Digital BW, The Print

----- Original Message -----

From: "hsitz" <hsitz@...>
To: <DigitalBlackandWhiteThePrint@yahoogroups.com>
Sent: Friday, August 02, 2002 9:48 PM
Subject: Re: [Digital BW] Do inkjets dither or not?


> I realize that inkjet spots may now be so small that their prints
> appear to be continous tone, but they still seem to fit the
> definition of "half-tone" printers that compose specific single-color
> pixels out of differently colored dots.  And, at least by the meaning
> of the terms as I learned them (from Blatner and Fraser), the classes
> of half-tone printers and continuous-tone printers are mutually
> exclusive.

They are not. In litho printing there was a range offset>screenless
offset>collotypie that represents that gradual change to continuous tone
printing quite good. Conventional rotogravure comes very near to continuous
tone printing, modern versions of it that rely on electronically engraved
cilinders adapted halftone screening partly in the process but never
totally. So these aren't exclusive systems. The same is possible in inkjet
printing.
Ink bleeding isn't always a negative issue in printing processes,
conventional rotogravure partly depended on that effect to get a continuous
tone result. No dot size differences in that proces but more or less ink per
spot. Transparent ink of course.

Ernst

Richard writes:

> My question - Aren't all inkjet printers required
> to use a dither by definition?

Not really.  Dithering is a process in which two or more pixels are assigned
two or more different colors, with the expectation that the human eye will
see the two pixels blend into some intermediate color that cannot be
directly represented in a single pixel.  For example, if you have only three
possible colors for your pixels (in the world's cheapest inkjet printer,
useful for hypotheses like this one), the only way to represent more than
three colors is to dither and hope that the image will be viewed from far
enough away that the dithered pixels will blend in the eye of the viewer,
producing a perception of some intermediate color.

So dithering is not required for an inkjet printer, but many printer drivers
will use it in some modes to produce intermediate tones that cannot be
represented exactly with a single pixel color.

Dithering is common in 256-color modes on computer monitors.  It's not
necessary in 24-bit color modes, as a general rule (although I think
Photoshop still dithers generated gradients in some cases).

Anthony -- While your "world's cheapest inkjet printer" may have only 
three possible pixel colors, there are lots of color inkjets that 
have only four possible pixel colors (CMYK) and lots more that have 
only six colors (like many of our Epsons).  

My understanding is that inkjets do indeed dither to create colors, 
but that the pixels are so small that for all practical purposes we 
can consider them to be "continuous tone" printers (rather than the 
dithering class of "half-tone" printers).

Dye sublimation printers can print a solid color other than CMYK and 
are truly continous tone, but I don't think inkjets are.  Inkjets 
could be, I suppose, if the inks they spray mixed to form a new 
color, but that's not what happens.  Or maybe it is; I'm sure if I'm 
wrong someone will correct me.

-- Herb


--- In DigitalBlackandWhiteThePrint@y..., "Anthony Atkielski" 
<atkielski@c...> wrote:
> Not really.  Dithering is a process in which two or more pixels are 
assigned
> two or more different colors, with the expectation that the human 
eye will
> see the two pixels blend into some intermediate color that cannot be
> directly represented in a single pixel.  For example, if you have 
only three
> possible colors for your pixels (in the world's cheapest inkjet 
printer,
> useful for hypotheses like this one), the only way to represent 
more than
> three colors is to dither and hope that the image will be viewed 
from far
> enough away that the dithered pixels will blend in the eye of the 
viewer,
> producing a perception of some intermediate color.

>> My question - Aren't all inkjet printers required
>> to use a dither by definition?
> 
> Not really.  Dithering is a process in which two or more pixels are assigned
> two or more different colors, with the expectation that the human eye will
> see the two pixels blend into some intermediate color that cannot be
> directly represented in a single pixel.

You are confusing pixel and dot here. Different dots, different inks are
combined to re-produce the pixel.

>For example, if you have only three
> possible colors for your pixels (in the world's cheapest inkjet printer,
> useful for hypotheses like this one), the only way to represent more than
> three colors is to dither and hope that the image will be viewed from far
> enough away that the dithered pixels will blend in the eye of the viewer,
> producing a perception of some intermediate color.

You don't need to be that far away, what ink jet printers have you been
using to see such output?  Even with a low end offset job, you see the
pattern arms length, but your brain still sees the colors that the screen
frequency and angles create with the four inks.

Carolyn

Herb writes:

> While your "world's cheapest inkjet printer" may
> have only three possible pixel colors, there are
> lots of color inkjets that have only four possible
> pixel colors (CMYK) and lots more that have
> only six colors (like many of our Epsons).

Nope.  You are confusing dots and pixels.  Machine dots can only have as
many colors as there are inks, on machines with opaque inks (such as
inkjets).  However, pixels are made up of multiple dots on such machines,
and so they can have quite a few colors, depending on how the individual
dots are mixed.

> My understanding is that inkjets do indeed dither
> to create colors, but that the pixels are so small
> that for all practical purposes we can consider
> them to be "continuous tone" printers (rather than the
> dithering class of "half-tone" printers).

Some do dither.  However, you can see individual dots and dithering on the
page with a loupe, so I wouldn't say that the pixels are small enough to be
call continuous tone.

If you want something closer to continuous tone, use dye sublimation.

> Dye sublimation printers can print a solid color
> other than CMYK and are truly continous tone,
> but I don't think inkjets are.

Yes.

> Inkjets could be, I suppose, if the inks they
> spray mixed to form a new color, but that's not
> what happens.

Yes.  But you need transparent inks and a way to control the amount of ink
deposited (not just the size of the dot) in order to accomplish that.
Dye-subs do it by varying the thickness of dye deposited on a spot and by
using dyes that are transparent.  This can't be done with inkjets for the
most part, particularly with pigmented inks, which tend to be highly opaque.

Carolyn writes:

> You are confusing pixel and dot here.

No, I'm not.  Dithering involves the use of more than one pixel to create
the impression of some color intermediate between those pixels.  A dithered
image contains adjacent pixels of slightly different colors that can create
the illusion of greater gamut when the image is examined at a distance.  In
dithered digital images, you can expand the image until the individual
pixels are readily visible, and you can see the dithering as alternating
colors of adjacent pixels.  It looks like grain from intermediate distances.

> You don't need to be that far away ...

Dithering is much more noticeable than ordinary non-dithered printing,
because dithers occupy several pixels, not just one pixel of several dots.
The difference in size can be an order of magnitude.

> 
> Carolyn writes:
> 
>> You are confusing pixel and dot here.
> 
> No, I'm not.  

I read you differently, the poster was asking "aren't all inkjet printers
required to use a dither by definition." and you responded with pixels not
dots. That's why I said the above.


>Dithering involves the use of more than one pixel to create
> the impression of some color intermediate between those pixels.  A dithered
> image contains adjacent pixels of slightly different colors that can create
> the illusion of greater gamut when the image is examined at a distance.  In
> dithered digital images, you can expand the image until the individual
> pixels are readily visible, and you can see the dithering as alternating
> colors of adjacent pixels.  It looks like grain from intermediate distances.

I'm aware of what dithering is on screen, and in print. I was just saying
that as a response to the posted question you responded with pixels rather
than dots. and I found that to be confusing. no problem.

>> You don't need to be that far away ...
> 
> Dithering is much more noticeable than ordinary non-dithered printing,
> because dithers occupy several pixels, not just one pixel of several dots.
> The difference in size can be an order of magnitude.

Of course it's more noticeable, I didn't say otherwise. I responded to your
thoughts that you had to view epson prints from far away in order to not see
the dither. I was saying you don't need to be that far away from current
epson prints in order for your eye to percieve good tone and basically zero
dither (from the printer).  If the original file is done well and correctly,
then the dithering introduced by the epson print driver is not noticeable
from arms length IMO... oh what the hell, I'm not cut out for this....


