Jason Bowdach CSI: Reverse Engineering Color With A Creative Mindset

Welcome to the Color Timer podcast. I'm your host, Vincent Taylor. This is the podcast where we speak to professionals who work with color. Today, I'm speaking to Mr. Jason Bowdach Jason is a freelance colorist. He comes from a color science and workflow background. He also has a very interesting company called Pixel Tools, which is what I'm going to focus on today. So I hope you enjoy it. Let's go. Take your seats because the hourglass is about to turn.

We are entering the world of the micro podcast. Explore the craft, creativity, and science of professionals who use color to tell stories. Welcome to the Color Timer with Vincent Taylor. Jason Bowdach, welcome. Welcome to the Color Timer podcast. Thank you so much for having me on. I'm really glad that we were able to get on each other's podcast. It's so much fun that there is a lot more color grading information and stuff for people to listen to. So thank you so much for having me on.

It's true. I did. I did kind of consider through my little brain this morning. Is this a version of nepotism? Because I'm on your podcast, you're on my podcast. And, you know, is it kind of, you know, I don't think so because, you know, I don't think so. I've I've I just started my first podcast. I have one season down. So I consider myself very much a beginner in creating podcasts. But one of the first things that I was told was work with other podcasters to spread the word about that podcast.

And that's how you discover new podcasts. I enjoy podcasts in a variety of different subjects. And I'm constantly looking for new ones. And so the fact that we're on each other's podcasts, we're spreading the word to people about that. Enjoy color grading that, by the way, if you're interested in this topic, here's something else you can watch. It's yeah. Yeah, it's it seems to.

And you think back to I'm going to say the old days, but you think back to years ago and it was just so hard to get any information. You know, so it's it's kind of nice. I feel I feel so you kids are spoiled. You kids are spoiled by all these, you know. But listen, I'm going to jump straight in. I do have my my little note here to remember to start the timer. So I've I've I'm kind of already so so let let's go. Let's let's I'm going to turn it over officially.

Let's let's I'm going to turn it over officially.

Yeah, flip that timer, man. And in a way and a way and your time starts now. Now, you are a colorist, you're a freelance colorist, but I am not going to talk to you about that. I'm not going to speak to you about that because I'm going to talk to you about your company, Pixel Tools, and about what? What is it? What is this company? Tell me all about it. Well, I think I probably should start off saying that I'm a huge nerd. I love color grading. I love filmmaking.

Everything about it is something that, as my wife says, "I wish I loved my job as much as you do." So the fact that I spend my days working and making pretty pictures and telling stories was something that I felt really lucky to do. But as a lot of artists, we have a set of tools that we're given or were taught to work with.

And as I started to work in color grading, I found myself limited is not the right word, but I found myself wanting a couple of additional tools, especially working with Resolve. And as I expanded my knowledge of color grading and especially color science, I found myself more interested and I found myself hitting walls in terms of what traditional color grading tools could do built into Resolve. And I decided to start to explore that area.

And that's where I found that there's a huge variety of tools that we can build for Resolve, whether it's plug-ins or power grades, which are presets, essentially, that weren't being done. And for me, that was a complete missing opportunity. And I sort of waited. As a beginning colorist when I was considering starting Pixel Tools, I didn't really consider myself the right person to do that. I can I thought there is somebody that has a lot more experience that should be building these, not mine.

So I waited about two years. Right. You always wait for somebody that knows better to build stuff. I'm still I'm still waiting and I'm glad people like you are doing it. But no, no, I hear you. So after about two years of waiting for for somebody that's a senior to step in and to build these tools, I decided, fine, I'll start building power grades for colors to make things easier. We don't all like to start from scratch. And so I started doing that at about four, wow, about four years ago.

And I started offering these power grades to other colorists. Now, I'm sure a lot of people build power grades, but these are specifically power grades designed for other people to use. They're labeled, they're documented. They're created in a way that other people can use them without having to know every single inch of it. And that's a little bit different than the current tools that are around. Maybe you work at a facility and someone shows you, oh, by the way, we created this internal tool.

