On 19 Jun 2013, at 11:26 AM, Vincent Wan <wan@...> wrote: > I am thinking of the .quad files as specifying how Lab L values correspond to density by controlling the amounts of ink. The .quad files contain a 16-bit number for each of 256 levels for each ink. I would love if Roy or anyone else would jump in here an clarify this for me. The getting started file in the CurveDesign folder of the current Mac distribution mentions a Curve-Info file but I can's find it so I'm guessing here. Yes, that's right. A .quad file can be looked at as a series of X-Y tables, with the inputs along the X axis, expressed as 8-bit integers (256 levels), and the outputs along the Y value, expressed as a 16-bit integers (65536 levels). The file consists of one or more "blocks" of 256 numbers. There is one block per channel (i.e., ink), and the blocks are listed in the order described in the first line of the file (like "## QuadToneRIP K,C,M,Y,LC,LM"; I believe the comments, starting with "#", are ignored by QTR itself.) The numbers in each block are the 16-bit Y values that correspond to the X value that's implied in the sequence number (e.g., line #1 of a block corresponds to an X value of 0). So if a block starts with: 0 17 34 ... then it could be interpreted as points in an X-Y table as such: 0,0 1,17 2,34 ... I can confirm by my own experimentation that you can indeed generate your own .quad files, without needing all the complexity (some may say power) of the QTR curve generator. You have to be careful the files are compatible -- make sure the line endings are correct (I'm seeing CR/LF as a line delimiter) and that the files have the right permission, and, of course, are in the right place. Hope this helps. --John
Message
Re: [QuadtoneRIP] Help calculating ideal Lab L values
2013-06-20 by John Labovitz
Attachments
- No local attachments were found for this message.