Olympus-OM
[Top] [All Lists]

Re: [OM] Bibble=>AfterShot [was Sagelight editor]

Subject: Re: [OM] Bibble=>AfterShot [was Sagelight editor]
From: Chuck Norcutt <chucknorcutt@xxxxxxxxxxxxxxxx>
Date: Wed, 18 Jan 2012 22:43:29 -0500
It was my understanding (from where I don't know) that ACR and some 
other converters try to go the manufacturer's converters one better by 
extrapolating beyond the clipped channel.  How they do that and how 
successful they are I don't know.

Chuck Norcutt


On 1/18/2012 5:12 PM, Ken Norton wrote:
> In Lightroom, it depends on which conversion algorithm you are using
> too. You can choose from the legacy ACR engine or the newer
> RawShooter-based engine. They do render things differently. Of course,
> there is the DCRAW based converters...
>
> Highlight recovery is an interesting animal. Over the past few months
> I've been really digging down into it trying to figure out what is
> going on as well as trying to figure out why the in-camera converters
> lose this headroom. Turns out that in every camera I've
> tested--including the 5D, 7D, D2X and my cameras, the in-camera
> converter clips right at the point where color integrity fails. When
> we perform highlight recovery, we lose color integrity.
>
> What essentially is going on is that the three (or four) sensels used
> in the Bayer conversion do not clip at the same point. Especially when
> working in an environment where we are in an off white-balance. A
> derived 100% "white" is 254.254.254 but a derived 100% red is 254.0.0
> and a derived blue is 0.254.0 and a derived green is 0.0.254. A
> derived 100% magenta is 254.0.254.
>
> Where things get really fuzzy, though, is when we work with the
> derived colors of magenta, cyan and yellow. The way these colors are
> mixed together requires that all sensels clip at the same point. If
> they don't, these three colors will shift when performing a highlight
> recovery. More often than not ACR will just turn it to gray. With most
> details we don't really notice because there are other pixels
> intermingled within that recovered area that provide color to the
> image.
>
> In regards to monochrome images, highlight recovery is just
> fine--highly desirable, in fact. This recovery does give us another
> stop to work with. But when working with color images the last stop of
> recovered highlights are rarely correct. The camera manufacturers just
> clip these off and thereby keep the color integrity intact.
>
> Why wouldn't the sensels all clip at the same time? They would if you
> are shooting only at the native WB of the sensor.
>
> AG
-- 
_________________________________________________________________
Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
Archives: http://lists.thomasclausen.net/mailman/private/olympus/
Themed Olympus Photo Exhibition: http://www.tope.nl/

<Prev in Thread] Current Thread [Next in Thread>
Sponsored by Tako
Impressum | Datenschutz