You touch here, here and here. Don't touch anywhere else. And I didn't really like that. If you didn't work at a facility, you didn't have the ability to get into a facility, you were stuck. And I wanted to sort of open that up. And that's really what Pixel Tools got started with.

Which I mean, you're talking about specific tools, which tool did you start with? What was the one thing? So you know what I really want? Really, it got started in a pretty pedestrian way. I loved. I liked third party LUTs, but I didn't like the black box that they were. And so as I started to learn Resolve and as sort of a petty excuse to get better at color grading and reverse engineering, I started to reverse engineer popular third party look up tables.

And I said, well, look up tables are nice, but they're not very helpful. And when they don't work, they don't work really hard. Yeah. And I can't really fix them. It's not like a plug in or a color grade where I can go in and find where it's broken and essentially fix it and get the project out the door. If it doesn't work, the door is slammed and it's locked. And I really didn't like that.

And so at least for me in my basic stage, I figured out, well, if we can recreate these LUTs inside Resolve with curves and lift gamma gain and secondaries as much as possible using the color warper and hue curves. Well, we have a lot more control over these tools. And so really the first popular three look up tables I started reverse engineering were just client LUTs. LUTs that they brought in that were like, hey, we shot with this. Can you grade with this?

Or I bought this popular third party LUT. I want to use it on my film. And it just broke on three or four scenes. And so the ability you've encountered that, I'm sure you have. And they love that LUT. They like something that it does, or they like the way that it makes the the scene or the movie feel. And they're attached to it because of course they used offline with it.

And so for me, the first sort of entry into this pond was I need to make this into a tool that I professional colorists can use. And so the first product we created was the volume one power grades collection, which was essentially a hundred different LUTs or common corrections that I reverse engineered. That's a lot, man. It - you can ask my wife for a lot of time we were watching TV. I just had my laptop out sort of reverse engineering them.

And then I would take them into the suite and really make sure that they were to to the proper level of precision because reverse engineering LUTs, you're just doing it on charts. You're you're it's a lot of boring work. And I get it. Nobody really wants to do that. But the result of it is so, so important. The ability to have now three or four different nodes. Here's the curves. Here's the hue changes. Here's the Luma changes is a really important task.

And so for me, I just want to jump in for a sec, because you're talking about the reverse engineering and looking at charts. Is it possible to explain in layman's terms? What is it when you reverse engineer? What where do you start with that? I don't quite understand. So this continually gets more complicated as you start to reverse engineer, more complicated looks. So obviously on the top end, you have film emulation, which is very nebulous. But let's start at the very beginning.

Let's just start at I have a a LUT that's changing blue skies to teal skies. And so you might be given this and the client's not going to tell you what it is. They're going to say, I really like what this is doing to my skies. Or this is what I like, the look of this. And you are in charge of saying, OK, what exactly is this doing?

And so you might pull out a common couple of common shots, whether they're from Ari, a couple of synthetic shots, and you're going to apply this and you're going to analyze it. You're going to go look at these color charts. You're going to say, what is it doing to red? What is it doing to yellow? What is it doing to different shades of hues, essentially, and trying to analyze it?

And one you can do that's the first step, which is just trying to understand, OK, what is it touching and what is it not touching? And then you start to get in. OK, now I can start to put points on my curve. Where are we starting to build our contrast curve? You can put it on a grayscale and start to see here is literally where it's expanding. I can see it on my waveform. I can see when I apply the LUT it pulls and stretches my image. And when I disable it does that.

And then you can start to rebuild that with Resolve's own tools to the level that allow that. Right. Right. And and are you almost you know, you're rebuilding it and you're checking back to those curves and those scopes to go. There we go. It's the same kind of world. And yeah, exactly. And it's it sounds like dull work because it is. And you're just trying to make your A/B'ing it to see, OK, my work, the LUTs work. And there's times where you're like, you know what?

I can see something that the LUT is doing that I don't like. So I'm going to rebuild everything except that. Yeah, that's going to be a problem. Yeah. Yeah. And that and then I'm back to what you're saying about it breaking or something like that. You're just kind of refining that and getting all those elements, but rebuilding in such a way that it's not going to break your image and it's not going to do those nasty things. Exactly.

A lot of the problem, at least now, a lot of us are working in scene referred workflows where we're not including the camera transform in our looks. A lot of the stuff provided by our clients has that baked in there. And so we have to figure out, well, how can I apply this creative transform without my technical camera transform? And so there's a lot of. I mean, it gets more complicated, but you are reverse engineering.

So you might you might apply their their applied camera LUT on a different project, pull stills, and then you're going to have to go into your scene referred workflow and try and match that. So it can get pretty complicated, but the gist of it is you're essentially trying to A/B things with synthetic charts and real charts and at different exposure levels. Because remember, sometimes these things will react different at bright scenes

and then at dark scenes. OK. Now, I know I kind of pulled you off track a little bit because you were talking about you started off by, you know, building all these power grades, right? And and and developing. Where'd you go from there?

So power grades for me, once I discovered that Resolve had a form of color grading presets that they call power grades, which essentially allows me to build a preset of notes that I can continually use on a variety of different projects, but also document and give out to other people. My next step was, well, what can I do with the. Essentially, a tool that I want to go beyond Resolve.

So, for instance, a very popular request is I want subtractive saturation or saturation that gets darker as it gets more saturated, not brighter. Really, really common sort of filmic saturation vibe. You can do that theoretically in Resolve. It's going to require a a node sandwich kind of thing. You're going to have a bunch of different nodes. You might compound them. But for me, I was getting distracted with that.

A big. Around that time, I decided that it was time for me to decide what I wanted to do with pixel tools. There's a lot of people that were starting to create color grading tools, and I really decided to define why am I creating color grading tools and what are going to make mine different? And what I decided to define at that time is a lot of plug ins, Resolve, the built in tools are highly complicated. They have more sliders and true.

Yeah, different different adjustable features that you can throw a stick at. For me, I'm not going to create a new plug in or tool unless there's a really definable benefit to it. So for me, pixel tools outside of the power grades going into what are called DCTLs or essentially a programming language that we can use to create plug ins. I wanted to make sure that I created color grading tools that were simple and easy for professional colorists.

We are jamming through sessions, and I find that a lot of us use tools that we ignore things on. So for me, as a working professional, we're paid per hour most of the time. I wanted to simplify the workflow. So there's a lot of tools out there that makes things more complex. I wanted to simplify them. So for me, it's all about making it so I can work on the panel. So that I can work quickly so that I don't have to leave my creative mindset. I don't want to go into working.

OK, what color space do I need to go into? I go to HSB, I go to HSL, I change the gain. OK, now what was my what did my client want again? They wanted punchy colors. I don't want to switch between technical and creative mindsets. And the requirement that resolve makes you do to do some of these creative solutions to me was a big problem. Certain colorists can do that. Not everybody can. And I don't think being highly technical is a requirement of being a good colorist.

So for me at this time, it was to create highly precise color grading tools that even just the creative colorists could use. I really love that. I really love that because, you know, I'm not I'm not a technical colorist, you know, and and the things that the tools that you've developed are really useful. And they are simple to use, you know, which which is a delight.

Do you have and I don't know if you can talk about this, but do you have some other tools coming up? I do. I'm - here's where I am right now. Right now, I'm in this interesting area of I'm examining the scene and I'm looking like it's easy for me to just go, you know what? We need a new contrast. Cool. We need a new split toning tool. But it's really crucially important to me that I'm not recreating the wheel for something that is already well done in Resolve.

It needs to be something that I'm having trouble with that is causing me slowdown because. There are a lot of people now that I think are are more - What's the right way to put this? I think are just trying to make money off plugins. And it's really easy to do that. I think it's really important to remember how very good resolve is a lot of color grading operations and only go outside the box if you really need to.

A lot can be done with power grades and creating an entire DCTL tool is not always necessary. So before I go, it's I'm sorry, but I'm just because you're speaking from a place of integrity and, you know, it's not you know, you've got to make a living and you've got to make money off these things. But it's really lovely that you're doing that. It's I've always been that guy that it took me a really long time to to to sell things. Even if it's a great product.

