Digital BW, The Print

Hello,

I have read and read again all the posts I could about the subject of
horizontal banding.

I print on a brand new Epson R1800 loaded with 3 Eboni MK and cleaning
fluid in the 5 remain color positions on Epson EEM and get horizontal
banding in the midtones. QTR, 2880, uni-directional, ordered.

With Epson OEM inks, color prints do not show any banding (with Epson
print driver). On Epson EEM.

I did 3 full cleaning cycles ( (3 cleans, one print, wait a day)x3 ),
print head alignment is A1 as are nozzle check patterns.

I did notice that I do not get banding when printing on Hahnemuhle
Museum Etching 350gsm.

I am wondering if this banding is simply just an ink/paper
combination. I am assuming that OEM color inks might have a dot gain
that is a little larger than MIS Eboni. Thus the OEM color prints
"hide" the banding with a bit of ink spread?

This would explain why I get banding on EEM and not on Hahnemuhle
Museum Etching, assuming that Hahnemuhle has different ink absorption
and drying characteristics.

Does this makes sense?

I tried changing the Ink Limit slider in QTR on EEM (0, +10, +20, -10,
-20) on an horizontal 21 step wedge. The banding changes zone with
different Ink Limit values. This seems to validate that the problem
might come from ink/paper/coating characteristics.

Has anyone experienced similar problems? Maybe I just need to find the
papers that do not show banding?

Apart from that the prints are AMAZING. Compared to Piezography NK7,
they appear more neutral (and perfectly warmer) to me, more sharp (dot
grain seems to add perceive sharpness), more luminous and much better
linearized when testing using circular gradients.

Thanks for any input, experiences or advice,

Denis

I also use 3-MK in an 1800, but I have not seen any midtone banding  
in my prints.  I use Red River Polar matte (EEM subsitute), Photo  
Rag, and Museo Portfolio Rag papers and all print fine with no  
banding.  Could it be your printer paper advance mechanism or platen  
gap?

Carl

On Sep 28, 2007, at 5:36 PM, the_des_bois wrote:

> Hello,
>
> I have read and read again all the posts I could about the subject of
> horizontal banding.
>
> I print on a brand new Epson R1800 loaded with 3 Eboni MK and cleaning
> fluid in the 5 remain color positions on Epson EEM and get horizontal
> banding in the midtones. QTR, 2880, uni-directional, ordered.
>
> With Epson OEM inks, color prints do not show any banding (with Epson
> print driver). On Epson EEM.
>
> I did 3 full cleaning cycles ( (3 cleans, one print, wait a day)x3 ),
> print head alignment is A1 as are nozzle check patterns.
>
> I did notice that I do not get banding when printing on Hahnemuhle
> Museum Etching 350gsm.
>
> I am wondering if this banding is simply just an ink/paper
> combination. I am assuming that OEM color inks might have a dot gain
> that is a little larger than MIS Eboni. Thus the OEM color prints
> "hide" the banding with a bit of ink spread?
>
> This would explain why I get banding on EEM and not on Hahnemuhle
> Museum Etching, assuming that Hahnemuhle has different ink absorption
> and drying characteristics.
>
> Does this makes sense?
>
> I tried changing the Ink Limit slider in QTR on EEM (0, +10, +20, -10,
> -20) on an horizontal 21 step wedge. The banding changes zone with
> different Ink Limit values. This seems to validate that the problem
> might come from ink/paper/coating characteristics.
>
> Has anyone experienced similar problems? Maybe I just need to find the
> papers that do not show banding?
>
> Apart from that the prints are AMAZING. Compared to Piezography NK7,
> they appear more neutral (and perfectly warmer) to me, more sharp (dot
> grain seems to add perceive sharpness), more luminous and much better
> linearized when testing using circular gradients.
>
> Thanks for any input, experiences or advice,
>
> Denis

Denis;

I have yet to run across an R1800 that prints perfectly in all 
channels. The OEM ink and driver are kicking out 6 channels of ink, and 
may hide the flaws in one or two channels. If you do a 'calibration 
mode' printout It will show you the weaker channels, and perhaps you 
can shift your inks to better-printing channels if you are stuck with a 
particulalry bad one now. 

Steven Karafyllakis


>

Steven,

Thanks for sharing this. I did print using single channels only and I
did notice some differences in the banding. Some are worse than others.

But before swapping MK PK and GLO, I did another test print on EEM
using the MIS ALPHA curve.

Strangely NO banding appears.

This banding issue is a real mystery.

I'll keep your suggestion in mind for when (if) it returns.

Denis


--- In DigitalBlackandWhiteThePrint@yahoogroups.com, "Steven
Karafyllakis" <stevekphoto@...> wrote:

>
> 
> Denis;
> 
> I have yet to run across an R1800 that prints perfectly in all 
> channels. The OEM ink and driver are kicking out 6 channels of ink, and 
> may hide the flaws in one or two channels. If you do a 'calibration 
> mode' printout It will show you the weaker channels, and perhaps you 
> can shift your inks to better-printing channels if you are stuck with a 
> particulalry bad one now. 
> 
> Steven Karafyllakis
> 
> 
> >
>

Hello Carl,

But this is a brand new printer that has not yet seen more than 100
8x10...

How can one know if the advance mechanism is at fault?

And what is the platen gap?

Thanks for your help and time,

Denis

--- In DigitalBlackandWhiteThePrint@yahoogroups.com, Carl Schofield
<list@...> wrote:

>
> I also use 3-MK in an 1800, but I have not seen any midtone banding  
> in my prints.  I use Red River Polar matte (EEM subsitute), Photo  
> Rag, and Museo Portfolio Rag papers and all print fine with no  
> banding.  Could it be your printer paper advance mechanism or platen  
> gap?
> 
> Carl
> 
> On Sep 28, 2007, at 5:36 PM, the_des_bois wrote:
> 
> > Hello,
> >
> > I have read and read again all the posts I could about the subject of
> > horizontal banding.
> >
> > I print on a brand new Epson R1800 loaded with 3 Eboni MK and cleaning
> > fluid in the 5 remain color positions on Epson EEM and get horizontal
> > banding in the midtones. QTR, 2880, uni-directional, ordered.
> >
> > With Epson OEM inks, color prints do not show any banding (with Epson
> > print driver). On Epson EEM.
> >
> > I did 3 full cleaning cycles ( (3 cleans, one print, wait a day)x3 ),
> > print head alignment is A1 as are nozzle check patterns.
> >
> > I did notice that I do not get banding when printing on Hahnemuhle
> > Museum Etching 350gsm.
> >
> > I am wondering if this banding is simply just an ink/paper
> > combination. I am assuming that OEM color inks might have a dot gain
> > that is a little larger than MIS Eboni. Thus the OEM color prints
> > "hide" the banding with a bit of ink spread?
> >
> > This would explain why I get banding on EEM and not on Hahnemuhle
> > Museum Etching, assuming that Hahnemuhle has different ink absorption
> > and drying characteristics.
> >
> > Does this makes sense?
> >
> > I tried changing the Ink Limit slider in QTR on EEM (0, +10, +20, -10,
> > -20) on an horizontal 21 step wedge. The banding changes zone with
> > different Ink Limit values. This seems to validate that the problem
> > might come from ink/paper/coating characteristics.
> >
> > Has anyone experienced similar problems? Maybe I just need to find the
> > papers that do not show banding?
> >
> > Apart from that the prints are AMAZING. Compared to Piezography NK7,
> > they appear more neutral (and perfectly warmer) to me, more sharp (dot
> > grain seems to add perceive sharpness), more luminous and much better
> > linearized when testing using circular gradients.
> >
> > Thanks for any input, experiences or advice,
> >
> > Denis
>

Another thing that can cause banding is the way the driver handles 
transitions across some parts of the tonal scale; particulalry the 
blend between K and the next lighter shades. This can be exacerbated 
by some curves, while others work fine. Next time you run into the 
problem, try using two or even 3 curves for similar papers, and 
mixing them in varying ratios. I'm not sure if it will help with the 
3MK setup, but it might.

Steve


>
> Steven,
> 
> Thanks for sharing this. I did print using single channels only and 
I
> did notice some differences in the banding. Some are worse than 
others.
> 
> But before swapping MK PK and GLO, I did another test print on EEM
> using the MIS ALPHA curve.
> 
> Strangely NO banding appears.
> 
> This banding issue is a real mystery.
> 
> I'll keep your suggestion in mind for when (if) it returns.
> 
> Denis
> 
> 
> --- In DigitalBlackandWhiteThePrint@yahoogroups.com, "Steven
> Karafyllakis" <stevekphoto@> wrote:
> >
> > 
> > Denis;
> > 
> > I have yet to run across an R1800 that prints perfectly in all 
> > channels. The OEM ink and driver are kicking out 6 channels of 
ink, and 
> > may hide the flaws in one or two channels. If you do 
a 'calibration 
> > mode' printout It will show you the weaker channels, and perhaps 
you 
> > can shift your inks to better-printing channels if you are stuck 
with a 

> > particulalry bad one now. 
> > 
> > Steven Karafyllakis
> > 
> > 
> > >
> >
>

Hi Denis,

>I print on a brand new Epson R1800 loaded with 3 Eboni MK 
>and cleaning fluid in the 5 remain color positions on 
>Epson EEM and get horizontal banding in the midtones. 
>QTR, 2880, uni-directional, ordered.

...
>I did notice that I do not get banding when printing 
>on Hahnemuhle Museum Etching 350gsm.

>I am wondering if this banding is simply just an 
>ink/paper combination. 

>I am assuming that OEM color inks might have a 
>dot gain that is a little larger than MIS Eboni. 
>Thus the OEM color prints
>"hide" the banding with a bit of ink spread?

...

>I did another test print on EEM
>using the MIS ALPHA curve.
>Strangely NO banding appears.

I agree with Carl with respect to both different channels being of different
qualities and paper transport being a suspect.  Others have also noted that
different 1800s are of different quality.

I also agree with your analysis that the Eboni dot gain is so little
compared to the OEM inks that it probably does not hide the defects as well.
(While it may not be the same, I've noticed on draw-down tests that the
Epson inks spread more than the MIS inks.)  And, I think different paper
coatings affect this.

I have found some banding at the top and bottom edges of 8x10 prints with my
1800.  My conclusion was that the paper transport is so critical with this
setup that it does best when both rollers are engaged.  So, I print more
like 9 x 7 on letter size printers.

I've also heard some comments that the 1800 3-MK approach seems to show
image file defects more.  I had an example of that yesterday.  I'd taken
some cloud shots with my Canon Rebel XT and saw what appeared to be
horizontal and vertical microbanding in the 3-MK print.  I tried the R260
with a full B&W inkset, and the visibility of the artifacts was less, but
they were there.  On examining the file at 100%, the artifacts were
definitely in the file.  The inkset with dilute inks just did not show them
as clearly, perhaps due to the dilute inks spreading more and hiding the
very fine details of the file.



>Apart from that the prints are AMAZING. Compared to 
>Piezography NK7, they appear more neutral (and perfectly 
>warmer) to me, more sharp (dot grain seems to add 
>perceive sharpness), more luminous and much better
>linearized when testing using circular gradients.

The R260 inkset that is about to be released with allow -- with QTR -- that
system to have a black only approach that is not quite as good as the 1800
3-MK, but still very good given how easy and cheap it is.  Oddly, multiple
channels of MK do not improve the 260, but a single channel on the 260 has
significantly less microbanding than a single 1800 channel -- but with a
bit more graininess.  

I've made some PKNs for the 260 RC inkset -- 2 levels of toning -- that make
very nice glossy prints.  The 260 will also have a full B&W inkset for
extremely smooth prints -- again with 2 levels of toning in addition to warm
carbon.  It'll be interesting to see how many go with the full inkset or
just use the BO approach with color inks installed.  

I might add that I expect the 260 inkset ("UT-RC" [R series printers with
Claria inks]) to run in the 1800 with QTR.  One or more PKN channels for the
FB glossy types of papers might be something I'll do with the extra channels
in my 1800. 

Paul
www.PaulRoark.com

Hi Steve,

(Sorry, I missed your previous comment in this thread.)

>Another thing that can cause banding is the way the driver 
>handles transitions across some parts of the tonal scale; ...
> Next time you run into the problem, try using two or 
> even 3 curves for similar papers, and 
> mixing them in varying ratios. I'm not sure if it will 
> help with the 3MK setup, but it might.

Most of the 3-MK profiles use the same underlying curves set.  So, it
probably will not make much difference, but who knows -- it's worth a try.
The primary differences among the profiles are the end points for dmax and
the linearization.

I tried to make minor differences in the level and slopes of the 3 curves
within the profile even where they seem to be about the same.  I noticed
that if 2 curves were at the same level and slope, vertical microbanding
could show up in smooth patches. 

Paul
www.PaulRoark.com

Hi Paul;

Having 2 or 3 different mulit-ink curves that mix different cross-
over points is  more effective at hiding flaws, no doubt about it. 
but a couple weeks back when Clayton was working out a 2K curve to 
get his R200 going again, we noticed that even if the 2 Ks were 
identical in slope, having them contribute a siginificantly different 
percentage to the overall quantity seemed to help  a bit with the 
microbanding. 

Of course if Epson had chosen to manufacture this printer to the 
standards of, say, the 780, ($79.00) or the R200, ($90.00), or even 
the R300 (180.00), we would not be having this discussion. It will be 
interesting to see how it goes with the R1400. 


Steven Karafyllakis


> 
> Most of the 3-MK profiles use the same underlying curves set.  So, 
it
> probably will not make much difference, but who knows -- it's worth 
a try.
> The primary differences among the profiles are the end points for 
dmax and
> the linearization.
> 
> I tried to make minor differences in the level and slopes of the 3 
curves
> within the profile even where they seem to be about the same.  I 
noticed
> that if 2 curves were at the same level and slope, vertical 
microbanding

> could show up in smooth patches. 
> 
> Paul
> www.PaulRoark.com
>