QTR-Quadtone RIP

I am in the process of running the ink limitation procedure and am having difficulty 
determining the first patch in the 21 step tablet "where it appears that all the paper is 
covered with ink". It seems to me when trying to follow that guidline that there is too 
much room for interpretation. My determinations can be all over the lot. 

Can anyone suggest a more precise guideline? Also how do you choose the ink limit to use 
if you get widely varying results for the different ink channels? 

Any suggestions will be appreciated.

Howard

Howard

First let me say that I don't think the patch you choose is that critical -
at least it hasn't seemed to be for curves I've created. That said, there
are some who want to squeeze the last bit of density out of each ink and try
to maximize each ink's cutoff point.

Other than using your eyeball, you can use a densitometer or scanner. Print
the separation page and then measure each patch along the area where the ink
seems to be reaching its peak density. The patches increase in density near
the start (lower levels) and then gradually flatten out or drop in density
then pick a patch at or near the tip of the curve.

With respect to using different limits for each channel, you certainly can
if you wish. Again, the idea is to maximize each ink and hopefully reduce
the dots near the transition to the next most dense ink. To measure the
relative density with different limits, you must reprint the inkseparation
chart with each unique limit. Unfortunately you cannot set individual limits
for each ink when printing the inkseparation page.

Steve Kale posted a number of messages regarding separate ink limits. I'm
not sure whether it was this list or the BW Printing list.

Tom Moore

> -----Original Message-----
> From: QuadtoneRIP@yahoogroups.com [mailto:QuadtoneRIP@yahoogroups.com] On
> Behalf Of steinbe2003
> Sent: Tuesday, August 02, 2005 2:09 PM
> To: QuadtoneRIP@yahoogroups.com
> Subject: [QuadtoneRIP] Hardware Ink Limit Calibration
> 

...

> Can anyone suggest a more precise guideline? Also how do you choose the
> ink limit to use
> if you get widely varying results for the different ink channels?
> 
> Any suggestions will be appreciated.
> 
> Howard
> 

...

Hi Howard,

I might just be repeating what others have written, but I'll send this anyway in case it 
might help.

I'm pretty new to this whole process and ran into the same situation.  So, in regard to 
being new at it, if anyone sees anything that I suggest, that might be a little off or could 
be clarified further, don't hesitate to do so.

What I've found is that if your assessing the ink limits by eye, you're going to want to look 
for an area where the paper is covered and subsequent patches level off as far as getting 
darker.  Granted that still leaves room for interpretation, but it might get you a little 
closer.

Now, if you have access to a densitometer, a spectrophotometer or at least a reasonably 
decent scanner with an image editing program capable of working in a Lab color space, or 
a scanner capable of reading L values (sometimes there is a "densitomer" feature capable 
of doing this, that can be used in the scanning process) you can approach it a little more 
"scientifically."  Basically what you would want to do is measure either densities or L values 
of the patches, then determine numerically, where the values level off.  I personally have 
been using my Epson 4990 scanner and Photoshop to do this.  I basically just scan the 
image into Photoshop, convert the image to the Lab color space, then do the reading with 
the color sampling tool set to average the value of a 5x5 pixel area.

As for setting different limits, I've only been able to do so, when creating the final profile. 
So in order to determine the relative values, ie: dark grey = 50% black, I've just taken 
whatever patch I determined to be the ink limit for the different gray values and compared 
that to a print out of a temporary, black only profile (created by indicating only one "gray" 
ink, and setting the rest as unused), since I haven't been able to do the separation page 
with a new general ink limit (actually if anyone could fill me in on that one, it would be 
helpful).  However, I should note that, following this procedure I've had great results, in 
that you can then determine whatever limits you want for each ink and maximize your ink 
usage.  For instance, I'm using an Epson 1280 with Piezotone inks, and have regularly 
found that the black ink, maxes out long before the other inks, so I create the black only 
profile with the black set to say 55%, print it, and then compare that (by scanning into 
photoshop) to the original ink separation printout to find the relative values with the inks 
that often max out around 85-90%..

Anyway, hope all that helps.

Regards,

Eric

I would just add two things:

1st is that almost anything you do gets adjuste,
d ie. undone by  the linearization, so don't sweat 
it too much.

2nd, that said, the adjustments of the linearization 
assume that adjustments to the ink produce 
linear changes to the density (to some extent). So, you don't 
want for example to choose an ink limit that leaves the 
65-90 steps practically the same. Even if 90 is better
than 65. You are better off making it  so that you can 
see a healthy change in density at each step. The boost on 
the black will usually get you most if not all of your dmax
back. 

The biggest goal is the smoothness of the pre-linearization
ramp. No flat spots (particularly in 90+ steps).

hope that helps.
Costa

--- In QuadtoneRIP@yahoogroups.com, "brigsby707" <brigsby707@c...> wrote:
> Hi Howard,
> 
> I might just be repeating what others have written, but I'll send this anyway in case it 
> might help.
> 
> I'm pretty new to this whole process and ran into the same situation.  So, in regard to 
> being new at it, if anyone sees anything that I suggest, that might be a little off or could 
> be clarified further, don't hesitate to do so.
> 
> What I've found is that if your assessing the ink limits by eye, you're going to want to 
look 
> for an area where the paper is covered and subsequent patches level off as far as 
getting 

> darker.  Granted that still leaves room for interpretation, but it might get you a little 
> closer.
>

Sorry, it's been a long day. Not sure anyone could decode
what I was saying..


> I would just add two things:
> 
> 1st is that almost anything you do gets adjusted
>  ie. undone, by  the linearization, so don't sweat 
> it too much.

fixed punctuation on that one. In general as long
as you pick a reasonable limit it just doesn't matter.
That's why Roy doesn't even offer individual limits
anymore.
 
> 2nd, that said, the adjustments of the linearization 
> assume that adjustments to the ink produce 
> linear changes to the density (to some extent). So, you don't 
> want for example to choose an ink limit that leaves the 
> 65-90 steps practically the same. Even if 90 is better
> than 65. You are better off making it  so that you can 
> see a healthy change in density at each step. The boost on 
> the black will usually get you most if not all of your dmax
> back. 
> 
> The biggest goal is the smoothness of the pre-linearization
> ramp. No flat spots (particularly in 90+ steps).
> 
> hope that helps.
> Costa

The linearization takes whatever you have produced (and
measured with the densitometer) and tries to scale it back
to have the proper (e.g., 2.2 gamma) curve. This might be
something like 1.7, 1.5, 1.4 for the first few values. Imagine
that you have instead, 1.7, 1.6, 1.5. Your values are a too 
high. So the linearization scales back the ink value some
for each. However, imagine that you have your ink limit very
high so that all the steps above 60% look the same. Now if the 
linearization tries to scale back by say 20%, it might have a very
small effect on density, (just as moving back from say the 80%
step to 60% shows only a small difference). So, better to
make sure that the linearization has something to work with-
ie stay in the mostly linear portion of the curve. 

You can use the highlight and shadows adjustments to similar
effect in the step after setting the limits.

hope that is at least slightly more useful.

Costa

--- In QuadtoneRIP@yahoogroups.com, "brigsby707" <brigsby707@c...> 
wrote:
> Hi Howard,
> 
> I might just be repeating what others have written, but I'll send 
this anyway in case it 
> might help.
> 
> I'm pretty new to this whole process and ran into the same 
situation.  So, in regard to 
> being new at it, if anyone sees anything that I suggest, that might 
be a little off or could 
> be clarified further, don't hesitate to do so.
> 
> What I've found is that if your assessing the ink limits by eye, 
you're going to want to look 
> for an area where the paper is covered and subsequent patches level 
off as far as getting 
> darker.  Granted that still leaves room for interpretation, but it 
might get you a little 
> closer.
> 
> Now, if you have access to a densitometer, a spectrophotometer or 
at least a reasonably 
> decent scanner with an image editing program capable of working in 
a Lab color space, or 
> a scanner capable of reading L values (sometimes there is 
a "densitomer" feature capable 
> of doing this, that can be used in the scanning process) you can 
approach it a little more 
> "scientifically."  Basically what you would want to do is measure 
either densities or L values 
> of the patches, then determine numerically, where the values level 
off.  I personally have 
> been using my Epson 4990 scanner and Photoshop to do this.  I 
basically just scan the 
> image into Photoshop, convert the image to the Lab color space, 
then do the reading with 
> the color sampling tool set to average the value of a 5x5 pixel 
area.
> 
> As for setting different limits, I've only been able to do so, when 
creating the final profile. 
> So in order to determine the relative values, ie: dark grey = 50% 
black, I've just taken 
> whatever patch I determined to be the ink limit for the different 
gray values and compared 
> that to a print out of a temporary, black only profile (created by 
indicating only one "gray" 
> ink, and setting the rest as unused), since I haven't been able to 
do the separation page 
> with a new general ink limit (actually if anyone could fill me in 
on that one, it would be 
> helpful).  However, I should note that, following this procedure 
I've had great results, in 
> that you can then determine whatever limits you want for each ink 
and maximize your ink 
> usage.  For instance, I'm using an Epson 1280 with Piezotone inks, 
and have regularly 
> found that the black ink, maxes out long before the other inks, so 
I create the black only 
> profile with the black set to say 55%, print it, and then compare 
that (by scanning into 
> photoshop) to the original ink separation printout to find the 
relative values with the inks 
> that often max out around 85-90%..
> 
> Anyway, hope all that helps.
> 
> Regards,
> 
> Eric

Seems we have the same settings : 1280 + Piezo. Besides, Eric sounds 
French : are you French. I am.

I've started experimenting QTR about 2 weeks ago (shame on me I did 
not yet pay my fee, but I intend to once I end the experimentation 
and start producing prints) with PiezoBW ICC inks, HWmatte and EEM 
(I'll go for Velvet once I feel I understand it all), 1280 and an X-
rite spectro dropping mesures in Colorshop X.

I have been so far relatively successful in creating a fairly 
reasonably acceptable curve. I've tested Epson Heavyweight matte : 
very easy to get a curve and EEM : far more complicated, but I admit 
even for color profiling this paper is a nightmare. I have not yet 
gone in to the blackboost (so I just apply the 10-15% more than ink 
limit rule), the highlight-shadow fine tuning (just use 8 and 4 as 
per Roy's recommendations) and I'm still struggling a bit 
undertsanding the latest ICC features (no problem to create the 
profile, just need to really understand the way it has been be 
applied). 

What I discovered on both papers is that the Black ink limit has to 
be fairly low if you want to get a smooth grayscale : the additional 
ink will only bring troubles to linearise as the black end of the 
separationink mesures will get very messy. I ended with a 1.62 
density which is enough for me and a great smoothness which was my 
initial goal. On Piezo NeutralWarm and HW matte general ink limit was 
measured at 55, with EEM at 70 (which I assume is still too high), 
then C is 35 and M 11, Y 5. Based on these values I have a fairly 
nice grayscale though in the 80-85% of the setpwedge I feel there is 
a small bump, I need to re-do the curve on EEM with general ink limit 
at 65 to see if the smoothness improves (the density varies all the 
time on this paper even though you allow long drying time : it's a 
real annoyance).

The only issue I got so far is to have the 6 heads firing on a single 
print : at the moment I have 1 curve KCMY and 1 KCLCLM. I have not 
been successful using the 2 same inks in 2 different positions for 
the same print and have an acceptable curve ; there is banding where 
I suspect the 2 same inks get into action (so 2 bands on the 
stepwedge). I've read you can either set M at the mesured density 
ratio and limit at 50 then set LC density at "copy from" M or simply 
have density ratio M at 11 and LC at 10 or 12 for instance : I have 
not tried yet (very time consuming and costful at the end of the day).

I'd be happy exchanging experiences if you wish so.

Olivier

Thank you Tom, Eric & Costa for helping me with the ink limit 
calibration. With your collective help I got over the initial  hurdle 
of curve making and produced my first acceptable QTR Profile. This is a 
beginning for me and I have much to learn. You got me off in the right 
direction and I am very appreciative.

Howard

On Aug 2, 2005, at 12:34 PM, Tom Moore wrote:

> Howard
>
>  First let me say that I don't think the patch you choose is that 
> critical -
>  at least it hasn't seemed to be for curves I've created. That said, 
> there
>  are some who want to squeeze the last bit of density out of each ink 
> and try
>  to maximize each ink's cutoff point.
>
>  Other than using your eyeball, you can use a densitometer or scanner. 
> Print
>  the separation page and then measure each patch along the area where 
> the ink
>  seems to be reaching its peak density. The patches increase in 
> density near
>  the start (lower levels) and then gradually flatten out or drop in 
> density
>  then pick a patch at or near the tip of the curve.
>
>  With respect to using different limits for each channel, you 
> certainly can
>  if you wish. Again, the idea is to maximize each ink and hopefully 
> reduce
>  the dots near the transition to the next most dense ink. To measure 
> the
>  relative density with different limits, you must reprint the 
> inkseparation
>  chart with each unique limit. Unfortunately you cannot set individual 
> limits
>  for each ink when printing the inkseparation page.
>
>  Steve Kale posted a number of messages regarding separate ink limits. 
> I'm
>  not sure whether it was this list or the BW Printing list.
>
>  Tom Moore
>
>  > -----Original Message-----
>  > From: QuadtoneRIP@yahoogroups.com 
> [mailto:QuadtoneRIP@yahoogroups.com] On
>  > Behalf Of steinbe2003
>  > Sent: Tuesday, August 02, 2005 2:09 PM
>  > To: QuadtoneRIP@yahoogroups.com
>  > Subject: [QuadtoneRIP] Hardware Ink Limit Calibration
>  >
>
>  ...
>
>  > Can anyone suggest a more precise guideline? Also how do you choose 
> the
>  > ink limit to use
>  > if you get widely varying results for the different ink channels?
>  >
>  > Any suggestions will be appreciated.
>  >
>  > Howard
>  >
>
>  ...
>
>
> YAHOO! GROUPS LINKS
>
> 	▪ 	 Visit your group "QuadtoneRIP" on the web.
>  
> 	▪ 	 To unsubscribe from this group, send an email to:
>  QuadtoneRIP-unsubscribe@yahoogroups.com
>  
> 	▪ 	 Your use of Yahoo! Groups is subject to the Yahoo! Terms of 
> Service.
>
>
>
Howard Steinberg
1585 Cordova Drive
San Luis Obispo, CA 93405
(805) 543-1392