QTR-Quadtone RIP

I'm a newbee and trying to print with the QTR process.  I am making icc profiles but it is not obvious what the functional difference is between the two programs.  Does not QTR require a gray scale image to print?  When is the RGB version used and why?  Thanks

QTR ICC-RGB allows but does not require embedding a Photoshop curve and is
compatible with more applications.  It's the only one I use now.

Paul
www.PaulRoark.com

On Mon, Dec 3, 2012 at 11:38 AM, rfsteckmeyer <steckmeyer@comcast.net>wrote:

> **
>
>
> I'm a newbee and trying to print with the QTR process. I am making icc
> profiles but it is not obvious what the functional difference is between
> the two programs. Does not QTR require a gray scale image to print? When is
> the RGB version used and why? Thanks
>
>  
>


[Non-text portions of this message have been removed]

Paul:

I recently noted on B&W-the-print that I had modified one of your curves to get really good results with the UT14 ink set on glossy paper. Some step-scale prints confirmed that the ramp of this is good, but not perfect. Could I use the QTRICC-RGB to perfect this?

Homer Shannon

--- In QuadtoneRIP@yahoogroups.com, Paul Roark <roark.paul@...> wrote:

>
> QTR ICC-RGB allows but does not require embedding a Photoshop curve and is
> compatible with more applications.  It's the only one I use now.
> 
> Paul
> www.PaulRoark.com
> 
> On Mon, Dec 3, 2012 at 11:38 AM, rfsteckmeyer <steckmeyer@...>wrote:
> 
> > **
> >
> >
> > I'm a newbee and trying to print with the QTR process. I am making icc
> > profiles but it is not obvious what the functional difference is between
> > the two programs. Does not QTR require a gray scale image to print? When is
> > the RGB version used and why? Thanks
> >
> >  
> >
> 
> 
> [Non-text portions of this message have been removed]
>

Yes, perfecting the linearization of a curve that is close is what should
be done.

Paul
www.PaulRoark.com

On Tue, Dec 4, 2012 at 8:44 AM, remononaz1 <homershannon@...> wrote:

> **
>
>
> Paul:
>
> I recently noted on B&W-the-print that I had modified one of your curves
> to get really good results with the UT14 ink set on glossy paper. Some
> step-scale prints confirmed that the ramp of this is good, but not perfect.
> Could I use the QTRICC-RGB to perfect this?
>
> Homer Shannon
>
> --- In QuadtoneRIP@yahoogroups.com, Paul Roark <roark.paul@...> wrote:
> >
> > QTR ICC-RGB allows but does not require embedding a Photoshop curve and
> is
> > compatible with more applications. It's the only one I use now.
> >
> > Paul
> > www.PaulRoark.com
> >
> > On Mon, Dec 3, 2012 at 11:38 AM, rfsteckmeyer <steckmeyer@...>wrote:
> >
> > > **
> > >
> > >
> > > I'm a newbee and trying to print with the QTR process. I am making icc
> > > profiles but it is not obvious what the functional difference is
> between
> > > the two programs. Does not QTR require a gray scale image to print?
> When is
> > > the RGB version used and why? Thanks
> > >
> > >
> > >
> >
> >
> > [Non-text portions of this message have been removed]
> >
>
>  
>


[Non-text portions of this message have been removed]

Paul:

I'll give this a go tonight. 

I read through the instructions posted at http://www.paulroark.com/BW-Info/Embedding_Photoshop_Curves_in_ICCs.pdf and this seems pretty straight forward. I have one question:

Once the RGB-ICC is generated do you apply a curve to the image and then print with the ICC or just take the un-altered B&W image and print it with the ICC?

Homer Shannon

remononaz1 <homershannon@...> wrote:

> **
>
>
> ...
> Once the RGB-ICC is generated do you apply a curve to the image and then
> print with the ICC or just take the un-altered B&W image and print it with
> the ICC?
>
>
The ICC both applies the curves you've embedded as well as the
linearization curve.  So, you just print with a GS image file.  I recommend
16 bit Tiff.  The image, as I understand the process, goes into the ICC as
16 bit (if RGB it's apparently converted to GS and then, presumably,
re-converted back to RGB), the curves within the ICC are applied in high
bit mode, and the output is 24 bit RGB to the Epson driver.  As long as the
PS curve you've embedded has it's R, G, and B curves as distinct,
non-identical curves, the original 16 bit image data, I would think, is
fairly well preserved in the final 24 bit pipeline.

(Of course, the printer dithering probably wipes out quite a bit of
information, but at least you know you're going to that stage with as much
as possible.)

Paul
www.PaulRoark.com


[Non-text portions of this message have been removed]

Stay on this thread, RF. I think we are both about to find out. 

I, too, have created standard ICCs but not ventured into RGB ICCs yet. I have a new ink/paper combo that uses a curve to get the results I want, but now I want to perfect the ramp, which is already pretty good. 

