Digital BW, The Print

Hi all

I've been redoing some QTR curves from 1440 to 2280 dpi lately and re-
encounter an effect I've was not sure it was there actually. My 
measured Dmax seems to vary quite wildy when I measure the same step 
wedges on different moments, also after drying properly.

Currently, I'm trying different ink limits/black boost/overlap QTR 
settings. But the differences between the settings seem to disappear 
in the noise of my measurements.

My setup: QTR/Windows/Epson 2200/UT3D/Enhanced Matte/Eye-One.

Most of the time I measure L*/Dmax values between 16.3/1.67 and 
17.0/1.64, but not seldom I get values of around 19/1.56. 
Occaisionally even higher. These variations do not occur within one 
measurement session: repositioning and measuring again gives some 
normal measurement noise (L* +/- 0.2). But when I repeat a 
measurement a or some days later, the varations are much larger.

I do take care that I allow for drying overnight. My feeling is that 
is has something to do with changes/contaminations of the surface of 
the eboni ink. For instance, if I print a second step wedge on the 
same paper, I tend to see a roller mark on the first step wedge. I 
can't substantiate this yet, but I also have the feeling that a layer 
with more ink (giving a better the Dmax at t=0) is more prone to this 
effect (and giving a worse dmax later).

Any clues here? 

Joost

Do you calibrate your i1 every time you use it? (every day)

Just wondering if it drifts enough without a fresh calibration to
cause this.

If your i1 is very old it may need to go back to be re-certified.

--- In DigitalBlackandWhiteThePrint@yahoogroups.com, "Greg" 
<dfaprinting@...> wrote:
>
> Do you calibrate your i1 every time you use it? (every day)
> 
> Just wondering if it drifts enough without a fresh calibration to
> cause this.
> 
> If your i1 is very old it may need to go back to be re-certified.
>

Hi Greg,

I use GretagMcBeth Profilemaker for doing my spot and chart 
measurements. That forces a calibration step before every measurement 
series. The Eye-one is only a few months old.

Joost

Joost,

It could be that you are putting too much ink down, but that normally
shows up right away and doesn't take days like you are seeing. Do you
have something that is fairly stable like a Macbeth chart that you can
measure? That way you can check for consistency between sessions.

First of all, a happy 2007 to you all! I hope in this year the world 
will not be as black and white as our pictures ;-)


--- In DigitalBlackandWhiteThePrint@yahoogroups.com, "Greg" 
<dfaprinting@...> wrote:

> Do you
> have something that is fairly stable like a Macbeth chart that you can
> measure? That way you can check for consistency between sessions.

The same thought occurred to me. I did checks with both the GMcB 
Digital Color Checker SG (a semi-matte plastic camera target) and the 
GMcB Eye-one Scan Target 1.4 (a glossy paper target). 

There seems to be nothing electronically wrong with my Eye-one. With 
the glossy scan target the measurements were immediately very 
reproducible (noise of L* = 0.1/0.3). With the semi-matte camera target 
I did have some performance issues on the start. Especially the blacks 
showed a similar problem (L* varying between 7.4 and 11). Upon closer 
inspection however it turned out to be the result of a very slight 
(frankly invisible) contamination. After cleaning the results were very 
reproducible, also the L* values dropped. Perhaps cleaning is not even 
the right word. I rubbed the targets with my finger after which they 
looked a tiny bit "smoother". The midtones and higlights were much less 
susceptible to this effect.

So this probably points to the source of my problems with the EEM/Eboni 
targets. I have noticed before that the eboni prints are quite 
sensitive (note my remark on the printer rollers in the original post). 
I have done nothing special with the targets, neither handled them 
roughly, but also did not protect them in a special way. The have just 
been laying around for a day or two. But that's maybe not the way to 
deal with them.

Going back to the semi-matte camera target, another, slightly related 
explanation could be that I did not actually clean the target, but put 
some tiny bit of grease on them with my finger, thereby "polishing" it. 
In that way, it might be on optical effect, perhaps similar to the 
difference between matte and glossy papers. Then again, it would be an 
inherent effect of the matte print.

Does this make sense to anyone? 
 
Joost

Joost Horsten wrote:
> First of all, a happy 2007 to you all! I hope in this year the world 
> will not be as black and white as our pictures ;-)

A happy 2007 for all.


> So this probably points to the source of my problems with the EEM/Eboni 
> targets. I have noticed before that the eboni prints are quite 
> sensitive (note my remark on the printer rollers in the original post). 
> I have done nothing special with the targets, neither handled them 
> roughly, but also did not protect them in a special way. The have just 
> been laying around for a day or two. But that's maybe not the way to 
> deal with them.
> 
> Going back to the semi-matte camera target, another, slightly related 
> explanation could be that I did not actually clean the target, but put 
> some tiny bit of grease on them with my finger, thereby "polishing" it. 
> In that way, it might be on optical effect, perhaps similar to the 
> difference between matte and glossy papers. Then again, it would be an 
> inherent effect of the matte print.
> 
> Does this make sense to anyone? 

It could be different on the Eye One but my SpectroCam is 
quite sensitive to the distance to the target and the angle. 
The automatic  measuring setup with an old HP plotter had the 
spectrometer slightly higher at first and the results were 
darker readings. A slightly different angle of the head on the 
patch showed similar changes. I had already made another strip 
guide that allows longer strips to be measured manually and 
that works good. The head doesn't touch the strip but is at 
the right height and a thin PET foil protects the rest of the 
target when the spectrometer is running over it.

I'm an advocate for multiple targets and multiple measurements 
+ averaging. The SpectroCam does that already on its way over 
the patches, between 7-15 readings per patch. But I use more 
targets where possible, the targets printed in another 
direction etc.

An interesting fact about the HP Z3100 and Z2100 is that the 
integrated calibration and profiling measures per patch and 
the dark patches get a longer time for the measurement. 
Something that is difficult to achieve with strip reading or 
one should increase the size of the dark patches.

Any change in the surface of matte black should give different 
readings.


Met vriendelijke groeten, Ernst


|  Dinkla Grafische Techniek  |
|     www.pigment-print.com    |
|             ( unvollendet )            |

Hi Ernst,

Thanks for your suggestions.

> It could be different on the Eye One but my SpectroCam is 
> quite sensitive to the distance to the target and the angle. 
> The automatic  measuring setup with an old HP plotter had the 
> spectrometer slightly higher at first and the results were 
> darker readings. A slightly different angle of the head on the 
> patch showed similar changes. I had already made another strip 
> guide that allows longer strips to be measured manually and 
> that works good. The head doesn't touch the strip but is at 
> the right height and a thin PET foil protects the rest of the 
> target when the spectrometer is running over it.

I checked this and it does not seem to explain what happens. The eye-
one is well constructed to keep the angle fixed. Also distance seems 
to be wel controlled, both with and without guide. Furhtermore, if I 
artificially increase the distance a bit, the meaused l* LOWERS (i.e. 
less light light reflected, which is logical to me) instead of 
increasing (what is what I find if the measurement seems not to be 
OK).  

> I'm an advocate for multiple targets and multiple measurements 
> + averaging. The SpectroCam does that already on its way over 
> the patches, between 7-15 readings per patch. But I use more 
> targets where possible, the targets printed in another 
> direction etc.

Given my recent experience, I'll will make this common practice as 
well. However, my current results seem to be obscured by systematic 
artifacts. No stochastic process will eliminate this.
> 
> An interesting fact about the HP Z3100 and Z2100 is that the 
> integrated calibration and profiling measures per patch and 
> the dark patches get a longer time for the measurement. 
> Something that is difficult to achieve with strip reading or 
> one should increase the size of the dark patches.

That's interesting indeed, confirming that black readings are more 
prone to errors. 

> Any change in the surface of matte black should give different 
> readings.

I come to that conclusion as well. It's not illogical. My explanation 
is that any contamination or damage increases the scattering at the 
surface and, by that, increases the reflectivity of the surface. Is 
that what you mean as well? But I'm surprised by the size of the 
effect.

Joost

Joost Horsten wrote:

>> Any change in the surface of matte black should give different 
>> readings.
> 
> I come to that conclusion as well. It's not illogical. My explanation 
> is that any contamination or damage increases the scattering at the 
> surface and, by that, increases the reflectivity of the surface. Is 
> that what you mean as well? But I'm surprised by the size of the 
> effect.

Given the up to 0.6 D differences between matte and gloss 
prints (with PK) it can be expected. A superb gloss top coat 
applied on a textureless matte print (with MK) will  show 
similar differences. It is the well known effect you get with 
spot varnishing on a matte or semi-gloss black offset print. 
Different visual density and a more selected view angle for 
the deeper black of the varnished area. So the grease 
(varnish) or marks (texture is reduced like in paper 
satination) reduce the scattering and the spectrometers 
light/sensor geometry results in less light reflectance to the 
sensor itself as it isn't in the the direct angle of reflection.

Once we had a discussion here on the differences between dye 
ink matte Dmax and pigment ink matte Dmax. That's a harder nut 
to crack.

Met vriendelijke groeten,   Ernst


|  Dinkla Grafische Techniek  |
|     www.pigment-print.com    |
|             ( unvollendet )            |

In a message dated 1/1/07 1:50:37 PM, j.h.j.h@... writes:


> That's interesting indeed, confirming that black readings are more
> prone to errors.
> 

Yes, reading darkness with a light sensor is less effective than reading 
light with a light sensor, this is true of all such devices.

C. David Tobie
Product Technology Manager
ColorVision Business Unit
Datacolor Inc.
CDTobie@...
www.colorvision.com


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