Best, Carolyn

Hi Herb,

> Anthony -- While your "world's cheapest inkjet printer" may have only
> three possible pixel colors, there are lots of color inkjets that
> have only four possible pixel colors (CMYK) and lots more that have
> only six colors (like many of our Epsons).
>
> My understanding is that inkjets do indeed dither to create colors,
> but that the pixels are so small

Dots...not pixels...printers don't print pixels, and there is really no
relationship between the pixels and the dots, that you will be able to see
after a (decent) dither.

> that for all practical purposes we
> can consider them to be "continuous tone" printers

The Piezo system is technically considered "continuous tone", even according
to Jon Cone, who has been in this very business for many many years.  The
accepted term "continuous tone" really means gives the "appearance" of
continuous tone...and I would also add, uses multiple inks, not just black.

So yes, absolutely inkjet printers DO dither, absolutely, unquestionably.
It's like asking if inkjet printers use ink...  They do not use cows though,
despite misinformation to the contrary ;-)

Austin

> > My question - Aren't all inkjet printers required
> > to use a dither by definition?
>
> Not really.  Dithering is a process in which two or more pixels
> are assigned
> two or more different colors,

Anthony, what on earth are you talking about?  Pixels have nothing to do
with this, and pixels aren't "assigned" anything.  A dither algorithm takes
pixels as it's input, and outputs DOTS.

There are many references on the web that describe dithering correctly.
One, in particular, used to be available on the Adobe web site....and had a
very good description of how the dither/halftone process worked.  I am not
sure if it's still there now...

> So dithering is not required for an inkjet printer, but many
> printer drivers
> will use it in some modes to produce intermediate tones that cannot be
> represented exactly with a single pixel color.

It absolutely IS required for photographic images.  No, to print TEXT or
other things that are solid color based like that, it is not...but this is a
group discussing B&W image processing...not printing text documents.

> Dithering is common in 256-color modes on computer monitors.

Do you REALLY mean "256 color" mode?  Who on earth uses 256 color mode?  I
HOPE no one in this group...as anyone doing any digital imaging should be
using a mode that is a LOT higher than that.

Austin

--- In DigitalBlackandWhiteThePrint@y..., "Austin Franklin" 
<darkroom@i...> wrote:

> > My understanding is that inkjets do indeed dither to create 
colors,
> > but that the pixels are so small . . .
> 
> Dots...not pixels...printers don't print pixels, and there is 
really no
> relationship between the pixels and the dots, that you will be able 
to see
> after a (decent) dither.

Yes, of course, I should have been more careful to distinguish 
between the pixels of the digital image and the dots the printer uses 
to reproduce each pixel.

> 
> > that for all practical purposes we
> > can consider them to be "continuous tone" printers
> 
> The Piezo system is technically considered "continuous tone", even 
according
> to Jon Cone, who has been in this very business for many many 
years.  The
> accepted term "continuous tone" really means gives the "appearance" 
of
> continuous tone...and I would also add, uses multiple inks, not 
just black.

Hmm, well I guess if the experts use "continuous tone" to mean "gives 
the appearance of continuous tone" then I can't quibble.  

But what terms are we then to use to differentiate between half-tone 
printers (those that create a colored pixel by breaking the color 
down into a bunch of differently colored spots) and continous tone 
printers (like dye-sub printers, which are able to lay down pixels of 
specific colors "by building up varying densities of primary colors 
in the same spot". (from Blatner and Fraser, Real World Photoshop).

I realize that inkjet spots may now be so small that their prints 
appear to be continous tone, but they still seem to fit the 
definition of "half-tone" printers that compose specific single-color 
pixels out of differently colored dots.  And, at least by the meaning 
of the terms as I learned them (from Blatner and Fraser), the classes 
of half-tone printers and continuous-tone printers are mutually 
exclusive.

-- Herb

Hi Herb,

> > > that for all practical purposes we
> > > can consider them to be "continuous tone" printers
> >
> > The Piezo system is technically considered "continuous tone", even
> according
> > to Jon Cone, who has been in this very business for many many
> years.  The
> > accepted term "continuous tone" really means gives the "appearance"
> of
> > continuous tone...and I would also add, uses multiple inks, not
> just black.
>
> Hmm, well I guess if the experts use "continuous tone" to mean "gives
> the appearance of continuous tone" then I can't quibble.