But like my split tone tool is there for convenience. You can do everything you can with our split tone tool in the curves. But not everybody wants to do that. To answer your question, what do we have coming next? Yeah, yeah, yeah. Yeah, yeah, yeah. Yeah, I'm looking at. Filmic contrast and some of the restrictions that we have with the current curves tool in resolve.

It is an amazing tool, especially now that they've given us our Gigantor Mode back or gigantic curves where you can pop out the curves mode. But I want to make it faster for you to make nice, soft, precise curves. I want to make you so you can make a nice soft S curve with sliders. And I haven't quite figured out how I want to do that yet.

But that's really my next goal is so that you can create those really nice soft film at contrast curves without having to be so fiddly with the built in curve. That's great. And with included with that, because obviously not everybody knows what a good contrast curve looks like. I'm going to try and include some pre built filmic contrast curves based off some scanned film emulations, which I think is a really important thing. That is cool.

Yeah, I think it's really important that we always have a basis to go off of. It's really easy to just like take a tool and go, OK, I'm going to make it look really cool. But for me, I enjoy working off something that has come from something. So for me, yeah, I enjoy making a contrasty image. But for me, the interest of now, once I have more precise controls, that gives me the ability to match stuff. I have our timer's run out. You know, I'm thinking I made this bed.

I made this 15 minute sand timer bed. And I wish I had a 20 minute sand timer because so quick. I've got all these questions for you. And and and I'm my own worst enemy because I keep interrupting you and asking you other things on top of it. But but listen, just quickly before you go, I want to get your take on AI.

About, you know, everyone's talking about AI this, AI that and how you think it is going to how it is impacting us as colorist or how you think it's going to impact us as colorists. Absolutely, I certainly have been helping that timer rambling on and on. But I have a pretty strong opinion about AI, but it's not one that I think is very usual. I actually feel that AI is going to help us quite a bit in taking away a lot of those. Boring tasks that you don't really want to do.

So for instance, the task that I talked about matching, we may be able to have AI help us do a match to a grayscale in the future or labeling clips or renaming stuff. Instead of being afraid of losing your job or having AI come in to replace you, reevaluate your job and say, what about my job do I hate doing or do I put on a podcast and listen to and just don't really pay attention to? And with that extra time, what could I do instead that I want to do or what could I learn to do instead?

Because the fact is, a lot of our job is just busy work, prepping, transcoding, renaming, organizing, labeling, clip coloring. This stuff is perfect for AI and we can do so much more precise work. The fact is, we just don't give ourselves the opportunity to look and say, what could I do instead? So I feel that we shouldn't be afraid of AI just like every tool that's come before us. It's nowhere near the point where it's going to be getting rid of all of our jobs, but our jobs will be changing.

And being open to that is the most important thing that I can recommend to people. Be open to your job changing and start to look at things that one, you enjoy doing and would like to do more of. Well, I'm I'm very glad that I'm not talking to an AI version of Jason Bowdach and that I'm talking to the real Jason Bowdach. Mate, thank you so darn much.

That was good. And don't you dare say say anything about rambling because I could keep talking to you for so long about all of this. It's really, really cool and really, really interesting. And my hat goes off to you 100 percent what you're doing. Well, thank you so much for having me on the podcast. It was truly a pleasure. And it went by way faster than I could have imagined. Thanks, mate. Cheers. Thank you so much. Have a good one.

Jason, thank you so much for being my guest on the Color Timer podcast. That was amazing. It does go painfully fast, but that was it was really, really good. Thank you, everybody, for listening. Like, subscribe to all that kind of stuff, because it does actually really, really help and keep the comments coming. They're really cool. Thank you to my executive producer, MixingLight.com. If you're watching this on the Mixing Light website, you already know what they do.

If you don't check them out, MixingLight.com, everything color and, anything else? No, that's it. See you on the next one. Bye. The Colour Timer, a micro podcast experience.