Adventures in digital push & pull (and possible ISO 50 on the R-D1?)

[rogermota]

Having recently bought an R-D1 and this being my major transition from film to digital, I took it upon myself to see what the CCD on the R-D1 was capable of in the areas of push & pull:

Shooting in RAW and processing in the EpsonPhotoRAW (mac), I did several tests shooting in, for example, ISO 800 and comparing that to ISO 200 underexposed by two stops and pushed two stops in EpsonPhotoRAW (effectively an ISO 800 end result).

Upon close inspection (100% view) there are some very subtle differences in how grain shows up. But basically ISO 200 pushed two stops is roughly the same as ISO 800.

I then tried to see what happens in pulling, shooting ISO 200 and comparing that to ISO 800 overexposed two stops and pulled two stops in EpsonPhotoRAW. About the same net result, with some very subtle differences.

But then decided to see if I could get an effective ISO 50 by overexposing ISO 200 two stops. And by my results, it does seem possible. Same sharpness is there, but basically no grain.

BTW, I tried to see what an ISO 6400 looks like -- it looks terrible 🙂

Anyway, click here to download the psd file with layers (1.15MB) and have a look for yourselves (its just a small crop of the bottom left corner of the original image, but you get the idea...)

Thoughts?

 

[zeos 386sx]

I've been trying to think of a way to transition from zone system film to digital. To do that I think we are going to have to do something like what you have just done - i.e. test for how much we can acceptably "under" and "over" expose, call that our contrast/density range, and then meter VERY accurately to get our most important picture elements within the exposure.

Digital clipping is a hard rock to deal with and underexposed digital, as you point out, isn't pretty.

 

[fgianni]

It looks that pulling is not really practical since highlights are burnt, on the other hand pushing iso 400 2 stops instes of shooting 1600 seems to preserve highlights better.
I can't see a significant effect on shadows fom your example.