The journey towards Universal Characterization Data (UCD) 1
Agenda 1. Why UCDs 2. Current approaches 3. Development of Fogra characterisation data sets 4. Technical background 5. "ISO 12647-2:2013+" versus "ISO 15339-2 2013" on coated paper 6. Outlook 2
1. Why UCDs Digital colour management solutions need characterization data Char.-data are basically CMYK LAB tables (if present, spectral data) Unified characterization data = internationally agreed upon aims are important for: unambiguous (and not regional) recommendations for print production (PDFXready, Ghent Working Group, Adobe CS Suites, etc) Easy communication of print content data across the world Reducing re-targeting (mostly DeviceLink) transformations many more 3
2. Current (and past) approaches WG3 meeting in Tokyo (2007) 2010, letter from David Zwang (Chairmen of GWG) toward IST TC130 Proposals: Basically char.-data that reflects the printing aims defined internationally (ISO 12647-2) or nationally (IDEAlliance, ANSI TR 21) 4
3. Fogra datasets Development of Fogra data: TVI Adjustment of: CIELAB Colours FOGRA1 1 Press Run FOGRA11 Averaging FOGRA27 t FOGRA39 C=M=Y ISO 12647-2:2004 FOGRA51 beta C=M=Y=K ISO 12647-2:2013 5
4. Technical background ISO 12647-2 approach??????? G7 calibration Creative work TR 16 approach ("G7") 6
4. Technical background Offset printing can be defined by knowing the solid coloration and the TVI (tone response) IDEAlliance is providing char.-data based on "G7", defined in TR 16 (CGATS. 21) since 2010 The NPDC (near neutral) definition results in char.-data where colorimetrical tone values can be extracted. The differences between "Equal-TVI" (for CMY or CMYK) ISO 12647-2 and Near Neutral often result in small colour differences (avg E00 <= 2) However there are printing conditions (such as non periodic, FM) where the differences are more significant There are systematic differences for some printing conditions which makes it impossible to find a consensus for all printing conditions (but not for a subset) for printing on PT1/2 the approaches can be averaged ("salomonic data") 7
4. ISO 15339 & ISO 12647-2 ("FOGRA51beta") ISO 12647-2/3 correspondents Source: McDowell ISO 12647-3 PC7 PC5 PC6 PC2 PC1 "FOGRA39+" PC= Printing Condition 8
5. Comparison of CRPC6 with FOGRA51beta 15339-2 (2013) 12647-2:2013 latest version Source: Dr. Bestmann 9
5. Comparison of ISO 15339-2 with ISO 12647-2 Colour differences for all 1617 patches (ISO 12642-2) 10
5. How to go from XYZ to densitometrical tone values? 1. Compare col. TVI with dens. TVI based from the same sample 2. Use a single scalar (value) to scale the col. TVI curve to match the dens. TVI curve 3. Analyse the values by means of different/typical substrates scalars CMYK: 1.25; 1.05; 0; 0 Dens. TVI of FOGRA39 Col. TVI of FOGRA39 11
5. Comparison of ISO 15339-2 with ISO 12647-2 Comparison of colorimetric TVI curves with adjustment factors estimated from ISO 12647-2 with its well known densitometric TVI 15339-2 Large differences in black. Acceptable differences in CMY. 12647-2 process control implication: - printing condition depending RIP corrections - Colour Server 12
5. Differences in the black channel CRPC4 - PC6 (SC) Large TVI differences! CRPC6 - PC1 (PT 1/2) 13
5. Comparison: Results G7 grey balance approaches (naturally) result in higher black curves (than CMY) Equal TVI has been updated with ISO 12647-2 revision from C=M=Y (and sometimes higher K) to C=M=Y=K to ease process control ("one value for Dot Gain (TVI)") Differences between ISO 12647-2 (Equal TVI) and TR21 (near neutral) approaches can be sometimes attributed to be within typical printing variations Characterization Data should reflect a process 1:1 (no need for colour servers clut transformations) Universal exchange data shall not benefit one over the other process calibration CRPC's favour TR21 printing conditions G7 calibrated print will result in a good proof/print match (providing a CRPC6 Proof) ISO 12647-2 calibrated presses needs additional RIP corrections and for high accuracy a colour server Outlook Discussion 14