QTR-Quadtone RIP

the downside being... if you really want highest possible linearity security, and blending your beautifully linearized curves leaves that a bit iffy.. you have to do the overall of any given settings combo with a real profile of the setup w/ settings. I have to do that for my StudioPrint setups that use spot channels as toners.
I made a 101 patch chart and text reference file for create ICC. The setup obviously has to be very well behaved in the first place so the profiler won't choke, but it works well. To even it out there are a few redundant patches at the dark end for averaging.
If you or anyone else here would like access to it let me know.
Tyler

--- In QuadtoneRIP@yahoogroups.com, Michael King <drmrking@...> wrote:
>
> Tyler, interesting idea. I will keep a more open mind as to how to leverage
> the blending.
> 
> Txs,
> 
> Mike
> 
> On 6 March 2010 18:20, tboleyyh <tyler@...> wrote:
> 
> >
> >
> > one more possibility Mike- have you thought of using the curve blending to
> > help with this? I don't know your inkset, but if you don't need the blending
> > in QTR for hue adjustment, you can use it to blend curves made with other
> > parameter differences. For example, one curve might be made with huge
> > overlaps, one medium, another none. Blend them so the highlights are 100%
> > huge overlaps, mids medium, shadows none... or some such variation..
> > Just a thought, the main idea being that the blend options in the driver
> > can be used in other creative ways than the hue variations for which they
> > were intended...
> > You could even blend a tritone with a hextone.. for more complex dotwork,
> > etc.. you get the idea..
> >
> > Tyler
> >
> > --- In QuadtoneRIP@yahoogroups.com <QuadtoneRIP%40yahoogroups.com>,
> > Michael King <drmrking@> wrote:
> > >
> > > Tyler,
> > >
> > > > Mike, you are up to some very cool things, really impressive. Very
> > > > interested in hearing about the ink.
> > > >
> > > I like it more than I expected. Paul Roark has done all the hard work, I
> > am
> > > just standing on his shoulders. Works great in R1900. I am using18% Eboni
> > as
> > > toner in 3/4 tones.
> > >
> > > > I think I know what you mean regarding my question... I'd import a
> > > > different toner curve based on tests, which requires another start
> > > > linearization, and then the proportion of toner to grays in various
> > tonal
> > > > areas has changed from that which I had used to base the curve edit
> > upon...
> > > > round and round. That's ok, I accept that for this particular project,
> > if
> > > > I'm getting the picture.
> > > >
> > > Yes that's exactly the issue, round and round we go and there is no way
> > to
> > > predict the outcome, though you can guesstimate better after the first
> > > linearization.
> > >
> > > > Let us know how your work goes, really intriguing. Can you overlap
> > > > progressively less moving down to the low end? So you have K only at
> > 100%?
> > > > Everything get's a bit jammed up down there and it's hard to control...
> > > >
> > > I am not sure, I think not for maximum smoothness. So I may just generate
> > > two curve sets one for maximum smoothness based on Cone style overlapping
> > > ink curves and one for maximum dmax based on QTR curves.
> > >
> > > Mike
> > >
> > > >
> > > > Tyler
> > > >
> > > >
> > > > --- In QuadtoneRIP@yahoogroups.com <QuadtoneRIP%40yahoogroups.com><QuadtoneRIP%
> > 40yahoogroups.com>,
> >
> > > > Michael King <drmrking@> wrote:
> > > > >
> > > > > Hi Tyler,
> > > > >
> > > > > I think the answer to your question is yes, but ...
> > > > >
> > > > > Having experimented over the last few months with HP-C6 aiming for
> > > > neutral
> > > > > and needing to warm it up in the 3/4 tones to achieve that, I got
> > very
> > > > > frustrated with the inflexibility of QTR to re-linearise after
> > editing a
> > > > > toning curve.
> > > > >
> > > > > The problem is that you can add a toner as you describe below, but
> > when
> > > > you
> > > > > linearise the tone curve will change and you can't edit it further,
> > you
> > > > can
> > > > > only start again with a new tone curve. There is no way, I can see,
> > to do
> > > > > sequential linearizations in QTR and that's a real pain.
> > > > >
> > > > > My solution was to write my own linearization program that takes a
> > set of
> > > > > QTR curves (actual ink values in Quad file) and linerarizes them.
> > This
> > > > let's
> > > > > me edit a toner curve ink values (often manually in Excel) and
> > > > relinearize
> > > > > them again. It even allows me to look at the predicted re-linearized
> > ab
> > > > > curves before I print them and decide if I need to tweak them
> > further.
> > > > >
> > > > > Through testing this HP-C6 approach I also discovered that the QTR
> > curve
> > > > > model that doesn't use many overlapping inks, is suspectable to
> > > > > posterization. I had a print that posterized and someone suggested is
> > was
> > > > > possibly a problem with QTR curve model.
> > > > > So I created another curve model that was more like Cone's model
> > where
> > > > the
> > > > > inks overlap much more and indeed the posterization problem was
> > solved.
> > > > BUT
> > > > > the cost of this overlapping ink approach is reduced dmax. On Canon
> > Rag
> > > > 310
> > > > > the dmax drops from L=15.3 to L=16 due to the other inks polluting
> > the
> > > > > darkest ink. So you've got to choose between the limitations of the
> > two
> > > > > approaches.
> > > > >
> > > > > Learn't a lot about QTR in the last few months.
> > > > >
> > > > > Mike
> > > > >
> > > > >
> > > > >
> > > > >
> > > > > On 5 March 2010 23:06, tboleyyh <tyler@> wrote:
> > > > >
> > > > > >
> > > > > >
> > > > > > after going through all the references I can find, it looks to me
> > like
> > > > the
> > > > > > Photoshop curve import function in the Windows version with curve
> > > > creation
> > > > > > is the most straightforward, current, and viable approach to making
> > > > unique
> > > > > > toner setups..
> > > > > > Would you QTR experts here agree?
> > > > > > Thanks,
> > > > > > Tyler
> > > > > >
> > > > > >
> > > > > >
> > > > >
> > > > >
> > > > > [Non-text portions of this message have been removed]
> > > > >
> > > >
> > > >
> > > >
> > >
> > >
> > > [Non-text portions of this message have been removed]
> > >
> >
> >  
> >
> 
> 
> [Non-text portions of this message have been removed]
>