Forum Replies Created
-
Replies
-
Another question is should I be changing my iso according to my camera in order to “shift” my dynamic range depending on the scene and what information I’d like to retain or not retain?
How common are exposure compensation LUTS (-1 stop, +1 stop, etc) or CDL’s used on set?
If you are shopping raw, changing your ISO is essentially the same as using a compensation LUT. All it does is change the brightness of your preview image (which will then change your exposure behaviour).
It’s quite common to lower ISO a stop, or even two, for low light scenarios if you want to retain more shadow detail. In very bright scenarios you might want to raise ISO to protect highlights from clipping. Some raise ISO for creative reasons, to get more of the camera’s specific noise/grain texture.
So if you’re shooting raw there is no need for compensation LUTs, shoot under the show LUT (the look for the project that has – hopefully – been established before shooting) and use ISO to shift more of the DR below or above middle grey if needed.
Shooting to a codec I think it gets a little more complex, as different cameras/codecs react differently to ISO changes.
I’m not sure if you’re asking for exposure or also contrast ratios ect.
I think there are two aspects of getting exposure right, a technical and a creative one. They kinda go hand in hand but with the quality of today’s cameras I believe the latter has more weight. I think sacrificing the benefits of a final in-camera look only to get a quarter stop more in the shadows or highlights isn’t a worthy trade off in most situations (unless your pipeline allows for visual compensation). Simply not screwing up the technical part is enough imo. Distributing a stop (or even two) of DR below middle grey for low light scenes or one above middle grey if you really need to protect some highlights should be sufficient for almost all scenarios with today’s technologies, both in production and post.
Afaik both Roger and David nail the look in camera and that has many benefits, e.g. all decisions on set can be made in context of the final vision or that there’s an unmistaken reference when it goes into post.
As for the dilemma with False Color pipeline, I think there’s no right or wrong or correct way. The more time you spent with tools the better your understanding of whether they’re actually benefitial to your personal way of working or not. That can vastly differ from person to person in my experience. E.g. personally I get confused by False Color (even more so since there are so many different versions out there).
Check out Juan Zambrano’s 2499 DRT color pipeline, it’s free and provides a more film like image transformation. You’ll need Davinci Resolve Studio.
There is a thread on it over at LiftGammaGain.
Points 2-6 are nothing to worry about imo.
About reliability, I did not have more issues with Blackmagic cameras than with any other manufacturer but haven’t worked with that particular camera yet.
Like Roger I operate on instinct, albeit on a very different level. But also like Roger (I believe) I learned a lot from starting with documentaries, where I would move around quite a few times within minutes to get different angles and shot sizes of a scene that did not wait for me, in order to have a variation that would work for the editor – who often was me! It was a very exhausting way of shooting but a great and rewarding exercise – no time to ponder over rules!
I see youtube videos of “Roger Deakin’s favorite setup” all the time where they proceed to make something that looks not even close to the reference shot.
Don’t worry about the quality of LEDs these days…it is better than HMI and on par with tungsten. YouTubers failing to recreate scenes is not about the quality of the light but the relationship in size, distance, power, placement, diffusion ect of the lights to the subject/scene. There’s also a huge skill (and probably experience) gap.
The same goes for cameras. The gap between latest generation semi-pro hybrids and traditional digital cinema cameras is already too small to matter much. What does matter is the transformation from sensor data into a display space. You can take the same data from a cinema camera and make the image feel and look terrible or gorgeous. We cinematographers don’t like to admit it but it does make a decisive difference. Roger has one of the best transformations ever made with his custom Alexa LUT, which basically is an extremely sophisticated film emulation. The trick is to have it in place before shooting (loaded into the camera). A solid transformation will not demand a lot of grading in post as the look is already established and all decisions during shooting have been made under it. That is different than shooting with a standard transformation (e.g. standard Rec.709 LUT from manufacturer) and then changing the look afterwards.
One more thing as to why YouTubers fail to replicate scenes: In my opinion Roger’s biggest strength is not even lighting but to put the audience exactly where they need to be. I think it is impossible to replicate his intuition for that. I also think it is more significant than his lighting, even if that sounds crazy. I’m sure if he would shoot a film only with available light, it would still have that capturing quality. Putting the audience exactly where they need to be to witness a story unfold is his superpower in my opinion.
Also, gutted to report that my Amazon pre-order (placed ~6 months ago) just got cancelled out of the blue.
Mine too. Amazon Germany.
Ordered April, yesterday it said it would arrive today, today it suddenly says ‘order canceled’.
Most of them have magnets on the back, you could try and attach a thin metal to them to slide behind the mirror and see if it holds. These tubes can be quite lightweight, especially the cheaper ones.
Once the light is on it might not look much different to a regular fluorescent tube light, depending on the model’s design and light intensity probably.
How about adding a LED tube light above the mirror as a practical? They most often offer remote control, daylight/tungsten or full RGB light, battery and magnets. Newest generation all have decent light quality, even cheap Astera/Aperture/Nanlite knockoffs.
You could remotely and seamlessly adjust its output (and color temp, if window light changes).
I would say both and add set design (see the walls’ colour being closer to Alejandro’s suit than the white of the water tank).
But happy to be corrected by Roger.
