I get asked regularly about my process...what applications I use, what gear I use, how I do what I do. So, I figured since I'm locked in my apartment by myself bc of COVID19, I might as well write my first blog post about it. I'm not going to go crazy making it spit-and-polished, it'll be more quick and dirty, and this will basically just be about what I do every night and why.
I'm going to presume a basic level of knowledge, and this is primarily aimed at basic to intermediate readers. Advanced users will not find much here maybe apart from my use of a waveform.
Let me also throw out all the standard provisos...this is not necessarily the best way, it's not the only way, it might not be the fastest way, it might not be the best way for you, you might not agree with what I do, etc.
Standard Everyday Gear:
Sony A9, set to Manual/Auto ISO
Sony GM 70-200mm f2.8
Sigma ART 24mm f1.4 + Sigma MC-11 Adapter
Macbook + some sort of Samsung external monitor
Ninja Recording Monitor
Applications:
Capture One
Adobe Bridge
Once we're done with the show, time is of the essence...I've got to get done with these images, they have to go to the show for various teams to choose and discuss and then post within an hour or an hour and a half.
I head to my room and start importing images into Capture One. As the images come in, I'll either delete or rate 5 stars. I'll do this for green room images, the interview images, any images of any special segments I'll have taken, or if I have any other shots I grabbed during the show that were relevant....was there a moment that they'd want to tweet about? Something that stood out? I use my judgement or they'll request something specific (and hopefully I caught it).
Once I've deleted everything I don't want and kept everything I do, I'll make another run through and re-rate everything: 5 star to submit, 4 star to keep just in case they ask for additional images in the coming days.
Next I run my presets. I get asked a fair amount about presets....I build my own, especially at the show. I went around with a Color Checker and shot in every possible room I could ever shoot in anywhere in the building, and anywhere where things changed significantly I made another preset and named it for the room in the building. That way when I shoot something in a certain room, I just click the drop down and apply that preset. The green room and backstage images are generally black and white, so the preset does that too. Easy. For me, it's all about time....I need to bang through a lot of images very quickly. I also have a thing about repeating motion; if I do something manually more than a few times, I automate it or build a preset.
Next is the only truly unusual step (at least as far as I know)...I look at everything on the recording monitor with a full-screen waveform scope.
What's a waveform scope? Briefly - and message me if you want to know more, perhaps I might write something up about this if there's enough interest - a waveform scope basically just shows me the luminance of what's in my photo.
Why do I do this? There are a few different reasons, but this is a good time for me to talk about my one big secret:
I don't look at the images when I'm at this stage of processing them.
dafuq does that mean??? Wait, I'm a photographer and I don't look at the image??
I do not look at the image.
Why is that? A bunch of reasons.
I work in a pitch black room and then all of a sudden I'm staring into lights and then I'm back into the dark again. I go from one color temperature to a different one to a third one. The studio has all kinds of accent light in blues and purples that bounce all over the place. Occasionally my external monitor does double duty as a prop elsewhere in the building and might come back to me with settings changed that I don't notice or have time to fix. My laptop might be cold, or warm, or on too long. I try not to update my software unless I absolutely have to, but if I do....what if some subtle setting was changed, like something that changes the color temp of the screen after a certain time? I know better than to trust my eyes or my equipment. I've got 30-45 minutes. I can't fuck around.
So if you don't trust your eyes or your equipment, what do you do? I do everything with numbers and charts and scopes. That way I don't need to worry about anything other than finishing the images. The numbers don't lie. My eyes can.
First, I make sure the color is correct by mousing over something neutral and making sure it's actually neutral....white shirt, grey surface....for me, I can use the mugs, which are a nice matte black with white writing and generally work out pretty well.
If you look at the image below, the 4 numbers are Red/Green/Blue values and luminance (this is Capture One, remember). 0 is black and 255 is white. The screenshot below doesn't show the cursor, but the mouse arrow was on the black of the coffee mug, and thus the values are fairly low.
