Yahoo Groups archive

QTR-Quadtone RIP

Index last updated: 2026-04-28 23:12 UTC

Thread

Photoshop Step-Wedges: Grayscale/HSB confusion ?

Photoshop Step-Wedges: Grayscale/HSB confusion ?

2012-01-17 by kenleegallery

Let's say we make a new image in 16-bit grayscale with a Gamma 2.2 profile, and create a step-wedge with 11 steps. We get an image whose K values go 0, 10, 20... 100. Fine.

However, if we mouse over the other patches, we see some discrepancies between the K value and the HSB B value. For example, the middle patch shows a K value of 50%, but the B value is 58%. Similarly, where K = 10%, B = 92%. 

Why aren't these two measures simply the compliment of one another ?  

Which system is appropriate for making a step wedge, when calibrating with QTR ?

Re: [QuadtoneRIP] Photoshop Step-Wedges: Grayscale/HSB confusion ?

2012-01-17 by Roy Harrington

Ken,

Using the Photoshop eye drop sampler has a lot more to it than one
would think.  In general if you "look"
at values that are the same format as the file you see the actual
values in the file (or a simple calculation).
In other works for an RGB file if you look at R G B values you do see
what's in the file.  For a grayscale
file, K values are a simple calculation.  (internally 8-bit files are
0=black to 255=white, 16-bit files are
0=black to 32768=white).  K = (255-value)*100/255.   This is pretty
simple and probably what you already
expected.   But whenever you look at values other than the same format
the calculation go though
color management.  This means that the embedded profile (if one) and
the settings under Color Settings...
are all involved in what the eyedropper reports.  For instance if you
look at R G B with a grayscale file
you see the icc conversion of the internal gray value from the
embedded profile (or working gray if its untagged)
to the working RGB profile.  It may appear that you a looking at the
internal 8-bit values but you are not.

I'm not very familiar with using HSB but like LAB you are seeing the
conversion from the grayscale profile
to those values.   So the reason you are seeing those B values is
because that's the "definition" of the
profile that is being used.  BTW, I thought I try this in PS and the B
values you mention actually
correspond to Dot Gain 20% not Gray Gamma 2.2 -- check out your file
and Color Settings...

That's the long answer -- the short answer is: for step wedge stuff
don't look at HSB or LAB or anything
but K values.   Stepwedge files are meant to be artificial -- exact
values that you want to send to the
driver, you don't want any color management involved.   The posterize
command that you probably
used is also completely profile ignorant so that's just fine for wedges.

--

This seems like a good place to mention -- the K vs R G B eyedropper
stuff is worth understanding
but its often annoying nevertheless.  So one thing that makes this
less of a problem is to pick
grayscale and RGB working profiles that are "compatible" i.e. have the
same gamma structure.
This makes the conversions "look" like you expected.

The most common pairs are:
  Gray Gamma 2.2 and AdobeRGB
  Generic Gray Gamma 2.2 and sRGB
  Gray Gamma 1.8 and ProPhotoRGB
(note that Gray Gamma 2.2 and Generic Gray Gamma 2.2 are distinctly different)

Roy


On Tue, Jan 17, 2012 at 9:28 AM, kenleegallery <kenlee333@...> wrote:
> Let's say we make a new image in 16-bit grayscale with a Gamma 2.2 profile, and create a step-wedge with 11 steps. We get an image whose K values go 0, 10, 20... 100. Fine.
>
> However, if we mouse over the other patches, we see some discrepancies between the K value and the HSB B value. For example, the middle patch shows a K value of 50%, but the B value is 58%. Similarly, where K = 10%, B = 92%.
>
> Why aren't these two measures simply the compliment of one another ?
>
> Which system is appropriate for making a step wedge, when calibrating with QTR ?
>
>
>
> ------------------------------------
>
> Yahoo! Groups Links
>
>
>



-- 
Roy Harrington
roy@...
www.harrington.com

Re: Photoshop Step-Wedges: Grayscale/HSB confusion ?

2012-01-17 by kenleegallery

You're right: my image was (accidentally) in Dot Gain 20%.

You mention Generic Gray Gamma 2.2, which seems rather important on the
Mac if we have a recent version of Photoshop and OS X 10.6 or higher. 
In a previous discussion http://www.dpug.org/forums/sitemap/t-2255.html
<http://www.dpug.org/forums/sitemap/t-2255.html%20> , Harlan said this
(emphasis mine):

"Roy Harrington noted that whatever gray space you use in CS5 and OS
10.6 that, on printing, it getsconverted into the "generic GG 2.2"
space. So unless your default gray space in CS5 is this one you will
have trouble onprinting.  The consequence for me was that my digital
negatives would plateau in the finaltwo steps causing blocked highlights
in my prints. Working in GG 2.2 I tested adozen modified QTR profiles
and many gray curves, getting more and moreextreme, to try to get rid of
this plateau without success. Wasted nearly aweek of darkroom time and a
lot of palladium. I thought it was due to somequirk in the spectrum of
my new light bulb in my Nuarc....
I see the take-home message as to either print in CS3 or work in CS5 in
the"generic gg 2.2" space that your print ends up in anyway
(haven'ttried that yet but plan to). From my experience it could be
really tough tore-calibrate for printing in CS5 without the gray space
coordination workingfor you."

Can we create images in Generic Gray Gamma ? Or rather, can we assign or
attach that profile  to a Grayscale image ? Perhaps I'm missing
something, but I don't see it as a choice on my copy of CS5.

Noting that sRGB is compatible, would it be wiser (for QTR purposes) to
work in that color space, even though the image is 16-bit grayscale ?
Another advantage is that the same image could be presented on the web
directly.

Thanks !!

(I've been trying to make a Pt/Pd profile - working in Gray Gamma 2.2 -
and finding it hard to get paper white without blowing the high values.
With so many variables in the process, I can't help but wonder if this
is one of the areas I have overlooked)

--- In QuadtoneRIP@yahoogroups.com, Roy Harrington <roy@...> wrote:
>
> Ken,
>
> Using the Photoshop eye drop sampler has a lot more to it than one
> would think.  In general if you "look"
> at values that are the same format as the file you see the actual
> values in the file (or a simple calculation).
> In other works for an RGB file if you look at R G B values you do see
> what's in the file.  For a grayscale
> file, K values are a simple calculation.  (internally 8-bit files are
> 0=black to 255=white, 16-bit files are
> 0=black to 32768=white).  K = (255-value)*100/255.   This is pretty
> simple and probably what you already
> expected.   But whenever you look at values other than the same format
> the calculation go though
> color management.  This means that the embedded profile (if one) and
> the settings under Color Settings...
> are all involved in what the eyedropper reports.  For instance if you
> look at R G B with a grayscale file
> you see the icc conversion of the internal gray value from the
> embedded profile (or working gray if its untagged)
> to the working RGB profile.  It may appear that you a looking at the
> internal 8-bit values but you are not.
>
> I'm not very familiar with using HSB but like LAB you are seeing the
> conversion from the grayscale profile
> to those values.   So the reason you are seeing those B values is
> because that's the "definition" of the
> profile that is being used.  BTW, I thought I try this in PS and the B
> values you mention actually
> correspond to Dot Gain 20% not Gray Gamma 2.2 -- check out your file
> and Color Settings...
>
> That's the long answer -- the short answer is: for step wedge stuff
> don't look at HSB or LAB or anything
> but K values.   Stepwedge files are meant to be artificial -- exact
> values that you want to send to the
> driver, you don't want any color management involved.   The posterize
> command that you probably
> used is also completely profile ignorant so that's just fine for
wedges.
>
> --
>
> This seems like a good place to mention -- the K vs R G B eyedropper
> stuff is worth understanding
> but its often annoying nevertheless.  So one thing that makes this
> less of a problem is to pick
> grayscale and RGB working profiles that are "compatible" i.e. have the
> same gamma structure.
> This makes the conversions "look" like you expected.
>
> The most common pairs are:
>   Gray Gamma 2.2 and AdobeRGB
>   Generic Gray Gamma 2.2 and sRGB
>   Gray Gamma 1.8 and ProPhotoRGB
> (note that Gray Gamma 2.2 and Generic Gray Gamma 2.2 are distinctly
different)
>
> Roy
>
>
> On Tue, Jan 17, 2012 at 9:28 AM, kenleegallery kenlee333@... wrote:
> > Let's say we make a new image in 16-bit grayscale with a Gamma 2.2
profile, and create a step-wedge with 11 steps. We get an image whose K
values go 0, 10, 20... 100. Fine.
> >
> > However, if we mouse over the other patches, we see some
discrepancies between the K value and the HSB B value. For example, the
middle patch shows a K value of 50%, but the B value is 58%. Similarly,
where K = 10%, B = 92%.
> >
> > Why aren't these two measures simply the compliment of one another ?
> >
> > Which system is appropriate for making a step wedge, when
calibrating with QTR ?
> >
> >
> >
> > ------------------------------------
> >
> > Yahoo! Groups Links
> >
> >
> >
>
>
>
> --
> Roy Harrington
> roy@...
> www.harrington.com
>



[Non-text portions of this message have been removed]

Re: Photoshop Step-Wedges: Grayscale/HSB confusion ?

2012-01-17 by kenleegallery

Correction - It is available on the Mac version of Photoshop CS5.

That being said, would it be best then to do all work in that setting ? 

--- In QuadtoneRIP@yahoogroups.com, "kenleegallery" <kenlee333@...> wrote:
Show quoted textHide quoted text
>
> You're right: my image was (accidentally) in Dot Gain 20%.
> 
> You mention Generic Gray Gamma 2.2, which seems rather important on the
> Mac if we have a recent version of Photoshop and OS X 10.6 or higher. 
> In a previous discussion http://www.dpug.org/forums/sitemap/t-2255.html
> <http://www.dpug.org/forums/sitemap/t-2255.html%20> , Harlan said this
> (emphasis mine):
> 
> "Roy Harrington noted that whatever gray space you use in CS5 and OS
> 10.6 that, on printing, it getsconverted into the "generic GG 2.2"
> space. So unless your default gray space in CS5 is this one you will
> have trouble onprinting.  The consequence for me was that my digital
> negatives would plateau in the finaltwo steps causing blocked highlights
> in my prints. Working in GG 2.2 I tested adozen modified QTR profiles
> and many gray curves, getting more and moreextreme, to try to get rid of
> this plateau without success. Wasted nearly aweek of darkroom time and a
> lot of palladium. I thought it was due to somequirk in the spectrum of
> my new light bulb in my Nuarc....
> I see the take-home message as to either print in CS3 or work in CS5 in
> the"generic gg 2.2" space that your print ends up in anyway
> (haven'ttried that yet but plan to). From my experience it could be
> really tough tore-calibrate for printing in CS5 without the gray space
> coordination workingfor you."
> 
> Can we create images in Generic Gray Gamma ? Or rather, can we assign or
> attach that profile  to a Grayscale image ? Perhaps I'm missing
> something, but I don't see it as a choice on my copy of CS5.
> 
> Noting that sRGB is compatible, would it be wiser (for QTR purposes) to
> work in that color space, even though the image is 16-bit grayscale ?
> Another advantage is that the same image could be presented on the web
> directly.
> 
> Thanks !!
> 
> (I've been trying to make a Pt/Pd profile - working in Gray Gamma 2.2 -
> and finding it hard to get paper white without blowing the high values.
> With so many variables in the process, I can't help but wonder if this
> is one of the areas I have overlooked)
> 
> --- In QuadtoneRIP@yahoogroups.com, Roy Harrington <roy@> wrote:
> >
> > Ken,
> >
> > Using the Photoshop eye drop sampler has a lot more to it than one
> > would think.  In general if you "look"
> > at values that are the same format as the file you see the actual
> > values in the file (or a simple calculation).
> > In other works for an RGB file if you look at R G B values you do see
> > what's in the file.  For a grayscale
> > file, K values are a simple calculation.  (internally 8-bit files are
> > 0=black to 255=white, 16-bit files are
> > 0=black to 32768=white).  K = (255-value)*100/255.   This is pretty
> > simple and probably what you already
> > expected.   But whenever you look at values other than the same format
> > the calculation go though
> > color management.  This means that the embedded profile (if one) and
> > the settings under Color Settings...
> > are all involved in what the eyedropper reports.  For instance if you
> > look at R G B with a grayscale file
> > you see the icc conversion of the internal gray value from the
> > embedded profile (or working gray if its untagged)
> > to the working RGB profile.  It may appear that you a looking at the
> > internal 8-bit values but you are not.
> >
> > I'm not very familiar with using HSB but like LAB you are seeing the
> > conversion from the grayscale profile
> > to those values.   So the reason you are seeing those B values is
> > because that's the "definition" of the
> > profile that is being used.  BTW, I thought I try this in PS and the B
> > values you mention actually
> > correspond to Dot Gain 20% not Gray Gamma 2.2 -- check out your file
> > and Color Settings...
> >
> > That's the long answer -- the short answer is: for step wedge stuff
> > don't look at HSB or LAB or anything
> > but K values.   Stepwedge files are meant to be artificial -- exact
> > values that you want to send to the
> > driver, you don't want any color management involved.   The posterize
> > command that you probably
> > used is also completely profile ignorant so that's just fine for
> wedges.
> >
> > --
> >
> > This seems like a good place to mention -- the K vs R G B eyedropper
> > stuff is worth understanding
> > but its often annoying nevertheless.  So one thing that makes this
> > less of a problem is to pick
> > grayscale and RGB working profiles that are "compatible" i.e. have the
> > same gamma structure.
> > This makes the conversions "look" like you expected.
> >
> > The most common pairs are:
> >   Gray Gamma 2.2 and AdobeRGB
> >   Generic Gray Gamma 2.2 and sRGB
> >   Gray Gamma 1.8 and ProPhotoRGB
> > (note that Gray Gamma 2.2 and Generic Gray Gamma 2.2 are distinctly
> different)
> >
> > Roy
> >
> >
> > On Tue, Jan 17, 2012 at 9:28 AM, kenleegallery kenlee333@ wrote:
> > > Let's say we make a new image in 16-bit grayscale with a Gamma 2.2
> profile, and create a step-wedge with 11 steps. We get an image whose K
> values go 0, 10, 20... 100. Fine.
> > >
> > > However, if we mouse over the other patches, we see some
> discrepancies between the K value and the HSB B value. For example, the
> middle patch shows a K value of 50%, but the B value is 58%. Similarly,
> where K = 10%, B = 92%.
> > >
> > > Why aren't these two measures simply the compliment of one another ?
> > >
> > > Which system is appropriate for making a step wedge, when
> calibrating with QTR ?
> > >
> > >
> > >
> > > ------------------------------------
> > >
> > > Yahoo! Groups Links
> > >
> > >
> > >
> >
> >
> >
> > --
> > Roy Harrington
> > roy@
> > www.harrington.com
> >
> 
> 
> 
> [Non-text portions of this message have been removed]
>

Move to quarantaine

This moves the raw source file on disk only. The archive index is not changed automatically, so you still need to run a manual refresh afterward.