On Feb 18, 2015, at 9:34 PM, kjrslr@... [datacolor_group] <datacolor_group@yahoogroups.com> wrote:
>
> Some quick background. Windows 64. Spyder3Print 4.2.2. Epson 4800 w/custon inkset from Cone. I use the DataColor Spectrocolormeter 1005. The last profile I made was Oct 2011 and I still use the profile today with Canson Platine Photo Rag-great consistant output from this profile. I generally print through Lightroom.
>
> Recently I have been wanting to create a color profile for Canson Baryta Photographique (which I use for B&W w/Piezography in a seperate printer). Last month, I made sure I had a good nozzle check & printed some images(on the Platine) for a club meeting. No problems. So I then opened up S3P and steped through the process of making the targets(on the Canson Baryta) to profile using the 729 EZ w/extended greys (9 letter size sheets). I set the sheets aside for a month(in a drawer) out of sunlight. Last night I opened up S3P and stepped through the process to read the targets to create a profile. When it came to making the print(from the image in S3P) to evaluate the image looked as it should. I stepped through the printer set up dialog settings w/No Color Adjustment, Photo Luster 260, etc but when I hit the print button, the image looked like there was a red gel over everything. I made a print and it printed that way. It had a title in the lower left coner: SPTemp730213.icm, Relative Colormetric (with the date) 2/18/2015, 4:46PM-but it was printed on the 17th.
The most likely mistake you've made is: you printed the target sheets for this new profile incorrectly.
From what you're saying:
1. In the past, you printed target sheets back in 2011 (correctly, we'll assume), measured, built a profile,
and that profile worked correctly. That's fine. It has nothing to do with your second paragraph, "Recently".
2a. Now on to "Recently". You did a nozzle check (good!), then printed the target sheets (but may have
printed them -incorrectly-, with color management enabled, and this is your logic flaw) and put them
in a drawer for a month.
2b. A month later, you took those sheets out of the drawer, measured and built a profile, which doesn't
print correctly (reddish). What you've missed is the possibility (probability!) that those target sheets weren't
printed correctly. It doesn't matter whether you protected them from sunlight or not. If you printed them
wrong a month ago (color managed), then you'll never be able to build a good profile from them and
every print you'd make from using them for a profile will be flawed.
>
> I thought I did something wrong and went back through the help file but didn't find anything. After I got home from work tonight, I re read the targets, gave it a new name w/todays(18FEB15) date and continued on to make another print of the S3P image. Same thing as before.
See above. It doesn't matter if you remeasure target sheets that were printed incorrectly. If the target
prints are wrong, you will NEVER BE ABLE TO MAKE A GOOD PROFILE FROM THEM no matter how many
times you measure. :-)
> When I chose the target (dated 18FEB15) to print, stepped through the printer set up process and once again the image changed from the normal look to one that was the same as last night. Red.
Same comments as above.
> I then went through the process of making adjustments and got the red from the image and named what I thought would be a profile. I then printed it out. The print came out as before, with a red tinge and with none of the corrections I had made
Same comments as above. The most likely problem is, you're continually trying to build a profile
from a CALIBRATED target print, which is wrong. It will never work.
Assuming I'm right about this, it is not, as Paolo Cavestro insists in a later response here, an
"unresolved DataColor problem". It would be from printing the target incorrectly. This same
situation exists in every printer profiling product on the market that has ever existed from
any company. If you print the target wrong, profiling from the incorrect target print will never
work properly.
> Oddly,the title for the profile that I named was not printed but the EXACT same title in the lower left corner (including the date,correct this time)as from the first print I made the night before. I feel like it's double profiling, but how can that be? Neither PS or LR was running.and I've never had this kind of a problem before.
You're mixing up 2 things in the profiling process: the MEASUREMENT file (which is the only
thing that SpyderPRINT stores and works with) and the PROFILE that you eventually build
from the measurements (which is an exported-only file and which is what shows up in the
LR and Photoshop popups). When you run SpyderPRINT again and go to its screen of
"existing measurement files" and select one: you're reloading the MEASUREMENTS you had
made for that target originally. You're not selecting or using any profile that you may have
actually built from them, with or without additional adjustments from the Advanced screen.
If you select a measurement file in SpyderPRINT, then step through to the softproof screen
(prior to actually saving a profile) and then print test images from there, you'll simply be
using an on-the-fly (stored in memory) temporary profile that's built from the loaded
measurements plus any tweaks you've made to the "recipe" if you go into the Advanced
editing screen. You won't be using any actual ICC printer profile that you may have made
previously in earlier sessions of using SpyderPRINT.
When you print the test images from SpyderPRINT that way, it's no surprise that you would
still be getting the red cast (just as you were getting, when using the profile you built from
these and then printing from LR or PS), because (see all of the above comments): if the
target sheets you've been using as the basis for all of this are printed incorrectly, nothing
is going to work properly, and you're consistently incorrect printing behavior (red) is
what I'd expect.
*****
So, let's get to the bottom of all of this. I'd like you to do the following and email me
a couple of things to look at so I can confirm my suspicions. I want you to send me
your MEASUREMENT FILES (not any actual ICC profiles) from the (a) 2011, and (b) "Recent".
Please follow my instructions below exactly and send me 2 measurement files (with
.XML extensions). Don't send me ICC profiles. :-)
1. Run SpyderPRINT
2. Use the File:Open Data command. That will open the folder that contains your
SpyderPRINT measurement files. They're all small text files with .xml extensions.
3a. Find the measurement file you created back in 2011, that you'd been using to
print successfully, and attach it to your email to me.
3b. Find your "Recent" measurement file, that one from this latest set of target prints
that's not working correctly, and also attach it to your email to me.
4. Send the email to me at dmiller@datacolor.com
5. After I've looked at these (should be "today", Friday 2.20.15, if you get them to me
within the next few hours), I'll report back my findings.
6. Assuming I'm right, the simple answer to this is going to be: you'll need to start
over, reprint the sheets (with color management disabled), remeasure, and build another
profile. And the way to prevent this from happening again will be simple: when you're
building printer profile, it's just as critically important to print with the correct driver
settings (no color management) as with doing a nozzle check; and you need to be
able to visually tell whether a target print was done "correctly", or not. A correctly
printed target (no color management) is going to be dark, "heavy", saturated, probably
color cast, particularly when using 3rd party inks. An INCORRECTLY printed target
(color managed) is going to look "nice", "pretty", lighter, with visibly adjusted colors
near the primaries. With the EZ targets (which you're using), on the first sheet, the
first row of patches are the primaries. The blue patch should be a fully inked blue,
not color managed, and it should be very dark and saturated. That's the easiest
one to use when looking for trouble. If the target print is color managed (calibrated)
and therefore wrong, the blue patch will be lighter, prettier, and not fully inked.
David Miller
Senior Software Developer, Digital Color Solutions
Datacolor