QTR-Quadtone RIP

Thanks Tom
Yes I noticed that the pre-linearisation step wedge was considerably 
darker than post- in the mid-ranges and the pre-linearised Lab line sags 
considerably beneath the ideal. Why do you think that is? Could it be to 
do with the high ink limit?

Incidentally, once I changed that penultimate value and created the 
linearised curve the resulting plot is almost perfect. The shape of the 
individual ink curves though is not very smooth but I think part of the 
reason for that is using 51 linearisation steps rather than 21.

Howard


Tom Moore wrote:
> Howard
> 
> I plotted your density values to have a look and the values are fairly
> linear (i.e. they are fairly far from the LAB curve). One thing you might
> try is to increase the gamma value of the gray curve. I'm not sure if you've
> changed it yet, but assuming it's at its default value of 1, you could try
> something in the 1.5 to 2 range. That would bring the curve closer to the
> target LAB values before linearizing and might make it easier for the
> linearizer to work its magic.
> 
> I'm not sure what others do, but I go through several iterations for each
> curve adjusting the highlight, shadow and gamma values to bring the curve as
> close as I can (reasonably close anyway) to LAB, before linearizing. The
> curveeval spreadsheet I posted makes it convenient to save and plot curves
> over several iterations and gives you a sense for the effect of changing
> values. It's set up for the 21 step wedge, but that's a lot more convenient
> to use until you get the curve close anyway.
> 
> At some point early on with QTR Roy told me that it was better to bring the
> curve closer to the target profile using the shadow and highlight (and gamma
> if necessary). This way the linearization would be a small adjustment of the
> curve rather than a major move. I've not seen much discussion of this
> recently, but I still do it out of habit. It's fairly quick once you get a
> feel for what the numbers do. The main thing to remember is that larger
> numbers make the density curve lower. That's true for all three values,
> although the gamma changes have larger effect.
> 
> Tom Moore
> 
>  > -----Original Message-----
>  > From: QuadtoneRIP@yahoogroups.com [mailto:QuadtoneRIP@yahoogroups.com] On
>  > Behalf Of Howard Shaw
>  > Sent: Friday, October 28, 2005 12:12 PM
>  > To: QuadtoneRIP@yahoogroups.com
>  > Subject: Re: [QuadtoneRIP] Error message: Invalid linearize curve - not
>  > constantly increasing
>  >
>  > I have ink limit at 90%, boost at 95%, shadow at 12 & overlap at 16. The
>  > paper is Permajet Alpha. The limit is quite high so I'm not sure how
>  > much effect black boost is having anyway. I chose the high limit as the
>  > dmax was at its greatest (1.72 absolute) on the 90% patch.
>  >
>  > I'm using x-rite 810 for measurements and I've re-checked them and they
>  > are consistent.
>  >
>  > I'll have a play around with reducing the boost & the overlap as I don't
>  > seem to get quite the dmax I did on the calibration strip.
>  >
>  > Howard
>  >
>  >
>  >
>  >
>  >
>  > John Moody wrote:
>  > > The last step has a much higher density increase than all others,
>  > > effectively asking the linearization to fix a sharp bend at the end of
>  > > the curve.
>  > >
>  > > Do you have black boost at a fairly high level?  If so, you may get
>  > > smoother results cutting back on black-boost, and upping the ink limit
>  > > to achieve the 1.69 Dmax you are getting.  A high shadow value can bend
>  > > the curve like that as well, but since it's only the last step, I'm
>  > > guessing black boost.
>  > >
>  > > Also, if you scanned the patches with an i1, you might go back and
>  > > re-read the last several in patch mode to see that you get consistent
>  > > values.  I have a hypothesis that the dark patch accuracy degrades with
>  > > increasing scan speed, but I have not seen others corroborate that.
>  > >
>  > >
>  > >
>  > > Best regards,
>  > >
>  > > John Moody
>  > >
>  > >
>  > >
>  > > -----Original Message-----
>  > > *From:* QuadtoneRIP@yahoogroups.com
>  > > [mailto:QuadtoneRIP@yahoogroups.com]*On Behalf Of *Howard Shaw
>  > > *Sent:* Thursday, October 27, 2005 5:51 PM
>  > > *To:* QuadtoneRIP@yahoogroups.com
>  > > *Subject:* [QuadtoneRIP] Error message: Invalid linearize curve - not
>  > > constantly increasing
>  > >
>  > >
>  > >
>  > > I was making a linearisation today with the following values from a
>  > > 51-step and got the above error message and the curve was not created:
>  > > 0.06
>  > > 0.08
>  > > 0.1
>  > > 0.12
>  > > 0.15
>  > > 0.17
>  > > 0.2
>  > > 0.23
>  > > 0.26
>  > > 0.29
>  > > 0.33
>  > > 0.36
>  > > 0.41
>  > > 0.45
>  > > 0.48
>  > > 0.53
>  > > 0.57
>  > > 0.61
>  > > 0.65
>  > > 0.69
>  > > 0.72
>  > > 0.75
>  > > 0.79
>  > > 0.83
>  > > 0.87
>  > > 0.91
>  > > 0.95
>  > > 0.98
>  > > 1.01
>  > > 1.04
>  > > 1.09
>  > > 1.12
>  > > 1.14
>  > > 1.18
>  > > 1.19
>  > > 1.22
>  > > 1.24
>  > > 1.28
>  > > 1.31
>  > > 1.34
>  > > 1.37
>  > > 1.4
>  > > 1.42
>  > > 1.46
>  > > 1.49
>  > > 1.52
>  > > 1.55
>  > > 1.57
>  > > 1.6
>  > > 1.62
>  > > 1.69
>  > >
>  > > All the values are increasing albeit not "constantly". If I change the
>  > > penultimate value from 1.62 to 1.63 the curve creation works. Any 
> reason
>  > > this should be so?
>  > >
>  > > Howard

>