[Blaise Douros]

With the FS7, when you bake in the LUT, it’s not just something that happens in metadata–it changes the footage that gets recorded. So yes, you are stuck, for now, with the baked-in look.

However, re-read my above. What I’m suggesting is that you create, via LUT Calc, a new LUT that does the reverse of the one baked into the footage. In theory, since the footage is recorded at 10-bit 4:2:2, the information is pretty much still there, but the baked-in LUT just changes the distribution of the luminance data.

So, what you MUST find out is the exact LUT that was baked into the footage. With that information, go to LUT Calc and see if you can create a new LUT that goes from the baked-in look to sLog3.

It will likely not be as good as if they recorded the footage in sLog3, but you may be able to get it looking pretty flat again. However, it is likely that you will not extract the full dynamic range from the footage, and the director’s just going to have to accept that–that is, simply, not your fault, and you can’t be held responsible for trying to reverse an irreversible mistake made on-set.

The other thing you need to let the director know is that, if the LUT that was baked in was just a basic Rec709 look or similar, then he hasn’t lost much–at some point during the coloring process, through a series of levels and curves adjustments, or through the application of a LUT, the colorist would have to get the footage to fit within the Rec709 color space anyway, so that it looks correct on consumer monitors. So working from a baked-in look is NOT the end of the world.

[Blaise Douros]

You might be able to use https://cameramanben.github.io/LUTCalc/ to generate a LUT that you can use to reverse the effects of the one applied in-camera. Normally you’d go sLog3 > Rec709, for example, but you might be able to generate a Rec709 > sLog3 LUT with this LUT Calculator.

What LUT was baked in? One of the presets, or a custom LUT?

[Blaise Douros]

30p will give you slightly smoother motion. Again, the delivery venue becomes important–if you’re projecting on a large screen, then a higher framerate might make sense. A bicycle moving horizontally will have a certain amount of judder at lower framerates; i.e. at lower framerates, the bike moves farther between frames than at higher framerates, making the motion more “jumpy.” On a computer screen, it’s doubtful that it will be noticeable but on a large, projected image, it might make a slight difference.

This is one of the reason that big-budget filmmakers are pushing for higher framerates like 48p–fast motion can be really difficult to follow on an IMAX screen at lower framerates.

However, most feature films are shot at 24 fps, and the 24 fps “flicker” is associated with movies. I would rather have the extra resolution and keep the 24p look; 30p looks too home-video to me. But that’s personal taste.

Bottom line–there is zero reason to shoot at 720p24. You’d just be robbing yourself of resolution. So your real choice is 720p30 (smoother motion, lower rez), or 1080p24 (filmic motion, higher rez).

TL;DR–tell the client that they can have either a higher-rez image with filmic motion, or a slightly lower rez image with smoother motion.

You can use this website to illustrate the motion piece.

[Blaise Douros]

Unfortunately, without knowing the output specifications, it’s going to be difficult for us to help you. For example, if your delivery format is 4K IMAX 48fps, then I would advise you to use none of these. Obviously, this isn’t your delivery format, but knowing how you need to deliver it IS important to get the best answers for your question.

If it was me, and the client just wants a usable web video, I would probably shoot the 1080p24. The extra rez will help you if you need to stabilize or crop on a 720p timeline.

[Blaise Douros]

The mp3 to wav analogy is spot-on.

I would definitely investigate Premiere–FCP6 may not even support ProRes LT, and then you’re right back where you started. You’ll find Premiere to be pretty similar in workflow to FCP6, except everything is faster, and it’s more powerful. And for $50 a month, you get the whole suite of Adobe apps!

[Blaise Douros]

ProRes is used to transcode footage for editing without a loss in quality–it’s easier for FCP to handle than h.264 footage. Each different flavor retains a different amount of information; ProRes 4444 will retain a huge amount of color information, and is useful for RAW footage like from RED cameras, or similar. But it’s easier for the computer to read ProRes 444 than decoding RAW.

However, since DSLR footage like what you’re working with is highly compressed, there just isn’t very much color information to begin with. So it’s a massive waste of space to transcode it to a format like 444, or even ProRes 422, which are designed to retain high levels of color detail. ProRes LT, which is a newer flavor, gets you better playback and footage handling in FCP, without ballooning the amount of data to obscene (and unnecessary) levels.

However, you’re saying that ProRes and the h.264 footage are being handled the same way in your timeline? Like, you’re having to render everything before you can see playback? If that’s the case, then don’t transcode at all! If you’re going to go through the misery of having to render every time you make a change, at least save yourself some hard drive space…

It would be entirely worth your consideration to look at getting into a more modern NLE. For all the crap people talk about it, FCPX is capable of a lot, and there are many professionals that use it. Premiere (which is what I use) is frankly awesome, and handles any and all footage natively, no transcoding needed. If a couple hundred bucks for FCPX saves you the enormous headache of rendering your timeline every few minutes, or transcoding any of the footage from modern cameras to suit your old NLE, you would make back the investment in man-hours alone.

