Tomasz Brodecki

In your other examples, there was shadow detail that got deteriorated because of the compressed output. In this example, there was no shadow detail whatsoever. As for the "why" — any and every step of the pipeline could have contributed to this result — from a potentially underexposed original negative, through timing, printing, scanning and eventually preparing this digital copy that you're sharing (with its own bitdepth and bitrate limitations). Usually when a copy throws away half the potential dynamic range of the scene, we don't call it "good", but I understand that everything is subjective.

In that screenshot you shared, shadow detail was completely replaced with absolute black, tossing away any information within almost half of the frame, histogram below (pasting a boosted version below to highlight the issue)

Sure, with careful placement and/or a bit of diffusion, the two 300s can be enough. I'm mostly going to advise lighting the background criss-crossed (right light shines on the left part of the background and vice versa), so that you avoid hot spots and provide a more even illumination.

It doesn't. You're probably describing output or preview files based on those RAW inputs. As long as you shoot uncompressed or lossless RAW, compressed blocks of pixels are not something that exists in your image, their existence depends solely on your export settings — and in that regard you should follow Phil's advice above.

I suppose Karim was referring to the clipped highlight on the actual tail light, which turned white-yellow-red, not the red ghosting above it. What's worse, after grading (and "extinguishing" the blown out part), the white area turned slightly cyan.

That's exactly the case if you're going for an effective 180° shutter angle, as I've said above

No, to the recording speed — e.g. shooting 1/200th at 100fps or 1/100th at 50fps and slowing it down (so that no frame is skipped) results in the same motion ratio as shooting 1/50th for 25fps. And it looks like that's exactly what you achieved with the slowed down versions of your videos. I still fail to see the problem. The footage is definitely shaky, as in, unstabilized, but the shutter angle aspect looks alright to me (once slowed down and not frame-skipping).

I don't see the issue with the slowed down version, can you describe what exactly "looks shit" to you in this clip? Shutter angle math remains the same between 1:1 (real time) recording:playback and slow-motion, as long as you remember to match your footage to the playback frame rate (so that every frame is played and none is skipped)

The first clip you embedded looks alright, just what 50fps slowed down to 25 should look like when shot at 1/100 (so you're left with a 180° shutter angle). With the two shorter clips, it looks like you forgot to slow them down, so at 25fps you're only watching every other frame of your 50fps recording (skipping the rest), and that results in the "juddery" 90° shutter angle look.

What Ed said. It's a newer Manfrotto 190, too large for BeFrees, too small for a 055. And the head on the bottom one is an older model of the 494RC2, not sure about the fluid head (although it's not a Manfrotto Live). I assume that on top of it is an iPhone, with what looks like Moondog Labs' anamorphic adapter.

Yes, there's always that third option, but it's not exactly the best one for the planet ? Just to be clear — we're going to need more light for any scene intended for single-color result compared to an equivalent full-color one anyway (if we're going for a believable look of a single-color light source, because of the values that we will be substracting), but by using white we're left with more choices. There's a time and place for everything, and unless you have the colorist on set and/or performing final grading on the entire footage as you shoot, you will be making the ultimate color decisions in post production anyway, so the time on set is probably better spent focusing on the task at hand. The choice is up to you (or your DIT or your colorist, whoever prepares your LUT) as to how much you want it to use the blue and green values to be turned into reds and how much to be filtered out, or any specific point in between.

Because of the Bayer filter array in front of your sensor, if you shoot using only blue or only red light, the effective sensitivity of the sensor is reduced by 75%, so in order to compensate for that difference, you'd have to bump your gain, along with the accompanying noise, by two stops (for green, that's 50% or one stop). In other words, if you're going for a monochromatic result and you're shooting digitally, use the most of the spectrum you can, shoot with a red LUT if you can, and save the final color decisions for grading.

I kind of expect the gaffer to be equipped with a light meter at all times, mostly so that he can take measurements without walking back and forth to the camera, but for me, from the image acquisition point of view, waveform monitor is the best light meter. And as for location scouting, everything there is to know about the luminance of the scene, you can learn from the scout photo exposure settings (after all, a DSLR or a mirrorless camera is a perfectly capable metering tool in itself, and I don't mean just its built-in metering that returns the f-stop value for a given shutter time and ISO, but also the millions of per-photosite measurements that it stores in every single photo).

Do you have any specific one in mind? Most of those are relatively simple effects and animations that can be achieved using both digital and traditional methods, but the execution of the trickier ones (as well as the potential cost savings) would suggest digital.

This was just sent my way, I think you guys may enjoy it as much as I have: And for those of you that speak German (or are willing to suffer through auto-translated subtitles), here's an interview with von BERG: