Re: X-Rite Passport Inaccuracies
I had sent this to Tom without CC'ing the list.....not sure what the protocol is in this case, but I thought I'd send it along in case someone else wanted to chime in as well. Thank you for all the input thus far. Tom Thank you very much for the detailed response. If I'm not mistaken there are two issues at play. 1) First, there is the problem of matching color across different light sources (including 100% CFL's), for which I understand you to say that's outside of the capabilities of the Passport and other DNG profiles. I'm better off using the HSL sliders to get the color right there. To do that I would probably set some color sampler points on swatches and get the RGB values close between different shots of a color checker under each light source. However, it seems like Lab would be a better color space, so I could look at color only, and that's not available in Lightroom. Otherwise I'm not sure of a reliable way to do that. 2) Then there is my test, which I first conducted in daylight (overcast) with the 60d and 5d. In that case the Passport proved to be no better than the Adobe Standard Profile in correcting colors or matching colors between cameras. Am I mistaken in thinking this is basically the test you were suggesting I do? Or is there something special about warm daylight? Here is the data from my test-- The color values after correction with the Passport (given in a,b values of Lab): 5d: green= -51,39 red= 69,41 yellow= 4,86 blue= 27,-65 60d: green=-42,34 red= 60,39 yellow = 4,79 blue= 25,-53 The color values prior to correction with the Passport, using Adobe Std (given in a,b values of Lab): 5d: green= -47,36 red= 61,37 yellow= 1,87 blue= 18,-59 60d: green= -38,31 red= 56,33 yellow= 2,78 blue= 16,-46 I cannot discern any appreciable change in the difference between the color values of the 5d and 60d using the Passport. Thanks, Justin Krug
Hi where are you getting your Lab values from if you are using Lightroom? Best Matthew Ward On 20 Sep 2013, at 20:43, Justin Krug <jkrug@hotmail.com> wrote:
I had sent this to Tom without CC'ing the list.....not sure what the protocol is in this case, but I thought I'd send it along in case someone else wanted to chime in as well. Thank you for all the input thus far.
Tom
Thank you very much for the detailed response. If I'm not mistaken there are two issues at play. 1) First, there is the problem of matching color across different light sources (including 100% CFL's), for which I understand you to say that's outside of the capabilities of the Passport and other DNG profiles. I'm better off using the HSL sliders to get the color right there. To do that I would probably set some color sampler points on swatches and get the RGB values close between different shots of a color checker under each light source. However, it seems like Lab would be a better color space, so I could look at color only, and that's not available in Lightroom. Otherwise I'm not sure of a reliable way to do that.
2) Then there is my test, which I first conducted in daylight (overcast) with the 60d and 5d. In that case the Passport proved to be no better than the Adobe Standard Profile in correcting colors or matching colors between cameras. Am I mistaken in thinking this is basically the test you were suggesting I do? Or is there something special about warm daylight? Here is the data from my test--
The color values after correction with the Passport (given in a,b values of Lab): 5d: green= -51,39 red= 69,41 yellow= 4,86 blue= 27,-65 60d: green=-42,34 red= 60,39 yellow = 4,79 blue= 25,-53 The color values prior to correction with the Passport, using Adobe Std (given in a,b values of Lab): 5d: green= -47,36 red= 61,37 yellow= 1,87 blue= 18,-59 60d: green= -38,31 red= 56,33 yellow= 2,78 blue= 16,-46
I cannot discern any appreciable change in the difference between the color values of the 5d and 60d using the Passport.
Thanks,
Justin Krug
_______________________________________________ Do not post admin requests to the list. They will be ignored. Colorsync-users mailing list (Colorsync-users@lists.apple.com) Help/Unsubscribe/Update your Subscription: https://lists.apple.com/mailman/options/colorsync-users/matthew%40matthewwar...
This email sent to matthew@matthewwardphotography.com
Lr 5 offers L*a*b* color readouts as an option. Right-click on the histogram and choose "Show Lab Color Values" from the context popup menu. The values are calculated with a D50 reference white. Eric On Sun, Sep 22, 2013 at 5:24 AM, Matthew Ward < matthew@matthewwardphotography.com> wrote:
Hi where are you getting your Lab values from if you are using Lightroom?
Lr 5 offers L*a*b* color readouts as an option. Right-click on the histogram and choose "Show Lab Color Values" from the context popup menu. The values are calculated with a D50 reference white.
Eric
Is it using RelCol (like Photoshop) from the source RGB? I sure wish Photoshop provided Absolute Colorimetric like ColorThink. Andrew Rodney http://www.digitaldog.net/ On Sep 22, 2013, at 9:27 AM, Eric Chan <madmanchan@gmail.com> wrote:
Andrew, there is no rendering intent involved in the L*a*b* color readouts. The image pixel values are simply converted from (Linear) ProPhoto RGB to L*a*b* using standard CIE conversion routines. You may be thinking of doing conversions to other spaces (to another RGB space), in which the rendering intent option is involved. On Sun, Sep 22, 2013 at 1:23 PM, Andrew Rodney <andrew@digitaldog.net>wrote:
Lr 5 offers L*a*b* color readouts as an option. Right-click on the histogram and choose "Show Lab Color Values" from the context popup menu. The values are calculated with a D50 reference white.
Eric
Is it using RelCol (like Photoshop) from the source RGB?
I sure wish Photoshop provided Absolute Colorimetric like ColorThink.
Andrew Rodney http://www.digitaldog.net/ On Sep 22, 2013, at 9:27 AM, Eric Chan <madmanchan@gmail.com> wrote:
_______________________________________________ Do not post admin requests to the list. They will be ignored. Colorsync-users mailing list (Colorsync-users@lists.apple.com) Help/Unsubscribe/Update your Subscription:
https://lists.apple.com/mailman/options/colorsync-users/madmanchan%40gmail.c...
This email sent to madmanchan@gmail.com
On Sep 22, 2013, at 12:19 PM, Eric Chan <madmanchan@gmail.com> wrote:
Andrew, there is no rendering intent involved in the L*a*b* color readouts. The image pixel values are simply converted from (Linear) ProPhoto RGB to L*a*b* using standard CIE conversion routines.
Sorry, I was in error about RGB to Lab values, I am referring to CMYK to Lab. And sorry if this is OT. If I convert ProPhoto RGB to SWOP V2, Photoshop reports the Lab values as 0/0/1, LR reports them as 0.3/0.3/0.0. ColorThink reports them as 13.5/-0.07/0.93. I asked Chris Cox about this, he told me the CMYK to Lab conversions were using RelCol, I believe ColorThink reports them as Abs Colorimetric. Just wondering why. Andrew
participants (4)
-
Andrew Rodney
-
Eric Chan
-
Justin Krug
-
Matthew Ward