[Blaise Douros]

Compressed video has a white balance value baked in, and it doesn’t have the color correction latitude that a RAW file does. If you take video with a DSLR, and the white balance is wrong, you are almost assuredly screwed, because the 8-bit 4:2:0 footage doesn’t have enough color information to allow for any but the most miniscule corrections.

Even when you shoot RAW, the camera is determining a white balance value to a certain extent; the difference is that the RAW file retains all of the color information, and can be easily, and usually losslessly, corrected.

These days, I am shooting with a Sony FS7 a lot, which shoots a 10-bit 4:2:2 XAVC codec. It has plenty of latitude for me to correct white balance later–but I usually still set one on set. Why? Because it gives me an image that is that much closer to “correct,” and so when I, or another editor, go to cut the footage, it’s one less area to have to worry about. No editor wants to f*ck around trying to fix a white balance that some idiot camera op didn’t bother to do on-set. That’s a sure-fire way to not get called back up for the next gig.

TL;DR–always set a white balance on set.

[Blaise Douros]

Yes, it’s normal. No, it’s not worth converting DSLR footage to ProRes–use the ProRes LT flavor if you must.

Is this FCP7? If it’s FCPX, it should be fine without transcoding.

[Blaise Douros]

I have a different take than Mark.

You need to first shoot the background plate of the running dog. Go through and find the take you like best, and then do your best to replicate the angle when you go to shoot the green screen; now, since you know what your focal length was when shooting the dog, you can use something similar for the kid. This may mean shooting on different days, in order to plan the second shot of the kid. You’re going to have to worry about scale afterwards–the key is going to be the angle.

If you can, have your selected take on an on-set monitor with you. Put the kid on a rocking horse or sawhorse or whatever, and put that on top of a turntable. Someone needs to turn the turntable so that the kid’s rotation matches the dog’s as closely as possible. Remember, once you rotoscope the background out, you can control scale, vertical, and horizontal movement, but rotation is the one thing you can’t control in post. I think it’s also key to have the kid on a turntable, and not the camera on a dolly–if the dog makes a sharp turn, you want your turntable operator to mimic that, and the kid will sway in reaction to the turn.

As you said, you can control the kid shot, but not the dog. So why go through the trouble of shooting every possible angle on the shoot you can control, when it might be completely different from the footage you get of the dog?

Don’t shoot this on a DSLR if you can help it. AVCHD or h.264 is going to make the chroma key difficult to pull; you need something that shoots 10-bit 4:2:2 in order to pull a clean key.

Make sure there are no shadows on the greenscreen, and that it’s placed far enough back from your subject that the green isn’t reflecting in their skin.

[Blaise Douros]

I can’t help you with the corrupted files, but I can tell you that there’s nothing wrong with your camera that’s causing the purple fringing. I pulled down one of your images and looked at the EXIF info–the one I looked at had you at 1/125, f5.6, ISO 4000, with the EF-S 55-250.

So your number one problem is that you were shooting in low light. I can follow your thought process–fast shutter speed of 1/125 in order to freeze motion, ISO 4000 to support that exposure, f5.6 because that’s as wide as your lens goes. And that’s your problem–the EF-S 55-250 is just not anywhere near fast enough to use in a low-light situation. You would need a f2.8 lens at the very least, to get a few extra stops of light.

See, what’s happening is that at ISO 4000, the sensor gets super noisy, and the dynamic range is lowered. If you zoom in on the images, you can see a ton of luma grain (the little black dots). The other effect of this is that super-saturated colors, like the colors coming from the purple light, start clipping and bleeding because the sensor is struggling to capture enough light.

At lower ISOs, the sensor won’t have trouble with those super saturated colors, because it has more dynamic range–but as you squeeze it upwards, you’ll get lower and lower image quality.

This is true for almost any camera except for the Sony A7S, because Sony sacrifices goats and kittens to Satan on a weekly basis to support the black magic that runs their newest sensors.

So: is your camera broken? No. But there are a couple things you need to do if you’re planning to do a lot of low-light shooting, especially at high speeds:

1) Invest in better lenses. The faster the lens, the less you have to crank that ISO up. If you just can’t afford it, consider renting them for these gigs where you know there’s going to be a lot of low-light.

2) Invest in off-camera flashes–pushing more light into the scene means you can afford to use lower ISOs. Using flash is nerve-wracking if you don’t know what you’re doing. Go to Strobist.com, grab some popcorn, and read like the wind. David Hobby explains how to use off-camera flashes better than just about anyone. There are some really inexpensive speedlights out there by Yongnuo; they’re not the greatest quality, but at 80 bucks, who cares? Buy five of them and rotate one in when another one fails.

3) If none of this sounds appealing, and you NEED to shoot a lot of low-light, then consider returning the Canon and looking at the Sony A7S. I hate to tell you to get out of a system you’ve invested in, but you do need the right tools for the right job. I’ve been a loyal Canon user for a long time, but right now Sony is that sexy little redhead waving at me from her convertible, and texting me that I should really come over and check out what she’s wearing…