Digital Light & Color

I use a Canon DSLR and a friend pointed me to the "Magic Lantern" f/w enhancement that adds
a bunch of features including one called "Dual ISO". In this mode, the camera (as I understand it)
sets adjacent pixels to different ISO values to grab a photo at two different ISOs simultaneously.
Here's a link:
http://www.magiclantern.fm/forum/?topic=7139.0
Has anyone played with this? The page mentions post processing s/w, I wonder if there is a way
to integrate or simulate the functionality in PWP.

It's amazing to see people reversing engineering camera and adding new features.

An interesting idea and great proof of concept. However changing your camera firmware in this way would then require you to use custom RAW processing software to process all your images, and you lose some vertical resolution to get the extra dynamic range. Ideally, this kind of feature would be integrated by the camera manufacturer into their standard firmware and the specialized pre-processing would be done in-camera.

A smiliar approach was Fuji's SuperCCD SR sensor introduced in 2003.
It had 2 photodiodes for each pixel, one high sensitive and one low sensitive.
http://www.dpreview.com/articles/685125 ... superccdsr
All the computation to combine the 2 images was done inside the camera.
Would be interesting to know if that two sensitivities are done with real physical change of the amplification of the signal,
or if it's done by software from the same database.
From my oppinion only a change of amplification would make sense, just to change the values of the existing data could be done in every raw converter too.

Re. the Fuji sensor, for which a quick intenet search turned up an interesting discussion of how this increase in sensor dynamic range actually performs in the Fuji S2 Pro DSLR (2006) (at the section "Exposure and Dynamic Range" far down the page). Quick superficial summary: it works for out-of-the-camera JPEGs, but determining an optimum exposure for the advantage is tricky.

mjdl, yes, i can confirm that. The S3 was my first DLSR.
It was damn slow when saving RAFs to the CF card (was a bit faster saving to the xD card, but i never liked those small thingies)
It was not possile to utilize the double pixels with PWP as RAW-converter, it was just possible with Fuijis own software,
which had a hillarious user interface and performance.
So I hardly ever used RAW with it, just when it was a picture that really needed a lot of tweaking.