Datacolor User to User Support Group.

I have been profiling Canon Pro9000 MK II printing on Kodak Matte Photo Paper, but not able to generate a correct profile. The results (softproof and printout) are always washout.

The washout effect is shown in the following links: softproof and no-softproof . I have double-checked media selection, disable color management at the printer level, and correctly take measurement data. All seem correct. Furthermore, even I use the measurement file "HP B9180 Matte-Art" provided by DataColor, the results are still washout.

Anybody has a clue?

I found something might help to understand why the softproof is washed out: the measured L value of the black patch is too high. 

I use spectro to measure the black surface of the spydercube which looks pretty dark black to me. The measured L value is 10, which is a gray shown on my screen. The L value of the middle gray is about 44.90.

It seems my spectro cannot correctly measure black patch.


--- In datacolor_group@yahoogroups.com, "kk325ic" <kk325ic@...> wrote:

>
> 
> I have been profiling Canon Pro9000 MK II printing on Kodak Matte Photo
> Paper, but not able to generate a correct profile.  The results
> (softproof and printout) are always washout.
> 
> The washout effect is shown in the following links: softproof
> <http://farm9.staticflickr.com/8149/7641568514_9a180e54d4_z.jpg>   and
> no-softproof
> <http://farm8.staticflickr.com/7250/7641569098_c570b63fe0_z.jpg>  . I
> have double-checked media selection, disable color management at the
> printer level, and  correctly take measurement data. All seem correct.
> Furthermore, even I use the  measurement file "HP B9180 Matte-Art"
> provided by DataColor, the results are still washout.
> 
> Anybody has a clue?
>

There may be nothing wrong at all.

If you're profiling matte paper: (just as you've seen with the correct
measurements and profile sample for the HPB9180): the spectro measures
(correctly) a lighter black. Inside SpyderPRINT, the softproof mechanism
softproofs it that way, and you end up with something that looks a bit
washed-out. That doesn't mean the measurements or the profile are "wrong",
however.

(As a rough guideline: for matte paper, the first patch in the target, which
is "perfect" black, usually measures with an L value in the upper teens to
the low 20's at most on matte papers. An L value above, say, 22 or 23 would
be a warning sign that something else is wrong. With some Epsons, the most
common mistake people make is to use photo black ink on matte paper, to save
the horrible ink-swap routine, and the result is a much-too-light black
that's significantly washed out. But with your Canon, that shouldn't be a
problem)

Inside Photoshop (or Lightroom 4), you can use their additional softproofing
controls to turn off black ink simulation; most of your contrast will
pop back; that's the recommended way to properly softproof with any of these
profiles.

If you'd like email me a copy of your measurement file and I'll take a quick
look to see if there are any problems. Use the File:Open Data command in
SpyderPRINT to open your measurement file folder; find the .xml file; then
.zip and email it to me at dmiller@...

David Miller


On Jul 25, 2012, at 6:09 PM, kk325ic wrote:

> I found something might help to understand why the softproof is washed out: the measured L value of the black patch is too high. 
> 
> I use spectro to measure the black surface of the spydercube which looks pretty dark black to me. The measured L value is 10, which is a gray shown on my screen. The L value of the middle gray is about 44.90.
> 
> It seems my spectro cannot correctly measure black patch.
> 
> --- In datacolor_group@yahoogroups.com, "kk325ic" <kk325ic@...> wrote:
> >
> > 
> > I have been profiling Canon Pro9000 MK II printing on Kodak Matte Photo
> > Paper, but not able to generate a correct profile. The results
> > (softproof and printout) are always washout.
> > 
> > The washout effect is shown in the following links: softproof
> > <http://farm9.staticflickr.com/8149/7641568514_9a180e54d4_z.jpg> and
> > no-softproof
> > <http://farm8.staticflickr.com/7250/7641569098_c570b63fe0_z.jpg> . I
> > have double-checked media selection, disable color management at the
> > printer level, and correctly take measurement data. All seem correct.
> > Furthermore, even I use the measurement file "HP B9180 Matte-Art"
> > provided by DataColor, the results are still washout.
> > 
> > Anybody has a clue?
> >
> 
> 
> 

David Miller
Senior Software Developer, Digital Color Solutions
Datacolor

Thanks David for taking your time to help out.

> (As a rough guideline: for matte paper, the first patch in the target, which
> is "perfect" black, usually measures with an L value in the upper teens to
> the low 20's at most on matte papers. An L value above, say, 22 or 23 would
> be a warning sign that something else is wrong. 
I get the measured L value of the first patch about 17. If I adjust the L of the reference black point to 3 in the advanced edit, the softproof result looks much better. 

It seems to me that the measure of the black patch is not very accurate enough. If this is true, SpyderPrint should be able to automatically adjust the black reference point to get the best result.

> 
> If you'd like email me a copy of your measurement file and I'll take a quick
> look to see if there are any problems. Use the File:Open Data command in
> SpyderPRINT to open your measurement file folder; find the .xml file; then
> .zip and email it to me at dmiller@...
Yes, I sent you the measurement file.

On Jul 25, 2012, at 7:32 PM, kk325ic wrote:

> Thanks David for taking your time to help out.
> 
> > (As a rough guideline: for matte paper, the first patch in the target, which
> > is "perfect" black, usually measures with an L value in the upper teens to
> > the low 20's at most on matte papers. An L value above, say, 22 or 23 would
> > be a warning sign that something else is wrong. 
> I get the measured L value of the first patch about 17. If I adjust the L of the reference black point to 3 in the advanced edit, the softproof result looks much better. 
> 
> It seems to me that the measure of the black patch is not very accurate enough. If this is true, SpyderPrint should be able to automatically adjust the black reference point to get the best result.
> 
> > 
> > If you'd like email me a copy of your measurement file and I'll take a quick
> > look to see if there are any problems. Use the File:Open Data command in
> > SpyderPRINT to open your measurement file folder; find the .xml file; then
> > .zip and email it to me at dmiller@...
> Yes, I sent you the measurement file. 
> 

I just responded. Your black patch measurement is fine, and there's no problem
with accuracy. What you're seeing is a lack of being able to turn off "black
ink simulation" in SpyderPRINT's softproof. It's not a bug; just a limitation
as to how our softproof works. It actually behaves exactly as Photoshop's
does, when "black ink simulation" is turned on in Photoshop.

SpyderPRINT's softproof isn't meant to be as advanced and full-featured
as Photoshop's. It's basically a quick way of getting a softproof configured
a particular way (showing the full effect of the black ink measurement) and
it's technically correct that way; as I said, Photoshop's will look the same
when you check the "black ink simulation" box there.

You have one incorrectly measured patch, which I've identified so that you
can fix it. That's not enough to throw off your prints through the profile
as a whole. The next issue is going to be: how you're actually using it when
you print, and we'll continue in our emails about that. If you're printing
through Photoshop, either you're not using the profile correctly, OR, you're
running into a problem with some Canon drivers in which they will absolutely
not work properly when printing with any externally specified profile (including
Canon's own profiles, so this is easy to prove if true) in Photoshop's print
dialog. We'll see....

David Miller
Senior Software Developer, Digital Color Solutions
Datacolor

David,

Thanks for your help. The problem solved. The print out looks good. 

Regards,
Kongbin
--- In datacolor_group@yahoogroups.com, David Miller <dm2363@...> wrote:

>
> 
> On Jul 25, 2012, at 7:32 PM, kk325ic wrote:
> 
> > Thanks David for taking your time to help out.
> > 
> > > (As a rough guideline: for matte paper, the first patch in the target, which
> > > is "perfect" black, usually measures with an L value in the upper teens to
> > > the low 20's at most on matte papers. An L value above, say, 22 or 23 would
> > > be a warning sign that something else is wrong. 
> > I get the measured L value of the first patch about 17. If I adjust the L of the reference black point to 3 in the advanced edit, the softproof result looks much better. 
> > 
> > It seems to me that the measure of the black patch is not very accurate enough. If this is true, SpyderPrint should be able to automatically adjust the black reference point to get the best result.
> > 
> > > 
> > > If you'd like email me a copy of your measurement file and I'll take a quick
> > > look to see if there are any problems. Use the File:Open Data command in
> > > SpyderPRINT to open your measurement file folder; find the .xml file; then
> > > .zip and email it to me at dmiller@
> > Yes, I sent you the measurement file. 
> > 
> 
> I just responded. Your black patch measurement is fine, and there's no problem
> with accuracy. What you're seeing is a lack of being able to turn off "black
> ink simulation" in SpyderPRINT's softproof. It's not a bug; just a limitation
> as to how our softproof works. It actually behaves exactly as Photoshop's
> does, when "black ink simulation" is turned on in Photoshop.
> 
> SpyderPRINT's softproof isn't meant to be as advanced and full-featured
> as Photoshop's. It's basically a quick way of getting a softproof configured
> a particular way (showing the full effect of the black ink measurement) and
> it's technically correct that way; as I said, Photoshop's will look the same
> when you check the "black ink simulation" box there.
> 
> You have one incorrectly measured patch, which I've identified so that you
> can fix it. That's not enough to throw off your prints through the profile
> as a whole. The next issue is going to be: how you're actually using it when
> you print, and we'll continue in our emails about that. If you're printing
> through Photoshop, either you're not using the profile correctly, OR, you're
> running into a problem with some Canon drivers in which they will absolutely
> not work properly when printing with any externally specified profile (including
> Canon's own profiles, so this is easy to prove if true) in Photoshop's print
> dialog. We'll see....
> 
> David Miller
> Senior Software Developer, Digital Color Solutions
> Datacolor
>

On Jul 26, 2012, at 7:37 PM, kk325ic wrote:

> 
> David,
> 
> Thanks for your help. The problem solved. The print out looks good. 
> 
> Regards,
> Kongbin

You're welcome! I was just going to respond to your email - your test
print of our test image set looks fine, so all should be well!

Best regards,

David Miller
Senior Software Developer, Digital Color Solutions
Datacolor