When the RGB values match or are at least within a few numbers of each other, the colors in the image can be presumed to be correct. When the red number is high, the image will be warmer. Blue? Cooler. Green? Then it's green, and you need to fix that shit. This is showing to be just a tick cool, but I know from experience that I can never get this mug to be perfect. The overall look of the image won't be affected.
Once the color is correct, I get to the waveform. Why do I use this weird and obscure and seemingly unnecessary piece of gear?
First, I need to match the look of the broadcast as closely as possible. This is not about my "look." I'm not interested in getting into any kind of crazy style with these images...they need to be correct, the skin tones and brightness values need to match the show and reality, and I need to not have to worry about my eyes or the equipment or anything else I spoke about above. Don't get me wrong, it needs to look like a good photograph, but above all it needs to look fairly close to the show.
All broadcast shows on television have - somewhere in the mix - a video engineer who is sitting in the dark with all their monitors and an array of the same scopes, making sure that their color and levels are correct. I sat with our video engineer and double checked all his settings, and the waveform allows me to nail his levels quickly and with confidence.
Second, remember when I said that I don't look at the images? I like to do everything with the RGB values along with levels and the histogram, but if you notice those are both great to tell you how many values exist at what points in the image and whether your image is exposed correctly as far as not clipping your highlights or your shadows, but nowhere I'm aware of in any of the most common image processing applications is there a way for you to quickly and confidently work with luminance levels in your images.
What does this mean? Well, you can make sure that your skin is correctly exposed, as far as not clipping your highlights or your shadows, but how do you know that the skin tones exist at the correct levels, in this case with "correct" referring to what they were for the broadcast? The lighting director set light levels and the video engineer set his cameras for those values, and I made my images and registered skin values, but how do I know that they're the same as the show? I can have my skin be properly exposed but still too dark or too bright, and all the things that I mentioned above can potentially affect whether I got that right or not. And again, I don't want to rely on my eyes...I've been staring into lights for an entire show and now I'm sitting in the dark trying to make sure my images are correct.
So, I use a waveform. Let's take a look:
Keeping in mind that this is a phone photograph of two screens, here's what I work with. The waveform here is showing me the luminance level of everything on the screen. On the "Y" axis of the graph are values from 0-100 IRE, with 0 IRE being black and 100 IRE being white. Happily, everything on the scope is where it exists on the screen. This is one of the true joys of it, to me. The histogram and levels tools will just tell you what the values are, but it takes a little work to find *where* they are in the image, and really, it's not all that useful for these purposes.
So here, if you look at the upper left-hand corner, you'll see that the white is registering at 100 IRE. That's the text in the Capture One settings box. Then to the right of that, I've circled where the arm of the person on the left appears. Here's the true joy of the waveform...there are relatively standard settings that will allow you to know for sure that your image is correct, and they all involve skin tones. Caucasian skin can safely be placed at between 65-67 IRE. If you look here, I put this particular skin tone at 65. I'll just run the exposure up or down to place it exactly where I want it, and then I'm finished. I copy those settings and paste them to the rest of the images. I double check that my color is correct (RGB values match, remember) and I'm done.
Do I *need* the waveform every night? I do not. The first time I put it into the mix, I made sure that my presets contained the correct exposures, and they'll be pretty close, within 3 IRE or so. But I like to be as precise as I can.
Additionally, there are variables every night...is one guest very dark? Are there sequins or rhinestones? Is the makeup reflective? Sometimes the meter in the camera decides not to play nice with all the monitor walls in the background I'm shooting into. So it's a nice confidence measure that allows me to close the computer and be confident that I'm not going to get a message when I'm halfway to the subway that something looks really off.
The rest is just cropping and straightening. Why do I have to crop? Because I always shoot wider than I need. Years and years ago I was taught "always leave room for them to put "National Geographic" at the top of your image." Obviously that's not happening here, but there might be a billboard, or a caption in a magazine, so I shoot with some latitude. In order to save some time on straightening, I always shoot with the "level" tool turned on in the Sony's EVF. Another confidence measure, on a set that's all curves.
Once everything is straight and the colors are correct and I know the levels match the show I export everything and upload it to the show. Now I can pack up to go home.