Yes I have an i1, but as a K7 printer my interest is in re-linearising existing K7 curves for an ageing printer. i haven't had the need to create new ones, nor to create any new K3 QTR curves either, although I've done it some years ago. I have a technique for K7 re-linearisation that works pretty darn well for me. So do you I gather, as have others on this forum. No doubt all variations on a theme.
There's no point debating how IJM relinearise their curves, since any opinions that any of us have cannot be either be validated or invalidated. But having compared the K7 master curves with the shipped K7 curves, you'll have a hard time convincing me that they're simply setting the black point and rescaling. If you want to share your own demonstration of how your way of doing this changes the shape of component curves, either here or on your blog, then feel free.
But I do have a couple of questions. In a Lula thread:
http://forum.luminous-landscape.com/index.php?topic=94806.msg797910#msg797910
you advocated using at least chart with at least 51 patches for doing re-linearisation / creating curves. As you noted, these show the real unevenness in certain printer / paper combinations. Even printing several of these and averaging them can still give a linearisation plot with a lot of twists and turns.
Qu 1: Do you average these linearisation plots, to smooth out the little bumps, or use them as-is in creating curves and linearising? I'm not just referring to a single set of 51 patches, but to an average of three of them, in order to average out printer and measurement device variation.
Qu2: Have you noticed any variation between papers as to whether the linearisation plots are smooth or bumpy? I.e., if you do the 51x3 test on three papers ** on the same printer**, do you find one smooth (although with some departure from linearity), another not all that smooth, and the third somewhere in between? If so, what is your interpretation of this?
---In QuadtoneRIP@yahoogroups.com, <richard@...> wrote :
There's no point debating how IJM relinearise their curves, since any opinions that any of us have cannot be either be validated or invalidated. But having compared the K7 master curves with the shipped K7 curves, you'll have a hard time convincing me that they're simply setting the black point and rescaling. If you want to share your own demonstration of how your way of doing this changes the shape of component curves, either here or on your blog, then feel free.
But I do have a couple of questions. In a Lula thread:
http://forum.luminous-landscape.com/index.php?topic=94806.msg797910#msg797910
you advocated using at least chart with at least 51 patches for doing re-linearisation / creating curves. As you noted, these show the real unevenness in certain printer / paper combinations. Even printing several of these and averaging them can still give a linearisation plot with a lot of twists and turns.
Qu 1: Do you average these linearisation plots, to smooth out the little bumps, or use them as-is in creating curves and linearising? I'm not just referring to a single set of 51 patches, but to an average of three of them, in order to average out printer and measurement device variation.
Qu2: Have you noticed any variation between papers as to whether the linearisation plots are smooth or bumpy? I.e., if you do the 51x3 test on three papers ** on the same printer**, do you find one smooth (although with some departure from linearity), another not all that smooth, and the third somewhere in between? If so, what is your interpretation of this?
---In QuadtoneRIP@yahoogroups.com, <richard@...> wrote :
The "problem" that needs solving is how to get from a single master curve to a finished linearized curve with just one printed target. I am not saying exactly how they are doing it, but generally what the method and result is—establishing a limit/black point and adjusting the y values up and down (which is what causes the shape to change).