> 00:18:19.210
00:18:19.210 --> 00:18:24.569
00:18:24.569 --> 00:18:33.689
00:18:33.689 --> 00:18:34.809
00:18:34.809 --> 00:18:39.849
00:18:39.849 --> 00:18:42.250
00:18:42.250 --> 00:18:42.730
00:18:43.049 --> 00:18:48.889
00:18:48.889 --> 00:18:54.889
00:18:54.889 --> 00:18:58.169
00:18:58.169 --> 00:19:04.329
00:19:04.649 --> 00:19:12.089
00:19:12.089 --> 00:19:13.609
00:19:13.609 --> 00:19:20.250
00:19:20.250 --> 00:19:22.409
00:19:22.409 --> 00:19:23.769
00:19:23.769 --> 00:19:26.730
00:19:26.730 --> 00:19:27.849
00:19:27.849 --> 00:19:28.970
00:19:28.970 --> 00:19:35.210
00:19:35.210 --> 00:19:45.369
00:19:45.369 --> 00:19:55.129
00:19:55.129 --> 00:20:01.049
00:20:01.049 --> 00:20:03.849
00:20:03.849 --> 00:20:09.450
00:20:09.450 --> 00:20:14.250
00:20:14.250 --> 00:20:22.490
00:20:22.490 --> 00:20:30.970
00:20:30.970 --> 00:20:36.169
00:20:36.169 --> 00:20:43.289
00:20:43.289 --> 00:20:49.129
00:20:49.129 --> 00:20:56.009
00:20:56.009 --> 00:21:03.929
00:21:03.929 --> 00:21:05.129
00:21:05.129 --> 00:21:08.490
00:21:08.490 --> 00:21:16.169
00:21:16.169 --> 00:21:34.329
00:21:34.329 --> 00:21:38.009
00:21:38.009 --> 00:21:53.289
00:21:53.289 --> 00:21:55.289
00:21:55.289 --> 00:22:04.490
00:22:04.490 --> 00:22:14.329
00:22:14.329 --> 00:22:19.769
00:22:19.769 --> 00:22:22.329
00:22:22.329 --> 00:22:32.329
00:22:32.329 --> 00:22:38.409
00:22:38.409 --> 00:22:51.369
00:22:51.369 --> 00:22:54.089
00:22:54.089 --> 00:23:05.289