Digital BW, The Print

Hi all,

This is my first post here so treat me gently ;o)

I've been researching the world of dedicated B&W printing for a 
little while now, getting ready to take the plunge with (almost 
definitely) a second 2100 to sit alongside my existing unit, which 
I'll keep for colour use. I've sifted through a lot of the discussion 
that has gone on here, and I think I've got a pretty good handle on 
most of the issues. I've also swapped a bit of direct email with Paul 
Roark which has been very informative in getting to grips with the 
finer points of UT7 (thanks Paul!)

I've got one question to ask now, one that springs to mind after 
reading Steve M's recent thread about using quadtone inks in a 7000. 
He implies that if you want "dotless" printing, then you want a true 
quadtone inkset. It seems logical enough that having a fine gradation 
of different ink shades would help hide the dots, but what I don't 
understand is why it would be much different for say UT7 using Paul's 
workflow. If you don't use the cold and sepia toners, then you still 
have pure black plus two shades of grey to work with -- essentially 
a "tritone" pure carbon inkset. Why should that be noticeably more 
dotty than a quadtone? Or is it more to do with the effect of the 
RIP? But in that case, why doesn't QTR give similar results?

Yours confused,

Mike Nunan

--- In DigitalBlackandWhiteThePrint@yahoogroups.com, "mike_nunan"
<mike_nunan@h...> wrote:
snip...
> He implies that if you want "dotless" printing, then you want a true 
> quadtone inkset.

Using a quad set alone doesn't gaurantee anything, the nature and
quality of the partitioning and the dither/screening of the driver are
two big factors.

>It seems logical enough that having a fine gradation 
> of different ink shades would help hide the dots, but what I don't 
> understand is why it would be much different for say UT7 using Paul's 
> workflow. If you don't use the cold and sepia toners, then you still 
> have pure black plus two shades of grey to work with -- essentially 
> a "tritone" pure carbon inkset. Why should that be noticeably more 
> dotty than a quadtone? Or is it more to do with the effect of the 
> RIP? But in that case, why doesn't QTR give similar results?

It may. There are too many variables. Printer dot size, best use and
partitioning of individual ink densities, the particular inkset, the
driver. As you suggest, a tritone from one printer, process and
materials may equal a quadtone from another.
Direct print comparisons would be good but are hard to come by. Some
of these processes offer different printing options, that's another
deciding factor.
Tyler

You have to think like the printer does. To make a tone, you put dots on
paper. To make a tone as dark as you can, the dots touch so that there
is no paper showing. 

To make a lighter tone you can go one of two ways. First, you can space
the dots apart, and the resulting tone is a combination of the color of
the ink, and the paper. If little paper shows, the tone is darker. If
mostly paper shows, the tone is lighter.

If you are using a single ink (black only printing, for example), to
make the lightest tones, you spread the dots very far apart. In very
light areas of the print, you can then see the dots, because the very
dark dots stand out in the sea of very light paper.

The second way, is to use dilutions of the black ink. This allows you to
bring the dots closer together to create the same tone. The lightest ink
in a quad set, for example, may be only 7% as dark as the black. That
is, for a tone that is 7%, the printer would make the dots touch, with
no paper showing (that's not really how it works - there are xovers and
most tones use more than one ink, but I'm keeping it simple for the sake
of argument). That's as dark as that ink can be. As the tones decrease
to 1% or less, the dots are still very close together - thus "dotless"
printing, yes?

Now we go to a variable tone set - one that has four dilutions of black
(a quadtone set), plus a warm toner and a cool toner. 

The two toners are single dilution inks. To tone the lightest areas of
the print, the printer uses the dark toner dots spread very wide apart.
Sometimes you can see the dots of toner, just like you can see the dots
of black when you do black only printing.

And that's why, for "dotless" printing, you need a fixed tone quad ink
set.


On Wed, 2004-04-21 at 13:00, mike_nunan wrote:

> Hi all,
> 
> This is my first post here so treat me gently ;o)
> 
> I've been researching the world of dedicated B&W printing for a 
> little while now, getting ready to take the plunge with (almost 
> definitely) a second 2100 to sit alongside my existing unit, which 
> I'll keep for colour use. I've sifted through a lot of the discussion 
> that has gone on here, and I think I've got a pretty good handle on 
> most of the issues. I've also swapped a bit of direct email with Paul 
> Roark which has been very informative in getting to grips with the 
> finer points of UT7 (thanks Paul!)
> 
> I've got one question to ask now, one that springs to mind after 
> reading Steve M's recent thread about using quadtone inks in a 7000. 
> He implies that if you want "dotless" printing, then you want a true 
> quadtone inkset. It seems logical enough that having a fine gradation 
> of different ink shades would help hide the dots, but what I don't 
> understand is why it would be much different for say UT7 using Paul's 
> workflow. If you don't use the cold and sepia toners, then you still 
> have pure black plus two shades of grey to work with -- essentially 
> a "tritone" pure carbon inkset. Why should that be noticeably more 
> dotty than a quadtone? Or is it more to do with the effect of the 
> RIP? But in that case, why doesn't QTR give similar results?
> 
> Yours confused,
> 
> Mike Nunan








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

Hi,

Thanks for that reply. I understand in principle why a quadtone 
should show very little dot pattern, and why the single-strength 
toning inks would lead to visible dots as soon as they are used in 
highlights or midtones; that's why I qualified, my statement 
saying "if you don't use the ... toners".

Tyler, in your reply you mention partitioning in the driver. I 
haven't heard the term used in that context before, what does it 
mean? Is it the proportioning of the inks according to a set of 
crossover curves? Regarding the rest of your post, you seem to be 
saying what I thought, intuitively, must be true: there is no reason 
why a UT7-based setup should be especially "dotty" compared to the 
quadtone Studio Print system.

Ok, you lose a bit by having three rather than four tones, but with a 
suitable driver and avoiding the toner inks, there's no reason why a 
print created on a 2200 using Paul's workflow shouldn't be very 
smooth indeed. Of in practice the Epson driver might well NOT be 
doing the "right" thing -- anyone know from firsthand experience what 
the case is? Steve M, if you're reading this would you care to 
comment?

TIA

-= mike =-

--- In DigitalBlackandWhiteThePrint@yahoogroups.com, hogarth 
<hogarth@s...> wrote:
> You have to think like the printer does. To make a tone, you put 
dots on
> paper. To make a tone as dark as you can, the dots touch so that 
there
> is no paper showing. 
> 
> To make a lighter tone you can go one of two ways. First, you can 
space
> the dots apart, and the resulting tone is a combination of the 
color of
> the ink, and the paper. If little paper shows, the tone is darker. 
If
> mostly paper shows, the tone is lighter.
> 
> If you are using a single ink (black only printing, for example), to
> make the lightest tones, you spread the dots very far apart. In very
> light areas of the print, you can then see the dots, because the 
very
> dark dots stand out in the sea of very light paper.
> 
> The second way, is to use dilutions of the black ink. This allows 
you to
> bring the dots closer together to create the same tone. The 
lightest ink
> in a quad set, for example, may be only 7% as dark as the black. 
That
> is, for a tone that is 7%, the printer would make the dots touch, 
with
> no paper showing (that's not really how it works - there are xovers 
and
> most tones use more than one ink, but I'm keeping it simple for the 
sake
> of argument). That's as dark as that ink can be. As the tones 
decrease
> to 1% or less, the dots are still very close together - 
thus "dotless"
> printing, yes?
> 
> Now we go to a variable tone set - one that has four dilutions of 
black
> (a quadtone set), plus a warm toner and a cool toner. 
> 
> The two toners are single dilution inks. To tone the lightest areas 
of
> the print, the printer uses the dark toner dots spread very wide 
apart.
> Sometimes you can see the dots of toner, just like you can see the 
dots
> of black when you do black only printing.
> 
> And that's why, for "dotless" printing, you need a fixed tone quad 
ink
> set.
> 
> 
> On Wed, 2004-04-21 at 13:00, mike_nunan wrote:
> 
> > Hi all,
> > 
> > This is my first post here so treat me gently ;o)
> > 
> > I've been researching the world of dedicated B&W printing for a 
> > little while now, getting ready to take the plunge with (almost 
> > definitely) a second 2100 to sit alongside my existing unit, 
which 
> > I'll keep for colour use. I've sifted through a lot of the 
discussion 
> > that has gone on here, and I think I've got a pretty good handle 
on 
> > most of the issues. I've also swapped a bit of direct email with 
Paul 
> > Roark which has been very informative in getting to grips with 
the 
> > finer points of UT7 (thanks Paul!)
> > 
> > I've got one question to ask now, one that springs to mind after 
> > reading Steve M's recent thread about using quadtone inks in a 
7000. 
> > He implies that if you want "dotless" printing, then you want a 
true 
> > quadtone inkset. It seems logical enough that having a fine 
gradation 
> > of different ink shades would help hide the dots, but what I 
don't 
> > understand is why it would be much different for say UT7 using 
Paul's 
> > workflow. If you don't use the cold and sepia toners, then you 
still 
> > have pure black plus two shades of grey to work with -- 
essentially 
> > a "tritone" pure carbon inkset. Why should that be noticeably 
more 

> > dotty than a quadtone? Or is it more to do with the effect of the 
> > RIP? But in that case, why doesn't QTR give similar results?
> > 
> > Yours confused,
> > 
> > Mike Nunan
> 
> 
> 
> 
> 
> 
> 
> 
> [Non-text portions of this message have been removed]

--- In DigitalBlackandWhiteThePrint@yahoogroups.com, "mike_nunan"
<mike_nunan@h...> wrote:
...
> Tyler, in your reply you mention partitioning in the driver. I 
> haven't heard the term used in that context before, what does it 
> mean? Is it the proportioning of the inks according to a set of 
> crossover curves?

