Making my first digital negs

[PVia]

Wow...there are so many methods I've read about, it makes one dizzy to comprehend all the information and techniques.

I'm using a new-to-me Epson 3800 and my goal is to make negs for pure Pd printing with no (or very little) contrast agent.

With Epson driver:
-----------------
My first neg used a process curve Kerik kindly sent me. I printed in RGB mode with PK and exposed for only two minutes; viewing through the split back frame told me it was ready. My final print was way too dark with low contrast, nothing like my screen, but I could see there was promise. When I printed the neg, I used the profile for Epson Premium Glossy...I probably should have used Same As Source in PS and then picked No Color Management in the Epson driver. That's something I have to look into if I print another one, maybe it'll print denser that way.

My next one tried adding a duplicate layer in Multiply mode at 50% opacity. The negative was definitely more robust. I haven't exposed this one yet.

My third neg was as above (with the Multiply layer) only this time I added +16% density. Once again, this gave me a thicker looking neg than the second one. I haven't exposed this yet, either.

With QTR:
----------
My fourth neg was using Ron's palladium curve for QTR. This uses the matte black ink. The neg is definitely much denser looking than the previous ones. Upon inspection with a loupe, it seems that some fine detail may be compromised, but I won't know until I expose it.

The last one I'd like to try uses Kerik's curve for QTR, found on this forum, which uses Photo Black ink. I'll try printing this one tomorrow if I have a chance.

------------------

It's amazing that there are so many different opinions as to Photo/Matte Black, Epson driver/QTR, AB&W, PDN, Chartthrob...it really confounds the beginner. I'm willing to do the work. I just wish there was a baseline that could be "in the ballpark". Maybe one of the ones above will do that for me.

I'll post back here with results. Hopefully it will help someone else on their journey as well...in the meantime I'd love to hear any comments y'all have.

It's truly humbling to browse through your individual websites and see the inspiring work you are all doing...

 

[sly]

Wow, right back! I gave up on trying to make diginegs for pt/pd. I realized I didn't know enough about alt printing, so I've spent the last year making 4x5 contacts, and am just venturing back to digi negs for cyanotype and gum. I'll be very interested in your results.

 

[bobwysiwyg]

At the risk of showing off my ignorance, what is a digital neg. and how is one made? :confused:

 

[PVia]

Bob,

A digital negative is made from a digital file, be it a scan or a digi-camera file. It's then printed, usually via inkjet printer, to a translucent/transparent substrate like an overhead projector transparency, in order to be used in traditional contact printing processes, ie, cyanotypes, platinum/palladium, van dyke, salt prints, gelatin silver, etc...

Paul

 

[bobwysiwyg]

Thanks Paul, it seems like I envisioned.. in my otherwise ignorant world. Do all transparency materials work just as well as another, or are there some better than others for this specific purpose?

 

[David A. Goldfarb]

Thanks Paul, it seems like I envisioned.. in my otherwise ignorant world. Do all transparency materials work just as well as another, or are there some better than others for this specific purpose?

Transparency materials for printers may be subbed for laser printers or for inkjet printers (or sometimes laser on one side, inkjet on the other). I tried some laser transparency material in my HP B9180 inkjet printer and got a runny mess, but I've gotten good results with Pictorico Premium OHP Transparency film, as have others.

 

[sanking]

I can understand why you are confounded. I have been making digital negatives since before there was PDN, Charthrob or QTR and am also confused at times as to the best overall method when you weigh ease of calibration. Here are some personal thoughts, others may or may not agree.

PDN --Overall PDN is still the best overall method because it is a complete system that can be done without sophisticated equipment. With PDN you have a method to choose the best UV blocking density and to produce a curve, without sophisticated equipment. All one needs is a simple flatbed scanner or reflection densitometer. However, the color scheme of PDN does not give sufficient UV blocking with current Epson photo printers for printing palladium with no contrast agent, unless you use the + Ink density setting.

Charthrob -- Charthrob is not a complete system but just a method for producing a curve. What is lacking in Charthrob is a method for finding the best color that will give the optimum UV blocking density for your process. Some people have used the color array of MKS that is available on the forum to find the best color, then use Charthrob to produce a curve. This works well enough, but the MKS color array does not give enought UV blocking density for printing palladium with no restrainer with current Epson printers.

