Mk III ISO 100 Black Frame -- Cleaner Than We Thought?

[Stephen Melvin]

Okay, opening this black frame in ACR shows a lot less read noise than I expected. I have to take the black clipping point to "0" to get it to show any noise, even when pushing it four stops/bringing the fill light way up. On my Mk II, just bringing the fill light up more than 30% can show banding noise. Is my sensor cleaner than we thought? Or am I doing something wrong?

Notice that the only difference is the black clipping point moving from "0" to "1."

And the raw black frame:
http://galleries.stevemelvin.com/100blackframe.CR2

 

[LetTheRightLensIn]

I looked at it in IRIS and it has about the same read noise as all the production camera ISO100 blackframes I've seen (which are all a trace worse than all of the pre-pro black frames actually). Same noise, no horizontal banding, but still one of the vertical banding types. There is a chance it MIGHT be possible to design a pre-RAW processor to filter out this type of vertical banding, not quite sure yet, possibly though.

The random, non-pattern read noise itself (ignoring banding) is just like with the 5D2, no better at all.

 

[Mt Spokane Photography]

I've been playing with images from the first samples available, the blacks are really locked down. For High ISO's that looks good.

My 5D III will arrive next Tuesday evening and I'll get to see for myself.