Oh, please DO quibble!  I've had a number of arguments about this very
subject with Jon...and others.

> But what terms are we then to use to differentiate between half-tone
> printers (those that create a colored pixel by breaking the color
> down into a bunch of differently colored spots) and continous tone
> printers (like dye-sub printers, which are able to lay down pixels of
> specific colors "by building up varying densities of primary colors
> in the same spot". (from Blatner and Fraser, Real World Photoshop).

Understood.  But, technically, there is no such thing as "continuous" tone,
as you are always subject to SOME resolution, meaning just how little can
you add that you can measurably make a distinguishable difference.  Believe
it or not, it's a dynamic range thing.  Yes, somehow I seem to work dynamic
range into just about everything, including dinner ;-)

> I realize that inkjet spots may now be so small that their prints
> appear to be continous tone, but they still seem to fit the
> definition of "half-tone" printers that compose specific single-color
> pixels out of differently colored dots.

At least in the B&W world, I would say that halftone specifically relates to
using black and not black, and that you are clearly moving more towards
continuous tone by providing more "shades" to work with...  But I understand
the issue you are raising, and I agree, BTW.

Regards,

Austin

Austin,

I've got a cow in one of my inkjet prints! It looks fairly well-used.

I liked the comment - sorry Anthony.

Bob Frost.

----- Original Message -----
From: "Austin Franklin" <darkroom@...>


> So yes, absolutely inkjet printers DO dither, absolutely, unquestionably.
> It's like asking if inkjet printers use ink...  They do not use cows
though,
> despite misinformation to the contrary ;-)

Austin Franklin wrote:

>Hi Tyler,
>  
>
>>>      
>>>
>>Austin, starting with a 16 bit grayscale file, then preparing it
>>for quad printing though a RIP remaining entirely in 16 bit in
>>very much the same way Piezography does for you behind the
>>scenes, yeilds a file that has 29,037 "unique colors" (levels).
>>    
>>
>
>Well, no, 16 bits is 64k unique levels.  Anyway, that isn't relevant.  The
>printer simply can not produce that many different levels, without
>sacrificing overall output resolution, and besides, the human eye can
>absolutely not discern them.
>  
>

64k Unique greyscale levels?  What would the point be to that?  

The human eye would be unable to distinguish between the VAST majority 
of them... AFAIK the human eye will only discern about 256 levels of 
luminance/grey..

I don't know anyone making prints for remote sensing detector 
devices....  Although it's definitely an untouched niche market..

>  
>
>>This is determined with a cool little utility written by Bruce Linbloom.
>>Even after converting it to 8 bit, it reports 1,495 levels.
>>    
>>
>
>8 bits only allows 256 levels.  It is PHYSICALLY impossible to get more than
>256 levels from 8 bits.
>  
>
EXACTLY

Of course if you add a chroma value things get all screwy...  

Tyler are you sure that those measurements are ONLY of 
luminance/greyscale..?

Remember, that if you read a quadset print in RGB, the inks will have 
some chroma tones as will the paper/media... That could easily give you 
the numbers of which you speak, but that isn't really greyscale..

Keith
 
 



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

--- In DigitalBlackandWhiteThePrint@y..., "Editor P.O.V. Image Service" 
<editor@p...> wrote:
snip
> the numbers of which you speak, but that isn't really greyscale..
> 
> Keith

I'm not sure why this is so hard to understand. There is nothing 
grayscale, or single channel, about quad tone printing except your 
original file.
Unless you are sending single channel info to a single channel driver 
to one ink (as in K only Epson driver printing), non of these 
arguements address what I'm talking about.
Tyler

> Remember, that if you read a quadset print in RGB, the inks will have
> some chroma tones as will the paper/media... That could easily give you
> the numbers of which you speak, but that isn't really greyscale..


I thought Tyler was talking four channels/ ink delivery/ four inks/ three
grays and a black - creating a grayscale image on paper from an image file.

You are both talking about a grayscale image file, a single channel
grayscale image file.

The original thought was that the peizo driver could not produce the number
of levels a statement from Cone stated no?  This is with respect to the
print, not the image file.

You and Austin are arguing Tyler's point, I don't understand why since it
seems to be a different thing you are thinking about.

Carolyn

tboleyyh wrote:

>--- In DigitalBlackandWhiteThePrint@y..., "Editor P.O.V. Image Service" 
><editor@p...> wrote:
>snip
>  
>
>>the numbers of which you speak, but that isn't really greyscale..
>>
>>Keith
>>    
>>
>
>I'm not sure why this is so hard to understand. There is nothing 
>grayscale, or single channel, about quad tone printing except your 
>original file.
>Unless you are sending single channel info to a single channel driver 
>to one ink (as in K only Epson driver printing), non of these 
>arguements address what I'm talking about.
>Tyler
>
>  
>
This was the problem I think... On the one hand.. we are speaking 
greyscale.. you have been speaking RGB, so people have been talking past 
each other..

Theoretically, it would be possible to get more than 256 RGB levels from 
an 8 bit file when dithering/overlapping values...

If you have RGB levels(colors) that can overlap or dither, surely you 
can get more than 256 values out of that combination..

BUT...

At the same time.. there would still be a max of 256 discrete greyscale 
or luminance values..

So, both you AND Austin are correct..

Of course, the question is, is there and advantage to this method?  How 
predictable would transitions between values be?  (I am guessing that is 
what the software has to be properly designed for to create these 
transitional values smoothly)

Remember that we are starting with a file with only 256 levels anyway.. 
 So those transitional levels are going to be pseudo-values or 
interpolations of a sort..  (It's analogously like colorized B&W film 
--  yeah we have colors that we can now perceive, but in many cases they 
represent something that never perceptibly existed for any human eye)

Keith
 




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

Carolyn Frayn wrote:

>
>I thought Tyler was talking four channels/ ink delivery/ four inks/ three
>grays and a black - creating a grayscale image on paper from an image file.
>  
>
Yes, but think Carolyn, the inks could have differing chroma -- just as 
they do in a traditional color inkset..  The VM or VM Sepia sets would 
be clear examples.. You have the ability to represent more than 256 
values by representing luminance and a sort of transitional 
Pseudo-luminance with chroma..

>You are both talking about a grayscale image file, a single channel
>grayscale image file.
>  
>
Yes,  which can have a maximum of 256 discernible levels.. That's all.. 
The human eye won't see more than that.

>The original thought was that the peizo driver could not produce the number
>of levels a statement from Cone stated no?  This is with respect to the
>print, not the image file.
>  
>
Right... and  if you start with 256 luminance values... and  4 inks of 
VARYING chroma...  imagine the 256 levels as a two dimensional  256 
swatch chart instead of the 256 levels of a one dimensional greyscale 
file...

Although your swatch chart may map to 256 discrete values, the 
combinations of those 256 values will certainly be "capable" (not 
assured of) giving more than 256 values..  You see why? 

>You and Austin are arguing Tyler's point, I don't understand why since it
>seems to be a different thing you are thinking about.
>
>  
>
Read the part of my earlier post on how people have been "talking past 
each other on this"...




Keith
 



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

--- In DigitalBlackandWhiteThePrint@y..., "Editor P.O.V. Image Service" 
<editor@p...> wrote:
snip
> This was the problem I think... On the one hand.. we are speaking 
> greyscale.. you have been speaking RGB,
No, I'm not. Again, four channels, four. Printed through a four channel 
driver, to four inks.
Sorry Keith, you're still not grasping what's happenning in a partioned 
quad ink four channel file. But really, none of us actually need to.

> Remember that we are starting with a file with only 256 levels anyway.. 
Not really, Austin was quoting Jon's statement about Piezo from a 16 
bit file, far more than 256 levels.

>  So those transitional levels are going to be pseudo-values or 
> interpolations of a sort..
nope.
Anyway, I'm done with this. But perhaps those who find this stuff 
fascinating may enjoy the following info from Bruce Linbloom, this is 
quoted from him, not about quads, but of interest.
Tyler


Note that any 16-bit image saved out of Photoshop will have, at a 
maximum, 32769 unique levels per channel, not 65536 as one might 
expect.  This is a 
limitation of Photoshop's internal architecture.  A true 16-bit image 
may therefore lose levels and number of unique colors simply by opening 
and saving 
with Photoshop.  The limited Photoshop style 16-bit encoding is 
detected and reported by Levels.  Another subtle, but interesting 
variant is that opening and 
saving can change the actual pixel values without reducing the number 
of levels or the number of unique colors.  An example of this would be 
a raw, 14-bit 
image from a BetterLight scanback.  There will be the same number of 
unique levels and colors both before and after Photoshop, but the pixel 
values in the 
saved file will be a little different from those of the original.

Note that converting from 16-bits to 8-bits in Photoshop adds noise to 
the image.  This may affect the number of levels and/or unique colors 
in the 8-bit 
image.  For example, if you start with a 16-bit image having all its 
pixels set to the same color, and then convert it to 8-bits, you may 
find that the 8-bit image 
has more than one unique color.

Note that passing an 8-bit image through profiles or mode changes in 
Photoshop causes noise to be added to the image, thus increasing the 
number of levels 
and/or unique colors.  This behavior may be controlled with the Color 
Settings setup,  in Advanced Mode, "Use dither (8-bit/channel images)".  
For example, 
if you start with an 8-bit RGB image having all its pixels set to the 
same color, and then convert it to CMYK, you may find that the CMYK 
image has more 
than one unique color.

> 
> This was the problem I think... On the one hand.. we are speaking
> greyscale.. you have been speaking RGB, so people have been talking past
> each other..

not RGB, four channels, not three.  The original question/doubt/response was
with regard to quad prints. Prints derived from four inks, from a file
partitioned into four channels... No one was speaking past anyone, the point
seemed to be misunderstood and a different point seemed to be introduced.


Carolyn

Going back to the original...

rmcooke wrote:

>
>In the August 2002 issue of the Digital Imaging Digest (which is
>written by the folks at the PMA - Photo Marketing Association) they
>cover quad tone printing among others and in the article they mention
>Jon Cone's Piezography system.  They attribute the following quote to
>someone from Cone's company:
>
>""Unlike the Iris process, PiezographyBW does not use a dithered
>pattern. It is as close to silver properties as possible.
>PiezographyBW's perceived resolution is greater than the human eye. It
>renders more than 1,000 gray values in 8-bit mode.
>
I think the problem here is that if Jon is making prints with 1000 
luminance values he's gone totally nuts..  He must be marketing prints 
to CCDs, PMTs, and other sensors..

> The new
>PiezographyBW Pro24 renders more than 4,000 gray values in 16-bit mode." 
>
>  
>

Even more ludicrous..

I think the fault probably lays with the publication in that the writer 
likely misunderstood Jon..

There is no conceivable reason on God''s green earth to use 4000 
greyscale values, unless you like proving you can create things the eye 
cannot perceive..  In that case I would stop at 257 and be done with 
it.. LOL

I would be willing to bet that Jon meant 4000 distinct RGB values on a 
print, when rendered from a 256 greyscale original...  That "more 
information is presented thereby to the human eye than one might get 
from a simple greyscale image"...  That would make sense..

The statement as it is.. well.....  it's nonsense..
Keith
 
 



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

Keith writes:

> Yes,  which can have a maximum of 256
> discernible levels.. That's all.. The
> human eye won't see more than that.

FWIW, luminance discrimination in the human eye varies with color and
position in the visual field.  I believe it is highest for blue overall.
But 256 levels is probably a good average figure.

> Read the part of my earlier post on how people
> have been "talking past each other on this"...

I see lots of young males burning off testosterone, and little else.  Have
fun.

Tyler,

> snip
> > This was the problem I think... On the one hand.. we are speaking
> > greyscale.. you have been speaking RGB,
> No, I'm not. Again, four channels, four. Printed through a four channel
> driver, to four inks.
> Sorry Keith, you're still not grasping what's happenning in a partioned
> quad ink four channel file. But really, none of us actually need to.

It doesn't matter how many inks, a grayscale file is ONE channel, period.
The file that is sent to the Piezo driver is ONE channel when using the
traditional Piezo software.  That isn't questionable.

Out of that 8 bit data, you can ONLY achieve 256 maximum steps, period.
That, also, is not debatable.  IF (and I doubt it) the driver decides to
MAKE UP more tones, they are NOT real tones.

> > Remember that we are starting with a file with only 256 levels anyway..
> Not really, Austin was quoting Jon's statement about Piezo from a 16
> bit file, far more than 256 levels.

No, I was quoting from the 1000 comment in this case, which is the
"traditional" Piezo claim, and that is only 8 bits, and therefore is only
256 levels.

> Note that any 16-bit image saved out of Photoshop will have, at a
> maximum, 32769 unique levels per channel, not 65536 as one might
> expect.  This is a
> limitation of Photoshop's internal architecture.  A true 16-bit image
> may therefore lose levels and number of unique colors simply by opening
> and saving
> with Photoshop.  The limited Photoshop style 16-bit encoding is
> detected and reported by Levels.

This is simply silly.  If this is the case, then simply save it as TIFF, and
you should have no loss problems.  Anyway, even if this does happen, it is
irrelevant to this discussion.  A 16 bit grayscale file has the ability to
have 64k possible values per pixel, period.  If Photoshop clips the file to
15 bits, then it isn't a 16 bit file.

It also appears that Bruce is talking about RGB files.  People who use
Piezo, don't use RGB files...

Austin

Tyler,

> snip
> > the numbers of which you speak, but that isn't really greyscale..
> >
> > Keith
>
> I'm not sure why this is so hard to understand.

Exactly!!!

> There is nothing
> grayscale, or single channel, about quad tone printing except your
> original file.

Yeah, and that is THE image...anything that varies from that, is simply not
the image...the driver doesn't go "making up" tones that weren't there!

> Unless you are sending single channel info to a single channel driver
> to one ink (as in K only Epson driver printing), non of these
> arguements address what I'm talking about.
> Tyler

I understand what you are trying to say, but I claim it's simply wrong.  The
original Piezo driver doesn't print 1000 tones, period.  It simply
can't...there isn't enough discernment in the ability to print separate and
distinguishable tones anyway...  Print a 100 step step wedge, and see how
many of the tones you can separate out.  There is no way you can do that
with 1000.

Anyway, if I have a file with 256 tones, I want the printer to print 256
tones...and to discern the tones the way I want them discerned...not adding
some MORE tones where none originally existed.  So even if it could print
1000 tones, I don't want it to.

Regards,

Austin

At 04:31 PM 8/2/2002 -0400, Austin wrote:

>Understood.  But, technically, there is no such thing as "continuous" tone,
>as you are always subject to SOME resolution, meaning just how little can
>you add that you can measurably make a distinguishable difference.  Believe
>it or not, it's a dynamic range thing.  Yes, somehow I seem to work dynamic
>range into just about everything, including dinner ;-)

Right.  Austin cooks on a dynamic range.



****************
James E. Martz
Milan, OH
jemartz@...
*****************

----- Original Message -----

From: "James E. Martz" <jemartz@...>
To: <DigitalBlackandWhiteThePrint@yahoogroups.com>
Sent: Thursday, August 08, 2002 8:49 PM
Subject: RE: [Digital BW] Do inkjets dither or not?


> At 04:31 PM 8/2/2002 -0400, Austin wrote:
>
> >Understood.  But, technically, there is no such thing as "continuous"
tone,
> >as you are always subject to SOME resolution, meaning just how little can
> >you add that you can measurably make a distinguishable difference.
Believe
> >it or not, it's a dynamic range thing.  Yes, somehow I seem to work
dynamic
> >range into just about everything, including dinner ;-)
>
> Right.  Austin cooks on a dynamic range.

But it is an electric stove with a finite number of temperature settings and
not a gas range with a continously variable flame. <G>

Sorry. Couldn't resist.

Martin