Essentially yes. I haven't been keeping up on all new developements
but these are the methods I'm aware of-

Editing a file converted to color to control individual inks-
Paul Roake's and others RGB curves, also can be done in CMYK with a
RIP, Conetech ICC profiles (essentially doing the same on the fly).

Driver partitions inks on the fly from a grayscale file-
ImagePrint in Quad mode, StudioPrint, QTR, IJC, R9 plugin, Pro24RIP,
Septone system.
Of these, ImagePrint, R9 plugin, Pro24RIP, and Septone system are
preconfigured to do this for you and use supplied profiles for
specific inks and/or media. No user controls for the specifics of
partitioning, but considering many user's great results this may not
be a drawback for most users. Others find canned approaches a major
drawback.
StudioPrint, QTR, and IJC on the other hand give a great deal of
control to the user for partitioning and allow hardware linearization.
The advantages are that many inksets, papers, and approaches can be
utilized and brought to a linear state to specifically meet the user's
needs. I haven't worked with IJC, but with the other two you do not
specifically work on curves, only set parameters that are in effect
like applying curves. The downside is that some knowledge and
experience are required to get results that equal or exceed some of
the preconfigured options. It should be mentioned that StudioPrinbt
will require some very creative thinking, profiling, etc. to use
monochromatic aproaches that are not straight quads or quad blends.

> Regarding the rest of your post, you seem to be 
> saying what I thought, intuitively, must be true: there is no reason 
> why a UT7-based setup should be especially "dotty" compared to the 
> quadtone Studio Print system.

I'm not familiar with the UT7 system, but you seem to be grokking this
whole deal so I suppose so. Keep in mind there are issues besides
partitioning with some of these advanced RIPs, dither and pass
functions that effect results. But everyone knows Epson's screening is
great.

> Ok, you lose a bit by having three rather than four tones, but with a 
> suitable driver and avoiding the toner inks, there's no reason why a 
> print created on a 2200 using Paul's workflow shouldn't be very 
> smooth indeed. Of in practice the Epson driver might well NOT be 
> doing the "right" thing -- anyone know from firsthand experience what 
> the case is? Steve M, if you're reading this would you care to 
> comment?

Others, like Paul, who have put more time in on this should respond. I
can say from my own many hours of working with RGB curves and the
Epson driver that it is ill suited to the task but can be tamed. I've
seen spectacular results from the Conetech 1160 icc profiles that
prove the point.
I hope there is no misinformation about particular products in the
above, please correct me if so.
Tyler

>...
>Now we go to a variable tone set - ...

>The two toners are single dilution inks. ...
>Sometimes you can see the dots of toner, just like you can see the dots
>of black when you do black only printing.

>And that's why, for "dotless" printing, you need a fixed tone quad ink
>set.

The hextone VM and UT inksets have light and dark cool toners.

The last time I used a flatbed scanner to get objective measures of
smoothness (measured by the standard deviation in Photoshop of 1600 dpi
scans), the hextone VM inkset in a 1280 made a smoother print than the Piezo
software and its monotone inkset.  The 1280 VM was the smoothest print I
measured.

In the 3000 I do notice the advantages of the monotone inkset over the
variable tone, but in modern printers with their small and variable-size
dots, the super-light inks are just not needed.  In my 7500, which is
arguably closer to a 3000 than a 1280 in its non-variable dot size, I do
stick with the UT1 inkset as opposed to the UT2 or UT7.  My 7500 has, then,
3 gray inks (like the full quad 3000 or 1160, or PiezoBW hextones prior to
the ICC approach) as well as light and dark cool toners.  In my view it is
dotless.

It is true that theoretically there can be some smoothness advantages to
monotone inksets, but if the dots are beyond what one can see in normal
viewing, is it worth trading theoretical smoothness for more flexibility?

In the UT2 & 7 inksets, I clearly opted for more flexibility.  I wanted to
move forward without the need to support old printers.  As such, the 1280
was the starting point.  The densities of the inks are about what Epson
uses.  However, the gamut is radically less.  So, if one thinks the 1280 or
2200 can make smooth color prints, be assured those printers can make even
smoother B&W prints.

In actual practice, with modern printers I find the defect that shows up
most often is microbanding.  Dots are really not a serious issue any more
with inksets that use gray inks (including cool and very warm gray inks) in
new printers.

When I first started to do digital output in 1994, I thought I needed to
hide the digital character of the output.  I explored all sorts of methods,
including imagesetter output using the latest stochastic (random) dot
patterns.  I would contact print these on 16x20 fiber paper.  These were,
thus, silver prints -- but they did still have a digital step.  Even with
these expensive digital negatives, if one took a loupe to the print, the
digital step could be detected.  

But even if the digital step could not be detected, what would I say if
people asked about that?  I was not about to lie.

So, if I wasn't going to lie about the digital step, the issue of what
digital output to use no longer turned on trying to hide the character of
the output from those who might use a loupe.  (It couldn't be done anyway.)
What output to use was a function of the quality of the images -- all things
considered, including cost, longevity, ease of doing multiple iterations of
prints to improve them, etc.  

With the advent of the Piezo system, the inkjet won the contest for me, and
the issue became optimizing the inkjet output, in all respects, including
the flexibility to print the tones that I like and profiling different
papers to print a more uniform tone.

Hiding the dots from the loupe users is not very high on my list of ways to
improve inkjet output, but different people have different preferences.

Paul
www.PaulRoark.com 

For UT2 information, curves, and settings see:
http://home1.gte.net/res09aij/

In a message dated 4/22/2004 8:21:34 AM Pacific Daylight Time, 
mike_nunan@... writes:

> Ok, you lose a bit by having three rather than four tones, but with a 
> suitable driver and avoiding the toner inks, there's no reason why a 
> print created on a 2200 using Paul's workflow shouldn't be very 
> smooth indeed. Of in practice the Epson driver might well NOT be 
> doing the "right" thing -- anyone know from firsthand experience what 
> the case is? Steve M, if you're reading this would you care to 
> comment?
> 
> TIA
> 
> -= mike =-
> 
Mike, my only comment is don't get too hung up on dots. From my persepective, 
keep in mind how much work knowledge and time you want to spend 
getting things tweaked. This can easily add up to a years time and maybe even 
a couple thousand $ in paper and ink. Thats what I went through.
Things are much more advanced now with many more users and workflows and 90% 
of them make great prints. In my own experience If seen great results with RGB 
curves and other images that RGB curves didn't print so great. You would need 
to know how to make these curves yourself, understand them, and be able to 
make adjustments to them. Me? No thankyou, I need to make great BW prints 
predictably and as simply as possible.  
First thing for me is to simplfly. No color inks, A program like Studio print 
for many control options, and thats it. Simple tools but expensive, but I 
spend my time making prints for sale. I don't print other peoples work, so I only 
need my own look.
If I have time and want to play, QTR, and image print look like wonderful 
options. I would use some sort of RIP software for control even on these new 
ultarchrome printers for better than standard epson output. If I wasn't making 
prints for sale I wouldn't worry about it because there seems to be all the time 
you want in a world like that.  The differance's are not obvious. Even Nash 
editions uses 9600's for there black and white work,  neutralized color  
ultrachome inks I believe. 
It has been stated that 3 inks(tritone) printing is enough for smooth BW 
work, and these statements come for those way more expert than I,so I believe 100%
There is a large range of acceptable quality on this list, and its boiling 
down to a 1% gain hear and there, and what you can see with a loupe. 
I wouldn't doubt for a minute that Guys like Paul Rourke and Roy Harrington 
have at least 1000 hours in working up thier workflows. It wasn't exactly 
childs play figuring all this stuff out.
I started these topics just to point out that there are very good simple pro 
options out there, starting with the non chipped 7000-9000 series printers. 
And also to point out that those who skimp and try to save money usually don't 
in the long run.
Best thing to do is jump in somewhere and get started. Great results are 
everywhere. 
I plan to spend free time learning how to change out the heads in my 7000 for 
3 more years of printing.
Steve M.



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

Hi Steve,

Thanks to you and also to Paul and Tyler for your helpful replies. Re 
your comment about not getting hung up on dots, I hear that, and I'm 
only really trying to be thorough in this research phase that I'm in. 
I'm worried about how prints look to the naked eye, and I'm not 
concerned about what people might see if they start examining my 
prints with a loupe.

Tyler, thanks for that nice summary of the range of options. There 
are some software packages there that I hadn't heard of before, so 
I'll read up on those a little bit. I can see that's an important 
distinction you make between systems that rely upon RGB or CMYK 
curves in PS (where effectively you're working with narrow-gamut 
colour images) and those which use a monochrome image and do all the 
work in the driver. The former gives the advantage that you can 
create split-tone images where the tone is linked to things other 
than the grey level, which opens up some creative possibilities. The 
second approach theoretically allows fine tuning of the way the ink 
patterns are applied, specifically to give better B&W quality. It 
seems from your and Paul's comments that the Epson driver is still a 
pretty good choice though.

Paul, your observations about hiding the origins of digital images 
are very interesting. I guess I'm slowly coming to accept that I have 
to work *with* the look of digital output a bit more, rather than 
fighting against it constantly. As an amateur, the main audience I'm 
trying to please is my own unaided eyes, but they are a fairly 
critical audience. For instance I've seen jaggies on an A4 colour 
2100 print made from a full-size Minolta 5400 scan. I guess I'm the 
type of guy who probably won't relax completely until the true on-
paper resolution exceeds 1000ppi. That's not to say that in the 
meantime I won't be able to make some prints that I'm happy with -- 
it's certainly been true in colour and as you point out, things 
should be better in some ways in B&W.

That said, UT7 certainly seems like the obvious starting point for 
me. I want the flexibility of being able to generate good quality 
gloss and matte prints from the same inkset, and I'm definitely 
interested in split-toning. I'll probably try some black-only prints 
as well, and if I've understood correctly that should also be 
possible with that setup.

Best regards,

-= mike =-

Friday, April 23, 2004, 7:53:29 AM, you wrote:

m> For instance I've seen jaggies on an A4 colour 
m> 2100 print made from a full-size Minolta 5400 scan. 

Assuming the neg was scanned properly in the first place, I would suspect
then that it was not printed properly. At a full scan resolution and
printed at at least 240 dpi there should not be any "jaggies".

-- 
Best regards,
 Richard                            mailto:richard@...

Mike,

>...
>Paul, your observations about hiding the origins of digital images 
>are very interesting. I guess I'm slowly coming to accept that I have 
>to work *with* the look of digital output a bit more, rather than 
>fighting against it constantly. As an amateur, the main audience I'm 
>trying to please is my own unaided eyes, but they are a fairly 
>critical audience.

Visible digital artifacts at normal viewing distance are unacceptable to
most of us, I would guess.  Under normal viewing conditions, I think people
will not see dots with any of the modern dedicated B&W solutions, at least
on matte papers.  My original criteria included the ability to hang a
display 16 x 20 inch digital print, framed and under glass, next to my 16 x
20 darkroom prints and have the quality be equal.  Even the old 3000 can do
this with, for example, Epson Enhanced Matte.

> For instance I've seen jaggies on an A4 colour 
>2100 print made from a full-size Minolta 5400 scan.

That does surprise me a bit.  However, one caveat to my statement above
about not seeing digital artifacts with current systems is that, to me, the
technology is still not at a point where I'm satisfied with the glossiest
print papers.  The pearl and semi-gloss surfaces look excellent, at least
when sprayed with something like Lyson Print Guard, but the high gloss
surfaces do still look less than perfect to me.

> I guess I'm the 
>type of guy who probably won't relax completely until the true on-
>paper resolution exceeds 1000ppi. ...

Some claim to be able to see the difference between 360 dpi and 720 dpi
printing.  Frankly, I cannot see it on the matte papers that I generally use
for serious display prints.  (I have not tested this on glossy papers.)

The machines are not perfect, and I expect them to get better.  However, I
clearly think they are good enough that, overall and on average, my digital
prints are better than my darkroom prints.  I guess for me, that was the
cross-over point.  I went with the approach that, for me, produced the most
satisfying final product.

Paul
www.PaulRoark.com

--- In DigitalBlackandWhiteThePrint@yahoogroups.com, "mike_nunan"
<mike_nunan@h...> wrote:
snip...
>...I can see that's an important 
> distinction you make between systems that rely upon RGB or CMYK 
> curves in PS (where effectively you're working with narrow-gamut 
> colour images) and those which use a monochrome image and do all the 
> work in the driver.

Actually that's not what I meant. I'm assuming special inks for any of
this stuff, not color inks. The partitioning you asked about, and
necessary to gain the most benifits from these inksets, is done in
photoshop on the file itself then printed with the Epson RGB driver or
a CMYK RIP. as opposed to having a special driver do that for you.

> The former gives the advantage that you can 
> create split-tone images where the tone is linked to things other 
> than the grey level, which opens up some creative possibilities. The 
> second approach theoretically allows fine tuning of the way the ink 
> patterns are applied, specifically to give better B&W quality.

The second approach allows for split toning as well depending on the
inkset. I do it all the time here with StudioPrint, the right inks,
and a whacked brain. The Septone system makes it easy to do, and I'm
sure others here can tell you how they do it with their inksets and
drivers.

> It 
> seems from your and Paul's comments that the Epson driver is still a 
> pretty good choice though.

Actually I believe I said it was ill suited to the task. But if Paul
or others have good curves available for the particular
ink/paper/printer you use you're ok.
Tyler

Hi Paul...

> Mike,
> 
> >...
> >Paul, your observations about hiding the origins of digital images 
> >are very interesting. I guess I'm slowly coming to accept that I 
have 
> >to work *with* the look of digital output a bit more, rather than 
> >fighting against it constantly. As an amateur, the main audience 
I'm 
> >trying to please is my own unaided eyes, but they are a fairly 
> >critical audience.
> 
> Visible digital artifacts at normal viewing distance are 
unacceptable to
> most of us, I would guess.  Under normal viewing conditions, I 
think people
> will not see dots with any of the modern dedicated B&W solutions, 
at least
> on matte papers.  My original criteria included the ability to hang 
a
> display 16 x 20 inch digital print, framed and under glass, next to 
my 16 x
> 20 darkroom prints and have the quality be equal.  Even the old 
3000 can do
> this with, for example, Epson Enhanced Matte.

That's good to know, it sounds like I'm going to be happy with the 
results on that basis. Any print I care about is going to be made on 
matte paper and displayed behind glass.

> 
> > For instance I've seen jaggies on an A4 colour 
> >2100 print made from a full-size Minolta 5400 scan.
> 
> That does surprise me a bit.  

In response to that and to Richard's comment, I should say that 
despite pushing 40 I'm lucky enough (although maybe in this context 
maybe I should say "unlucky" ;o) to have very good eyes. When last 
tested a couple of years ago, I was told my vision was still rated 
60/20, and I can focus down to under 8" without squinting. That said, 
most people could *easily* see the jaggies on 300ppi laser prints 
(remember when those seemed high tech?!) and even after the jump to 
600ppi I could still see them on the curves of large characters. Now 
obviously that's pure black against white, which is maximum contrast 
and therefore the worst case. In the 2100 print I referred to there 
was a straight, sharp boundary between a deep red area and a white 
area, running at a shallow diagonal across the page. It doesn't 
surprise me that I could see the pixels on a 360ppi print in a 
situation like this, but to be fair it doesn't happen often. I do 
wonder how much difference I would see between a 360ppi print and a 
720ppi print of a "good" image, however. It's a test I'd like to make 
at some point when I have the opportunity. FWIW I'd come to these 
conclusions of my own accord, even though the mantra of "300ppi is 
enough" get repeated so often that it's take as a fact by many, so 
it's heartening to see there's at least one other person who agrees 
with me:

http://clarkvision.com/imagedetail/printer-ppi/

So maybe I'm not crazy after all! <g>

Oh, and regarding the scan, it was done at "full whack" on the 5400 
from a 35mm frame of AstiaF and printed as a full frame, so the final 
scaling gave about 600ppi of information to print from. I guess some 
would say I should have down-rezzed that to 360ppi, but I tend to 
think you'll get less aliasing in the driver if it has more data to 
work from in the first place. It seems to me that scaling from say 
365ppi to 360 would give more artifacts than scaling from 600 to 360 
for example. If someone can tell me a way of getting the Epson driver 
to give *exactly* 1:1 scaling in the final print, I'd love to hear 
about it. It looks to me that there's enough looseness in the way it 
sets margins etc that you can never be sure it's bang on.

> However, one caveat to my statement above
> about not seeing digital artifacts with current systems is that, to 
me, the
> technology is still not at a point where I'm satisfied with the 
glossiest
> print papers.  The pearl and semi-gloss surfaces look excellent, at 
least
> when sprayed with something like Lyson Print Guard, but the high 
gloss
> surfaces do still look less than perfect to me.

It's nice to have decent quality glossies to hand around, but I tend 
to look at them as disposable items. As I said in my earlier mails to 
you, I think the R800 can fill this gap quite nicely. Anything "fine 
art" will go onto matte.

> > I guess I'm the 
> >type of guy who probably won't relax completely until the true on-
> >paper resolution exceeds 1000ppi. ...
> 
> Some claim to be able to see the difference between 360 dpi and 720 
dpi
> printing.  Frankly, I cannot see it on the matte papers that I 
generally use
> for serious display prints.  (I have not tested this on glossy 
papers.)
> 

Doesn't that mean that you're never going to get the out-and-out 
sharpness on matte paper? But then again, already EEM looks pretty 
freakin' sharp to me with the standard inkset so maybe it is more to 
do with the way the ink settles onto the surface of the gloss media.

> The machines are not perfect, and I expect them to get better.  
However, I
> clearly think they are good enough that, overall and on average, my 
digital
> prints are better than my darkroom prints.  I guess for me, that 
was the
> cross-over point.  I went with the approach that, for me, produced 
the most
> satisfying final product.

So many people are saying that, I'm ready to be wowed by the results. 
Here's hoping! I just heard my offer has been accepted for the place 
I'm trying to buy, so that's another step towards having room to 
house my full B&W production suite. Exciting times...

All the best,

-= mike =-

--- In DigitalBlackandWhiteThePrint@yahoogroups.com, "Paul Roark" 
<paul.roark@v...> wrote:
>...  The pearl and semi-gloss surfaces look excellent, at least when 
> sprayed with something like Lyson Print Guard, but the high gloss 
> surfaces do still look less than perfect to me.

Agreed. I immediately fell in love with the Ilford SP paper after 
printing on it with the MIS Ultratone FS inkset for the first time. 
But for some reason all other micro-ceramic type papers don't fare 
too well except the inkjetart gloss lite version. At least this is my 
experience with the 1160 printer. My C80 printer gives good results 
for all these papers by mixing color ink with black at a slightly 
reduced dmax but still good. 

I've just purchased the Hawkmountain Merlin matte paper to try with 
the MIS Eboni Ultratone FS/FSN inkset. Have you had any experience 
with this combination? If this new paper gives me an acceptable dmax 
I might forget about coating altogether except for a few special 
enlargements.

--nick

Hi Tyler,

--- In DigitalBlackandWhiteThePrint@yahoogroups.com, "Tyler Boley" 
<tyler@t...> wrote:
> --- In DigitalBlackandWhiteThePrint@yahoogroups.com, "mike_nunan"
> <mike_nunan@h...> wrote:
> snip...
> >...I can see that's an important 
> > distinction you make between systems that rely upon RGB or CMYK 
> > curves in PS (where effectively you're working with narrow-gamut 
> > colour images) and those which use a monochrome image and do all 
the 
> > work in the driver.
> 
> Actually that's not what I meant. I'm assuming special inks for 
any of
> this stuff, not color inks. The partitioning you asked about, and
> necessary to gain the most benifits from these inksets, is done in
> photoshop on the file itself then printed with the Epson RGB 
driver or
> a CMYK RIP. as opposed to having a special driver do that for you.

Sorry, I wasn't being clear. I'm talking about the toner inks as 
being effectively colour inks, albeit very unsaturated ones. I can 
see how by using the RGB or CMYK curves, you can control the ink 
proportions that are used for each grey level in the image. That's 
what I understand to be the meaning of partioning. Correct?

> 
> > The former gives the advantage that you can 
> > create split-tone images where the tone is linked to things 
other 
> > than the grey level, which opens up some creative possibilities. 
The 
> > second approach theoretically allows fine tuning of the way the 
ink 
> > patterns are applied, specifically to give better B&W quality.
> 
> The second approach allows for split toning as well depending on 
the
> inkset. I do it all the time here with StudioPrint, the right inks,
> and a whacked brain. The Septone system makes it easy to do, and 
I'm
> sure others here can tell you how they do it with their inksets and
> drivers.
> 

Yes, but you can only split based upon grey level in the original 
image, true? AIUI with the other approach you can do clever stuff 
such as applying one curve set to one part of the image and a 
different curve set to another, in order to get a different split.

> > It 
> > seems from your and Paul's comments that the Epson driver is 
still a 
> > pretty good choice though.
> 
> Actually I believe I said it was ill suited to the task. But if 
Paul
> or others have good curves available for the particular
> ink/paper/printer you use you're ok.
> Tyler

You did, but you also said that you'd seen great Cone ICC prints, so 
maybe my natural optimism was running away with me a bit there! Can 
you elaborate on the shortcomings of the Epson driver?

Cheers,

-= mike =-

--- In DigitalBlackandWhiteThePrint@yahoogroups.com, "mike_nunan"
<mike_nunan@h...> wrote:
snip...
> Sorry, I wasn't being clear. I'm talking about the toner inks as
> being effectively colour inks, albeit very unsaturated ones. I can
> see how by using the RGB or CMYK curves, you can control the ink
> proportions that are used for each grey level in the image. That's
> what I understand to be the meaning of partioning. Correct?

Partitioning, the way the term tends to be used here, and how I think
of it, has more to do with traditional quad inksets, and is expanded
from there to account for other inksets. So in it's more basic form,
it has to do with sending the appropriate part of the tonal scale in
the image to the appropriate gray ink, and making the transitions
between the inks linear. So for a normal, non-tinting quad inkset, you
are sending highlights to the light ink, mids to the middle gray ink,
dark tones to the dark gray ink, and blacks to the black ink, to put
it very simplistically. In inksets that also use toners, other issues
come into play to adjust them, but usually there are some inks to
partitioned as mentioned, and others to effect hue (they may also have
dark and light components that require partitioning).

snip...
> > The second approach allows for split toning as well depending on
> the
> > inkset. I do it all the time here with StudioPrint, the right inks,
> > and a whacked brain. The Septone system makes it easy to do, and
> I'm
> > sure others here can tell you how they do it with their inksets and
> > drivers.
> >
>
> Yes, but you can only split based upon grey level in the original
> image, true? AIUI with the other approach you can do clever stuff
> such as applying one curve set to one part of the image and a
> different curve set to another, in order to get a different split.

Yes that's true, though the newest version of ImagePrint is supposed
to have more flexibility to do what you suggest.

> > > It
> > > seems from your and Paul's comments that the Epson driver is
> still a
> > > pretty good choice though.
> >
> > Actually I believe I said it was ill suited to the task. But if
> Paul
> > or others have good curves available for the particular
> > ink/paper/printer you use you're ok.
> > Tyler
>
> You did, but you also said that you'd seen great Cone ICC prints, so
> maybe my natural optimism was running away with me a bit there! Can
> you elaborate on the shortcomings of the Epson driver?

Since it works with RGB input, everything must be done with RGB
curves. There isn't a straight through path from the file to the
specific inks, which are not three channels of R, G, and B. What the
driver does after the fact, while possibly good for standard color
printing with Epson inks and papers, may not be ideal for what we'd
like to do with these inksets, and sometimes is very difficult to
precisely predict. Not only do you have no direct control of the black
ink (you have to control it with specific RGB numbers) but some RGB
values will cause density to go down instead of up the way we are
attempting to control it. This is because at certain RGB levels the
driver is pulling out CMY to replace it with K, and this may not work
well with the inkset we are using. The way it works in general is
sometimes unpredictable, usually counter intuitive, occasionally
outright maddening as some curve change you made causes a totally
unexpected result.
It can obviously be done, but it is difficult. Even with a good
profile, which helps you predict the behavior, I wound up doing a lot
of trial and error. Way back before there was Piezography, Lyson
quads, or 1/10th the knowledge we have now, I threw in the towel and
went with a RIP, which gives direct control of each ink with no
intervention.
Every once in a while I have reason to go back and try it again
thinking I know more and have better tools now, and encounter the same
obstacles.
I believe somewhere in the files section of this list there may be a
more extensive write-up of partitioning with the Epson driver.
What Cone is doing is very interesting, but proprietory and little is
known of their method though I have a few guesses and have made
similar profiles with difficulty. I've seen the profiles and they do
indeed do all of this partitioning and preview well.
Tyler

Tyler wrote:

>...
>Partitioning, the way the term tends to be used here, and how I think
>of it, has more to do with traditional quad inksets, and is expanded
>from there to account for other inksets. So in it's more basic form,
>it has to do with sending the appropriate part of the tonal scale in
>the image to the appropriate gray ink, 

The Epson driver regularly "partitions" the inks between the light and dark
magenta and cyan, the colors and black, and now the light and dark black.

One reason I prefer to put my value-added on top of the Epson driver is that
I believe Epson does an outstanding job of partitioning its inks. 

I believe Jon Cone's ICC system is also using the Epson driver.

>Since [the Epson driver] works with RGB input, 
>everything must be done with RGB curves.

Yes, and I find this rather convenient, open, flexible, and, of course,
cheap.  With the newest inksets, because I've made inksets that basically
follow Epson's lead, the Epson driver does most of the heavy lifting with
respect to partitioning.  In fact, the inks can be printed with no curves at
all.  Writing curves for the newest inksets is much easier than it used to
be.

>There isn't a straight through path from the file to the
>specific inks, ...

Being able to print with no curves and no special software seems like a very
direct path.

However, it's true that the Epson driver does not give me total control of
every jet.  There might, indeed, be circumstances where I'd take advantage
of such control, and, overall, I'd probably prefer CMYK to RGB.

If I were a pro shop doing color work, I'd probably have a 9600 with a good
RIP so that I could do both B&W and Color display prints with the machine.  

However, my medium is B&W.  With a dedicated B&W inkset, including both
monotone and variable-tone approaches, I think the Epson driver can deliver
excellent results.

>...
>Not only do you have no direct control of the black
>ink (you have to control it with specific RGB numbers) ...

Yes, but this really is not much of a problem.  

> some RGB values will cause density to go down instead of up the 
>way we are attempting to control it. ...

This happens when the total ink load has reached the driver's limit for the
paper type designated.  If I then try to darken the image with a relatively
light ink, the driver puts in more of the light ink but pulls out an equal
amount of the dark ink.  So, the image gets lighter instead of darker. 

Where there are 2 inks, using the darker one or moving them both tends to
avoid the problem.  With the UT2 & 7 inksets I don't run into this issue
very often.  

I'm not saying most will want to get into the curves, but experienced users
will find the job has gotten easier.  That is part of what I'm doing with
these inksets.

>... I wound up doing a lot of trial and error...

Yes, so did I, and I still do some.  I generally find it takes 6 test strips
to make a new profile for a new paper.  That's one letter-size sheet or
paper.  Of course, those who have not done this might find it very
frustrating, but for those who are accustomed to the curves approach, the
job has gotten easier.

Again, however, if I was a business that could pass the cost onto customers,
I might find there is an easier way that would be more efficient for me.  As
a B&W printer for my own uses, however, the cost effectiveness and openness
of the Epson driver is going to be hard to beat.

Paul
www.PaulRoark.com 

For UT2 information, curves, and settings see:
http://home1.gte.net/res09aij/

Hi All,

  Just A quick Post. I invite everyone to visit my website. and check out
the 3 new Hahnemuhle products that Have just hit the market..
  1. Art Proof Paper
  2. Pro Kit  (A Little of Everything.. Including Hahnemuhle profiles on CD
  3. Certificates of Authenticity (With a website by Hahnemuhle)

  If anybody has questions Please email or Call anytime

  Jim Doyle

  J.Doyle Enterprises L.L.C.
  114 Old Orchard Rd.
  Cherry Hill, NJ 08003
  856-424-8660
  http://www.shadesofpaper.com


  Please visit the Group Homepage to check the Files, and other resources as
they are often being updated.

  http://groups.yahoo.com/group/DigitalBlackandWhiteThePrint

  If you wish to receive no emails or just a daily digest, or you wish to
unsubscribe, please edit your Membership preferences by visiting this same
page.

  Please follow these basic guidelines:
  - As threads develop, trim off excess portions of earlier messages to keep
them short.
  - Good manners are required at all time. No personal attacks or flames.
Hostile, aggressive or argumentative users may be removed from the
membership without notice.
  - Keep your posts and threads related to the group topic of digital B&W
printing. Users who persistently make off-topic posts may be removed from
the membership.
  - By posting on this forum you agree to abide by the group rules and
guidelines, and to abide by the actions and decisions of the group Owner and
Moderators. See “Group Topic, Rules and Guidelines” in the Files section:
  http://groups.yahoo.com/group/DigitalBlackandWhiteThePrint/files/

  BY PARTICIPATING IN AND/OR POSTING MESSAGES TO THE DIGITAL BW, THE PRINT
YAHOO! GROUP YOU EXPRESSLY UNDERSTAND AND AGREE THAT THE “OWNER” AND
“MODERATORS” OF DIGITAL BW, THE PRINT YAHOO GROUP SHALL NOT BE LIABLE TO YOU
FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, CONSEQUENTIAL OR EXEMPLARY
DAMAGES, INCLUDING BUT NOT LIMITED TO, DAMAGES FOR LOSS OF PROFITS,
GOODWILL, USE, DATA OR OTHER INTANGIBLE LOSSES (EVEN IF THE  “OWNER” AND
“MODERATORS” OF DIGITAL BW, THE PRINT YAHOO GROUP HAVE BEEN ADVISED OF THE
POSSIBILITY OF SUCH DAMAGES), RESULTING FROM: (i) THE USE OR THE INABILITY
TO USE THE DIGITAL BW, THE PRINT YAHOO GROUP; (ii) UNAUTHORIZED ACCESS TO OR
ALTERATION OF YOUR TRANSMISSIONS OR DATA; (iii) STATEMENTS OR CONDUCT OF ANY
THIRD PARTY ON THE DIGITAL BW, THE PRINT YAHOO GROUP; OR (iv) ANY OTHER
MATTER RELATING TO THE DIGITAL BW, THE PRINT YAHOO GROUP.




----------------------------------------------------------------------------
--
  Yahoo! Groups Links

    a.. To visit your group on the web, go to:
    http://groups.yahoo.com/group/DigitalBlackandWhiteThePrint/

    b.. To unsubscribe from this group, send an email to:
    DigitalBlackandWhiteThePrint-unsubscribe@yahoogroups.com

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



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

Sunday, April 25, 2004, 6:03:06 AM, you wrote:




JD> \ufffd Hi All,

JD> \ufffd Just A quick Post. I invite everyone to visit my website. and check out
JD> the 3 new Hahnemuhle products that Have just hit the market..
JD> \ufffd 1. Art Proof Paper
JD> \ufffd 2. Pro Kit\ufffd (A Little of Everything.. Including Hahnemuhle profiles on CD
JD> \ufffd 3. Certificates of Authenticity (With a website by Hahnemuhle)

JD> \ufffd If anybody has questions Please email or Call anytime

JD> \ufffd Jim Doyle

I see Art Proof priced at "Art Proof (8.5 x 11) - $48.00"

How many sheets is that?  50?

Anyone out there tried this stuff as a more accurate proofer to PR
than EEM?

-- 
Best regards,
 Richard                            mailto:richard@...

Rich

Thyanks For Pointing That Out!
50 SHTS Box!!

Jim Doyle

J.Doyle Enterprises L.L.C.
114 Old Orchard Rd.
Cherry Hill, NJ 08003
856-424-8660
http://www.shadesofpaper.com

  -----Original Message-----
  From: Richard Sintchak [mailto:richcontaxg@...]
  [Jim Doyle]
  Sent: Sunday, April 25, 2004 10:24 AM
  To: Jim Doyle
  Subject: Re[2]: [Digital BW] 3 New Products from Hahnemuhle



  Sunday, April 25, 2004, 6:03:06 AM, you wrote:




  JD>   Hi All,

  JD>   Just A quick Post. I invite everyone to visit my website. and check
out
  JD> the 3 new Hahnemuhle products that Have just hit the market..
  JD>   1. Art Proof Paper
  JD>   2. Pro Kit  (A Little of Everything.. Including Hahnemuhle profiles
on CD
  JD>   3. Certificates of Authenticity (With a website by Hahnemuhle)

  JD>   If anybody has questions Please email or Call anytime

  JD>   Jim Doyle

  I see Art Proof priced at "Art Proof (8.5 x 11) - $48.00"

  How many sheets is that?  50?

  Anyone out there tried this stuff as a more accurate proofer to PR
  than EEM?

  --
  Best regards,
  Richard                            mailto:richard@...



  Please visit the Group Homepage to check the Files, and other resources as
they are often being updated.

  http://groups.yahoo.com/group/DigitalBlackandWhiteThePrint

  If you wish to receive no emails or just a daily digest, or you wish to
unsubscribe, please edit your Membership preferences by visiting this same
page.

  Please follow these basic guidelines:
  - As threads develop, trim off excess portions of earlier messages to keep
them short.
  - Good manners are required at all time. No personal attacks or flames.
Hostile, aggressive or argumentative users may be removed from the
membership without notice.
  - Keep your posts and threads related to the group topic of digital B&W
printing. Users who persistently make off-topic posts may be removed from
the membership.
  - By posting on this forum you agree to abide by the group rules and
guidelines, and to abide by the actions and decisions of the group Owner and
Moderators. See “Group Topic, Rules and Guidelines” in the Files section:
  http://groups.yahoo.com/group/DigitalBlackandWhiteThePrint/files/

  BY PARTICIPATING IN AND/OR POSTING MESSAGES TO THE DIGITAL BW, THE PRINT
YAHOO! GROUP YOU EXPRESSLY UNDERSTAND AND AGREE THAT THE “OWNER” AND
“MODERATORS” OF DIGITAL BW, THE PRINT YAHOO GROUP SHALL NOT BE LIABLE TO YOU
FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, CONSEQUENTIAL OR EXEMPLARY
DAMAGES, INCLUDING BUT NOT LIMITED TO, DAMAGES FOR LOSS OF PROFITS,
GOODWILL, USE, DATA OR OTHER INTANGIBLE LOSSES (EVEN IF THE  “OWNER” AND
“MODERATORS” OF DIGITAL BW, THE PRINT YAHOO GROUP HAVE BEEN ADVISED OF THE
POSSIBILITY OF SUCH DAMAGES), RESULTING FROM: (i) THE USE OR THE INABILITY
TO USE THE DIGITAL BW, THE PRINT YAHOO GROUP; (ii) UNAUTHORIZED ACCESS TO OR
ALTERATION OF YOUR TRANSMISSIONS OR DATA; (iii) STATEMENTS OR CONDUCT OF ANY
THIRD PARTY ON THE DIGITAL BW, THE PRINT YAHOO GROUP; OR (iv) ANY OTHER
MATTER RELATING TO THE DIGITAL BW, THE PRINT YAHOO GROUP.




----------------------------------------------------------------------------
--
  Yahoo! Groups Links

    a.. To visit your group on the web, go to:
    http://groups.yahoo.com/group/DigitalBlackandWhiteThePrint/

    b.. To unsubscribe from this group, send an email to:
    DigitalBlackandWhiteThePrint-unsubscribe@yahoogroups.com

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



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

In a message dated 4/25/2004 7:25:22 AM Pacific Daylight Time, 
richcontaxg@... writes:

> Anyone out there tried this stuff as a more accurate proofer to PR
> than EEM?
> 
> -- 
> Best regards,
> Richard               mailto:richard@...
> 
Richard, I've been using it for the last few days, YES it is a lot more 
accurate than EEM.
It prints very acurate to PhotoRag, and Turner. You could sell the prints 
except its painfully thin, see through even. I plan on using it regularly. Across 
the board its 1/2 the price of the real Hanna papers. Its amazing that this 
paper can take the ink being so thin but it can. Did I say its  Thin, ;0) It 
shows how crappy EEM really is. BW printing of course, probably equal in color 
though. Only suitable for accurate proofing. Not sure if its the type of proof 
you would show clients(too flimsy) but great for getting an accurate density 
contrast check as well as tone before going to finals on the thick stuff. Dmax 
is equal to photorag.
Steve Meyers



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

>   2. Pro Kit  (A Little of Everything.. Including Hahnemuhle
> profiles on CD

there were some of the Hahnem\ufffdhle made profiles online -0 anyone remember
where?

tim

--- In DigitalBlackandWhiteThePrint@yahoogroups.com, "Paul Roark"
<paul.roark@v...> wrote:
snip...
> One reason I prefer to put my value-added on top of the Epson driver
is that
> I believe Epson does an outstanding job of partitioning its inks.

I have no idea what "value-added" is. However, as you say the driver
does an excellent job of partitioning ITS inks. Though I'm sure this
can be utilized for our purposes, I've also worked with inksets not
perfectly designed for it, therefore impossible to perfectly linearize
since they relied on some of the driver's partitioning.

> I believe Jon Cone's ICC system is also using the Epson driver.

Yes, as I said earlier in the thread.

> >Since [the Epson driver] works with RGB input,
> >everything must be done with RGB curves.
>
> Yes, and I find this rather convenient, open, flexible, and, of
course,
> cheap.  With the newest inksets, because I've made inksets that
basically
> follow Epson's lead, the Epson driver does most of the heavy lifting
with
> respect to partitioning.  In fact, the inks can be printed with no
curves at
> all.  Writing curves for the newest inksets is much easier than it
used to
> be.
>
> >There isn't a straight through path from the file to the
> >specific inks, ...
>
> Being able to print with no curves and no special software seems
like a very
> direct path.

That's not what I meant. With respect to what you say above though, it
applies to that inkset only.

snip
>...With a dedicated B&W inkset, including both
> monotone and variable-tone approaches, I think the Epson driver can
deliver
> excellent results.

Also as I said earlier in the thread. Within the context of this
thread which discussed various inkset approaches and drivers, I said
the Epson driver can be used well, but is not well suited to the task.
In the next breath I said if you have curves available for them they
are in luck.
Then I was asked more specifically what shortcomings the Epson driver
may have for some of this work, hence my comments.
I have no personal interest in whether people choose to work with the
Epson driver or not. I began working with if for B&W work beginning in
early 1999, with all kinds of inksets, on many Epson models. I've made
RGB curves for everything from many different MIS sets, Lyson quads,
several Cone iterations. I'm fully aware of what can and can't be done
with it. Also heavily involved in color, I can say
without reservation that most agree the driver is even ill suited for
it's intended use- color.
After working with drivers that impose no limitations, require no
work-arounds, and let me do whatever I want with any inkset I want, my
preference moves there.

However I make no particular recommendations based on those
preferences, and fully acknowledged solutions using the Epson driver
including yours and Cone's.
Tyler

Hi Tyler,

Thanks for that very helpful explanation, that's made things another 
step clearer. Foretunately I don't foresee a need for me to get into 
crafting my own curves, as I'm likely to select a two or three media 
types from among the usual suspects and simply use Paul's curves as-
is. Based on the feedback I've seen from users here, I'm expecting to 
be fairly well pleased by the results, but if not at least I won't 
have spent much money finding out.

That said, I'm still toying with the idea of switching to ImagePrint 
for my colour work. I've tried the demo but it's a shame that they 
chose such a heavy "DEMO" watermark as I found it interfered with 
assessment of the images and of course it puts you off making too 
many of them. I'm also troubled by comments I've seen saying that it 
sacrifices some resolution in comparison to the Epson driver due to 
the larger dither cell size. Anyway, at the risk of veering off 
topic, I'd love to hear your thoughts on any of that, as well as the 
ill-suitedness of the Epson driver for colour work in general.

Best regards,

-= mike =-

--- In DigitalBlackandWhiteThePrint@yahoogroups.com, "Tyler Boley" 
<tyler@t...> wrote:
> --- In DigitalBlackandWhiteThePrint@yahoogroups.com, "mike_nunan"
> <mike_nunan@h...> wrote:
> snip...
> > Sorry, I wasn't being clear. I'm talking about the toner inks as
> > being effectively colour inks, albeit very unsaturated ones. I can
> > see how by using the RGB or CMYK curves, you can control the ink
> > proportions that are used for each grey level in the image. That's
> > what I understand to be the meaning of partioning. Correct?
> 
> Partitioning, the way the term tends to be used here, and how I 
think
> of it, has more to do with traditional quad inksets, and is expanded
> from there to account for other inksets. So in it's more basic form,
> it has to do with sending the appropriate part of the tonal scale in
> the image to the appropriate gray ink, and making the transitions
> between the inks linear. So for a normal, non-tinting quad inkset, 
you
> are sending highlights to the light ink, mids to the middle gray 
ink,
> dark tones to the dark gray ink, and blacks to the black ink, to put
> it very simplistically. In inksets that also use toners, other 
issues
> come into play to adjust them, but usually there are some inks to
> partitioned as mentioned, and others to effect hue (they may also 
have
> dark and light components that require partitioning).
> 
> snip...
> > > The second approach allows for split toning as well depending on
> > the
> > > inkset. I do it all the time here with StudioPrint, the right 
inks,
> > > and a whacked brain. The Septone system makes it easy to do, and
> > I'm
> > > sure others here can tell you how they do it with their inksets 
and
> > > drivers.
> > >
> >
> > Yes, but you can only split based upon grey level in the original
> > image, true? AIUI with the other approach you can do clever stuff
> > such as applying one curve set to one part of the image and a
> > different curve set to another, in order to get a different split.
> 
> Yes that's true, though the newest version of ImagePrint is supposed
> to have more flexibility to do what you suggest.
> 
> > > > It
> > > > seems from your and Paul's comments that the Epson driver is
> > still a
> > > > pretty good choice though.
> > >
> > > Actually I believe I said it was ill suited to the task. But if
> > Paul
> > > or others have good curves available for the particular
> > > ink/paper/printer you use you're ok.
> > > Tyler
> >
> > You did, but you also said that you'd seen great Cone ICC prints, 
so
> > maybe my natural optimism was running away with me a bit there! 
Can
> > you elaborate on the shortcomings of the Epson driver?
> 
> Since it works with RGB input, everything must be done with RGB
> curves. There isn't a straight through path from the file to the
> specific inks, which are not three channels of R, G, and B. What the
> driver does after the fact, while possibly good for standard color
> printing with Epson inks and papers, may not be ideal for what we'd
> like to do with these inksets, and sometimes is very difficult to
> precisely predict. Not only do you have no direct control of the 
black
> ink (you have to control it with specific RGB numbers) but some RGB
> values will cause density to go down instead of up the way we are
> attempting to control it. This is because at certain RGB levels the
> driver is pulling out CMY to replace it with K, and this may not 
work
> well with the inkset we are using. The way it works in general is
> sometimes unpredictable, usually counter intuitive, occasionally
> outright maddening as some curve change you made causes a totally
> unexpected result.
> It can obviously be done, but it is difficult. Even with a good
> profile, which helps you predict the behavior, I wound up doing a 
lot
> of trial and error. Way back before there was Piezography, Lyson
> quads, or 1/10th the knowledge we have now, I threw in the towel and
> went with a RIP, which gives direct control of each ink with no
> intervention.
> Every once in a while I have reason to go back and try it again
> thinking I know more and have better tools now, and encounter the 
same
> obstacles.
> I believe somewhere in the files section of this list there may be a
> more extensive write-up of partitioning with the Epson driver.
> What Cone is doing is very interesting, but proprietory and little 
is

> known of their method though I have a few guesses and have made
> similar profiles with difficulty. I've seen the profiles and they do
> indeed do all of this partitioning and preview well.
> Tyler

DigitalBlackandWhiteThePrint@yahoogroups.com, "mike_nunan"
<mike_nunan@h...> wrote:
snip
> That said, I'm still toying with the idea of switching to ImagePrint 
> for my colour work. I've tried the demo but it's a shame that they 
> chose such a heavy "DEMO" watermark as I found it interfered with 
> assessment of the images and of course it puts you off making too 
> many of them. I'm also troubled by comments I've seen saying that it 
> sacrifices some resolution in comparison to the Epson driver due to 
> the larger dither cell size. Anyway, at the risk of veering off 
> topic, I'd love to hear your thoughts on any of that, as well as the 
> ill-suitedness of the Epson driver for colour work in general.

I've heard no negative comments about the IP dither, this is the first
I've heard about a larger cell size. On the other hand I don't use it
so have not been very attentive to discussions about it's particulars
quirks. We recently did a comparison of B&W output and IP's quad
output looked great in terms of dither and dot size, but it was
non-linear with no linearization capability. That many or may not be
relevant to it's color output. No specific resolution test was done.
The problem with the Epson driver for color is that it is extremely
non-linear and gray balance is completely whacked in it's unprofiled
state using the No Color Adjustment setting. Other settings for quick
and easy use like Automatic or PhotoRealistc are much better but at
the expense of significant gamut loss. So when profiling over the
accepted NCA setting, the software has to do so much wrangling of
color to bring things into line that it's always at some expense. As
Nick Wheeler says "Profiling packages were designed to provide that
last nth of characterization to an already well calibrated system." So
many critical printers use RIPs that can be linearized first, then
profiled.
The now legendary and monumental efforts of Bill Atkinson to make his
profiles for the x600 printers, literally making his own software,
combining other software packages, and measuring  many thousands of
patches, just to come up with profiles that could make the Epson print
reasonably neutral is probably the best illustration of the problems
with the driver. A friend recently got and profiled one of the newer
little Canon's, and it was far superior in it's unprofiled state,
there's no reason why Epson couldn't accomplish the same feat.
Tyler

Hi Tyler,

I wish I could remember where I saw those comments about the dither 
cell size. It may have even been in a discussion in this group, but 
wherever it was, it was a few months ago now. To be accurate, it was 
a comment made by one person only, with others agreeing that they 
could understand why it might be so. The logic was something along 
the lines of this: IP provides finer tonal gradation than the Epson 
driver, but does it by using a larger dither cell size.

> We recently did a comparison of B&W output and IP's quad output 
looked great 

I didn't realise IP could use quadtone inksets or is that not what 
you meant?


Thanks for the extra details about the Epson driver. I just read 
through Bill Atkinson's PDF file about the creation of his profiles --
 that's an eye-opener in itself. I can see how having perceptual 
linearity is ideal, and otherwise it becomes a guessing (or 
measuring) game in order to determine what it costs you. I suppose it 
would all depend upon the slope of the "worst" parts of the transfer 
curve, plus the number of bits of accuracy that are being used to 
represent the values being sent to the driver (out of interest, do 
know the answer to that second point?). Bill's success with his 
profiles tends to indicate that the Epson driver is somewhere on the 
right side of pathalogical in those respects. Still, as you say it's 
a shame Epson didn't remove this hurdle in the first place.

Colour printing is still an open issue for me, as I'm getting very 
variable results. While I'm getting good proportion of satisfactory 
prints, I've learned to dread any image with a strong dominant tone 
(sunsets, shots taken under coloured artificial light, etc). I guess 
I'm changing tack to B&W for a while because a change is as good as a 
rest. Can you suggest any groups or forums where I could find some 
solid information about the different colour RIP options? I've been 
lurking in the IP group but I think I'd be better off somewhere 
that's not devoted to one particular product. Also, is there a decent 
free RIP out there, like QTR but for colour?

Anyway, thanks for all the help so far, and I'll try to go bug 
someone else with my questions after this ;o)

-= mike =-

--- In DigitalBlackandWhiteThePrint@yahoogroups.com, "mike_nunan"
<mike_nunan@h...> wrote:
snip...
> I didn't realise IP could use quadtone inksets or is that not what 
> you meant?

Yes, it does, or did, support quad inksets and profiles were available.

> Thanks for the extra details about the Epson driver. I just read 
> through Bill Atkinson's PDF file about the creation of his profiles --
>  that's an eye-opener in itself. I can see how having perceptual 
> linearity is ideal, and otherwise it becomes a guessing (or 
> measuring) game in order to determine what it costs you. I suppose it 
> would all depend upon the slope of the "worst" parts of the transfer 
> curve, plus the number of bits of accuracy that are being used to 
> represent the values being sent to the driver (out of interest, do 
> know the answer to that second point?).

I believe out of necessity nearly all printer drivers have 16 bit
internal ramping. Is that what you meant?
But the problems lies less in individual channel linearity then it
does with gray balance, and the manner the Epson driver does its GCR.

snip
>...Can you suggest any groups or forums where I could find some 
> solid information about the different colour RIP options? I've been 
> lurking in the IP group but I think I'd be better off somewhere 
> that's not devoted to one particular product.

That's a problem, I've run across little objective comparison. Most
user forums get a bit testy about it, we all spent so much on our
software we tend to defend it too adamantly to justify the expense <G>.

> Also, is there a decent 
> free RIP out there, like QTR but for colour?

Not that I'm aware of, however there is an alpha version of GimpPrint
that looks very promising. A lot of individual channel tuning and K
generation controls. I'm dying to see if I can create an RGB tuned
setting with it that behaves much better, then profile over it. The
problem is that I haven't been able to make it print at all. When it
does work though, it will still be a lot of work just to get it to the
point of reasonalbe evaluation.
If Adobe hadn't killed PressReady we'd have something to talk about.
Even though it has no linearization or limiting, with good profiles it
out-performs the Epson driver for color and was very affordable.
Photoprinter version of many of these RIPs are coming out as
manufacturers are recongizing the market, that are much more
affordable, so there's hope. There's also scuttlebutt that Epson has
been listening and may address some of the driver issues in coming models.
Tyler

Hi Tyler,

--- In DigitalBlackandWhiteThePrint@yahoogroups.com, "Tyler Boley" 
<tyler@t...> wrote:
> --- In DigitalBlackandWhiteThePrint@yahoogroups.com, "mike_nunan"
> <mike_nunan@h...> wrote:
> snip...
> > I didn't realise IP could use quadtone inksets or is that not 
what 
> > you meant?
> 
> Yes, it does, or did, support quad inksets and profiles were 
available.
> 

I guess not in the Lite version though, that probably explains why I 
wasn't aware of it.

<snip>
> > I suppose it 
> > would all depend upon the slope of the "worst" parts of the 
transfer 
> > curve, plus the number of bits of accuracy that are being used to 
> > represent the values being sent to the driver (out of interest, 
do 
> > know the answer to that second point?).
> 
> I believe out of necessity nearly all printer drivers have 16 bit
> internal ramping. Is that what you meant?
> But the problems lies less in individual channel linearity then it
> does with gray balance, and the manner the Epson driver does its 
GCR.

I would expect at least 16-bit precision in all internal calculations 
and data tables in the driver, probably 32-bit in fact. If you're 
talking about a RIP such as IP which works directly from the TIF 
file, then by using high-bit files you should also get 16-bit 
precision for the input values. I'm more suspicious regarding the 
precision used to pass values from PS via the OS to the Epson driver, 
especially since Adobe have been so tardy about implementing high-bit 
support even in the places where it's immediately obvious. It's this 
lack of precision that could make it impossible to compensate for non-
linearity upstream of the driver, whether by curves or profiling, 
without introducing banding. I appreciate that the grey balance is an 
issue too, but I'm taking the long view that it *can* be fixed via 
profiling provided you have enough precision to tweak the values 
accurately enough. I do realise that this is non-trivial to do (an 
understatement!) but the other problem is worse still.

> 
> snip
> >...Can you suggest any groups or forums where I could find some 
> > solid information about the different colour RIP options? <snip>
> 
> That's a problem, I've run across little objective comparison. Most
> user forums get a bit testy about it, we all spent so much on our
> software we tend to defend it too adamantly to justify the expense 
<G>.

Your candour is refreshing sir =)

I guess I'll just have to stick around here and pester you a bit 
more, sorry!

> 
> > Also, is there a decent 
> > free RIP out there, like QTR but for colour?
> 
> Not that I'm aware of, however there is an alpha version of 
GimpPrint
> that looks very promising. A lot of individual channel tuning and K
> generation controls. I'm dying to see if I can create an RGB tuned
> setting with it that behaves much better, then profile over it. The
> problem is that I haven't been able to make it print at all. When it
> does work though, it will still be a lot of work just to get it to 
the
> point of reasonalbe evaluation.
> If Adobe hadn't killed PressReady we'd have something to talk about.
> Even though it has no linearization or limiting, with good profiles 
it
> out-performs the Epson driver for color and was very affordable.
> Photoprinter version of many of these RIPs are coming out as
> manufacturers are recongizing the market, that are much more
> affordable, so there's hope. There's also scuttlebutt that Epson has
> been listening and may address some of the driver issues in coming 
models.

I think I'll wait to see what becomes of GimpPrint then. I'm 
reluctant to spend big money on a commercial RIP at this stage, and 
I'm not in a rush. I'm also in favour of using something that's a bit 
more open, so I can really know and control what's going on under the 
covers.

TTFN

-= mike =-

--- In DigitalBlackandWhiteThePrint@yahoogroups.com, "mike_nunan"
<mike_nunan@h...> wrote:

snip

> I would expect at least 16-bit precision in all internal calculations
> and data tables in the driver, probably 32-bit in fact. If you're
> talking about a RIP such as IP which works directly from the TIF
> file, then by using high-bit files you should also get 16-bit
> precision for the input values.

It still depends. THat may be the case for IP, but though it and some
other drivers/RIPs can accept 16 bit input, whether or not it's
actually utilized remains uncertain. I've seen very few definitive
statements from any of these people that they actually do. There have
been indications or hints that these drivers/RIPs do some manipulation
of input data to their needs. Certainly some kind of ppi remapping is
done, and bit downsampling could be happening as well.

> I'm more suspicious regarding the
> precision used to pass values from PS via the OS to the Epson driver,
> especially since Adobe have been so tardy about implementing high-bit
> support even in the places where it's immediately obvious.

It has more to do with the Epson driver than Adobe, and may be OS
related as you suggest (but I know little about that). Some time ago
Photoshop let you hit the print command with files that were high bit
(it didn't used to) but all it was doing was changing to 8 bit on the
fly as a convenience since that's what the driver accepts.

> It's this
> lack of precision that could make it impossible to compensate for non-
> linearity upstream of the driver, whether by curves or profiling,
> without introducing banding. I appreciate that the grey balance is an
> issue too, but I'm taking the long view that it *can* be fixed via
> profiling provided you have enough precision to tweak the values
> accurately enough. I do realise that this is non-trivial to do (an
> understatement!) but the other problem is worse still.

This all gets into a huge issue about where the best place for massive
correction is best done. You can send 8 bit pure color gradients (not
neutral) through the epson driver with no profile conversion and see
good gradients on paper. In my experience most color banding comes
from converting to profiles, and doing so in high bit did not help.
C.D.Tobie once explained it well to me but I've lost the email. It has
to do with large remapping of lab values, and best guesses. When the
driver is so out of balance with massive crossovers etc., remapping
color based on LUTs becomes problematic. You can wind up jerking close
colors in multiple directions. So it still makes sense to have the
driver as linear as possible in the first place, then profiles can do
their jobs somewhat coherently. Others with more day to day experience
with profiling in different environments/drivers/RIPs etc. would have
more valuable opinions about how much practical difference it makes.

> I think I'll wait to see what becomes of GimpPrint then. I'm
> reluctant to spend big money on a commercial RIP at this stage, and
> I'm not in a rush.

The fact is, despite this somewhat techie discussion, people make
beautiful prints every day within these limitations. Profiles are like
darkroom materials, you get the best one you can, learn what it can
and can't do, and make it work for you. Softproofing makes this much
easier. But it's like type C paper, neither Kodak nor Fuji was "right"
but people decided which one they liked and made great prints based on
it's strengths and weaknesses. Sometimes I think the high degree of
technical discussion around digital raises our expectations too high,
we think technical perfection may be attainable, but it ain't so.
Which brings us back to art, thankfully.
Tyler

Hello again Tyler,

--- In DigitalBlackandWhiteThePrint@yahoogroups.com, "Tyler Boley" 
<tyler@t...> wrote:
> --- In DigitalBlackandWhiteThePrint@yahoogroups.com, "mike_nunan"
> <mike_nunan@h...> wrote:
> 
> snip
> 
> > I would expect at least 16-bit precision in all internal 
calculations
> > and data tables in the driver, probably 32-bit in fact. If you're
> > talking about a RIP such as IP which works directly from the TIF
> > file, then by using high-bit files you should also get 16-bit
> > precision for the input values.
> 
> It still depends. THat may be the case for IP, but though it and 
some
> other drivers/RIPs can accept 16 bit input, whether or not it's
> actually utilized remains uncertain. I've seen very few definitive
> statements from any of these people that they actually do.

Yes. I think this is why I would prefer an open-source RIP. At least 
you can find out for sure where you stand on issues like this.

> There have
> been indications or hints that these drivers/RIPs do some 
manipulation
> of input data to their needs. Certainly some kind of ppi remapping 
is
> done, and bit downsampling could be happening as well.

That's another hobby-horse of mine: how do you pass data to the RIP 
or driver in a way that you can be sure that it will be printed at 
exactly 100% reproduction. With the Epson drivers inexact approach to 
margins, I don't think it's possible there, for example. It's just 
one step of processing that it would be nice to eliminate (or do 
better in PS).

> 
> > I'm more suspicious regarding the
> > precision used to pass values from PS via the OS to the Epson 
driver
<snip>
> 
> It has more to do with the Epson driver than Adobe, and may be OS
> related as you suggest (but I know little about that). Some time ago
> Photoshop let you hit the print command with files that were high 
bit
> (it didn't used to) but all it was doing was changing to 8 bit on 
the
> fly as a convenience since that's what the driver accepts.

If I had to put money on it, I'd bet they're still doing that.

> 
> > It's this
> > lack of precision that could make it impossible to compensate for 
non-
> > linearity upstream of the driver
<snip>
> 
> This all gets into a huge issue about where the best place for 
massive
> correction is best done. You can send 8 bit pure color gradients 
(not
> neutral) through the epson driver with no profile conversion and see
> good gradients on paper. In my experience most color banding comes
> from converting to profiles, and doing so in high bit did not help.
> C.D.Tobie once explained it well to me but I've lost the email. It 
has
> to do with large remapping of lab values, and best guesses. When the
> driver is so out of balance with massive crossovers etc., remapping
> color based on LUTs becomes problematic. You can wind up jerking 
close
> colors in multiple directions. So it still makes sense to have the
> driver as linear as possible in the first place, then profiles can 
do
> their jobs somewhat coherently. Others with more day to day 
experience
> with profiling in different environments/drivers/RIPs etc. would 
have
> more valuable opinions about how much practical difference it makes.

I think we're at cross purposes as little bit on this point. What I'm 
trying to say is that if you can get the full range of 
colours "through" the driver without any perceptible steps between 
neighboring colours, then *in theory* you can get a good result. It's 
just a matter of knowing what mappings to do in order to hit the 
right colours, and a LUT ought to do that job just fine. Of course, 
this relies heavily upon Adobe's implementation of profile mapping, 
but the fact that very good profiles have been created for the x600 
printers does add some weight to the argument that this is all 
possible. There's no doubt that life would be much easier without all 
this hoop-jumping though!

<snip> 
> The fact is, despite this somewhat techie discussion, people make
> beautiful prints every day within these limitations. Profiles are 
like
> darkroom materials, you get the best one you can, learn what it can
> and can't do, and make it work for you. Softproofing makes this much
> easier. But it's like type C paper, neither Kodak nor Fuji 
was "right"
> but people decided which one they liked and made great prints based 
on
> it's strengths and weaknesses. Sometimes I think the high degree of
> technical discussion around digital raises our expectations too 
high,
> we think technical perfection may be attainable, but it ain't so.
> Which brings us back to art, thankfully.
> Tyler

Yeah, I hear that. I'm just trying my best to create a setup that 
will give me results that are "perfect enough". As I said before, I'm 
actually pretty satisfied with a good majority of the colour images 
I'm getting from the Epson driver. Of course it would be nice to 
increase the percentage even further =)

TTFN

-= mike =-

There has been a thread on 8/16 bit handling by the driver/RIP on 
the colorsync list 10 months ago. About 10 messages that I could 
stack in one message if there is an interest in it.  In essence 
the drivers accept 8 bit, internally 12/16 bits is used in all 
drivers for the computation (tables), the printer is driven with 
8 bit. Mac, Windows, Linux alike. There are some exceptions but 
it is vague what happens in that case.

Ernst

Hi Ernst,

Yes please, I'd be very interested to see that. If you prefer to mail 
them to me directly, then by all means do that. You can reach me on 
mike dot nunan at csfb dot com.

Thanks,

-= mike =-

--- In DigitalBlackandWhiteThePrint@yahoogroups.com, Ernst Dinkla 
<E.Dinkla@c...> wrote:

> There has been a thread on 8/16 bit handling by the driver/RIP on 
> the colorsync list 10 months ago. About 10 messages that I could 
> stack in one message if there is an interest in it.  In essence 
> the drivers accept 8 bit, internally 12/16 bits is used in all 
> drivers for the computation (tables), the printer is driven with 
> 8 bit. Mac, Windows, Linux alike. There are some exceptions but 
> it is vague what happens in that case.
> 
> Ernst

--- In DigitalBlackandWhiteThePrint@yahoogroups.com, "mike_nunan"
<mike_nunan@h...> wrote:
snip...
> > It has more to do with the Epson driver than Adobe, and may be OS
> > related as you suggest (but I know little about that). Some time ago
> > Photoshop let you hit the print command with files that were high
> bit
> > (it didn't used to) but all it was doing was changing to 8 bit on
> the
> > fly as a convenience since that's what the driver accepts.
>
> If I had to put money on it, I'd bet they're still doing that.

But they really have no choice if 8 bit in is required by the driver.

snip
> I think we're at cross purposes as little bit on this point. What I'm
> trying to say is that if you can get the full range of
> colours "through" the driver without any perceptible steps between
> neighboring colours, then *in theory* you can get a good result. It's
> just a matter of knowing what mappings to do in order to hit the
> right colours, and a LUT ought to do that job just fine. Of course,
> this relies heavily upon Adobe's implementation of profile mapping,
> but the fact that very good profiles have been created for the x600
> printers does add some weight to the argument that this is all
> possible. There's no doubt that life would be much easier without all
> this hoop-jumping though!

I think we're talking about the same thing, but just disagree a tad.
I'll just wrap up my end of this by saying that the lengths Bill
Atkinson had to go to to achieve his results border on the absurd, and
by all reports he'll never do it again for any amount of money. Even
then there are things I'm still not thrilled with about his profiles.
Additionally, you won't rise to some of his levels of performance with
any profiling method currently available.
This makes me question whether or not it is in fact possible, and
really have to go back and look at the driver and alternatives to it.
Again, someone with a lot more experience profiling over liearized
RIPs and seeing the real difference could offer more insight than I,
Ernst?<G>?

Also, mapping during profile conversions has little to do with Adobe,
unless you are using the ACE engine. You do have several alternatives
to Adobe's ACE, however the difference I've seen between them are very
slight with ACE being as good or better than any of them. Also, the
profiling software itself is probably the most responsible for how
colors map to the device space and here (I think) is really the larger
problem. I've been hoping someone would have their algorithms refer to
Lindbloom's perceptually uniform LAB space just to see what happens,
but no one has stepped up to the plate as far as I know. The different
packages I've used all seem to characterize a device just fine and
similarly, but the decisions they make about LAB to device conversion
remains an "art" they all do differently.
I'm probably taking up too much of B&W list space with this color
stuff, and I'm on the edge of spouting BS, I highly recommend the
Colorsync list, many discussions to wade through, but much there to learn.

> Yeah, I hear that. I'm just trying my best to create a setup that
> will give me results that are "perfect enough". As I said before, I'm
> actually pretty satisfied with a good majority of the colour images
> I'm getting from the Epson driver. Of course it would be nice to
> increase the percentage even further =)

I've seen a few posts here and there about the current (official, not
alpha or beta) Gimp drivers being more well behaved, and someone just
asked on the Colorsync list about profiling over them, you might want
to check it out.
Tyler

Hi Tyler,

Ok, the light is definitely growing at the end of the tunnel here. I 
don't know which part of my brain was switched off when I was somehow 
assuming that 16-bit values could ever get to the driver, but of 
course under Windows (at least) there's no way to do this. The OS's 
bitmap structure can handles a maximum depth of 8 bits, end of story. 
Thanks to you Ernst for the digest of the past discussion on that 
subject, it's good to see that the later versions of IP5 do actually 
use the full 16 bits. Also, the discussion of dither in that thread 
sorta tallies with my previous hearsay about them trading smoother 
tonality in exchange for a larger dither cell.

Anyway, this realisation about the 8-bit limitation in the OS has 
changed my feelings about the non-linearity of the Epson driver. It's 
possible to get a smooth tonal scale with just 256 values, provided 
those values are reasonably evenly spaced in terms of perceived 
lightness (L*). Roy's approach to linearisation guarantees exactly 
that. If the driver response curve varies from this, there will be 
some larger gaps somewhere in the range, and it only takes the wrong 
image to make them obvious. Paul mentioned "microbanding" being an 
occasional problem with prints generated via the Epson driver, which 
I would strongly assume is a symptom of what I'm talking about. Paul 
if you're still following this thread, what d'you make of this? How 
do you define microbanding, as opposed to any other kind of banding?


A few small comments inline...

--- In DigitalBlackandWhiteThePrint@yahoogroups.com, "Tyler Boley" 
<tyler@t...> wrote:

<snip>
> 
> I think we're talking about the same thing, but just disagree a tad.
> I'll just wrap up my end of this by saying that the lengths Bill
> Atkinson had to go to to achieve his results border on the absurd, 
and
> by all reports he'll never do it again for any amount of money. Even
> then there are things I'm still not thrilled with about his 
profiles.

If the comments in my second para above hold water, then they will 
apply here too, so I can see why there might be unwanted visible 
effects even with his profiles.

<snip> 
> Also, mapping during profile conversions has little to do with 
Adobe,
> unless you are using the ACE engine. You do have several 
alternatives
> to Adobe's ACE, however the difference I've seen between them are 
very
> slight with ACE being as good or better than any of them.

I'm still in Windows-land, so ACE is best of a bad lot (AFAIA).

> Also, the
> profiling software itself is probably the most responsible for how
> colors map to the device space and here (I think) is really the 
larger
> problem. I've been hoping someone would have their algorithms refer 
to
> Lindbloom's perceptually uniform LAB space just to see what happens,
> but no one has stepped up to the plate as far as I know. The 
different
> packages I've used all seem to characterize a device just fine and
> similarly, but the decisions they make about LAB to device 
conversion
> remains an "art" they all do differently.
> I'm probably taking up too much of B&W list space with this color
> stuff, and I'm on the edge of spouting BS, I highly recommend the
> Colorsync list, many discussions to wade through, but much there to 
learn.

I'm headed there now. Just let me confirm we're talking about the 
same place; I'm assuming it's the one with the homepage at:

http://lists.apple.com/mailman/listinfo/colorsync-users

> 
> > Yeah, I hear that. I'm just trying my best to create a setup that
> > will give me results that are "perfect enough". As I said before, 
I'm
> > actually pretty satisfied with a good majority of the colour 
images
> > I'm getting from the Epson driver. Of course it would be nice to
> > increase the percentage even further =)
> 
> I've seen a few posts here and there about the current (official, 
not
> alpha or beta) Gimp drivers being more well behaved, and someone 
just
> asked on the Colorsync list about profiling over them, you might 
want
> to check it out.

Will do, thanks for all the help.

Mike