PDA

View Full Version : How to improve the dark details on my Pt/Pd print



Andrew ren
15-Oct-2009, 03:34
Hello all,

Got some problem with my Pt/Pd printing.

I am using the digital negative approach to make my contact prints.

now, the same file, scanned from 57 tmax, is printing fine by my epson 3800 on silver rag, all sorts of nice deep dark details showing on the inkjet print.

but the same file, got pitch black on the Pt/Pd print, I tried to open the shadow a little bit in the Photoshop and re-adjust the level, the curve, just can't get it right, at least to my eye. i mean the black is blocky, ugly looking.

is that hard to get a nice deep dark details on Pt prints? I knew that Pt is good on the middle/upper tone..

Jim.K, my friend on this forum suggests me to over-expose, say, zone 3 to zone 5, then N-2 to get all the dark details i need on the Pt print, and meanwhile you can adjust them in the Photoshop to get a darker detail on the digital print...

any ideas?

Cheers

Andrew

Steve Sherman
15-Oct-2009, 04:03
Hello all,

Got some problem with my Pt/Pd printing.

I am using the digital negative approach to make my contact prints.

now, the same file, scanned from 57 tmax, is printing fine by my epson 3800 on silver rag, all sorts of nice deep dark details showing on the inkjet print.

but the same file, got pitch black on the Pt/Pd print, I tried to open the shadow a little bit in the Photoshop and re-adjust the level, the curve, just can't get it right, at least to my eye. i mean the black is blocky, ugly looking.

is that hard to get a nice deep dark details on Pt prints? I knew that Pt is good on the middle/upper tone..

Jim.K, my friend on this forum suggests me to over-expose, say, zone 3 to zone 5, then N-2 to get all the dark details i need on the Pt print, and meanwhile you can adjust them in the Photoshop to get a darker detail on the digital print...

any ideas?

Cheers

Andrew

Hi Andrew,

I am fairly new to PT / PD printing but do remember that the dark values of a Pt / Pd print will dry down considerably from what you see while still in a wet state. I think much of what you need is to balance the high value and dark values through your developer contrast knowing the dark values will darken considerably when dry.

Also, I remember that for larger areas of dark tone many printers use a solution of "Tween" to even out the dark values you describe.

Hope this helps

Jim Michael
15-Oct-2009, 04:25
Doesn't pt have a different curve that pt/pd? If so I think the digital negative production process should be recalibrated for pt and include your step wedge in the test print in order to establish that any issues are with the negative rather than the chemistry.

Andrew ren
15-Oct-2009, 04:40
Steve,

you knew me, I am doing things in Photoshop :-0 but not too much though.

the problem is the "file" looks good on my calibrated screen and print well on the silver rag on the epson 3800.

Jim,

yes, I did the test pretty consistently, the ratio of Pt/Pd, the temperature of the Potassium developer.. and so on.

the only thing i noticed is that the uncovered the coating area is still slightly lighter than the covered area(by OHP) even @ 275units on my nuarc 26-1k. so you are right, I have to redo everything, start from 200units, then rebuild the curve. but this will only fix the solarizing problem, not the dark detail.

Andrew

Colin Graham
15-Oct-2009, 05:05
I'm not really clear what your workflow for digital negatives is, so it's hard to tell how to help. How did you arrive at your curve for pt? Are you using the epson driver and composite black ink, or a color negative system like PDN, or a RIP like Quadtone Rip?

You might want to check out this site- hybridphoto.com (http://www.hybridphoto.com/forums/)- there is an entire subforum devoted to digital negatives.

Andrew ren
15-Oct-2009, 05:15
Colin,

first and foremost, is the base exposure, as i mentioned, 275unit on my nuarc.

then print the step wedge with QIP, with no curve modification, raw scanned. and build the correction curve based on the scanned file.

Andrew ren
15-Oct-2009, 05:17
next, print the step wedge with QIP with this newly built correction curve. scan it and double check the file, the input and output value.

Colin Graham
15-Oct-2009, 05:20
If your using QTR and need more shadow separation, you might want to adjust the LK inkset (density and limits) to give more of a boost in the 90-100 step range. That's often the most difficult area to curve. I usually try to get a much more linear response in the 0-20 steps and 90-100 steps just with ink densities and limits before I even start to curve.

Andrew ren
15-Oct-2009, 05:22
to me, the I/P and O/P are pretty much matched and seems quite linear.

so the correction curve is a working curve. to me.

and also, any value higher than 95% ON THE STEPWEDGE(from 95-100) is rendered as pitch black on my print. but not on my digital print. same "file"

Andrew

Colin Graham
15-Oct-2009, 05:28
I'm a little confused about the two separate curves. Are you using them in sequence somehow? Did you change the ink profile between them to get a more linear response?

But maybe try printing a wedge with a longer range of tones in the problem areas. Smaller wedges with 5 step increments might be trickier to tweak with small inconsistancies of sensitizer, etc. If the step is printing correctly but similar tones in your image file aren't, they may be blocking up in areas hard to nail down with shorter step tablets.

Andrew ren
15-Oct-2009, 05:36
Colin,

thanks for the link.

my 95% is already "noticeable" lighter than the 100% pure black. I think. how does it compare to your view, does the 95% still need to be lighter? that I might have to increase LK, LC, and LM to certain degree...

what do you mean precure? is the one after setup the base exposure yet before print stepwedge with QTR(first time printing the step wedge)?

the first stepwedge i print and developed is non-curve version. do you mean increasing the LK, LC, and LM before printing the first step wedge? if that's the case, how far should I go?

Cheers

Andrew

Colin Graham
15-Oct-2009, 05:38
OK, I get it, the second curve is the reading from the corrected step wedge, not a correction curve.

What I mean by tweaking the inks to get a more linear response is to avoid a dramatic curve like the one you posted. I think you might be sacrificing shadow detail on that steep low end curve. That's the beauty of QTR, the light and dark inks are indepentedly adjustable so you can avoid lossy curves. . The downside is it takes some testing to get there.

Andrew ren
15-Oct-2009, 05:39
yes! sorry Colin.

:-)

Andrew

Colin Graham
15-Oct-2009, 06:10
If your max-black base exposure time is good, maybe tweak the LK limits up until you show good linear separation between 90 and 100 before you apply any curves. As I mentioned, a longer range of tones in your step wedge in this area would be helpful, maybe one that also has some redundant steps so you can compare against any inconsistencies of coating.

(I wouldn't mess with the LK density unless there's a big problem with the crossover of light to dark inks, like odd grain in transition steps in the midtones- doesn't sound like you have that problem. Generally that number is supposed to remain constant with the 3800.)

The snapshot is for a carbon profile, so I don't know how well these numbers would work for PT/PD.

clay harmon
15-Oct-2009, 06:27
Your curve looks like you have done a pretty good job in linearizing the negative. If the problem is really just in the 95-100% black area, you can start tweaking the QTR profile to add some ink, but the fundamental issue with this approach (and all the other digineg approaches share this) is that since you a printing a negative, that means there is very little ink being deposited for the 90-100% tones. So getting separation in the dark areas of your print is an exercise in very small adjustments that make some pretty large differences on your negative.

This also makes it extremely sensitive to exposure time. If you are using a nuarc unit or amergraph UV unit, then you have a light integrator, but even with the integrator there can be minute differences in exposure that will not affect midtones so much as the dark areas. Just a tiny bit too much exposure can turn a 95% black to a 100% black.

Getting it perfect can be time consuming. If all you want is just a good print, you might try just making a levels adjustment to your image prior to inverting and printing. Try keeping your input levels the same, and set your output level at something like 96-97%. This will guarantee that you get a little more ink even in the thin areas of the negative once it is inverted and printed. Another approach is to just slightly reduce your exposure time, by maybe a 2-5% and see if that opens up your shadows.

Good luck!

cjbroadbent
15-Oct-2009, 06:58
Cheat. Scan twice, do an HDR on your picture - print large - re-shoot the print - over-develop. The new negative now has some bite for PT/Pd.

Andrew ren
15-Oct-2009, 08:35
clay,

thanks for the input.

I will start from the scratch, redo the base exposure time, and LK value, rebuild curve..

What a FUN!

Cheers

Andrew

PViapiano
15-Oct-2009, 08:50
Cheat. Scan twice, do an HDR on your picture - print large - re-shoot the print - over-develop. The new negative now has some bite for PT/Pd.

Bite your tongue...time out for you, Christopher! :eek:

PViapiano
15-Oct-2009, 08:51
clay,

thanks for the input.

I will start from the scratch, redo the base exposure time, and LK value, rebuild curve..

What a FUN!

Cheers

Andrew

Andrew...show us the print.

It's never going to look exactly like the inkjet print, though.

Andrew ren
15-Oct-2009, 17:47
Paul,

I know it will be different...
check your flicker, please. I had a post on your work -"Black bird, Switzerland"



ok, there is something weird, check this out.

this is the no curve text, run the install command QTR, got the nocurve.quad.

see the difference of LC/LM/LK between the graph and text?

Andrew ren
15-Oct-2009, 17:48
same as my newly built correction curve.

hmm, my LK and LLK does look very steep.

Colin Graham
15-Oct-2009, 19:07
That profile is so different from my approach I'm honestly not sure if any of my suggestions would help or hurt. You're already at double the density and limit of LK I use, and I don't use the LLK ink at all.

Very interesting about the high levels of LLK, I'm really curious about it's function. It's so cool to see other profiles and approaches!

PViapiano
15-Oct-2009, 21:25
Andrew...

Saw your comment on my Flickr image. That was my first "larger" image after printing many 4x5s to get the process under control. I miscalculated the amount of NA2 for the negative and got the dark shadows as well as bright highlights. The print could be softer in contrast, but it worked so well that I decided to leave it. In person, it's a very dramatic print. Since then, I've tweaked to adjust my neg, exposures and amount of NA2 to dial it in better. I'm very happy now.

It all depends on what you're looking for and what you want to express in the print.

Andrew ren
16-Oct-2009, 14:20
Colin,
that confused me too.
I got this from a workshop, normally its a good start point. but now i don't know where I should start to tweaking it, as the LK number is so high already.

Paul,
yes, that is what I want as well, but my eye just is not satisfied with the current result. so still long way to go.

but, I at least figured out the "Pizza Wheel" problem. ha-ha.

Andrew