Digital BW, The Print

Paul, 
Thanks for the quick answers to those questions.  

I suppose for me the only questions left are whether I have the skills needed and time needed to do the "workflows" dance.  

As it is, I have a lack of time for tweaking/experimenting (any more than I do) and generally playing around.  Too many computers, too many printers, too many variables...

I am tempted by the VM because I too would like to have control over the warmth/coolness.  Your VM solution and curves sounds like a great advance.  

However, I am sure I don't have your level of Photoshop skill and pshop experience.  At this time, I am trying to "master" many things.  IF the VM solution is one that requires a lot of print/modify curve/print again/ modify again/print again/ modify... I probably can't use it for my production work.  I need a pretty straightforward streamlined (ok...it would not hurt if it was just plain ole easy) workflow so I can concentrate on the frontend, marketing, shooting, customers, making money etc...the usual stuff. 

I know you have spent many hours on this.  I just don't have many hours to spend on it.  

So, for someone like me...is it VM/Curves  easy enough? Or should I stick to the turnkey Piezo?

I appreciate your opinion and guidance on this.  

Steadman
  ----- 


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

Steadman,

You wrote:

>I suppose for me the only questions left are whether
>I have the skills needed and time needed to do the
>"workflows" dance.

I hope there is no "dance" needed.  The basic workflow with the
variable-tone inkset is: edit in grayscale, convert file to RGB, apply
adjustment curve, print.

A Photoshop "action" is easy an easy-to-make macro that sets up a key that
can handle the conversion to RGB and application of the appropriate curve.
It also then calls up the printer driver.

So, once one sets up an action, the printing function is virtually a
one-button procedure.  I suppose I ought to summarize how to do that one
this forum, and perhaps MIS should have that procedure on the its
workflow/settings page.

Frankly, I don't bother with a macro/action because I find the manual
workflow rather simple.  On the other hand, I admit I've forgotten to
convert to RGB at least once.

>As it is, I have a lack of time for tweaking/experimenting ...

None should be required.  While the curves are written for Archival Matte
paper, that profile works for many papers.  It is my hope that others will
contribute and share Transfer Function profiles for different papers, as
needed.  In fact, there are all sorts of interesting things that can be
easily done and shared that way, including "canned" split-toning approaches.

One of the many advantages of the Epson driver is the availability of that
easy Transfer Function capability.

>I am tempted by the VM because I too would like to have
>control over the warmth/coolness.  Your VM solution and
>curves sounds like a great advance.

I came up with this approach because the existing inks, as good as they are,
were just not producing the results I wanted.  Once a mix and curves are
done for oneself, I just don't see any reason not to share.  All the work is
done once the curves or Transfer Function file is saved.  I hope others
using this inkset adopt the same attitude.

>IF the VM solution is one that requires a lot of print/
>modify curve/print again/ modify again/... I probably can't
>use it for my production work.  ...

I sure hope that isn't necessary.  I don't really know the
printer-to-printer variances within Epson printer types.  It is true that
each printer family needs it's own set of curves, but those curves will all
be available from me or MIS soon.  Once that is done, they should work
without further modification.

Then, if and when profiles are needed, I hope others share their solutions.
I just ordered some Legion Matte, for example.  It's gotten such good
reviews that I wanted to try some.  I'll test -- both for it's looks and
light fastness with the variable-tone inkset -- and see if it needs a
profile.  If it does, I'll write one and make it available.

Obviously, not all papers are going to be of interest to me, so this won't
happen with all the papers out there.  But if we all share, there will be
many more profiles available and more quickly than any other system.

>So, for someone like me...is it VM/Curves  easy enough? ...

I think you'll find it easy to use.  I'm getting comments from first-time
users who seem to be getting great results immediately.

Good luck.

Paul
http://www.PaulRoark.com

--- In DigitalBlackandWhiteThePrint@y..., "Paul Roark" 
<paul.roark@v...> wrote:
<I suppose I ought to summarize how to do that one
> this forum, and perhaps MIS should have that procedure on the 
<its workflow/settings page.

I would think that if you can "export" a Transfer File, you could 
easily upload it to the FILES section of this website. That way, 
we, as the real users, can control what's seen and used, rather 
than by any manufacturer. After all, you created these workflows; 
you should control how they're seen and downloaded.

You'd probably have to test it a couple of times to see what 
format they upload and download best. They might need to be 
stuffed/compressed for best usability over the web. But that's the 
power of a site like this; we can trade/swap workflows that best 
suit the printer, as long as Martin/Antonis agree.

If you want to write a short summary on any of these topics, you 
could easily stick it on a webpage, and then list it in the 
BOOKMARKS section of this site as well.

Paul, 
Thanks for all the perfect answers to my questions (again). 

Your level of confidence in the ease of use is contagious.  Now I am really itching to buy those inks and try those Roark Curves!

(I hope no one thinks I am a shill for you or MIS.)  (grin)

On another note...your responses make it seem so easy.  I wonder why some of the other discussions on this list leave me in a fog when people talk about transfer curves and related techo speak stuff?

While you are "curving" the Legion Photo Matte (LPM), I would like to put in a request for you to check out the Hahnemuhle Watercolor Rag Smooth (HWRS) which is very similar (but a touch warmer in tone) in surface finish (ultra smooth) but also is 100% rag. 

Regards again...and I will try your ink/method/curve. 

Steadman

  ----- Original Message ----- 
  From: Paul Roark 
  To: DigitalB&WPrint 
  Sent: Sunday, August 19, 2001 10:35 AM
  Subject: Re: [Digital BW] 4 Paul Roark: MIS VM and Piezo Questions


  Steadman,

  You wrote:

  >I suppose for me the only questions left are whether
  >I have the skills needed and time needed to do the
  >"workflows" dance.

  I hope there is no "dance" needed.  The basic workflow with the
  variable-tone inkset is: edit in grayscale, convert file to RGB, apply
  adjustment curve, print.

  A Photoshop "action" is easy an easy-to-make macro that sets up a key that
  can handle the conversion to RGB and application of the appropriate curve.
  It also then calls up the printer driver.

  So, once one sets up an action, the printing function is virtually a
  one-button procedure.  I suppose I ought to summarize how to do that one
  this forum, and perhaps MIS should have that procedure on the its
  workflow/settings page.

  Frankly, I don't bother with a macro/action because I find the manual
  workflow rather simple.  On the other hand, I admit I've forgotten to
  convert to RGB at least once.

  >As it is, I have a lack of time for tweaking/experimenting ...

  None should be required.  While the curves are written for Archival Matte
  paper, that profile works for many papers.  It is my hope that others will
  contribute and share Transfer Function profiles for different papers, as
  needed.  In fact, there are all sorts of interesting things that can be
  easily done and shared that way, including "canned" split-toning approaches.

  One of the many advantages of the Epson driver is the availability of that
  easy Transfer Function capability.

  >I am tempted by the VM because I too would like to have
  >control over the warmth/coolness.  Your VM solution and
  >curves sounds like a great advance.

  I came up with this approach because the existing inks, as good as they are,
  were just not producing the results I wanted.  Once a mix and curves are
  done for oneself, I just don't see any reason not to share.  All the work is
  done once the curves or Transfer Function file is saved.  I hope others
  using this inkset adopt the same attitude.

  >IF the VM solution is one that requires a lot of print/
  >modify curve/print again/ modify again/... I probably can't
  >use it for my production work.  ...

  I sure hope that isn't necessary.  I don't really know the
  printer-to-printer variances within Epson printer types.  It is true that
  each printer family needs it's own set of curves, but those curves will all
  be available from me or MIS soon.  Once that is done, they should work
  without further modification.

  Then, if and when profiles are needed, I hope others share their solutions.
  I just ordered some Legion Matte, for example.  It's gotten such good
  reviews that I wanted to try some.  I'll test -- both for it's looks and
  light fastness with the variable-tone inkset -- and see if it needs a
  profile.  If it does, I'll write one and make it available.

  Obviously, not all papers are going to be of interest to me, so this won't
  happen with all the papers out there.  But if we all share, there will be
  many more profiles available and more quickly than any other system.

  >So, for someone like me...is it VM/Curves  easy enough? ...

  I think you'll find it easy to use.  I'm getting comments from first-time
  users who seem to be getting great results immediately.

  Good luck.

  Paul
  http://www.PaulRoark.com


        Yahoo! Groups Sponsor 
              ADVERTISEMENT
             
       
       

  If you do not wish to belong to Digital B&W, The Print, you may
  unsubscribe by sending an email to:
  DigitalBlackandWhiteThePrint-unsubscribe@yahoogroups.com



  Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service. 



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

Hi Again Steadman!

You don't have to "Modify" Pauls Curves. They are ready to use. Its a very easy workflow with superb results. You could easily run an action that would:

1. Flatten your Grayscale image
2. Turn your Grayscale image into an RGB image
3. Load the Curve of your choice and close the curve dialogue box...

After that all you have to do is size your print, and make sure everything is set up in Page setup as you want it, and Print.

Nothing more Complex than that.

You could actually have an action for each curve.

You don't HAVE to flatten your image, it just saves space if you're short on it.

Jerry

on 8/19/01 12:32 PM, Steadman Uhlich wrote:

> On another note...your responses make it seem so easy.  I wonder why some of
> the other discussions on this list leave me in a fog when people talk about
> transfer curves and related techo speak stuff?

Steadman,

I would suggest some of this talk, which I have been guilty of, is due to
the sudden opportunity to discuss some of these *possibilities* now that
this list exists. Some of the workflow conversations of the Piezo system
have a FAQ and website @ inkjetmall. The other systems have this list. You
are just seeing a flurry of discussions regarding possible ways to expand
the system and push it's boundaries.

This is esoterica, and not a requirement of the system except in some
circumstances. In the Roark method you use curves, whereas in the Piezo
system you use profiles. As you've probably noticed in the Piezo system, one
or two profiles will go a long way. You could probably do most of your Piezo
"production" work with the Orwell profile, an Somerset Enhanced. Likewise
Paul's EAM curve covers EAM, LPM, and the Hahnemuhle papers quite well. It
does fall down for Hawk Mountain, and Brightcube papers for instance. If you
are committed to using these papers you'll need to modify the curves or the
workflow. But Piezo doesn't include profiles for these papers either, so
mucking about is required there too. However I believe more of us have
facility with curves than we do with monochromatic profiling packages, and
methodologies, so I believe as a group we can push through such bottlenecks
faster on our own, than pleading our case to Cone. :-)

In the end, you'll just have to jump in and test the water, but the price is
right for a casual swim.

Todd

I wonder if would be possible for someone (perhaps Paul Roark) to write a tutorial on curve development. Paul is a busy guy, but perhaps this could be a long term project and be placed in the "FILES" section in installments.

I know Paul is gonna kill me for the suggestion! ;)

Regards,

Michael J. Kravit, AIA
Architect/Photographer
www.kravit.net/photography
  This is esoterica, and not a requirement of the system except in some
  circumstances. In the Roark method you use curves, whereas in the Piezo
  system you use profiles. As you've probably noticed in the Piezo system, one
  or two profiles will go a long way. You could probably do most of your Piezo
  "production" work with the Orwell profile, an Somerset Enhanced. Likewise
  Paul's EAM curve covers EAM, LPM, and the Hahnemuhle papers quite well. It
  does fall down for Hawk Mountain, and Brightcube papers for instance. If you
  are committed to using these papers you'll need to modify the curves or the
  workflow. But Piezo doesn't include profiles for these papers either, so
  mucking about is required there too. However I believe more of us have
  facility with curves than we do with monochromatic profiling packages, and
  methodologies, so I believe as a group we can push through such bottlenecks
  faster on our own, than pleading our case to Cone. :-)

  In the end, you'll just have to jump in and test the water, but the price is
  right for a casual swim.

  Todd


        Yahoo! Groups Sponsor 
              ADVERTISEMENT
             
       
       

  If you do not wish to belong to Digital B&W, The Print, you may
  unsubscribe by sending an email to:
  DigitalBlackandWhiteThePrint-unsubscribe@yahoogroups.com



  Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service. 



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

Steadman,

I can attest to the fact that the Roark workflow and curves are a simple matter even for the uninitiated.  Even for  a rather pedestrian Architect like me.

BTW, the EAM curves that Paul ahs developed work very well with all of the Hahnemuhle papers that I tested. I even tweaked a transfer curve in the Epson printer driver to punch up the 80% values a bit with great success.

The biggest difference that I see with the MIS Vari-tone inks is that the curves developed by Paul are very linear. Cones profiles seem to compress the higher densities to make the blacks appear more dense. Images that I print on the same paper with the two inksets look identical, except that the shadow areas of the MIS Vari-tones are more open and hold detail far better. An added bonus I suppose.

Regards,

Michael J. Kravit, AIA
Architect/Photographer
www.kravit.net/photography


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

The Variable curves take a long time to write, and it must be fine tuned many times. I don't think the average user here would want to tackle it.

But what about transfer curves? How do They work? What would they offer that the rgb curves wouldn't?

Jerry

Michael,

I assume you write a transfer curve to expand two or more tones a bit when they print too similar?

How do you actually write the transfer curve? How do you know what numbers to type in, or what shape to make the curve? Could you make this more understandable?

Thanks,

Jerry

"Michael J. Kravit" wrote:

> Steadman,
>
> I can attest to the fact that the Roark workflow and curves are a simple matter even for the uninitiated.  Even for  a rather pedestrian Architect like me.
>
> BTW, the EAM curves that Paul ahs developed work very well with all of the Hahnemuhle papers that I tested. I even tweaked a transfer curve in the Epson printer driver to punch up the 80% values a bit with great success.
>
> The biggest difference that I see with the MIS Vari-tone inks is that the curves developed by Paul are very linear. Cones profiles seem to compress the higher densities to make the blacks appear more dense. Images that I print on the same paper with the two inksets look identical, except that the shadow areas of the MIS Vari-tones are more open and hold detail far better. An added bonus I suppose.
>
> Regards,
>
> Michael J. Kravit, AIA
> Architect/Photographer
> www.kravit.net/photography
>
> [Non-text portions of this message have been removed]
>
> If you do not wish to belong to Digital B&W, The Print, you may
> unsubscribe by sending an email to:
> DigitalBlackandWhiteThePrint-unsubscribe@yahoogroups.com
>
>
>
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/

Jerry wrote:

>The Variable curves take a long time to write, and it
>must be fine tuned many times. I don't think the
>average user here would want to tackle it.

True.

>But what about transfer curves? How do They work?
>What would they offer that the rgb curves wouldn't?

I think the Transfer Function curves are much simpler to manipulate because
there are just boxes that one enters numbers into to change the curves.
They didn't give good enough control to use as the primary method of
controlling the variable-tone inks (I tried).  However, they do allow one to
make changes simply that can, in effect, profile the default Archival Matte
curve for another paper.  I've done this for Somerset enhanced, and Mike
Kravit just mentioned that he'd done it for some Hahnemuhle papers that he
uses.

It think that the Transfer Function, which is accessed though the "Transfer"
button on the first Epson driver dialog box (I'm not sure this is the same
for all printers and the same for PC and Mac), is easy enough for most to
use.  And, using it does not risk messing up the main Photoshop image
adjustment curves that will still be used to control the inks.  The Transfer
Function will be, in effect, an overlay on the primary curve.

Paul

--- In DigitalBlackandWhiteThePrint@y..., "Paul Roark" <paul.roark@v...> 
wrote:

 I suppose I ought to summarize how to do that one
> this forum,
>>>>>>>>>>>>>>>

Paul,

I applaud the thought! <g> And I second Mark's remarks. Of course Martin and 
I have no problem with the notion of uploading printing resources for the 
members of this forum. So, please let me know when you have something to 
contribute. 

Also, FYI, I have asked Mike Kravit to provide us a summary of comparisons 
between Piezo and MIS VT. If you would like to add your comments, I can 
circulate it through to you, or we can have two separate accounts.

Thanks for all your help here.

Antonis

Jerry,

Transfer curves are accessed at the print menu that first opens. When it come up you get a series of boxes where you can place values to adjust that density of the print. One important thing to remember is that small changes in density value can affect the print significantly.

Mike




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

Hi Paul,

But how do you know what numbers to type in?

Jerry

>
>
> >But what about transfer curves? How do They work?
> >What would they offer that the rgb curves wouldn't?

--- In DigitalBlackandWhiteThePrint@y..., Jerry Olson 
<jerryolson@r...> wrote:
> Hi Paul,
> 
> But how do you know what numbers to type in?
> 
> Jerry
> 
> >
> >
> > >But what about transfer curves? How do They work?
> > >What would they offer that the rgb curves wouldn't?

Take a look at the Nevins method (on his web site, available 
through MIS workflow page). The greyscale you create and 
adjust gives you a great understanding of how the transfer 
curves effect the output.

Dave

Jerry wrote:
>But how do you know what numbers to type in?

There is definitely some trial and error.  After printing a step wedge test
file, scan it.  Use levels to put the 0% and 100% where they should be.
Then read the values of the other patches.  If, for example, the 95% patch
reads 97%, try putting a 94 in the 95% box of the Transfer Function.  (It
tends to overshoot.)  Then print with it and see what happened.

Don't expect to get the exact values all the way.  If you make too many or
too radical adjustments in the Transfer Function the ramp get ripply.  So,
go easy and slow, saving the files as you go, and marking the test prints so
you can conclude after you've gone too far which made the best print.

Paul

>
>
> >But what about transfer curves? How do They work?
> >What would they offer that the rgb curves wouldn't?

> 
> There is definitely some trial and error.  After printing a step wedge test
> file, scan it.  Use levels to put the 0% and 100% where they should be.
> Then read the values of the other patches.  If, for example, the 95% patch
> reads 97%, try putting a 94 in the 95% box of the Transfer Function.  (It
> tends to overshoot.)  Then print with it and see what happened.

A couple of questions about transfer functions.

Once a transfer function is set in the page setup dialog box, does it get
saved with the file, or does it stay active for all other prints as well.

One disadvantage I see to a transfer function is that it is buried away on a
second page the Page Setup window, so one may forget if it is on or off. In
light of that, is it in anyway preferable to a luminosity adjustment layer
curve? I like the idea of an adjustment layer because of it's visibility,
and with PS6 we can group such a layer with one of Paul's tone curves as a
layer set. Such a set can be aptly named, and easily monitored, in a way I
find difficult with transfer curves.

I guess it's really moot in the end as a transfer curve can be loaded as an
adjustment layer curve, and vice versa. So people could apply the same curve
through either method. I just wonder what is the preference for the transfer
approach?

Todd

Don't forget, you can also type in decimals for less extreme adjustments. ie; 96.7

Regards,

Michael J. Kravit, AIA
Architect/Photographer
www.kravit.net/photography

  ----- Original Message ----- 
  From: Paul Roark 
  To: DigitalB&WPrint 
  Sent: Sunday, August 19, 2001 10:59 PM
  Subject: Re: [Digital BW] 4 Paul Roark: MIS VM and Piezo Questions




  Jerry wrote:
  >But how do you know what numbers to type in?

  There is definitely some trial and error.  After printing a step wedge test
  file, scan it.  Use levels to put the 0% and 100% where they should be.
  Then read the values of the other patches.  If, for example, the 95% patch
  reads 97%, try putting a 94 in the 95% box of the Transfer Function.  (It
  tends to overshoot.)  Then print with it and see what happened.

  Don't expect to get the exact values all the way.  If you make too many or
  too radical adjustments in the Transfer Function the ramp get ripply.  So,
  go easy and slow, saving the files as you go, and marking the test prints so
  you can conclude after you've gone too far which made the best print.

  Paul



  >
  >
  > >But what about transfer curves? How do They work?
  > >What would they offer that the rgb curves wouldn't?



        Yahoo! Groups Sponsor 
              ADVERTISEMENT
             
       
       

  If you do not wish to belong to Digital B&W, The Print, you may
  unsubscribe by sending an email to:
  DigitalBlackandWhiteThePrint-unsubscribe@yahoogroups.com



  Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service. 



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

Paul, in making a print of a scenic today, with an absolutely grainless sky, I
got a very obvious dot/grain problem evident all over the whole print, that
looks horrible to the eye, and extremely horrible with a loupe.

The before and after images I printed, same printer, (different photographs)
same curve (the new 16 cool curve) showed no grain or dots at all. Then I
reprinted this image, and once again got the terrible grain. Have you any idea
at all of what could be causing this to happen to just the one file?

Thanks,

Jerry





Paul Roark wrote:

> Jerry wrote:
> >But how do you know what numbers to type in?
>
> There is definitely some trial and error.  After printing a step wedge test
> file, scan it.  Use levels to put the 0% and 100% where they should be.
> Then read the values of the other patches.  If, for example, the 95% patch
> reads 97%, try putting a 94 in the 95% box of the Transfer Function.  (It
> tends to overshoot.)  Then print with it and see what happened.
>
> Don't expect to get the exact values all the way.  If you make too many or
> too radical adjustments in the Transfer Function the ramp get ripply.  So,
> go easy and slow, saving the files as you go, and marking the test prints so
> you can conclude after you've gone too far which made the best print.
>
> Paul
>
> >
> >
> > >But what about transfer curves? How do They work?
> > >What would they offer that the rgb curves wouldn't?
>
> If you do not wish to belong to Digital B&W, The Print, you may
> unsubscribe by sending an email to:
> DigitalBlackandWhiteThePrint-unsubscribe@yahoogroups.com
>
>
>
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/

Paul, I found out the trouble. I somehow had the black only set instead of the color inks. Don't know how it happened, but it did. Gawd!
And some people actually LIKE this effect?

Jerry



Jerry Olson wrote:

> Paul, in making a print of a scenic today, with an absolutely grainless sky, I
> got a very obvious dot/grain problem evident all over the whole print, that
> looks horrible to the eye, and extremely horrible with a loupe.
>
> The before and after images I printed, same printer, (different photographs)
> same curve (the new 16 cool curve) showed no grain or dots at all. Then I
> reprinted this image, and once again got the terrible grain. Have you any idea
> at all of what could be causing this to happen to just the one file?
>
> Thanks,
>
> Jerry
>
> Paul Roark wrote:
>
> > Jerry wrote:
> > >But how do you know what numbers to type in?
> >
> > There is definitely some trial and error.  After printing a step wedge test
> > file, scan it.  Use levels to put the 0% and 100% where they should be.
> > Then read the values of the other patches.  If, for example, the 95% patch
> > reads 97%, try putting a 94 in the 95% box of the Transfer Function.  (It
> > tends to overshoot.)  Then print with it and see what happened.
> >
> > Don't expect to get the exact values all the way.  If you make too many or
> > too radical adjustments in the Transfer Function the ramp get ripply.  So,
> > go easy and slow, saving the files as you go, and marking the test prints so
> > you can conclude after you've gone too far which made the best print.
> >
> > Paul
> >
> > >
> > >
> > > >But what about transfer curves? How do They work?
> > > >What would they offer that the rgb curves wouldn't?
> >
> > If you do not wish to belong to Digital B&W, The Print, you may
> > unsubscribe by sending an email to:
> > DigitalBlackandWhiteThePrint-unsubscribe@yahoogroups.com
> >
> >
> >
> > Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
>
>
> Please visit the Group Homepage to check the Files, Bookmarks, Polls and other resources as they are often being updated. The page is at:
>
> http://groups.yahoo.com/group/DigitalBlackandWhiteThePrint
>
>
>
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/

Afternoon Paul,

Did you get the latest batch of prints? If so, what did you think?


Jerry