[argyllcms] Re: Soft-proofing in Photoshop with Argyll - correction to my last post!

  • From: <robert@xxxxxxxxxxxxxxxxxx>
  • To: <argyllcms@xxxxxxxxxxxxx>
  • Date: Wed, 4 Jun 2014 15:01:31 +0100

Hi Bill,


<<I gather that you are converting in Photoshop, from whatever RGB to your
Custom Printer profile, created with Argyll.>>

No, I'm not doing any conversions; I'm just using Photoshop's soft-proofing.
Actually, I've just posted a reply to Graeme giving a (hopefully) clearer
and fuller breakdown of my workflow.


You can download the batch files I'm currently using here: http://
www.irelandupclose.com/customer/softproofing/batch-files.zip  (assuming
you're using Windows).


If you want to try them out you'll have to do a bit of editing otherwise
you'll have things like HPZ3100 as the printer description (not that this
will do much harm).  Also, if you look at p4-make-profile.bat you will see
that I'm generating a whole bunch of profiles, which you most likely won't
want to do!


You can run p-make-full-profile and this will guide you through the whole


I can send you the profiles I'm using, but even though you're using a Solux
lamp this won't be at the same color temperature as mine (there seems to be
quite a variation between the bulbs . have you found that?); also, do you
use Canson Platine?  If not, it's a great paper!


For the Photo Hi Gloss . the Argyll FWA compensation is not working (for me,
at least), so before doing anything further I would like to see if Graeme
can help me out there.  Have you tried it out, and does it work for you?  If
it does, could you check my commands to see if I'm doing anything wrong?


I'm getting a reading of a:12 b:27 off the Hi Gloss with a CCT of 4250K.
With the Platine I'm getting a reading of a: 12, b:31 with a CCT of 4080K,
so there isn't much difference between them. The Solux lamp is giving a CCT
of 4070K, which confirms (to me :-)) that the Platine has no FWA whereas the
Hi Gloss does. The Canson spec says that the Platine has no optical
brightening agents, but it says nothing about this for the PhotoHighGloss
Premium RC, so clearly it does have some.


As you say, with Solux lamps there won't be much fluorescence - however
there will be some.  When I say that the Hi Gloss soft-proof is a bit bluer
than the Platine soft-proof, I do mean just a little bit. But that little
bit should presumably be cancelled out by the FWA compensation, which it


I've thought about modifying the monitor to make it fit the soft-proof
better (the Eizo ColorNavigator has a lovely feature to allow one to do
that, by moving the white point manually), but that would be the last option
for me, only if I can't get the color management to do it for me.  Part of
the reason is that ColorNavigator modifies the calibration/profile when you
move the white point, so the soft-proofing would now be using this new
profile, presumably cancelling out the change!  What I could do is to
abandon soft-proofing altogether and calibrate/profile the monitor to my
Solux lamp and paper - but then I would be using the display way out of its
comfort zone, which doesn't seem a good idea.











From: argyllcms-bounce@xxxxxxxxxxxxx [mailto:argyllcms-bounce@xxxxxxxxxxxxx]
On Behalf Of Roger Breton
Sent: 03 June 2014 21:02
To: argyllcms@xxxxxxxxxxxxx
Subject: [argyllcms] Re: Soft-proofing in Photoshop with Argyll - correction
to my last post!


Wow! Your setup seems to work quite well. 

I wish you would use more descriptive "profile name".


I gather that you are converting in Photoshop, from whatever RGB to your
Custom Printer profile, created with Argyll.


I'm curious to "see" what you are seeing. (I would be interested in
obtaining some of your files, to test them out here).


But, if you are getting a good visual match to your monitor using Canson
Platine, that is no guarantee that the same recipe will yield the same
quality using the Canson Photo HiGloss. That's just the nature of FWA.


One approach that would be worth trying is to calibrated and profiled your
Eizo to the Canson Photo HiGloss visual appearance, as illuminated by the
Solux lamp instead of calibrating and profiling to native. Using your eye
might help take into account some of the fluorescence into consideration.


One thing to ponder, though. Solux is an incandescent light source. As such,
it does not excite much fluorescence from any paper. 


Now, the fact that your paper simulation with Canson Photo HiGloss appears
"bluish" is a good indication that the i1pro is getting an impression from
the substrate that it is VERY blue. Is that the case? Are you getting a
relatively high negative b* value, like -5 or -6 or -7? If that is the case
then it explains why the output profile is fooled into thinking that it
needs to sprinkle so much yellow on the print, to neutralize the bluish
appearance of the substrate.


With FWA, the devil is in the measurement. Always. And measuring with a UV
filter is not the solution.


Best / Roger



From: argyllcms-bounce@xxxxxxxxxxxxx [mailto:argyllcms-bounce@xxxxxxxxxxxxx]
On Behalf Of robert@xxxxxxxxxxxxxxxxxx
Sent: 3 juin 2014 15:23
To: argyllcms@xxxxxxxxxxxxx
Subject: [argyllcms] Re: Soft-proofing in Photoshop with Argyll - correction
to my last post!


Hi Roger,


I'm not sure about the success!  I'm still trying to get this right, and the
reason for my post was to ask whether or not I'm doing the soft-proofing
correctly.  At the moment I have an issue with the FWA compensation which
doesn't seem to work (I have another post about this).


But at any rate, the soft-proofing certainly looks pretty good (especially
on papers with no FWA), whether or not I have the technical side right
(hopefully you or Graeme or someone can set my mind at ease :-)).


Regarding the SoLux lamp.  I don't have your issue because I'm only viewing
an area of around 16"x12" (because I want the print to be the same size as
the soft-proof . as I think that if it is larger or smaller it will affect
how it looks).  So I have the lamp about 1m away, directed downwards, with
'blinkers' on it to focus the beam onto the print.  You can see my (very
experimental!) setup here:


To view large prints I think you would need to use multiple lights arranged
in a square, perhaps, with a diffuser in front to get as constant an
illumination as possible.  Of course there's bound to be some variation
between the lamps. On the positive side, you wouldn't be viewing the print
side-by-side with the monitor in this case, so containing the light wouldn't
be a problem.


I'm really trying hard to get this right (and it isn't easy as there's the
paper, the light, the ink, the monitor, the profiles, Photoshop, Argyll,
ColorNavigator .. enough to give anyone a headache :-)), so any and all
advice would be highly welcome!!





From: argyllcms-bounce@xxxxxxxxxxxxx [mailto:argyllcms-bounce@xxxxxxxxxxxxx]
On Behalf Of Roger Breton
Sent: 03 June 2014 17:15
To: argyllcms@xxxxxxxxxxxxx
Cc: robert@xxxxxxxxxxxxxxxxxx
Subject: [argyllcms] Re: Soft-proofing in Photoshop with Argyll - correction
to my last post!


Hi Robert,


One quick question, about your success with the Solux lamp. 

I have been a faithful user of the 4700K lamp for years.

But I have never achieve a uniform beam?

In my humble experience, the lamp has to be placed quite a distance from the
viewing area to form a reasonably wide uniform lighting area, something like
one meter is not far-fetched. As you know, the fall-off is very rapid,
quickly shift down to 3000K in a span for a few inches from the center beam.
And the new generation of "coated" inside bulb has not alleviate the problem
of beam uniformity for me.

BTW, thank you for the detailed account of your information. Makes tracing
your logic a breeze.


Best / Roger Breton


From: argyllcms-bounce@xxxxxxxxxxxxx [mailto:argyllcms-bounce@xxxxxxxxxxxxx]
On Behalf Of robert@xxxxxxxxxxxxxxxxxx
Sent: 1 juin 2014 18:55
To: argyllcms@xxxxxxxxxxxxx
Subject: [argyllcms] Soft-proofing in Photoshop with Argyll - correction to
my last post!




I'm trying to set up a good soft-proofing workflow using Argyll and I would
really appreciate a bit of help!


To start off with, my work area is set up as follows:

-       Monitor: good quality wide gamut (Eizo CG277), calibrated and
profiled to Native, 6500K, 80cd/m2, Gamma 2.2 (default values for the

-       Profiling: i1Pro2 / i1D3 / CG277-built-in-colorimeter. Eizo
ColorNavigator is used for the monitor (hardware) calibration and profiling
(i1Pro2 as reference).

-       A 4700K SoLux lamp (which actually has a temperature of around 4200K
according to Argyll/i1Pro2) for viewing the image. I adjust the luminance by
moving the lamp forward or backward. The lamps have a good SPD with low UV. 


What I'm doing for soft-proofing (and as I'm a novice at colour management
and Argyll, I have no doubt I'm doing things wrong) is as follows:


-       I run illumread -S -c1 -H PROFILENAME.sp

-       I create two profiles as follows:

o       colprof -v -A"HP" -M"Z3100" -iPROFILENAME.sp -DPROFILENAME -qh

o       colprof -v -A"HP" -M"Z3100" -DPROFILENAME -qh

-       I print using the PROFILENAME_NoSP.icc profile

-       I soft-proof using the PROFILENAME.icc with Simulate Paper White on.


This seems to work really well and the monitor and print match very closely.
HOWEVER . I have no idea if this works just by luck and so can't be relied
on, or if it's technically correct.  Softproofing with PROFILENAME.icc gives
the right paper white and the colors seem to match very well (on two test
prints, one with paper which has no FWA: Canson Platine; and one that does:
Canson Photo HiGloss).  Using the PROFILENAME_NoSP.icc to soft-proof (the
profile used to print) does not give the right paper white or colours (too
blue) . perhaps because the SoLux lamp is well below D50?


My reasoning is that PROFILENAME.icc has the correct illuminant / paper
information so it will soft-proof correctly, but it should not be used to
print (doing a Convert to Profile in Photoshop gives a bluer image for the
Canson Photo HiGloss image (presumably because of the FWA compensation?)).


Many thanks, once again!  




(BTW - the profiles generated using Argyll are really excellent, better than
the Canson-supplied profiles and better than the Z3100-generated profiles).


Other related posts: