Datacolor User to User Support Group.

On Jan 2, 2013, at 2:59 PM, ronlangara <ronshizu@...> wrote:

> Systems:
> Mac OSX 10.8.2, 
> Aperture 3.4.3, 
> Colorsync 4.8.0, 
> Adobe Color Printer Utility V 1.0
> Adobe Photoshop CS4
> Epson R1800 driver 9.00, connected directly via firewire
> Epson Printer Utility 4
> Datacolor SpyderPrint 4.2.3
> Paper Epson Exhibition Fibre Gloss F 13X19
> 
> 1) Looking at the 729 patch target in either Aperture or the Adobe Color Printer Utility the image is incorrectly read/displayed, showing mostly B+W bars with patches displaced horizontally from where they should be. In SpyderPrint and Photoshop they display correctly. Is there a way to get Aperture to open/display the target correctly? Failing that is there a way (in Photoshop or some other program) to translate it to a different file format that Aperture can handle? I would like to be able to do this so as to complete the entire calibration workflow in the same manner that I use for my photos to ensure proper application of the colour management tools and profiles.
> 

The target image files are nothing special. They're compressed TIF files to keep the size down.
They were saved from an earlier version of Photoshop that way and there's nothing technically
wrong with the files (they've always opened in Photoshop correctly).

(SpyderPRINT itself doesn't use these files; inside the software, it always dynamically generates
the target in the Target window instead of relying on a file)

There's no reason to open those target files in Aperture, in any case - you can't use it to print
them without color management. (So: don't do that...:-)

Adobe Color Print Utility should be able to open those files without a problem (remember, they
were saved from Photoshop). I've never heard of anyone else running into this problem. If you
want, you can open a target file from inside Photoshop and resave it - just make sure that you
save it as UNTAGGED and without applying any color conversions to it.

> 2) I have printed the 729 patch target twice on 13X19 paper above. The first time in the SpyderPrint utility itself,

That method is fine.

> the second time in Photoshop CS4.

That's also fine. CS4 was the last version of Photoshop to internally support target print (the
"No Color Management" setting in the Print dialog. This is vastly preferable to attempting to
print the targets through Adobe Color Print Utility because you can adjust print sizing and
placement to get the largest possible patches.

>  Each time the settings were set to turn off the colour management and the printer's internal management capabilities. Both prints look identical in terms of colour hue, saturation and density.
> 

That's as expected. (good). They should look the same.

> 3) I read the patches in the patch mode as opposed to the strip mode as I found the strip mode more difficult with the longer strips on the 13X19 inch paper. I added some low friction tape to the base of the device to help it slide across the paper more smoothly with less risk of marring the paper surface. I got good color gradations first time round with each. I found it helpful to crop the target edges a little in Photoshop so as to print the patches a little larger and thus have less risk of catching the edge of the target with the patch reader. It wold be nice if the plastic body of the colorimeter were a lighter colour so as to allow it do be visually differentiated from the target. It would also be good if it had the center lines marked on all four sides, as opposed to the two it does currently. This would be especially advantageous when reading the target in a low light environment with aging eyes.
> 

That all sounds good to me.

> 4) The resulting profiles gives substantially improved colour over the one provided by Epson. However I find that it is still a little on the dark side, perhaps in the order of 0.25 to 0.5 stops and perhaps a little lower in contrast (some of this is most likely the calibrated LCD display versus paper print difference that can only be approximated). Is it recommended to adjust this in the profile or should I be redoing the profile?
> 

Most likely reason is that you're using a modern LCD display which is going to always
look brighter than the print, even when calibrated, because it's running at or near
full brightness. Unless you use a very bright proofing light with the proper color temperature
to preview your print, you're going to run into this. (Think of the paper white vs. the screen
white, which on a bright LCD is always going to be much brighter than how the print appears)

The best way to get closer screen-to-print matching is to calibrate your display at a lower
luminance - 120 cd/m2 or so is about right.

If you want, you can tweak the brightness on the Advanced Editing screen in SpyderPRINT before
building the profile and make less-dark prints, but this still doesn't completely address what the
real problem probably is - a properly calibrated, but too-bright, display.

> 5) The ways I currently make the managed profiles work with Aperture is:
> a)tell Aperture to let the printer manage the colour. I then tell Colorsync to use the profiles I created and the R1800's internal colour management is then automatically turned off by the Epson driver (it is perhaps more correct to say it just passes the data to the print engine modified only by the Spyderprint profile).
> 
> or
> 
> b)tell Aperture to manage the colour and use the profile, which in turn tells Colorsync and the Epson driver to not manage the colour. This is currently my preferred approach. However I have had occasions when it seemed to give incorrect results, i.e. not applying or double application of the profiles. That may however have been operator error.
> 
> Which of these is the "recommended" or preferred approach?
> 

It's hard to say. If the workflow is correct, you should always get the same results. (b) is probably
more correct and it should also be consistent - if it's not, it implies that either Aperture isn't
doing things consistently or that it's difficult for you to consistently set up the controls in the
printing workflow. (a) is more of an unknown, and I think you've got the logic in how it would
work backwards. If you tell the printer driver to manage color, then the data being passed to
the driver is supposed to be completely unmodified by anything, and it gets converted at the
driver level (using ColorSync) because you've told ColorSync to use the SpyderPRINT profile
as a custom output configuration. The disadvantage of (a) is that you don't get to choose
a rendering intent and you don't really know which one ColorSync is using.


David Miller
Senior Software Developer, Digital Color Solutions
Datacolor