Eric Stevens
On Fri, 18 Apr 2014 19:43:46 -0400, PeterN <peter.newnospam@verizon.net>wrote:

PeterN
On 4/18/2014 6:39 PM, Eric Stevens wrote:

Eric Stevens
On Fri, 18 Apr 2014 11:06:59 -0400, PeterN <peter.newnospam@verizon.net>wrote:

PeterN
On 4/17/2014 8:32 PM, Eric Stevens wrote:

Eric Stevens
On Thu, 17 Apr 2014 18:15:50 -0400, nospam <nospam@nospam.invalid> wrote:

nospam
In article <2gi0l99msbj34q772pfa1jr8slbbdls2i0@4ax.com>, Eric Stevens <eric.stevens@sum.co.nz>wrote:

PeterN
But since you like quick and dirty processing, why don't you use the many color curves in LAB. Ater all you can apply a curve into a channel in LAB and get far more precise results than you can in ACR.

nospam
i don't like working in lab and do not find any advantage whatsoever, both in results and workflow.

dan marguilis is one of the biggest proponents of the workflow, and after reading his book, i was very unimpressed. it all can be done in rgb just as easily and likely with better results because you skip two conversions (which are not lossless).

Eric Stevens
But doesn't the Adobe color engine work in Lab anyway?

nospam
internally, but that's not the same as making a conversion of the image twice.

Eric Stevens
What are the two conversions?

PeterN
While it's rare that one can be certain what he's referring to: my vote would be converting the image to LAB & back to RGB. Two or three quick keystrokes in each direction.

Eric Stevens
As opposed to the following example where an RGB input is converted to CMYK for output to a printer and back to RGB for review on a screen. This requires numerous trips backwards and forwards through the Adobe (Lab) colour engine.

RGB(source) ->Lab (Adobe engine space) ->CMYK(printer) ->Lab ->RGB(screen).

PeterN
I usually have only one round trip, two at the most.

Eric Stevens
That applies to most ordinary situations.