Let's say you're trying to work on a project, And it would be extremely helpful to you if you're editing software could do some specific job.
But it doesn't.
Well there's always been a solution to this. All you really need to accomplish your custom task is a high level programmer fluent in a half dozen computer languages. It's as easy as going to your bank taking out a loan for $500,000 of development money, And Voila! you now have a program that automatically cuts all of the super blurry or shaky footage out of your project directory for example.
But what if you could just do this yourself for $0?
That would open up a remarkable number of possibilities all of a sudden.
Recently chat gpt added a new feature called Canvas. It's simply a new split screen interface that allows you to work on a document on the right side of the screen while talking to the robot on the left side. Sounds boring.
Here's why it's not. Producing a sophisticated document such as a functional program was extraordinarily unwieldy in the standard chat bot interface. In addition both you and the robot having a mutual reference screen solves many problems that centered around memory, Simply by including a consistent shared reference.
What's the big deal? What's the practical use here for the regular person? You can now write functioning computer programs in languages YOU DON'T KNOW.
There are pretty large number of very sophisticated languages now, And even remarkably intelligent people rarely learn them all, So this new ability really does open up a lot of new options for indie film makers.
Maybe you picked up a lot of background noise during your recordings, But it appears kind of randomly in all these hundreds of different shots. Maybe you should create a python program that accesses the Cuda architecture to rapidly scan for noise patterns and dynamically suppress them without reducing the quality of your dialog or Foley. If you take the time to sit down and really focus on hammering the design into shape, That's now possible for someone who has never written a program before.
Personally I know some code and have written a few programs, But it's such an immense world that's constantly changing that it's really impractical to stop in the middle of a film project and learn Javascript Or Pytorch in an effort to automate the process of finding all the shots that used that one lens you didn't realize had a big scratch on it until after you got home from the shoot.
Get creative, This is a process you could learn in a matter of days, And can be used to make custom tools specific to your needs whether that's dealing with scripts or storyboards, finance or image recognition, etc.
I spent 20 Bucks a month on chat gpt pro, and the first time I used this, I built a program that saved me three or four hours a day of work. I'm now expanding that tool into a kind of Swiss Army knife of film image management.
(Below I asked the alpha version of my program to look through a directory with 56,000 files in it and grab me a picture of the cat that was aesthetically pleasing)
Last edited: