In the past, I’ve been measuring and reporting the spectra of the read noise in various cameras that I’ve tested. I decided to revisit those calculations and see if I could improve the presentation of the results. I’ve been using ChatGPT to make me more efficient at writing my camera analysis and color science software for about six months. It’s changed the way I code, and for the better. I comment my code a lot better. I am more willing to undertake projects where the probability of success is not high. I am willing to go back and clean up old code. And I will take the time to make the presentation of the data clearer. The code that I used for this in the past was based on a Matlab Fourier transform function created by a DPR member about ten years ago and modified and refactored by me over the years. It was getting a little long in the tooth, so I started over. One thing that I found out late in the project was that Adobe DNG Converter doesn’t necessarily preserve the CFA ordering…
No comments yet. Log in to reply on the Fediverse. Comments will appear here.