Color Corrective Masking with Digital Negatives

Color corrective masking has been practicted for decades in relation to color carbon, color carbro, dye-transfer, printing for press and heck, even negative film.

In case you're unfamilliar, the idea is to compensate for the inadequacies of the dyes/pigments used in these processes.

For example, the cyan dye for dye-transfer has "unwanted absorption" of green (IIRC) and the magenta has unwanted absorption of blue. In other words, they deviate from the theoretically perfect dyes that transmit all blue/green and red/blue respectively.

Masking works by superimposing a weaker (less dense) positive from the green separation with the red separation negative (the cyan printer). This effectively makes up for the unwanted absorption by subtracting out green information in the cyan printer, compensating for the unwanted absorption

(it's possible I've gotten some of the specifics mixed up, but that's the general idea. there is no shortage of literature that explains this better than I)

Anywho, I'm wondering if anyone is doing this with digital negatives? Particularly since there are so many people doing tri-color gum, it seems like this is something that should be discussed more, but as far as I can tell it is not. The result is much purer and more realistic rendering of color. It is considered a necessary step in these high-end color processes.

So, what would be the best way to go about doing this? I'm relatively unfamiliar with digital processing, and don't even own photoshop. Heresy, I know, but I think that this technique would be of interest to many people, and I suspect that the effort would be trivial in comparison to the pay off.

I am playing with silver negatives from a lambda , currently making normal negs, highlight negs and shadow negs.
we also are making 4 colour negs and will use blocking masks for various purposes and yes it could include weak positives to block some transmission.
Today I am making a couple of silver gelatin fibres from one of these negatives, and will compare it to an inkjet version.

Lots of fun and experimentation,, nothing to hang on gallery walls,,, yet.

For tri-color gum I use different pigments, pigment concentrations & exposures to achieve a similar end. No positive masks.

I would be reluctant about getting another variable in my tri-color gum.

I would think you could achieve something similar to a mask in photoshop prior to printing the negative.

I have a different opinion from Bob, and from the way holmburgers seems to be thinking, at least when it comes to DT and color carbon. I know that many gummists don't strive for perfect color, so that's a different thing.

In my opinion, once you've made the commitment to use digital intermediaries (as opposed to trying to preserve a "pure" analog workflow) there's no technical reason to rely on positive masks, color control masks, or multiple negs for shadow/mid/highlight. In a properly-made digital negative, everything from paper white to d-max (or max chroma, for color seps) is completely under your control. And any desired aim can be accomplished in a single negative (or set of negatives, in the case of color).

If and when I get around to color carbon printing, I plan on treating it like any other 4-color output "device", using color management tools to calibrate, linearize and profile my output, and to generate separations based on that profile. I can't remember who, but I think someone is doing that with DT. holmburgers probably knows, I think he's spent a lot of time reading about DT.

--Greg

I think my comments are directed to making negs and alt printing possibilities in general, I am thinking more like a silkscreener and building up images.
So it is quite possible in different darkrooms and processes a lot of what I am talking about is unnecessary and overkill. blocking masks or neg/pos are vital for some
workers, second or third exposure negatives for pt pd with over hits can be helped by using blocking masks.
The list goes on in my head but at the end of the day , if you are trying to lay down tone , all you need is what you need and a lot extra work may not be required.

For example, yesterday for the very first time, we made a contact silver print from out negatives being generated out of the lambda using real silver gelatin film. We did not need
to split print this, I was able to dodge and burn to bring out more details, and we used a specific grade, although the negative was originally made for pt pd , we were able to
drop the grade to make a very lovely print, today I get to compare it to the enlarger print made on the same paper.....I will see differences in the image , but the tone, and overall balance will
look like it is in the same family group.
To date we have made lith prints, carbons, pt pd, gum, cyanotype and various over coats using film generated from the lambda , not using Inkjet negatives.
We also have printed these files directly onto fibre paper using the paper in the lambda.
but until yesterday never a contact silver gelatin print.... the results.... just as I thought a very nice silver gelatin fibre print on Ilford Warmtone paper... applications could be endless
for those wanting to make prints in a simple darkroom with contact only, those shooting only digital but wanting silver gelatin prints.. the Azo crowd now have
another method of making negs and for those not wanting to go ULF but still make larger prints this is a excellent option.

Can these negs go in an enlarger,,, I do not think so and am not going to try.
Can these files be outputed on Pictorico and printed silver gelatin,,,, Lith yes,,, solarization yes,,, silver gelatin Ilford Warmtone- not sure but I will give this a go over the next week and report back if
there is any interest.....

Blocking masks are required for the inkjet because the light bleeds through the ink on the film. Secondary masks on detailed highlight negative can be used to bring up highlight detail in a print.

Yes technically one could say that with PS all can be brought out on any file,,,,,,, but for those who have used PS a lot will tell you that you can tailor make a file for the mids, output to film, tailor make a file to reveal shadow detail only with sharpening , output to film, tailor make a file for highlights only , output to film.
And then much like multiple filter printing with contrast filters, create an image on paper that combines the best of all areas of the image, No single negative and laydown can do this in my darkroom, maybe others.

bob carnie said:

Blocking masks are required for the inkjet because the light bleeds through the ink on the film. Secondary masks on detailed highlight negative can be used to bring up highlight detail in a print.

Click to expand...

That hasn't been my experience. The only silver prints I made with digital negatives was while I was trying to get the hang of building digital negatives with something quicker to print & less expensive than carbon or Pt/Pd. But with both silver and with carbon, which can have very long exposure scales (2.4 to 3.0), I haven't had any problem putting down enough ink to get to paper white.

But really, with a Lambda, a processor, and yards of flm, why should you even bother with inkjet negatives?

Thanks for sharing your experiences...

--Greg

I think in a commercial world , I will be able to offer inkjet negs faster and cheaper, keeping the silver film for historical and show work where having negatives in a couple of hundred of years is of value to some people.
I like the fact I can work with both output technoloy's.
Greg how did you find making silver prints from inkjet negs compare to enlarger prints?

I just looked at my output onto Ilford Warmtone from my silver neg and enlarger print, and basically I would not be able to tell which print was from either if they were put in matts.
This was a very pleasant surprise.

gmikol said:

That hasn't been my experience. The only silver prints I made with digital negatives was while I was trying to get the hang of building digital negatives with something quicker to print & less expensive than carbon or Pt/Pd. But with both silver and with carbon, which can have very long exposure scales (2.4 to 3.0), I haven't had any problem putting down enough ink to get to paper white.

But really, with a Lambda, a processor, and yards of flm, why should you even bother with inkjet negatives?

Thanks for sharing your experiences...

--Greg

Click to expand...

Hey guys, a lot of great information and insight, the majority of which I can only read and learn from, not comment on.

But to some of the points raised, what I had in mind was to integrate the mask into the negative and then print it out. I'm definitely not talking about making a negative and a positive and then registering them in the "physical world".

Now Rob, I'm not sure that varying pigment concentration, etc. will have or can have the same effect as color correcive masking. The very important thing here is that there is information in one separation negative that needs to be integrated into the other negative (as a positive). No amount of manipulation on a solitary negative without this information from the other negative can achieve the result.

(And of course I'm talking about color corrective masking, which is different in aim than all these other contrast & density modulating masks)

The proper amount of masking is dependent on the dyes & pigments one uses, and ideally you'd want to do spectral analysis on a step wedge of the colors to see how far they deviate from the theoretically perfect pigments. Perhaps there are some subjective techniques that could get you in the ballpark as well.

I understand that perfect color rendition might not be the goal of a lot of alt printers, but it might be and I think that this kind of thing, done digitally, could be built into one's workflow pretty effortlessly.

Am I explaining color corrective masking well? I might have an example I can dig up....

bob carnie said:

Greg how did you find making silver prints from inkjet negs compare to enlarger prints?

Click to expand...

I can't say I was evaluating the prints critically. It was solely for the purpose of being able to learn digital negative making and optimization techniques with quick turnaround. I also don't have a ton of experience in darkroom B&W printing...I've been a process & scan kind of person.

I hope to do some more serious B&W digital negative printing in the new year.

BTW--What kind of inkjet were you using for your digital negative test?

--Greg

holmburgers said:

But to some of the points raised, what I had in mind was to integrate the mask into the negative and then print it out. I'm definitely not talking about making a negative and a positive and then registering them in the "physical world".
...
The proper amount of masking is dependent on the dyes & pigments one uses, and ideally you'd want to do spectral analysis on a step wedge of the colors to see how far they deviate from the theoretically perfect pigments. Perhaps there are some subjective techniques that could get you in the ballpark as well.
...
Am I explaining color corrective masking well? I might have an example I can dig up....

Click to expand...

Chris, you're explaining color corrective masking correctly, at least to me.

If I understand what you're after, you're looking for a set of digital negatives, that have been adjusted such that when printed "straight", yield accurate color.

One of the points in my original reply is that there is no "color corrective masking" per se. With a properly created ICC profile, the separations generated using that profile have accounted for all the non-linearities in the printing process and all of the spectral characteristics of the pigments/dyes used.

Now, if you don't have access to an ICC-aware application (like Photoshop or others) that is capable of generating separations from a profile, then the process becomes more complex. But even then, I can't picture what the digital equivalent of the type of masking done, e.g. for dye-transfer, would be, short of physical masking positives.

I hope I'm explaining my thinking clearly.

--Greg

pictorico with an Epson 7800

gmikol said:

I can't say I was evaluating the prints critically. It was solely for the purpose of being able to learn digital negative making and optimization techniques with quick turnaround. I also don't have a ton of experience in darkroom B&W printing...I've been a process & scan kind of person.

I hope to do some more serious B&W digital negative printing in the new year.

BTW--What kind of inkjet were you using for your digital negative test?

--Greg

Click to expand...

gmikol said:

If I understand what you're after, you're looking for a set of digital negatives, that have been adjusted such that when printed "straight", yield accurate color.

One of the points in my original reply is that there is no "color corrective masking" per se. With a properly created ICC profile, the separations generated using that profile have accounted for all the non-linearities in the printing process and all of the spectral characteristics of the pigments/dyes used.

Now, if you don't have access to an ICC-aware application (like Photoshop or others) that is capable of generating separations from a profile, then the process becomes more complex. But even then, I can't picture what the digital equivalent of the type of masking done, e.g. for dye-transfer, would be, short of physical masking positives...

Click to expand...

ICC profiles are a brand new concept to me (as in I just googled it). In the simplest incarnation of C.C.-masking for digital negatives I'm thinking along these lines: You will have a color picture in your editing suite and by the click of a button you will make the separation negatives. You then take one of the negatives (the green let's say), invert it to make a weaker positive and then overlay it on the appropriate negative (the red) as a transparent layer and then save it. This is your masked red separation negative; print it out and use it. Any sort of manual registration is going to be cumbersome of course, but is it possible to define points or something and have the program match it all up?

Now, I realize that I'm still thinking analog. I dont' understand these ICC profiles well enough to know what's capable.

Hmm, this is a brain buster for me...

The important aspects as far as I can see are (a), what "filters" are used to make the separations. I assume this might change whether your image is from a digital camera or a film scan, and each one might require its own ICC profile.

(b) How these separations are integrated past the point of separation. What's mixing me up is that I don't think you can effectively correct for the inadequacies of the printing colors in separation alone. So there has to be integration between negatives, that is, information from one separation as defined by the density of a weak positive has to be subtracted from the density of the target negative. Can an ICC profile achieve this?

wheww...

holmburgers said:

ICC profiles are a brand new concept to me (as in I just googled it). In the simplest incarnation of C.C.-masking for digital negatives I'm thinking along these lines: You will have a color picture in your editing suite and by the click of a button you will make the separation negatives. You then take one of the negatives (the green let's say), invert it to make a weaker positive and then overlay it on the appropriate negative (the red) as a transparent layer and then save it. This is your masked red separation negative; print it out and use it. Any sort of manual registration is going to be cumbersome of course, but is it possible to define points or something and have the program match it all up?

Click to expand...

Yes you use what is know as cut marks. There are also other ways of printing alignment marks. It's really a no brainer.

holmburgers said:

Now, I realize that I'm still thinking analog. I dont' understand these ICC profiles well enough to know what's capable.

Hmm, this is a brain buster for me...

The important aspects as far as I can see are (a), what "filters" are used to make the separations. I assume this might change whether your image is from a digital camera or a film scan, and each one might require its own ICC profile.

(b) How these separations are integrated past the point of separation. What's mixing me up is that I don't think you can effectively correct for the inadequacies of the printing colors in separation alone. So there has to be integration between negatives, that is, information from one separation as defined by the density of a weak positive has to be subtracted from the density of the target negative. Can an ICC profile achieve this?

wheww...

Click to expand...

The short answer is, IMO, you are really way off track. ICC profiles are used in this case for soft proofing. That's what Keith Taylor does.

My personal advice is to forget analog techniques completely and learn how to use digital negatives. You can purchase an older version of PS, like CS3 or CS4 for peanuts use it just fine. Anyway you are way off course, IMO (oops sorry I'm repeating myself.)

Even though the processes are different for DT, Color Carbon or 4 color gum the underlying methods for digital negatives will be very similar or essentially the same. Of course, as always there are many ways to skin the cat with digital negatives.

If I could afford Bob's Lambda negatives I would use those but, inkjet negatives with a modern printer ain't chopped liver either as far as quality goes.

Don Bryant said:

The short answer is, IMO, you are really way off track. ICC profiles are used in this case for soft proofing. That's what Keith Taylor does.

My personal advice is to forget analog techniques completely and learn how to use digital negatives. You can purchase an older version of PS, like CS3 or CS4 for peanuts use it just fine. Anyway you are way off course, IMO (oops sorry I'm repeating myself.)

Even though the processes are different for DT, Color Carbon or 4 color gum the underlying methods for digital negatives will be very similar or essentially the same. Of course, as always there are many ways to skin the cat with digital negatives.

Click to expand...

^^ What Don said.

The spectral characteristics of the pigments/dyes used are fundamentally taken into account in the generation of the ICC profile. So that when separations are made (in an ICC-aware environment), the negatives that are generated are for that particular pigment set, with all it's imperfections, as opposed to some idealized CMY set, for which the deviations of the actual CMY from the ideal need to be corrected.



Needless to say, it is a pretty complex process, and one which I have not personally attempted, but hope to some day. IMO, it is the only guaranteed way to get accurate color from a digital negative workflow and multi-color printing.

--Greg

Chris

Think of ICC profiles as the vocabulary of any given printed material.. All inkjets, RA4 prints, tri colour anything have their own specific language.

Therefore with a file with known set of colours and grey patches , printed on each different material to a neutral balance , one can see the failings or abilities of any paper or ink , or material you are trying to work on.

Using a spectrometer and a set of colour values, one can read what your paticular balance is and plot them so that when you want to go back and use that material / inks, your reference will be the ICC profile which gets you into the ballpark very quickly.

Profiles and their use can be very complicated or very simple, I prefer to just find out a simple way of laying down predictable tones on paper .

Bob

I have been working on this a little and so far have decided on the following approach. I would be intersted in others thoughts before I start. I have an i1 and ArgyllCMS can do the profiling and seperating (thanks Greg)
1. Measure C,M,Y,K patches on a colour test chart thing with a spectrophotometer. I intend using the chart that came with my i1, any better ideas?
2. Print with C,M,Y,K tissues on Yupo to achieve similar Lab values. To arrive at exposure and dichromate concentration. The L value is most important as the pigment colour is probably off a little.
3. create linear profiles for C,M,Y,K tissue. I intend using QTR, so will be unable to use icc profiles, instead the profile can be applied directly to the tiff or the QTR curve can be linearised directly.
4. Generate a test wedge with not too many patches (not to make life complicated at the start, they talk about 3000 but maybe 300 or less)
5. seperate into CMYK (important that it is repeatable and need to do some research on this)
6. print digital negatives with previously linearised profiles.
7. make a 4 colour print on Yupo (I hate it but its white and uniform)
8. measure with a spectrophotometer
9. create a 'printer profile' which you can in future apply to a file before seperation into CMYK
10. Print a test wedge to check the efficacy of printer profile.

Interested in your thoughts. This is a lot of work, so need to minimise mistakes.

David

davosproject said:

1. Measure C,M,Y,K patches on a colour test chart thing with a spectrophotometer. I intend using the chart that came with my i1, any better ideas?
2. Print with C,M,Y,K tissues on Yupo to achieve similar Lab values. To arrive at exposure and dichromate concentration. The L value is most important as the pigment colour is probably off a little.
3. create linear profiles for C,M,Y,K tissue. I intend using QTR, so will be unable to use icc profiles, instead the profile can be applied directly to the tiff or the QTR curve can be linearised directly.
4. Generate a test wedge with not too many patches (not to make life complicated at the start, they talk about 3000 but maybe 300 or less)
5. seperate into CMYK (important that it is repeatable and need to do some research on this)
6. print digital negatives with previously linearised profiles.
7. make a 4 colour print on Yupo (I hate it but its white and uniform)
8. measure with a spectrophotometer
9. create a 'printer profile' which you can in future apply to a file before seperation into CMYK
10. Print a test wedge to check the efficacy of printer profile.
David

Click to expand...

David...you've either spent at least as much time thinking about this as I have, or I'm just reeeeallllly good at explaining things.

I'll number my responses to correspond to your points. Keep in mind that these are just my opinions and thoughts on the subject. They could be way wrong. I apologize in advance for a really long post. There are a lot of considerations buried below the surface.


1) Not sure what the purpose of this is. The CMYK pigments you're using will likely have different coloration than those used in (I assume) a ColorChecker SG. It may give you an idea if you're just starting your pigment search, but that's about it. If you already have a pigment set in mind then this is not really necessary, IMO.

2) If you have a reasonable pigment set (like PB15:3, PR122 and any number of yellows like PY3, PY97, PY151/154 or some arylide yellows I don't have the # for in front of me), there's no point in trying to match some other set of LAB values, unless you're trying to use a canned profile made for different pigments. That's not what we're after here. In fact, the L isn't so critical, from my reading. What we're interested in is achieving max chroma (saturation) of our color pigments. That's sqrt(a^2 + b^2) if you're measuring Lab, and C if you're measuring LCH. Achieving maximum chroma means we're likely going to get the largest gamut possible. Use less exposure than to get to max chroma, and we have to use more black to get really dark colors, limiting the gamut. Printing with more exposure might actually result in lower chroma. There are probably some trade-offs and optimizations to be made, but since we're talking about a process for which iterations are expensive, in terms of time, max chroma might be a good target.

3) I agree. Remember, the negative we're outputting from QTR will be B&W, so it's just like linearizing any other digital negative. Not sure whether it's more important to linearize the against the chroma or the L.

4) Again, costs, both in terms of time and material need to be considered. I think 1 sheet would be a good starting point. I can fit 400+ patches on an 8.5x11. If no underlying changes need to be made (exposure times, pigment loads, sensitizer, etc.) after the profile is made, a second sheet with additional patches can be made and read in to improve profile accuracy.

5) If you open a CMYK TIFF in photoshop, and tell it to not color-manage the image, then "split channels" will give you the 4 layers, un-touched. The ArgyllCMS developers are working on a tool to do this, but it's still in very early development.

6) This is the "easy" part, right?

7) Keep in mind, a profile will only be truly valid for a given support material. Therefore, if you move to a natural white paper, or a bright white (OBA heavy) paper, or an ecru paper, there may be subtle color shifts. This is not necessarily a bad thing, but it is something to keep in mind.

8) Oh, wait...this is the easy part.

9) If you've read through the Argyll documentation, or are familiar with CMYK profiling in general, you know that this is not as straight-forward as profiling a pseudo-RGB printer. There is the issue of "black generation", e.g. the manner in which black is used to replace equal-parts-CMY. This relates more to printing presses, where this method reduces "total ink", but may be valuable here, too, to prevent extreme relief in dark areas where 3 or 4 pigments are all being laid down.

10) I would print an image with both synthetic (e.g. step wedges) and real image data. Again, the consideration of how long it takes to make a single print.

If you get around to doing this before I do, then I hope you share your results here, and tell us what you learned that might make it easier for others.


Best of luck--

Greg

Have to dust off my colour printing book again. I believe it is saturation we are linearising though. The single colour prints I made were definately varying in saturation. I have some tissue I poured in the winter so may have a go next week, I dont have any registration equipment yet so we will see what the results are like. I have to make a new indian ink black profile with a better dmax first and then I am bound to go off on a tangent making chocolate and mahogany and aubergine toned tissue.

Just as a matter of interest, here are th Lab values for the process colours. LAB Values of Process Inks for Comercial Printing

Thanks for the explanations guys. It's what I need to obtain some semblance of a starting point for understanding this. In theory I get it, but in practice I don't know the first step.

Don Bryant said:

The short answer is, IMO, you are really way off track. ICC profiles are used in this case for soft proofing. That's what Keith Taylor does.... My personal advice is to forget analog techniques completely and learn how to use digital negatives.

Click to expand...

I'm not going to forget analog techniques. If the two can't be discussed in similar terms, or at least related to one another in a way, then perhaps there's a gap in understanding or something. I suspect the ICC profile could be doing something similar to my analog description of C.C. masking. As far as I can reason, there's no other way to make up for printing color deficiencies when your input is based off 3 additive color separations and the output is 3 subtractive color inks/dyes.

gmikol said:

Needless to say, it is a pretty complex process, and one which I have not personally attempted, but hope to some day. IMO, it is the only guaranteed way to get accurate color from a digital negative workflow and multi-color printing.

Click to expand...

If it's complex and a hindrance, why don't we apply something much more simple to begin with? You make a 10% (or whatever) positive from 1 sep and overlay it on another negative. This is correcting for color and it's easy. Furthermore, it's worked for many decades in the analog trade and rarely do people complain about the color rendering of dye-transfer prints.

Am I being stubborn? Yes, probably... I'm entering unchartered territory and it's frustrating to be off base at every turn. But I also don't see a reason for such a massive chasm in discussions & vocabulary between analog & digital techniques.

Plus, I hate acronyms!

(had to get that out of my system)

Chris

Profiles have been around since the first colour print that I ever made. We used Kodak Shirley's to balance our papers before we analyzed various neg's for printing. Today we still use the Shirley's but in a different way. The 21 step grey scale is king in my operation and I am constantly running them and balancing the lasers to produce a good clean step wedge.

We always have had to balance to a known set of grey scales, then observe how different colours reproduce for each material. In a nutshell that is all I am talking about.

Today we can use a program to read how known colour's react on different papers, balance the grey values, then produce an action that you can put on your screen and with soft proofing see how your colour will change.

I use Adobe 1998 for all my screen viewing and then put the profile of a known paper and ink onto the file and print, the voodoo starts when one try's to explain basically what happens next.
I think Don Hutcheson google Hutchcolour has about the best handle on colour management in North America and has lots of articles that touch on some of the issues you are interested in.

For every process you ever work with, when you place a step wedge in, neutralize it through printing , using various means , you are basically making a profile. What you do with that information, and how you apply it is the difficult part.
My lambda will not operate properly without 21 steps neutralized , which means the Red Green Blue lasers are balanced to give basically a neutral step wedge in a rising scale of density between two aim points I give it. Once all steps are balanced then I can start imaging files that I send to the printer... This is one type of profile... Ron Reeder will create a digital negative profile for inkjet by exposing a step wedge onto paper then reading it and he has created an action that within one or two repeats has a good balanced step wedge on the material he wants to work with. This again is one type of profile... the list goes on and on .

So the more you learn about profiles the better, as you will certainly need the info for some of those crazy ideas you have.


regards

Bob

holmburgers said:

If the two can't be discussed in similar terms, or at least related to one another in a way, then perhaps there's a gap in understanding or something. I suspect the ICC profile could be doing something similar to my analog description of C.C. masking. As far as I can reason, there's no other way to make up for printing color deficiencies when your input is based off 3 additive color separations and the output is 3 subtractive color inks/dyes.

If it's complex and a hindrance, why don't we apply something much more simple to begin with? You make a 10% (or whatever) positive from 1 sep and overlay it on another negative. This is correcting for color and it's easy. Furthermore, it's worked for many decades in the analog trade and rarely do people complain about the color rendering of dye-transfer prints.

Am I being stubborn? Yes, probably... I'm entering unchartered territory and it's frustrating to be off base at every turn. But I also don't see a reason for such a massive chasm in discussions & vocabulary between analog & digital techniques.

Click to expand...

There are a huge number of tools available in the realm of digital imaging / manipulation that are simply not available to the analog user. In the digital world, you have the option to manipulate the data numerically, not just tonally. A lot of analog techniques have equivalents in the digital world, but there are a lot of digital techniques which simply don't translate into the analog world. A lot of the digital color management language comes from commercial 4-color printing and prepress, which had a long history before it collided with the 150+ year history of photography in the form of digital still cameras. There is a gulf in vocabulary, and because the digital side's vocabulary is tied to international standards, we're stuck with it.

If I understand correctly, you (holmburgers) have very little to zero digital photography / image manipulation experience. It can be frustrating. But the great part is DPUG has a great community of people who understand these things who are willing to help.

There is no reason you can't do as you describe...calculate matrix corrections and make masking positives. You might even be able to figure out a way to combine the masking image with the base image in PS so you don't have to print the masking positives separately and have to worry about registering the negative and the mask. My point has always been that since you've already committed to using digital intermediaries, i.e. you're not preserving a completely analog workflow, why not take advantage of all the tools available to you in the digital realm?

Bob did a great job of explaining that the presence of profiles in digital imaging is everywhere, and can be very simple and subtle, to very overt. Heck, even the 40Y 60M you dial into your color head for RA-4 printing can be thought of as a "profile" in some respects...you're changing the image source (albeit in an analog fashion) to account for the response of the image receiver (RA-4 paper, in this case). An ICC profile achieves the same ends as masking and matrix corrections, that is, getting the colors you want on the page, but does it in a very different way. The profile actually knows nothing of the spectral impurities of the dyes. All it knows is that a certain recipe of C,M,Y,K pigments makes a certain color. When you make a profile, you measure somewhere between a couple hundred and a couple thousand of such individual color patches. Making a profile allows the computer to numerically guess at what the recipe should be for a color that falls in-between any of the actual colors that you measured.

Hope that helps things to make sense a bit.

--Greg

Not bad Greg

Ok, you're right guys. I best buck up and put on my learning hat.