QTR -- The advantage of QTR is that you can adjust the profile to give as much UV blocking density as needed. The problem is that in order to carry out the tests and make the adjustments you really need a UV transmission densitometer, which very few people have.

I am still working on a perfect QTR profile for the Epson 3800. Last summer I worked with Ron Reeder and developed a very good profile for my carbon process using the matte black inks. However, on closer inspection I am getting more grain with this combination than with other methods I have used so I am hoping to create a good curve using the PK ink.

Meanwhile, I am still printing most of my digital negatives with an old 2200 that is still fairly new in terms of use because it has been used only for making digital negatives. To print I use QTR with a profile that was developed for pure palladium, and then with the profile installed I run Charthrob, which develops a curve. I apply the curve to the positive file, invert, and then print using QTR with the palladium curve installed. Not as elegant as adjusting the curve within the QTR profile, but this method has worked fine for me for several years.

Sandy King

It's amazing that there are so many different opinions as to Photo/Matte Black, Epson driver/QTR, AB&W, PDN, Chartthrob...it really confounds the beginner. I'm willing to do the work. I just wish there was a baseline that could be "in the ballpark". Maybe one of the ones above will do that for me.

...

 

[Ron-san]

QTR -- The advantage of QTR is that you can adjust the profile to give as much UV blocking density as needed. The problem is that in order to carry out the tests and make the adjustments you really need a UV transmission densitometer, which very few people have.

Sandy King

Sandy

I respectfully disagree that you need a UV densitometer to make QTR profiles for digital negatives. A UV densitometer is a convenience, especially if you are making several of these profiles as I sometimes do. But in my QTR Manual I go through the whole process without using a densitometer. Instead I use a contact print of a step wedge negative to assess UV blocking ability of a particular ink setting (not too different than what the PDN system does).

My own guess is that a beginner would find the PDN system easier to wrap his/her head around than the QTR approach. In the long run, however, I think that QTR offers the ultimate control over a printers inks. And thus offers the chance, at least, of making a superior negative (although you might not be able to see any difference on a forgiving surface such as Pt/Pd).

Ron Reeder

 

[sanking]

Ron,

I am not in disagreement with you as to the fact that one can make QTR profiles without a UV densitometer. It just seems to me that it would be much more complicated to do so, which is why I remarked one "really needs a densitometer." But I recognize that it is not absolutely necessary. But since I have a UV densitometer I have never had to try to make a profile without one. In fact, even when I experiment with PDN or the MKS color array the convenience of using a UV densitometer is great compared to contact printing and then examining the step wedge densities. Ultimately the contact print is necessary because UV reading of the colors is not entirely reliable.

Also, I am in agreement with you in that QTR offers the ultimate control over the printer's ink. However, even though I have a good understanding of how to create and modify profiles I still have not found something to put my hat on in terms of accurately predicting how a change to the profile will modify some of the tonal values, especially those in the middle. Hitting the maximum UV blocking density is pretty easy, but getting the tones in the middle right has been very tricky for me. And very time consuming too, I might add.



Sandy

Sandy—

I respectfully disagree that you need a UV densitometer to make QTR profiles for digital negatives. A UV densitometer is a convenience, especially if you are making several of these profiles as I sometimes do. But in my QTR Manual I go through the whole process without using a densitometer. Instead I use a contact print of a step wedge negative to assess UV blocking ability of a particular ink setting (not too different than what the PDN system does).

My own guess is that a beginner would find the PDN system easier to wrap his/her head around than the QTR approach. In the long run, however, I think that QTR offers the ultimate control over a printer’s inks. And thus offers the chance, at least, of making a superior negative (although you might not be able to see any difference on a forgiving surface such as Pt/Pd).

Ron Reeder

 

[pschwart]

Ron,

I am not in disagreement with you as to the fact that one can make QTR profiles without a UV densitometer. It just seems to me that it would be much more complicated to do so, which is why I remarked one "really needs a densitometer." But I recognize that it is not absolutely necessary. But since I have a UV densitometer I have never had to try to make a profile without one. In fact, even when I experiment with PDN or the MKS color array the convenience of using a UV densitometer is great compared to contact printing and then examining the step wedge densities. Ultimately the contact print is necessary because UV reading of the colors is not entirely reliable.

