The General Production & Filmmaking Technical Thread

[Dementeleus]

So, a lot of times I'll come across some interesting bit of technical information which doesn't really fit into an existing thread, or isn't about a particular movie, but nonetheless is interesting and could be impactful on many future films. Other times people will bring up technical questions that others can answer -- again, usually this doesn't pertain just to a specific film, but a general technique or technical minutae.

 

I thought it might be nice to have a thread for these questions. If you come across some new piece of equipment that's being discussed, mention it here (with a link to the article/whatever). If you have a question about some aspect of filmmaking, ask it here. Hopefully there'll be enough to keep the thread going.

[Dementeleus]

To start things off, I've been reading up on Arri's new digital camera, the Alexa 65. It's really impressive. First of all, some basics: we laypeople get caught up in resolution, because it's sexy and an easy way to compare specs... but what resolution a camera can shoot at is just part of the question. Another key part is the size of the camera's sensor -- literally, the size of the chip that captures the light and registers the image. The bigger the better. It's easiest to think in terms of the size of a frame of film: the same reasons that IMAX offers a better image than 35mm are the same reasons why a larger digital sensor is "better" (there are some occasional rare exceptions, but they tend not to be related to movies): larger sensors mean shorter depth of field (which we register as "more cinematic"), they mean the captured image needs to be magnified less on projection (leading to a greater sense of reality), and there's typically more precision and accuracy in capturing the image.

 

The breakthrough in digital cinema cameras wasn't just 4K, it was sensors that were equivalent in size (more or less) with Super35 and 35mm film. That meant that DCCs were capturing images in very similar ways to film cameras, and combined with first-rate professional lenses, meant that DCCs could approximate the resolution and "look" of film.

 

Fast forward to today. There are plenty of films being shot on the RED Epic, the Arri Alexa, and (recently) the RED Dragon. These are cameras with 35mm-sized sensors shooting less than 4K (Alexa) to 6K (the Dragon). But for our purposes, they're basically equivalent to the workhorse film cameras before the digital revolution. There's a huge gap between that and the large-format cameras (we tend to mainly think of IMAX 15/70, but 65mm film is also considered large-format).     [note: check out the chart at the bottom of the post for sensor size comparisons]

 

The Alexa 65 changes that. It may well be the first realistic contender as a Nolan-approved large-format digital camera.

 

The Alexa 65 has a huge sensor -- in fact, as its name suggests, it's as big as traditional 65mm film sensors. It also shoots in 6K. Arri's digital cookery has a very film-like look, that's why Alexas are so popular among DPs, despite their low-ish resolution. (The Alexa is the camera that convinced Roger Deakins to go digital).

 

So now the Alexa 65 is about to be released. It has a beautiful, rich film-like image. It shoots extremely high resolution and (more importantly) with an extremely large sensor. It's not for everyone. It's expensive as hell -- so expensive it's rental only. It's a beast. It probably won't shoot well in low-light situation. It shoots an insane amount of data -- you'll spend a small fortune just on media for storing it. And yet, with the right tools (i.e., a high-end Hollywood production) it may yield the closest digital equivalent we have to IMAX film yet.

 

Here's the Alexa 65 sensor compared to others on the market:

And here it is compared to even more (including IMAX 15/70, which clearly it's still far from matching in size):

Edited October 5, 2014 by Telemachos

[Alpha]

Nice thread Tele.

[lilmac]

To start things off, I've been reading up on Arri's new digital camera, the Alexa 65. It's really impressive. First of all, some basics: we laypeople get caught up in resolution, because it's sexy and an easy way to compare specs... but what resolution a camera can shoot at is just part of the question. Another key part is the size of the camera's sensor -- literally, the size of the chip that captures the light and registers the image. The bigger the better. It's easiest to think in terms of the size of a frame of film: the same reasons that IMAX offers a better image than 35mm are the same reasons why a larger digital sensor is "better" (there are some occasional rare exceptions, but they tend not to be related to movies): larger sensors mean shorter depth of field (which we register as "more cinematic"), they mean the captured image needs to be magnified less on projection (leading to a greater sense of reality), and there's typically more precision and accuracy in capturing the image.

 

 

 

 

Those principles apply to telescopes as well.  Larger mirrors gather more light (you can see further) with resolution being another matter.

[Dementeleus]

Here's a great article that explains the key differences between film and digital cinema and why film might possibly be better (for now):

http://www.redsharknews.com/technology/item/2089-we-may-have-solved-the-mystery-of-why-film-just-looks-better

[Dark Jedi Master 007]

Tele, this was a question that some of us were discussing a while back, and I thought I'd ask it here:

Why is it bad to use too many shots to convey information? Ok, that might be insanely vague, so let me explain. In general, there's a rule in cinema that says the less shots you use to convey the same info the better you are as a filmmaker. So if I can convey the same info in 20 shots as opposed to 30, I should do it in 20. My question is why? Why is it bad to use 30 shots rather 20? How do the 10 extra shots affect the viewer?

One answer I got was that using more shots is like using more pages in a book. But IDK if that comparison works. When you add pages to a book, you add time, and after a while, if you spend more time than you need to on something, then I'll notice, and that will take me out of the book. The thing is that 30 short shots could take as long as 20 long shots, so the time might not actually change, which means I wouldn't be taken out of the film by the use of 30 short shots.

So again, we come back to that question: why is it bad (or ill-advised) to use 30 shots when 20 will do? How do the 10 extra shots affect the viewer?


Note: This discussion centered around the café scene in Inception, if that helps. The main argument was that Nolan used too many shots to film the scene and violated "film economy."









 

[Dementeleus]

Tele, this was a question that some of us were discussing a while back, and I thought I'd ask it here:

Why is it bad to use too many shots to convey information? Ok, that might be insanely vague, so let me explain. In general, there's a rule in cinema that says the less shots you use to convey the same info the better you are as a filmmaker. So if I can convey the same info in 20 shots as opposed to 30, I should do it in 20. My question is why? Why is it bad to use 30 shots rather 20? How do the 10 extra shots affect the viewer?

One answer I got was that using more shots is like using more pages in a book. But IDK if that comparison works. When you add pages to a book, you add time, and after a while, if you spend more time than you need to on something, then I'll notice, and that will take me out of the book. The thing is that 30 short shots could take as long as 20 long shots, so the time might not actually change, which means I wouldn't be taken out of the film by the use of 30 short shots.

So again, we come back to that question: why is it bad (or ill-advised) to use 30 shots when 20 will do? How do the 10 extra shots affect the viewer?

Note: This discussion centered around the café scene in Inception, if that helps. The main argument was that Nolan used too many shots to film the scene and violated "film economy."

 

I've never heard that particular argument before. The principle(s) I've always worked with is that each new setup should be used for a purchase: it's revealing new information (a character's motivation, change in motivation, some new physical aspect of the scene, etc). If you're cutting simply to cut, it's wasted and redundant. However, that doesn't stop people from doing it all the time, and the reason for the cut may be more prosaic: the actor flubbed the next part of their line, or another line read is preferred, or whatever. If the director has shot enough coverage, and have broken down the scene carefully enough, then each angle is specific and necessary. So the question becomes not about how many shots, but is each shot used for an actual purpose?

 

There's a blog called "Every Frame a Painting" run by a guy named Tony Zhou, and he does a great job of exploring visual information on film. His most recent entry is about Fincher, and it focuses (somewhat) on your question: one of the distinguishing marks of a good director is how effectively they can tell a story in a purely visual manner (even if it's on a subconscious level), even if it's a scene that's not "active" or "exciting" -- a few people sitting and talking, for example.

 

[dashrendar44]

Interesting piece of information from Gone Girl post-production behind the scenes in terms of editing softwares and pipeline solution (also reframing techniques without quality loss):

 

[Dementeleus]

Interesting piece of information from Gone Girl post-production behind the scenes in terms of editing softwares and pipeline solution (also reframing techniques without quality loss):

 

 

Yeah, GG's the first movie edited on Premiere CC. Very interesting... I assume the same team will continue to use it for Fincher's future movies, but I wonder how many other features will start using it (probably some but not many). The new Will Smith movie was cut on FCPX -- now that's pretty wild.

[Dementeleus]

Wanna get a sense of how much work a colorist does? Just watch this 3-min summary of color-grading a L'Oreal spot:

t http://vimeo.com/108466337

[Dementeleus]

A way forward for the VFX industry? http://www.fxguide.com/featured/a-way-forward-for-the-vfx-industry/

Great, great long form article breaking down all the current issues within the VFX industry, and one new approach that's being explored as a possible solution.

[grim22]

As requested by Tele, posting this here

 

Transformers 4: The Final Touches from Michael Bay Dot Com on Vimeo.

[grim22]

This is interesting to read. Making fake money for movies: http://priceonomics.com/the-business-of-fake-hollywood-money/

[Dementeleus]

The complications of digital archiving:

 

http://www.vulture.com/2014/12/perils-of-an-all-digital-movie-future.html

[Dementeleus]

 

So true...