If I understand this right, the difference between ICCs and RGB-ICCs is that the latter uses RGB information to control which ink 'colors' are being used to create the B&W grayscales. With the RGB-ICC the monotone image is effectively converted to an RGB image with the required curve applied within the printer driver, eliminating the need to convert the image and apply a curve before printing. Not that applying a curve is a big deal, but you wouldn't be able to do it in Lightroom, you'd have to use Photoshop. With an RGB-ICC you could print the image right out of LR using the appropriate RGB-ICC in your printer setup. 

Paul Roark has made a couple of suggestions and using his instructions, http://www.paulroark.com/BW-Info/Embedding_Photoshop_Curves_in_ICCs.pdf, I expect that I can give this a whirl tonight. I have some questions but I'll see if I can work past them before looking for help. 

If this works, I'll post my results and any comments that clarify or assist with working through the instructions. 

Homer Shannon  

--- In QuadtoneRIP@yahoogroups.com, "rfsteckmeyer" <steckmeyer@...> wrote:

>
> I'm a newbee and trying to print with the QTR process.  I am making icc profiles but it is not obvious what the functional difference is between the two programs.  Does not QTR require a gray scale image to print?  When is the RGB version used and why?  Thanks
>

Paul:

I gave this a shot as soon as I got home. Having read your instruction several times, I was quite confident of the process and it proceeded very quickly. 

However, the results are not good. There are no surface artifacts, indicating that the blue curve is still holding down the Eboni ink, but the tones are too dark. I'm reading 72 at 50 and 85 at 65. 

Any ideas why the .icc created would yield such dark levels? When I print the curve with printer controls, G2.2, no icc, 50 reads 48 and 85 reads 82. Overall with that ramp I'm off 0 to 3 but it is linear. 

Homer Shannon

--- In QuadtoneRIP@yahoogroups.com, Paul Roark <roark.paul@...> wrote:

>
> remononaz1 <homershannon@...> wrote:
> 
> > **
> >
> >
> > ...
> > Once the RGB-ICC is generated do you apply a curve to the image and then
> > print with the ICC or just take the un-altered B&W image and print it with
> > the ICC?
> >
> >
> The ICC both applies the curves you've embedded as well as the
> linearization curve.  So, you just print with a GS image file.  I recommend
> 16 bit Tiff.  The image, as I understand the process, goes into the ICC as
> 16 bit (if RGB it's apparently converted to GS and then, presumably,
> re-converted back to RGB), the curves within the ICC are applied in high
> bit mode, and the output is 24 bit RGB to the Epson driver.  As long as the
> PS curve you've embedded has it's R, G, and B curves as distinct,
> non-identical curves, the original 16 bit image data, I would think, is
> fairly well preserved in the final 24 bit pipeline.
> 
> (Of course, the printer dithering probably wipes out quite a bit of
> information, but at least you know you're going to that stage with as much
> as possible.)
> 
> Paul
> www.PaulRoark.com
> 
> 
> [Non-text portions of this message have been removed]
>

Homer,

I can't tell with the information I have what might have gone wrong.

Maybe off list you could send me screen grabs of each step, the paper and
inkset information, and the curve you're printing the linearization 21-step
with and then embedding in the ICC.

Paul
www.PaulRoark.com

On Wed, Dec 5, 2012 at 3:35 PM, remononaz1 <homershannon@...> wrote:

> **
>
>
> Paul:
>
> I gave this a shot as soon as I got home. Having read your instruction
> several times, I was quite confident of the process and it proceeded very
> quickly.
>
> However, the results are not good. There are no surface artifacts,
> indicating that the blue curve is still holding down the Eboni ink, but the
> tones are too dark. I'm reading 72 at 50 and 85 at 65.
>
> Any ideas why the .icc created would yield such dark levels? When I print
> the curve with printer controls, G2.2, no icc, 50 reads 48 and 85 reads 82.
> Overall with that ramp I'm off 0 to 3 but it is linear.
>
> Homer Shannon
>
> --- In QuadtoneRIP@yahoogroups.com, Paul Roark <roark.paul@...> wrote:
> >
> > remononaz1 <homershannon@...> wrote:
> >
> > > **
>
> > >
> > >
> > > ...
> > > Once the RGB-ICC is generated do you apply a curve to the image and
> then
> > > print with the ICC or just take the un-altered B&W image and print it
> with
> > > the ICC?
> > >
> > >
> > The ICC both applies the curves you've embedded as well as the
> > linearization curve. So, you just print with a GS image file. I recommend
> > 16 bit Tiff. The image, as I understand the process, goes into the ICC as
> > 16 bit (if RGB it's apparently converted to GS and then, presumably,
> > re-converted back to RGB), the curves within the ICC are applied in high
> > bit mode, and the output is 24 bit RGB to the Epson driver. As long as
> the
> > PS curve you've embedded has it's R, G, and B curves as distinct,
> > non-identical curves, the original 16 bit image data, I would think, is
> > fairly well preserved in the final 24 bit pipeline.
> >
> > (Of course, the printer dithering probably wipes out quite a bit of
> > information, but at least you know you're going to that stage with as
> much
> > as possible.)
> >
> > Paul
> > www.PaulRoark.com
> >
> >
> > [Non-text portions of this message have been removed]
> >
>
>  
>


[Non-text portions of this message have been removed]