Digital BW, The Print

Martin,

You wrote:

>... the MIS Sepia-VM set...
>This is a very interesting set and the differences achieved
>by applying different curves seem much stronger than with the regular VM.

Yes, the range is rather extreme.

> This set is most interesting at the extreme ends ...

I like the tones best *near* the ends of the range as opposed to at the
ends.  For example, the old "warm" curve, which is at the neutral end of the
VM-sepia, is too cool and uneven.  This is the end where you can use just a
little toner to balance the inkset to right-on neutral through the
mid-tones.  This will take some fine tuning, but for those who want a true
neutral tone, I think this does the best job I've seen.

>... On the 1280 I was able to get a modest sepia tone with the
>vmp8-c14 curve. I have seen Paul's prints from an 1160 and he was
>able to get a much stronger sepia tone. Perhaps with the older
>"cc" curve from last summer I could get a better sepia effect.

Yes, the old "cc" curve is now useful.  The prints done with it closely
match some of the old photo tones that I found. (My conclusion was that the
old photos when new were very sepia-toned.  They fade to a lighter yellow
tone.)  For landscapes, however, I back off one step and use the "cold"
curve that is out there for most of the printers.

>For my own taste, only the print with the vmp8-mw16 curve is
>something I might use occasionally. All of the other effects
>are nice but are not the direction I personally want to go.
>They will be heaven for many others though.

That is why the market needs different inksets.  I think that the standard
MIS VM inkset is the one most should start with.  I suspect, however, that
the VM-sepia is the one that will stay in my printer the most.  In addition
to my old photo reproductions, I've simply found that the sepia does very
well when in competition with color photos.  For my own tastes, the dead
neutral end is the one I like.

>I tried some combination layers of a "medium warm" curve over a
>"warm" (actually neutral with these inks) and set the "medium warm"
>upper layer's opacity to 50% and then 25%. Now these are looking
>like my selenium toned silver! This is in line with Paul's earlier
>post about adding some of the toner to the black to warm up the
>ink set and damp down the slight blue tint of the grays.

>The one difficulty here for me is that the VM curves for the
>6 ink printers are not so smooth in the extreme positions,
>especially the warm. The curves have to become too radical ....
>Gentler curves like the "neutral cool" and "medium warm" are
>very smooth and I would like to be printing this new VM ink
>with curves in that range.

The 1280-PC is the most troublesome with respect to the warm curve.  The
medium warm is where one should start in making a true neutral curve for the
1280.  There will be some toner added throughout the range, and the slope
will be sufficiently mild that the artifacts should be under control.

Martin, if you, or anyone else, is interested in this, I'll take a shot at
making a neutral curve for the 1280-PC.  I'll use the X-Rite
spectrophotometer to get accurate tones, which means I'll need the actual
test strips.  When it's done, it will also serve (hopefully) as a smoother
"warm" curve for the 1280-PC -- although less warm than the one that tries
to get all the cold toner out of the midtones.  Let my know if you're
interested.

> I am considering mixing up some Sepia-VM with a smaller gamut ...

I may be doing that anyway.  I'm currently testing a version of the sepia
toner that does not warm.  For this test batch, the gamut had to be lowered
a bit.  However, it'll be a while before I know what the final non-warming
formula will be -- if anything.

>... Give up a full sepia tone in exchange for more subtle control near the
>"neutral" point of the ink set. Maybe reduce the yellow and magenta to 1/3
or >1/4 the current amounts.

This is where the ability to manipulate the curves pays off.  You can
currently get fine control of the inkset as is.  At least on my printers it
can give a great neutral and the full-on sepia.  (Whether the 1280 can do
this is an open question, which is why I'd like to give it a shot.)

The pre-made curves may not hit the exact tone you want.  Layering with
different curves or the sliders give some control, but nothing is a good as
going into the curves yourself and tweaking them as needed to achieve the
exact tone you want.  I obviously want to encourage all serious printers to
get familiar with these curves.  It's not that hard once you dive in, and
the amount of control it gives is most rewarding.

>Paul, in the past you were mixing inks by measuring them with syringes
>but I recall you mentioned you had switched to an electronic balance.
>Do your earlier formulas where you specified it as "x parts of a to y
>parts of b" translate from volume to weight directly?

Yes, the densities of the inks does not give an exact volume/weight ratio,
but its so close I ignore the differences.

>What make and model are you using?

It's an Ohaus Corp. Scout II scales.

>Back in November you were looking at the Sepia-VM and posted this formula:

>"The sepia toner formula is 4 parts Yellow pigs (the newer,
>much more fade-resistant version sold by MIS as "FS yellow"),
>2 parts magenta pigs, 3 parts MIS "25" and 3  parts MIS clear base."

The current toner is 50% color pigs and 50% gamut/density controls.  For the
color half, it is still 2:1 Enhanced/FS yellow:Arc Magenta.  For the
gamut/density control half it's essentially 1:4 clear base:MIS quad "25."

>I want to try a Y:M:MIS25:Clear of 4:2:9:9 or even 4:2:21:21
>which should give me a tone with the VM neutral curve that
>is close to the tone I got with the medium warm and the warm
>in combination.

Getting the density right for the curves is tricky.  It's really easier to
manipulate the "software" (the curves) than the "hardware" (the inks).
That is what is behind the entire VM approach and ability to run on so many
machines.  Given the small size of our quad market, I would never be able to
convince MIS or anyone else to make inks that would just be compatible with
one printer model.

>Do you know if a mix of 8 parts toner to 2 parts clear is the
>proper dilution for the photo position of the toner?

The photo toner is 50% toner, 50% MIS "25."

>Of course a six channel RIP could make for the ultimate VM ...

Yes, you can go both cool and warm/sepia with a single hextone.  With an
8-tone, you can get perfect quads and photo-quality color from the same
machine.  I have no doubts at all that Epson will get there.  I doubt the
new 7-ink machine will quite make the grade, however.  Only one mid-tone
gray is not quite enough, in my view.

Paul
http://www.PaulRoark.com

At 11:35 PM 3/11/2002 -0800, you wrote:
>Well the B&W inkjet world is even broader with the availability of the MIS
Sepia-VM set. This set is not listed on their website but can be ordered in
4 oz or larger bottles if you call them. I purchased a 4 oz. set and loaded
them into virgin cartridges. I currently have it my 2nd 1280 and did some
13x19 and 11x17 prints on EAM and Photo Rag today using a variety of Paul's
and Tyler's curves for the standard MIS VM and some combinations.
>
>This is a very interesting set and the differences achieved by applying
different curves seem much stronger than with the regular VM. This set is
most interesting at the extreme ends with maximum toner or minimum toner
where the standard set is most often used with the middle curves. On the
1280 I was able to get a modest sepia tone with the vmp8-c14 curve. I have
seen Paul's prints from an 1160 and he was able to get a much stronger sepia
tone. Perhaps with the older "cc" curve from last summer I could get a
better sepia effect.
>
>With the warm, vmp8-w21 curve applied, the print looked exactly like MIS
Neutral FS as far as I can tell. A very slightly cool neutral.
>
>With the neutral cool, vmp8-nc16 curve I got print that fell in between
warm and light sepia. This was my least favorite.
>
>With the medium warm vmp8-mw16 curve I got a print that was very smooth
tonally and considerably warmer than anything I have seen with regular MIS
VM, regular MIS FS or Piezo inks. If you like warm tone prints this would
seem to be the ink set to go with.
>
>For my own taste, only the print with the vmp8-mw16 curve is something I
might use occasionally. All of the other effects are nice but are not the
direction I personally want to go. They will be heaven for many others though.
>
>Next I tried some combination layers of a "medium warm" curve over a "warm"
(actually neutral with these inks) and set the "medium warm" upper layer's
opacity to 50% and then 25%. Now these are looking like my selenium toned
silver! This is in line with Paul's earlier post about adding some of the
toner to the black to warm up the ink set and damp down the slight blue tint
of the grays.
>
>The one difficulty here for me is that the VM curves for the 6 ink printers
are not so smooth in the extreme positions, especially the warm. The curves
have to become too radical to overcome the Epson driver crossovers for the
photo cyan and photo magenta position inks. Gentler curves like the "neutral
cool" and "medium warm" are very smooth and I would like to be printing this
new VM ink with curves in that range.
>
>So with that in mind, I am considering mixing up some Sepia-VM with a
smaller gamut approach for the 1280. Give up a full sepia tone in exchange
for more subtle control near the "neutral" point of the ink set. Maybe
reduce the yellow and magenta to 1/3 or 1/4 the current amounts.
>
>Paul, in the past you were mixing inks by measuring them with syringes but
I recall you mentioned you had switched to an electronic balance. Do your
earlier formulas where you specified it as "x parts of a to y parts of b"
translate from volume to weight directly? What make and model are you using?
>
>Back in November you were looking at the Sepia-VM and posted this formula:
>
>"The sepia toner formula is 4 parts Yellow pigs (the newer, much more
>fade-resistant version sold by MIS as "FS yellow"), 2 parts magenta pigs, 3
>parts MIS "25" and 3  parts MIS clear base."
>
>Is this close to the Sepia-VM I just got from MIS?
>
snip

>Martin Wesley
>

Thanks Martin,

You answered my last post before I posted it!

AZ
Maker of Lookaround panoramic camera.

http://www.lookaroundcam.com/
         or
keyword.com lookaround

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

From: "Paul Roark" <paul.roark@...>
To: "DigitalB&WPrint" <DigitalBlackandWhiteThePrint@yahoogroups.com>
Sent: Tuesday, March 12, 2002 8:26 AM
Subject: [Digital BW] MIS Sepia-VM First Impressions


> Martin,
>
(snip)
>
> > This set is most interesting at the extreme ends ...
>
> I like the tones best *near* the ends of the range as opposed to at the
> ends.  For example, the old "warm" curve, which is at the neutral end of
the
> VM-sepia, is too cool and uneven.  This is the end where you can use just
a
> little toner to balance the inkset to right-on neutral through the
> mid-tones.  This will take some fine tuning, but for those who want a true
> neutral tone, I think this does the best job I've seen.

Yes there is a very big difference between the tones achieved with the
medium warm and warm curves. I think some intermediate curves would be very
useful to many people.
>
(snip) Perhaps with the older
> >"cc" curve from last summer I could get a better sepia effect.
>
> Yes, the old "cc" curve is now useful.  The prints done with it closely
> match some of the old photo tones that I found. (My conclusion was that
the
> old photos when new were very sepia-toned.  They fade to a lighter yellow
> tone.)  For landscapes, however, I back off one step and use the "cold"
> curve that is out there for most of the printers.

That sounds very good.
>
> >(snip)  All of the other effects
> >are nice but are not the direction I personally want to go.
> >They will be heaven for many others though.
>
> That is why the market needs different inksets.  I think that the standard
> MIS VM inkset is the one most should start with.  I suspect, however, that
> the VM-sepia is the one that will stay in my printer the most.  In
addition
> to my old photo reproductions, I've simply found that the sepia does very
> well when in competition with color photos.  For my own tastes, the dead
> neutral end is the one I like.

I think you are right. I would like access to a dead neutral and slightly
warm neutral. The Sepia-VM has the potential.
>
> >(snip) The curves have to become too radical ....
> >Gentler curves like the "neutral cool" and "medium warm" are
> >very smooth and I would like to be printing this new VM ink
> >with curves in that range.
>
> The 1280-PC is the most troublesome with respect to the warm curve.  The
> medium warm is where one should start in making a true neutral curve for
the
> 1280.  There will be some toner added throughout the range, and the slope
> will be sufficiently mild that the artifacts should be under control.

That was the approach I was thinking of.
>
> Martin, if you, or anyone else, is interested in this, I'll take a shot at
> making a neutral curve for the 1280-PC.  I'll use the X-Rite
> spectrophotometer to get accurate tones, which means I'll need the actual
> test strips.  When it's done, it will also serve (hopefully) as a smoother
> "warm" curve for the 1280-PC -- although less warm than the one that tries
> to get all the cold toner out of the midtones.  Let my know if you're
> interested.

I would be very interested. Just let me now what I should send you or e-mail
me the target file for the X-Rite.

As you know Tyler Boley did some curves for me for the standard VM on the
1280 with Photo Rag using Dan Culbertson's RGB soft proof method with
Profiler Pro and a Spectrocam photospectrometer. He ran into similar
problems in that using a mid point of ink distribution, as you would with a
non-variable ink set, gave very nice smooth curves but as he attempted to
move the toner ink in or out of the mix the smoothness of the curves became
harder to achieve.

One of the problems is that applying the RGB separation curves can cause
combing in the histogram. If you start out with a nice smooth image you are
okay but anything that already has some combing you can really loose the
image. In 16-bit this is probably not a problem.

If you haven't already, you might want to try Dan's settings on the Epson
driver too. He sets the driver to "No Color Adjustment". I wonder if this
does not make the "black box" of the Epson driver a bit smaller and allow
your separation curves to address the printer a little more directly. You
give up the slider adjustment but with the 1280 that seemed to not be very
usable since it tended to make the curve response more non-linear.
>
> > I am considering mixing up some Sepia-VM with a smaller gamut ...
>
> I may be doing that anyway.  I'm currently testing a version of the sepia
> toner that does not warm.  For this test batch, the gamut had to be
lowered
> a bit.  However, it'll be a while before I know what the final non-warming
> formula will be -- if anything.

I will be interested in your results.
>
> >... Give up a full sepia tone in exchange for more subtle control near
the
> >"neutral" point of the ink set. Maybe reduce the yellow and magenta to
1/3
> or >1/4 the current amounts.
>
> This is where the ability to manipulate the curves pays off.  You can
> currently get fine control of the inkset as is.  At least on my printers
it
> can give a great neutral and the full-on sepia.  (Whether the 1280 can do
> this is an open question, which is why I'd like to give it a shot.)

Is it just the 1280 or do the other six color printers have similar
problems?
>
> The pre-made curves may not hit the exact tone you want.  Layering with
> different curves or the sliders give some control, but nothing is a good
as
> going into the curves yourself and tweaking them as needed to achieve the
> exact tone you want.  I obviously want to encourage all serious printers
to
> get familiar with these curves.  It's not that hard once you dive in, and
> the amount of control it gives is most rewarding.

Once again I agree. Too many of use are sitting around waiting for you to
tweak the curves for us. You gave a brief talk on curve adjustment at the SF
area meeting but time was short and I only absorbed a bit of it.
>
> >(snip)
> >Do your earlier formulas where you specified it as "x parts of a to y
> >parts of b" translate from volume to weight directly?
>
> Yes, the densities of the inks does not give an exact volume/weight ratio,
> but its so close I ignore the differences.
>
> >What make and model are you using?
>
> It's an Ohaus Corp. Scout II scales.

Thanks for the info!
>
> >Back in November you were looking at the Sepia-VM and posted this
formula:
>
> >"The sepia toner formula is 4 parts Yellow pigs (the newer,
> >much more fade-resistant version sold by MIS as "FS yellow"),
> >2 parts magenta pigs, 3 parts MIS "25" and 3  parts MIS clear base."
>
> The current toner is 50% color pigs and 50% gamut/density controls.  For
the
> color half, it is still 2:1 Enhanced/FS yellow:Arc Magenta.  For the
> gamut/density control half it's essentially 1:4 clear base:MIS quad "25."

Great! I have mixed my own darkroom chemicals for years so I feel like I
should take a shot at this. Maybe play with the Y to M ratio of the toner
too.
>
> >I want to try a Y:M:MIS25:Clear of 4:2:9:9 or even 4:2:21:21
> >which should give me a tone with the VM neutral curve that
> >is close to the tone I got with the medium warm and the warm
> >in combination.
>
> Getting the density right for the curves is tricky.  It's really easier to
> manipulate the "software" (the curves) than the "hardware" (the inks).
> That is what is behind the entire VM approach and ability to run on so
many
> machines.  Given the small size of our quad market, I would never be able
to
> convince MIS or anyone else to make inks that would just be compatible
with
> one printer model.

The density is the problem if you want to stay compatible with existing
curves. I generally agree with you on this and certainly don't expect MIS to
offer inks in 31 flavors. But there may be a small number of people who
would like to try hand rolling some ink sets and with all the work you have
done in this area you knowledge in invaluable.
>
> >Do you know if a mix of 8 parts toner to 2 parts clear is the
> >proper dilution for the photo position of the toner?
>
> The photo toner is 50% toner, 50% MIS "25."

Thanks for that!
>
> >Of course a six channel RIP could make for the ultimate VM ...
>
> Yes, you can go both cool and warm/sepia with a single hextone.  With an
> 8-tone, you can get perfect quads and photo-quality color from the same
> machine.  I have no doubts at all that Epson will get there.  I doubt the
> new 7-ink machine will quite make the grade, however.  Only one mid-tone
> gray is not quite enough, in my view.

I don't expect that it will satisfy me or lot of the other quad printers but
I do suspect it will satisfy a great number of people with less exacting
requirements. This is likely to reduce the market for MIS, Conetech and the
other ink manufacturers unfortunately.

Martin