Also, I am in agreement with you in that QTR offers the ultimate control over the printer's ink. However, even though I have a good understanding of how to create and modify profiles I still have not found something to put my hat on in terms of accurately predicting how a change to the profile will modify some of the tonal values, especially those in the middle. Hitting the maximum UV blocking density is pretty easy, but getting the tones in the middle right has been very tricky for me. And very time consuming too, I might add.



Sandy

I would agree that QTR offers the most control, but I find this the most difficult method. Creating QTR profiles is an art. Personally, I have had much more success making paper prints using QTR than negatives. I'm sure I could get this to work for me if I invested more time. The simplest process I have found is:
- find a printer/inkset that *does* provide sufficient UV blocking
- print negatives using all inks (composite black) with the native printer driver
- linearize with a Photoshop curve (transmission densitometer required)

I am currently using an Epson R1800 to print my digital negatives. This is only a 13" printer, but I make small prints so this works for me

 

[pschwart]

I would agree that QTR offers the most control, but I find this the most difficult method. Creating QTR profiles is an art. Personally, I have had much more success making paper prints using QTR than negatives. I'm sure I could get this to work for me if I invested more time. The simplest process I have found is:
- find a printer/inkset that *does* provide sufficient UV blocking
- print negatives using all inks (composite black) with the native printer driver
- linearize with a Photoshop curve (transmission densitometer required)

I am currently using an Epson R1800 to print my digital negatives. This is only a 13" printer, but I make small prints so this works for me

linearize with a Photoshop curve (transmission densitometer required)

Sorry, this should be reflection, not transmission (for reading step wedges). Get an X-Rite 810 and you get both in one unit

 

[sanking]

I think the ideal way for QTR to work to our advantage is to make our profiles available to others, say by putting them in a file section on this forum or on Ron's web site.

However, the profile should have a thorough, and very precise, description of the working procedures and I believe this can be included as text within the profile itself.

Sandy

 

[Ben Altman]

Sandy has a couple of great ideas in this thread:

- Get the QTR profile close (and smooth), so it is doing most of the work by adjusting the way the printer lays down ink. Then use a routine like Charthrob to make a fine-tuning curve in Photoshop, which can adjust for the minor variations between machines, batches of wet materials, photographers, etc. This should be quicker than trying to get the QTR profile perfect.

- Put a lot of info. in the QTR profile text file itself so self and others know how to use it, and what for. Maybe we could figure out a standard format for this - here's a starter list of what should be in there:

Printer
Negative substrate
Ink-set or subset of inks
Embedded ICC profile of image file
Whether profile prints from positive or inverted image file
Process (i.e. Carbon, Ziatype, Gum, whatever)
Intent (long or short scale, shadow or highlight detail, whatever)
Mixture (contrast agents)
Paper type and preparation
Light source
Other process controls (humidity, whatever)
Post-processing (toning etc.)

What am I forgetting?

Ben

 

[Colin Graham]

I'm curious, has anyone ever successfully used someone else's QTR profile for digital negatives?

I've posted a few curves on various threads, but quickly became breathless with all of the particulars of my workflow, and felt a little silly in the bargain.

Case in point- I recently had to return a 3800 and was surprised that I had to start over when the new printer arrived. I got head banding with the old inkset profile and had to tweak the k boost and dark inks to get around it on the new printer. It seems they are subtle enough variations even within the same printer to make curve sharing ultimately unfriendly, to say the least.

 

[Ron-san]

I'm curious, has anyone ever successfully used someone else's QTR profile for digital negatives?

Colin--

Good question to which I do not know the answer. I give curves away and post them on my web site because lots of folks seem to want them. But I have always assumed that if you are particular about getting good results, any curve I give you is just a decent starting point. Maybe printers vary from one unit to the next. People certainly vary from unit to the next. And I think it is easier and faster to fine tune a profile to your own specifications rather than to try and exactly duplicate someone else's working conditions.

For QTR, fine tuning means fiddling the dark ink limits up and down until they just give you pure white at their darkest. Then redo the gray curve, if needed, to get things linear. Doing a gray curve is no harder and takes no more time than making a regular Photoshop .acv correction curve (well, alright, it does take maybe 30 seconds more fiddling with pencil and paper).

With regard to Sandy's stated problem with getting the midtones correct in QTR -- if your profile gives you a linear response with a step tablet negative -- and I usually do not find it very hard to get to that point -- then problems with the midtones lies somewhere else in your workflow. So seemeth to me.

Or maybe I have just made too many of these things and no longer recognize the problems others have. Sigh.

Cheers, Ron Reeder

 

[Colin Graham]

Thanks Ron, I agree. Glad my question made sense, I've been printing since 5 am this morning, and I'm a little goofy. I've always thought your method quite intuitive and wonderfully efficient once you dive it. I wouldn't give up the control it gives for anything. Being able to 'out-profile' a case of pizza wheels and head banding is truly a wonderful thing! Even though I had to begin anew with the replacement 3800, it was a fairly straightforward affair to get back to where I was before.

I'm reminded I could never get matte black to work for beans, which make me wonder all the more about inconsistencies between printers, workflows and myriad other variables, and if sharing profiles leads to frustration more than success.

 

[sanking]

Colin,

Yes, I have. Quite a long time ago Clay Harmon posted a profile for pure palladium for the Epson 2200. I installed this profile in QTR and have used it for printing with vandyke, palladium and carbon. However, before printing I use Chartthrob to develop an .acv curve for each process, which I then apply to the positive before inverting.

Naturally, every time I make a new batch of carbon tissue I run a test with Charthrob to fine tune results. I don't have to do anything for vandyke or palladium because I always work with the same chemistry, but carbon needs the constant updates because of variation in tissue, most of which are deliberate. In all honesty, this has been one of the most trouble-free methods of printing with digital negatives I have ever used. Sometimes I feel a bit guilty about the method because people tell me it would be better to develop the perfect QTR profile. But damn, I want to make prints, not spend all my time working on profiles.

Sandy King

I'm curious, has anyone ever successfully used someone else's QTR profile for digital negatives?

I've posted a few curves on various threads, but quickly became breathless with all of the particulars of my workflow, and felt a little silly in the bargain.

Case in point- I recently had to return a 3800 and was surprised that I had to start over when the new printer arrived. I got head banding with the old inkset profile and had to tweak the k boost and dark inks to get around it on the new printer. It seems they are subtle enough variations even within the same printer to make curve sharing ultimately unfriendly, to say the least.

 

[sanking]

Just to clarify, when I originally used the pure palladium QTR profile with palladium and vandyke it did not give perfectly linear results in my working conditions so I created very early on .acv curves for vandyke and palladium which I apply to the positive file, as I do in carbon.

Sandy

Naturally, every time I make a new batch of carbon tissue I run a test with Charthrob to fine tune results. I don't have to do anything for vandyke or palladium because I always work with the same chemistry, but carbon needs the constant updates because of variation in tissue, most of which are deliberate.
Sandy King

 

[sanking]

Ron,

My problem is definitely not workflow. I have enough control of the carbon process to make repeat prints of a Stouffer TP-45 step wedge and not have variation of more than log 0.05 in any step.

The problem with adjusting the internal curve in QTR is that if you don't geet it right the first time it is necessary to step back and create an entirely new curve. There is no mechanism for adjusting the internal curve to fine tune results, or is there?

I may be mistaken as to how this work, but my recollection from working with you on several dozen of these profiles is that if the internal QTR profile is not exact you have to throw it away and go back to the original profile without curve to make adjustments.

Sandy

For QTR, fine tuning means fiddling the dark ink limits up and down until they just give you pure white at their darkest. Then redo the gray curve, if needed, to get things linear. Doing a gray curve is no harder and takes no more time than making a regular Photoshop .acv correction curve (well, alright, it does take maybe 30 seconds more fiddling with pencil and paper).

With regard to Sandy's stated problem with getting the midtones correct in QTR -- if your profile gives you a linear response with a step tablet negative -- and I usually do not find it very hard to get to that point -- then problems with the midtones lies somewhere else in your workflow. So seemeth to me.


Cheers, Ron Reeder

 

[Ron-san]

Ron,
The problem with adjusting the internal curve in QTR is that if you don't get it right the first time it is necessary to step back and create an entirely new curve. There is no mechanism for adjusting the internal curve to fine tune results, or is there?

Sandy

Sandy--

You have hit upon the weakest link in the system. But there is a way to tweak the gray curve. One way to insert a gray curve into a QTR profile is to use the number pairs for the gray curve to make a Photoshop .acv curve. Save this .acv curve someplace convenient. Then, grab the icon for the .acv curve and drag it in front of the = sign in the line that reads
GRAY_CURVE=
QTR will automatically insert the address of the .acv curve in that place and it will thereafter function as a gray curve for that particular QTR profile.

The advantage of this method is that you can always call up the relevant .acv curve (since it now exists outside of the QTR profile) and look at it as a curve. With the curve reading in percent ink, pulling the upper part of the curve up lightens the highlights of the positive print. Pulling downward darkens the highlights. Pulling the middle of the curve up lightens the midtones of the print and pulling down darkens the midtones. Same thing for the shadow values.

The drawback of this method is that you have no numerical indication of how much to pull up or down. You will have to try something, insert the altered curve, and make a step wedge print. But it is a tweaking method.

This method is described at the end of my QTR manual V2 that is on my website.

Maybe this will be of help??

Cheers, Ron Reeder

 

[timbo10ca]

I'm overwhelmed. I don't even know what a .acv curve is. Ron, your book is the only one on my xmas list but I think I am going to need alot more! I think I need another year or 2 before I even contemplate this. I hope your book is really dumbed down..... I've only just started getting repeatable results with ziatype so I think I'll take a year to get the process down then look at digi negs again. Hopefully by then you'll have a near-perfect profile for me ;-) Then another year of photoshop and computer programming to figure out what you guys are talking about!

 

[pschwart]

Ron,

My problem is definitely not workflow. I have enough control of the carbon process to make repeat prints of a Stouffer TP-45 step wedge and not have variation of more than log 0.05 in any step.

The problem with adjusting the internal curve in QTR is that if you don't geet it right the first time it is necessary to step back and create an entirely new curve. There is no mechanism for adjusting the internal curve to fine tune results, or is there?

I may be mistaken as to how this work, but my recollection from working with you on several dozen of these profiles is that if the internal QTR profile is not exact you have to throw it away and go back to the original profile without curve to make adjustments.

Sandy

My experience has been that it is impossible to predict exactly how a tweak in the QTR profile would impact output. For me, this results in endless poking and retesting, and not much printing. I also find that it is far easier to tweak using a Photoshop curve.

 

[Colin Graham]

What about simply adjusting the curve within the QTR profile? I find those curve adjustments very predictable and easy to work with, just like an .acv curve in PS. In the curve creation tool GUI, just enter the number pairs in the curve field under the gray curve tab. Then there's no need to make a whole new curve when tweaking. Sorry if I've misunderstood the problem.

 

[pschwart]

What about simply adjusting the curve within the QTR profile? I find those curve adjustments very predictable and easy to work with, just like an .acv curve in PS. In the curve creation tool GUI, just enter the number pairs in the curve field under the gray curve tab. Then there's no need to make a whole new curve when tweaking. Sorry if I've misunderstood the problem.

The gray curve tab in the QTR GUI provides a field for importing a Photoshop acv. As far as I know this field doesn't accept number pairs.

 

[Ben Altman]

It seems to me that adjusting the QTR profile with a Photoshop curve (.acv) for different peoples' machines will work well if the variations between machines are linear or at least smooth. My instinct is that one does not want a Photoshop curve that jiggles around a lot. If the adjustments are to linearize a bumpy output, the detailed ink control that QTR makes possible would be the way to go.

It would be interesting to see how much these machines vary. Sandy and Ron-San have printed QTR calibration sheets for me from their 3800's, for a project I'm working on (slowly). If anyone else would like to print one and send it to me I'll measure that too. PM me for the specs.

Ben