SWOP Proof Certification, TRxxx Characterization Data
SWOP Proof Certification, TRxxx Characterization Data
- Subject: SWOP Proof Certification, TRxxx Characterization Data
- From: Klaus Karcher <email@hidden>
- Date: Mon, 27 Oct 2008 02:49:01 +0100
Hi all,
I'm sorry, but I have to vent my anger about the trouble to obtain
reliable information about SWOP/SNAP/GRACoL Characterization Data, proof
evaluation criteria, tolerances and measurement conditions.
Tender-minded people should better skip this posting as it might offend
their sensibilities ;-)
I am writing proof evaluation software at the moment and thought it
would be a good idea to draft it as universal and maintenance-low as
possible.
Therefore I considered to use the ICC Characterization Data Registry
<http://www.color.org/registry2.xalter> as a authoritative,
well-maintained online repository for characterization data, the more so
as the "technical requirements" listed in the Paragraph
"Characterization data structure" suggest that it might be even possible
to parse the registry automatically.
This attempt succeeds in most cases: when one follows the "Reference
name" links in <http://www.color.org/drsection1.xalter>, one will get a
HTML page containing a brief description and at least one link to a
"electronic data source": a simple, CGATS-like text file. Parsing these
files is a piece of cake.
But when one follows the links to current[1] SWOP, SNAP or GRACoL data,
one will *not* find CGATS files (note the "CGATS ..." in the reference
names!). The "electronic data source" links for *all* those "CGATS..."
files leads to the *same* NEPES page titled "TOOLS, REFERENCES & BEST
PRACTICES": <http://www.npes.org/standards/tools.html>. It contains
references to all the world and his brother, amongst others also the
sought-after CGATS files ... one might think ... only to be
disappointed: There are separate zip files for every printing condition,
each with a PDF and an CSV file. I am taking on any bet that that no one
of the responsible parties tried to open them with any *non-English*
EXCEL version: MS EXCEL as well as Apple Numbers fail miserably to
interpret them in any locales where the comma is the decimal separator.
One has to use dirty, lengthy tricks to read the files correctly or has
to use Open Office (which is more clever in this regard) to convert
them. Common CGATS parsers fail to interpret these files as well as a
matter of course. Even after the files are converted from CSV to CGATS,
some parsers might dash already against the wrong "shebang" in the first
line of TR 3, 5 and 6 ("CGATS.17" instead of "CGATS17").
I remember that there once /were/ ordinary CGATS files -- I guess I
found them at <http://gracol.org/resources/>, but most of the links on
this page are dead now. The same applies to the links to
<http://www.npes.org/standards/CGATS-SNAP-TR002-2007.zip> in the ICC
Profile Registry.
Therefore I had to hit the dirt road, translate the CSV files to CGATS
and bid farewell to the idea of a machine-readable authoritative online
repository.
The next topic on my list was to obtain information about criteria and
tolerances for contract proof evaluation. No problem for the Ugra/Fogra
media wedge: My open questions concerning the new MediaWedge v3 were
clarify by Andreas Kraushaar in the ECI mailing list within less than
half an hour.
And for SWOP/SNAP/GRACoL proofs? Well, there are nice "streaming
WebeXes" (recorded web tutorials) where on can "learn why SWOP proofs
are G7 proofs" or "how to make a SWOP proof" and there is a link to the
/free/ IDEAlliance Control Strip -- but if one simply wants to know
which patches have to meet which criteria, one gets wrapped up with
obfuscation, misinformation and a confusing number of "ADS" (Application
Data Sheets) for all kinds of proofing systems. Why the heck should I
use different evaluations for different proofing systems when the proofs
are made for the same printing condition?
The IDEAlliance Control Strip seems to be a big step forward, but
apparently not even the proofing system providers seem to know how to
number the patches nor their exact CMYK and Lab values[2].
No wonder in view of this search results:
<http://www.google.com/search?q="all+fifty-four+(54)+patches"+site:http://www.swop.org/newADS/&filter=0>
Open any of them, scroll to the page about the IDEAlliance ISO12647-7
Control Strip: "This ADS Attachment provides the CIELab data for all
fifty-four (54) patches contained in the new strip" -- but all
subsequent tables list only 52 Lab values!
How about extracting the missing patches from the characterization data
I painfully converted? -- Nope, since 5 of the IDEAlliance Control Strip
patches are not part of the characterization files!
So back to the IDEAlliance Control Strip: The ZIP archive
<http://www.gracol.org/resources/IDEAlliance_ISO12647-7_2007_Color_Control_Strip.zip>
contains a ReadMe.pdf. Quote: "The file 'ISO12647-7_ControlStripV5.txt'
contains the exact CMYK percentages typed into the original Illustrator
document." -- but there is no file ISO12647-7_ControlStripV5.txt in the
archive. "The actual reference values will normally be expressed in
CIELab units, and determined within suitable software by interpolation
of an ICC profile or characterization data set." hmmm... so the result
of the proof evaluation depends on the quality of the ICC profile or
interpolation procedure used to determine the aim values? "Colorimetric
reference files are not provided by IDEAlliance." And what are these
"TRxxx..._AimData.csv" files in the archive good for if they aren't
reference files?
I think this was more then enough trouble for one weekend. Maybe the
best idea is not to give a damn about the SWOP/IDEAlliance proof
evaluation, stick to the Fogra MediaWedge instead and assume that a
proof validated with the Fogra MediaWedge and the ICC characterization
Data for the corresponding printing condition is also a valid
SWOP/SNAP/GRACoL contract proof. Eventually both methods refer to ISO
12647-7.
Klaus Karcher
----
[1] I don't get steamed up about the "ransom" for CGATS TR 001 any
longer since it has been outdated for a long time.
[2] see <http://www.swop.org/newADS/ADS_GMG_Epson_x880_SWOP5.pdf>:
GMG Table:
- Patch names: A1, B1, ... Z1, 1A1, A2, B2, ... 1A2
- CMYK values of last Patch: 78.82 70.45 70.45 100
- Lab values for last Patch: 9.94 0.37 0.48
IDEAlliance Strip:
- Patch names: A1, A2, ... A27, B1, B2, ... B27
- CMYK values of last Patch: 80 70 70 100
- Lab values for last patch (B27): ???
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Colorsync-users mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden