New article on color management

Started Apr 26, 2013 | Discussions thread
Detail Man
Detail Man Forum Pro • Posts: 16,816
Re: Does Melissa "Gloss Over" Image Noise ?

The raw processor's working space varies from one processor to another. In ACR, for example, you have a choice among sRGB, Adobe RGB, ProPhoto RGB, or ColorMatch RGB, although a linearized version (gamma = 1.0) of the very broad ProPhoto RGB appears to be its native intermediate space. In others, such as LightRoom or PhotoNinja, there is no choice: they use a linearized version ProPhoto RGB (although the image as displayed on the monitor and the histograms have a 2.2 gamma applied to make them appear more in line with human perception).

"Color Management - a Walkthrough", Section 3b., Raw Processor:

It occurs to me that since the "Melissa" gamma-corrected color-space (used only for user-interface previewing, histograms, and color readouts) contains the same linearized region at low levels that sRGB does (at an input value existing 5.767 EV below maximum input):


From quoted text attributed by Sean McCormack to Adobe's Jeff Schewe:

Lightroom (and Camera Raw) uses a "working space" (meaning the processing
color space) of Pro Photo chomaticities (colors) and a linear gamma (1.0 gamma).
Martin is correct to call LR's working space as linear, but (and this is
where it gets confusing), "Melissa RGB" (the color space coined by Mark
Hamburg) isn't the processing space of LR/CR but the display space in Develop.
Melissa RGB does indeed have the Pro Photo chomaticities but the gamma is
actually that of sRGB and the ONLY place where Melissa RGB is used is for
the Develop histogram and the color readouts.
So, Lightroom's internal processing space is ProPhoto RGB in a linear gamma
(1.0), but Melissa RGB is ProPhoto RGB with an sRGB tone curve. Also note
that the sRGB tone curve is not a simple gamma curve but is a tweaked curve
based on gamma 2.2 but with the toe adjusted.
Jeff Schewe

- Sean Mccormack, 11 May 2008, 1:55 am:


... unlike Adobe RGB (which does not utilize any low-level linearized region at all), and ProPhoto RGB (which by definition does at an input value existing 7.105 EV below maximum input):

... and which is stated here by Eric Chan to precisely comport with that ProPhoto RGB specification, but has been measured by Bill Janes using Imatest to appear to not comport:

While it is the case that the perceivable range of human vision when viewing preview displays may be limited to around a 100:1 ratio or so (around 6.64 EV), it seems that the use of the sRGB low-level linearized region (at an input value existing 5.767 EV below maximum input) in Adobe Melissa RGB previews appears to provide something of an over-optimistic picture of image-noise that exists in the lowest (around) 1 EV of perceptability.

Thus, if and when the user happily renders a previewed and processed image in either Adobe RGB 1998 or ProPhoto RGB (particularly as it was measured when implemented in ACR 7.2 by Bill Janes using Imatest) color-spaces, they will find that "Melissa" has (at least to some extent) glossed-over and "fibbed" to them regarding the amount of image-noise present in such images (when those images are presented in either Adobe RGB 1998 or ProPhoto RGB color-spaces).

They will also discover that the numerical values of RGB color readouts at low levels are erroneous.

The only apparent work-around would be to use sRGB color-space as the working color-space (if using Adobe Camera RAW, whereas Lightroom users are not afforded such color-space options).

DM ...

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