I didn't want to do the ol' feature request for a non possible feature.
And in a way this plays back on the hdr in camera (although I'm not sure if the hdr in camera in 650D and what not is fake from 1 picture or is from multiple pics).
The question is:
Would it be possible to tone map in camera before it saves as a jpeg (or video)?
So say it takes the raw sensor data, applies the white balance, applies the picture style, (as usual) -
Then on top of that it applies a tone mapping with a selectable contrast compression from 1.0 to 2.0 (although I find anything over 1.5 to be a bit much) -
And then the jpeg be saved from that.
First you need to realize that a local tone mapping operation needs to analyze the neighbouring pixels over a quite large area. With a 10x10 kernel, the processing power will be 100x higher than for a normal image. With a 100x100 kernel... 10000x higher (unless you can optimize that particular operation with some clever FFT math or something like that).
On a PC, just time enfuse and compare with some trivial brightness change in a video editor.
Second, remember that we don't know how to program Canon's image processing chip.
Third, good luck designing the math formulas to do that.
LOL. Ah, ok. :P I was just curious.
It's easy to do on RAW. It doesn't work so well on 8bit images though, that's why I wondered if it could be done before hand.
But thanks for the reply.