Timeline for How does a modern digital camera process to jpg?

Current License: CC BY-SA 4.0

4 events

when toggle format	what		by	license	comment
Jun 12, 2022 at 4:35	comment	added	Michael C		@Stuart444 You're making the incorrect assumption that the number of bits must equal the number of steps of dynamic range in the scene, or even the number of steps of the DR of the sensor. That is no more the case than that Ansel Adams' Zone System which had 11 zones had to equate to 11 stops of DR in the scene captured. Nothing could be further from the truth. The entire point of the Zone System was to capture scenes with more DR and squeeze details from all of it into the more limited DR of the photo papers Adams had to work with.
Jun 11, 2022 at 17:48	comment	added	Steven Kersting		@Stuart444, nothing is necessarily "outside the 8-bit range"... but the more you try to encode with 8-bit the more compression (rounding) is required, and it eventually results in banding. E.g. you have 256 numbers available to describe 256 tonal values; those tonal values can be very close together (smooth transitions) or they can be far apart (banding), it doesn't matter. BUT, most of the jpeg banding results are due to less accurate 8-bit post processing math (which is why HDR post processing is typically done in 32-bit).
Jun 11, 2022 at 14:59	comment	added	Stuart444		Thanks. I'm preparing a talk for our photographic club on HDR. The process of tone mapping is well described on the internet. However in the course of my work I realised that when a camera processes its data to a jpg, it must do something with the darker tones and lighter tones that are outside the 8-bit range of the jpg. It seems to me that it can either throw away those out-of-range tones, or it must tone map . My Olympus takes HDR brackets and will merge them in camera, and the result is quite different to what it will produce when set to taking a single frame to jpg.
Jun 11, 2022 at 12:33	history	answered	szulat	CC BY-SA 4.0