Do you shoot in Adobe RGB?

Started Mar 14, 2008 | Discussions thread
gollywop
gollywop Veteran Member • Posts: 8,301
Re: Thanks, Iliah . . .

Iliah Borg wrote:

Question for Iliah, if he's reading this: I achieved something very
close to UniWB (as contrasted to Near UniWB). Are, however, the
stats I just gave (256, 224, 260), which resulted from doing exactly
your steps, the Near UniWB figures, and should that WB be used for
Near UniWB instead of the one that lead to the (256,256, 255) stats?
(thanks)

I use R=256, G1=G2=256, B=257 as NearUniWB. In my testing it is
enough not to trick ACR/LR into believing this is a multi-exposure
shot.

Now, a question to you if you please:) Do you know why this methodic
of obtaining approximate 1's for white balance works?

Iliah: Your question has an ambiguous context, so I can find interpretations leading to any of the following:

Why the process you outlined produces nearly unitary WB coefficients?
Why Near UniWB prevents ACR/LR from believing one has a multi-exposure shot.
Why using the resulting UniWB works (i.e., produces the desired result)?

I'll give my notions regarding the last interpretation first, then the first. Please forgive my not knowing the proper and accepted words for describing various concepts; I can but do the best I can. I hesitate to play this game because I don't know that I can trust you to play fair and not indulge in your proclivity towards being a bully. That will end it quickly, and that would be a shame. But, what the heck -- here goes.

My understanding is that to achieve so-called white during the in-camera transformation of sensor site values, actual relative values of the R and B sensor information must be altered relative to G. The result is that the color-channel histogram information reflects these altered values and does not accurately reflect the relative intensities of the color information actually in the sensor sites. By setting the WB as the UniWB, this distortion of the actual sensor information is removed (the alteration is neutralized) and the histograms provide better information about the actual distribution of luminance in each color channel.

As to interpretation 1: taking the WB of a known neutral gray allows one to determine just what alterations in R and B relative to G are being applied by the camera, let's call it a base adjustment. Taking the shot of the gray with this WB setting provides a camera determined image of that gray. Altering its WB in NX (or whatever) in the inverse of the base adjustment should provide an image which, when used to set the WB, will just compensate for the base adjustment.

As to the issue with multi-exposures, I fear really don't know. But I would like to.

-- hide signature --

gollywop

-----------

Post (hide subjects) Posted by
Mjs
Mjs
JBP
JBP
(unknown member)
(unknown member)
(unknown member)
JBP
JBP
JBP
JBP
JBP
JBP
JBP
JBP
JBP
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow