Choosing Colours and Curves: The HSL Array

[Katharine Thayer]

Print Values works for me. You touched on something I think is important; although you've been talking about the hue and saturation of color with regard to choosing a blocking color for negatives, the same qualities apply of course when talking about print values, and it would be helpful, especially for tricolor printers I think, but really for anyone who wants to think about color values and discuss them with others intelligently, to be able to settle on a convention for identifying where those values fall.

I've been thinking a lot the last few months about choosing pigments for tricolor gum printing, while working on my website section on tricolor gum. In thinking about this, I've settled into a habit of identifying the hue of pigments using the color model used by Bruce MacEvoy of handprint.com, which is the CIELAB a-b plane. He has placed all the watercolor pigments on this plane (averaging the hues of all the brands of paint containing a particular pigment) which makes it easy to estimate where a pigment falls in hue, and get a sort-of idea of where it falls on saturation.

The CIELAB model that he uses is different from the HSB/HSL model we're using here. For one thing, the hue values are shifted considerably. The primaries yellow cyan and magenta are at approximately 90, 240 and 0, for example, rather than 60,180 and 300. (For a complete explanation of why the primaries aren't empirically equidistant see the handprint.com site).

I wonder if you have thoughts on the different models and how (if) they are related to each other?

P.S. I printed the HSL array on gum last night and wanted to post the result for your consultation, but my scanner isn't responding properly and will have to go to the shop before I can do that. I tried shooting it with my crummy digital camera, but couldn't get the picture sharp enough to be useful.
Katharine

Print Tonal Range works for me. We could also just say "Print Values". It hard to use one term because of the differences between colour (where we need to discuss hue and saturation) and Black & White (where we just are concerned mostly with brightness).

~m

 

[Katharine Thayer]

P.S. I printed the HSL array on gum last night and wanted to post the result for your consultation, but my scanner isn't responding properly and will have to go to the shop before I can do that. I tried shooting it with my crummy digital camera, but couldn't get the picture sharp enough to be useful.
Katharine

Okay, curiosity trumps presentation, as usual with me; I've decided to go ahead and post this crummy digital shot for your consideration. My digital camera does weird things with pictures, and the print wasn't flat to start with, which made it do even weirder things, so I wouldn't put too much faith in this being an accurate representation of the print overall. But the crucial areas, the edges where color meets white, are pretty accurately represented I think, though blurred by the bad photo. When I used the density patches you sent me before, it was fairly straightforward what patches to pick, because there were only two rows, the red ones, where there was white next to a value. I don't have a clear idea of where to start choosing a color from this, so I guess I haven't learned enough yet from your examples, though I've studied them carefully several times. Help? :--)
Katharine

P.S. I printed this twice with the same emulsion and got the same pattern both times, so I'm reasonably satisfied this is the pattern I'll be getting with this array.

 

[mkochsch]

I would need a better scan to make a full judgment....BUT... I'd say look for any obviously grainy areas that look like they'd print rough and avoid them. Next I'd say the diagonals on either side of the lobe in the middle will probably give nice starting points...the closer to the top the better. Maybe even on the extreme right hand side...hard to tell...You'll probably have to refine/tweak the colour choice as some point similar to what I did.
It a very interesting pattern. What is your light source, time, printer + inks, I'm guessing it's gum not cyano right? Mix?

 

[mkochsch]

I wonder if you have thoughts on the different models and how (if) they are related to each other?

I've thought of modeling colours in Lab too, but I didn't think it would be as intuitive as the HSL (HSB,HSV) wheel. There will be plus and minuses to each model I'm guessing.

~m

 

[Katharine Thayer]

Thanks, Michael, for giving a stab at it in spite of the poor reproduction. I'll take your comments under advisement as I proceed.

Since then I've printed a couple more mixes, since this mix was atypical (the mix had gone sticky and gluey, which sometimes happens with Daniel Smith Premium gum, so I added a little water to make it stirrable and spreadable. It was still much thicker and more viscous than a normal gum mix, but at the same time the color saturation was diluted by the water. This is a mix I ordinarily use for one-coat gums, and it normally prints very dark, almost black. But with the added water, it printed more like what I would use for a multiple gum print. Yes this is gum (I've never printed cyanotype in my life and probably never will) but it's Prussian blue (iron blue) pigment, which I believe is the same compound that comprises a cyanotype print.

The normal mixes (PR 209 and PY 97) printed with less contrast than the watered Prussian mix so the pattern isn't quite so stark, but the basic structure of it is still very much the same, with the downward-thrusting arrowhead shape centered at cyan (180) with verticals at each side of the lobe, and a pronounced horizontal line from 240 or so to 300 or higher. But there's also some lighter tone around the edges of the darker pattern, which does soften the pattern a bit. I'll scan one of them for comparison when I get my scanner fixed.

My light is an EBV-type photoflood. Printer: Epson 1280 with standard Epson dye inks. The exposure, established by separate step wedge, was 5 minutes for the watered Prussian, which is unusually long for my exposures. I thought at first maybe it's that it's drier in my workroom here than at my previous place, but when I unearthed my hygrometer, I found that wasn't the case. But at any rate, the other two mixes needed significantly less exposure.
kt

I would need a better scan to make a full judgment....BUT... I'd say look for any obviously grainy areas that look like they'd print rough and avoid them. Next I'd say the diagonals on either side of the lobe in the middle will probably give nice starting points...the closer to the top the better. Maybe even on the extreme right hand side...hard to tell...You'll probably have to refine/tweak the colour choice as some point similar to what I did.
It a very interesting pattern. What is your light source, time, printer + inks, I'm guessing it's gum not cyano right? Mix?

 

[Katharine Thayer]

Gum and mixes

Sorry, you asked about the mix. My basic mix is always the same: 1 unit pigment/gum: 1 unit saturated ammonium dichromate. I never add water to it, so the mix I posted yesterday, with water added, was way atypical. I was tired; usually when a mix goes sticky and gluey like that, I just throw out the mix, but this time it made me cranky to think I was going to have to throw out all that paint, so I tried adding some water to see if I could get it to the point where I could stir it and brush it. It was too far gone to add gum, although sometimes if a mix has gone a little stiff but not too much so, a little fresh gum can revive it. As I said, I've never seen any other gum behave this way, and I've come to the conclusion that I need to give up on it and start mixing gum from powder, because I'm really tired of this gum doing this, and I haven't been able recently to find a consistent supply of commercially prepared gum that I like and that's readily available to me. The Daniel Smith premium gum, like the girl with the curl in the middle of her forehead, prints nicely when it's good, and that's why I've been sticking to it (in more ways than one) the last year or two in spite of this really annoying fault: when it's bad, it's really really bad. At any rate, the mix did print, although with a higher contrast than the other mixes I printed, but it wasn't by any means a normal mix.

 

[Katharine Thayer]

Adding water to gum/pigment mix

Weird, I seem to be having trouble with the edit function lately; this time, when I entered the reason for editing, it crashed my (Safari) browser.

I just wanted to show how this mix usually prints for me, if anyone is interested, so you can see how much adding water dilutes the color saturation. I added around 10 drops of water for about 20 ml very stiff gum/pigment.

(Warning: another bad digital shot coming up).

 

[mkochsch]

Silver Tests Continued VC 2-and-a-half

I wanted to continue my tests using silver paper this time back away from using the number #5 VC filter and try it with something closer to the middle contrast ranges. I pulled the #2.5 filter out and calibrated a curve to that. The exposure test indicated that I need to add significant more density. I decided to stay with the hue I'd established in the first test (140 degrees, green) but I had to drop the Brightness from 46 down to 20 to get the density (RGB 0,51,17). This is definitely picking up some of the density from the black ink cart. I curved this and noticed that the curve drops faster and harder right from the beginning but the shadow transition is somewhat less dramatic than the first curve. The result was not bad overall, slightly "noisier" than the first and under a 10X loupe is can see some horizonal patterning (but only under the loupe). So I'd conclude that it's probably better to calibrate Black and White to a higher grade (#4 or #5) and get slightly smoother tones from the negative rather than chase after tones in the lower numbers which probably won't look any different once the curve has made its adjustment.
View attachment 114 View attachment 115
Original and Digital Neg Version (right).
~m

 

[mkochsch]

The #2.5 Curve

Slightly lifted "toe" but you get the idea.
View attachment 116

~m

 

[mkochsch]

The colour choice

As I mentioned in the previous posts I decided to stick with the initial colour choice of green (Hue 140 denoted at point "A") but I did have many other choices. I noticed what I thought was "grain" around the 180 to 270 degree areas, which, I thought, if I can avoid that area I will. I used the smart "sharpen" function in PS to exaggerate the image to show me where the grain was the worst. But now back to choosing a colour....the next choice seems like it might be a better one as marked by point "B". H285,S100,B45 nearly the same brightness value as my initial colour with the #5 paper (H140,S79,46B). So stay tuned to see if that curve charts a little smoother.
View attachment 117
~m

 

[Katharine Thayer]

Print size for arrays?

Michael, what size are you printing your HSL arrays?

I really appreciate your continuing to post examples; with each example my little dim light bulb gets a few lumens brighter. Some of the insights you've shared during this thread, I think, are especially important and useful, such as the observation that the light source is more important than the chemistry in determining the pattern printed by a color array. But it's all good. Thanks!

Katharine

 

[mkochsch]

Michael, what size are you printing your HSL arrays?

I really appreciate your continuing to post examples; with each example my little dim light bulb gets a few lumens brighter. Some of the insights you've shared during this thread, I think, are especially important and useful, such as the observation that the light source is more important than the chemistry in determining the pattern printed by a color array. But it's all good. Thanks!

Katharine

I'm changing my page size to 8X10 and printing the HSL at that size since I'm testing silver at the moment. Ends up be about 8" by 4". It might seem big but when you consider it's really 72 step wedges in one it covers a lot of ground. I realize I'm almost blogging to this thread at the moment, that's partly to share the info and partly to record what ground I've covered already. I hope people find it interesting and useful. Judging by the number of views it seems that more people than just those who've posted are reading it. Currently I'm painting my house so I get more house painted on sunny days and more darkroom time on rainy days. Today it's a little of both.

~m

 

[Katharine Thayer]

Thanks, I'm printing mine at 5 x 8.5, not that different.

Keep on bloggin' :--)
kt

 

[themichael]

I'm one of those reading. Almost have my set up complete. A little at a time.

One quick question: Where can I get a 31 step wedge? The photo store by me went from a photo store to a some stuff in a corner to make room for the 1 hour lab. Nobody in the store knew what I was talking about. Major change from last year......

 

[mkochsch]

31-Step Wedge

They're invaluable for establishing exposure times. Maybe a pro photo store would still stock them, but your best bet is just to order two or three from Stouffer Industries. Get the 21 step and the 31 step (T2115 and T3110). The reason for this is if you're like me you eventually put one on wet emulsion by accident and it'll be toast. Another reason is you can lay two on top of each other and record densities over 3.0 logD.

 

[donbga]

They're invaluable for establishing exposure times. Maybe a pro photo store would still stock them, but your best bet is just to order two or three from Stouffer Industries. Get the 21 step and the 31 step (T2115 and T3110). The reason for this is if you're like me you eventually put one on wet emulsion by accident and it'll be toast. Another reason is you can lay two on top of each other and record densities over 3.0 logD.

I presume you mean two of each kind, 21 step or 31 step. If I had to choose just one go for the 31 step tablet.

I also have a 41 step from Stouffer with 0.05 steps. The 0.05 steps are too small to produce distinct steps and more or less becomes a gradient.

Don Bryant

 

[Katharine Thayer]

Yes different printers and inks are going to produce different results. Add to that different light sources. Seems that printing digital negatives runs a close second in its individuality to gum printing in general. It's possible (what you're inferring) that a particular manufacturer puts "a little something" into its yellow ink but not into its cyan ink that inadvertently blocks more UV.

Just to clarify, I wasn't suggesting that at all, I was simply saying that a pigment's opacity to UV light, just like its opacity to visible light, is inherent to the pigment and isn't highly tied to the color of the pigment. I can't speak too much to UV opacity, since I don't have any way to measure that, but I know from experience that opacity to visible light is largely independent of the color of the pigment. Ivory black can print very black, but is transparent even at its very darkest, for example. Lamp black, on the other hand, is very opaque, even printed at charcoal grey. There are transparent blues and opaque blues, transparent reds and opaque reds, and transparent yellows and opaque yellows. Rather than saying that a particular manufacturer puts "a little something" into its yellow ink but not into its cyan ink that inadvertently blocks more UV, as you've inferred above, I would say simply that the yellow pigment used by that manufacturer is more opaque to UV than the cyan pigment used by that manufacturer. I personally suspect that the pigments that are opaque to visible light would tend to be the pigments that are also opaque to UV, but I don't know that; someone with those kinds of measuring tools would have to confirm or disconfirm that hypothesis.

How's that painting going?
Katharine

 

[Katharine Thayer]

New HSL array on gum

I promised that when I got my scanner fixed, I'd scan an HSL array so you could have a better look at it. The earlier one I scanned was printed on an mix that had got thick and tarry and I'd added water to it to make it spreadable, so was not typical of my usual mix. This one is with my standard mix: 1 unit gum/pigment: 1 unit saturated ammonium dichromate. I'd be interested in any thoughts you might have.

Katharine

 

[mkochsch]

So I'm guessing you're going to use a blocking colour of H0 S65 B100 or the equivalent in RGB R255 G89 B89 or there abouts. I'm also predicting a very gentle shallow curve because it's so high above the median.
What's also interesting is you can clearly see that the densest part of the neg, where very little light hit the array, you get a fogging from underexposure with gum. The infamous tonal inversion no doubt?

~m

 

[Katharine Thayer]

So I'm guessing you're going to use a blocking colour of H0 S65 B100 or the equivalent in RGB R255 G89 B89 or there abouts. I'm also predicting a very gentle shallow curve because it's so high above the median.
What's also interesting is you can clearly see that the densest part of the neg, where very little light hit the array, you get a fogging from underexposure with gum. The infamous tonal inversion no doubt?

~m

That (color) seems about right, thanks.

Yes, this looks like tonal inversion but I didn't mention it because I don't have energy to deal with it, and because when I start making observations about tonal inversion, people get really nasty to me, for reasons I have yet to fathom. Not here, but elsewhere. At any rate, this one doesn't fit any of the usually proffered explanations. I would call it stain rather than fog; there's no reason in the world to suppose it is fog, since gum just doesn't fog like that. But it doesn't fit my usual explanation of being stain due to overpigmentation either, since the mix isn't heavily pigmented. Nor does it fit your explanation of "fogging from underexposure" because by the step wedge, if anything this print was overexposed rather than underexposed at 5 minutes, and in fact I let it develop longer than usual to allow the tones to open up. At any rate, since I've already spent quite a lot of time establishing that underexposure doesn't do this by itself, this isn't an explanation that makes any sense to me. If anyone had ever shown that underexposure produces stain or "fog" if you want to call it that, that would be one thing, but I have yet to see actual evidence of this theory. It also doesn't fit the explanation that some others have suggested, that it's a matter of not choosing the right negative color, since it's present across the entire spectrum. So I'm just trying to ignore it for the moment, because I'm too tired to think about it. I'm trying out a new gum, and maybe this gum behaves differently somehow. Who knows... I don't.

 

[dwross2]

Hi Guys:

I've been trying to follow this thread. Y'all are in a higher digital plane of existence than I normally frequent, but it's fascinating. A question: What is tonal inversion? Anything that makes people turn nasty is probably important.

Thanks,
Denise

p.s. Take care of yourself, Katherine.

 

[Katharine Thayer]

Hi Guys:

I've been trying to follow this thread. Y'all are in a higher digital plane of existence than I normally frequent, but it's fascinating. A question: What is tonal inversion? Anything that makes people turn nasty is probably important.

Denise, this made me laugh and go out into the living room and watch some tugboats moving a big ship around (in the river, not in my living room) for a while. I do miss the beach, a bit, but there's always something interesting going on in the river, and I absolutely love my house and my new spacious darkroom, painting space, and matting/framing workshop. It's the best working space I've ever had, and I can't wait to really get back into it.

Tonal inversion is something that happens in gum printing once in a blue moon (I'd been printing gum full time for many years before I ever saw it once). What happens is that on a step wedge, for example, the tones start dark and get lighter and lighter, as you would expect, and then after there is white for a while, then there's tone again. It's kind of like solarization, but it's not really solarization. Everyone has a different opinion about what it is and what causes it; every time I've seen it, it was a result of using a deliberately or inadvertently overpigmented mix, and it was easily resolved by cutting back the pigment/gum ratio. If you're interested in seeing some examples, I have a page about it here:

http://www.pacifier.com/~kthayer/html/tonalinversion.html

P.S. believe me, not everything gum printers turn nasty over is important; there are gum printers that turn nasty over the least important things.
Katharine

 

[mkochsch]

Nah Denise! It's just good 'ol colours and numbers. "Tonal inversion" as it was tagged, is a phenomenon observed in gum printing where the gum starts gaining density when it really shouldn't. I've see it myself on Stouffer step wedges, but I haven't printed this HSL wedge enough to see it there yet. I think Katherine's test is a big piece of the puzzle. I think what she has shown (or proven) is that gum needs some light to react in a proper photographic (read: logarithmically exposed) way. She has a properly exposed print here. The fuzzy shadow area at the bottom is where the black ink cart has kicked in and is strongly blocking light. She's going to choose a colour for blocking in her neg' in the 1.05 logD to 1.35'ish UV density range. Normal for gum. But waaaay down on the bottom those black densities are in the 1.8 to 2.7 or higher range. Useless for gum printing. Almost no light is getting through, but maybe just a little to cause the effect we're seeing. Staining, fogging, inversion. We can debate the word, but the way it has affected the paper is undeniable. Personally, I think staining or fogging is best because inversion would denote that the tones progress in a linear fashion which they don't really appear to do. The dMax never really shows up in end of that area. Like Katharine I have no interest in debating it either. Not many gum printers would want this in a print (although, being gum, someone might ). I think it's footnote material myself. An interesting footnote.

~m

 

[Katharine Thayer]

Michael, I love the way you explain things. There's only one problem with this theory, as I see it; if this were true, (if it's a matter of density that's too great for gum that's causing this) I should get the same effect at the bottom every time I print the HSL array, shouldn't I? And yet I've printed this about ten times, with the same and different pigments, without getting this effect. And besides that, in my years of printing gum, I've printed a great many negatives that were much too dense, theoretically, for gum, without ever seeing this effect. It was only the ones I did yesterday, with the new mix with the new gum, that exhibited this effect.

I said I was too tired to deal with this, but my curiousity got the best of me and I went down and did some experiments. Cutting back on the pigment a little bit eliminated the effect altogether, without attenuating the brilliance of the color saturation, so it still seems to me that it's about pigment, not about density, and that this new gum mixes differently for some reason; less pigment required for the same color saturation. I've noticed that before; once when I was choosing a new gum, I tried a whole lot of samples of different gums, and it was surprising how with the same pigment in the same ratio, the different gums printed very differently; some drab and dull, others brilliant and lively. Just another variable...

Besides, if this density thing were the right explanation, then you should have got this effect in the gum example you posted, shouldn't you? but I don't remember seeing it. (I'm afraid to go back and look for fear of losing what I've written here). Just some thoughts to consider...
Katharine

Nah Denise! It's just good 'ol colours and numbers. "Tonal inversion" as it was tagged, is a phenomenon observed in gum printing where the gum starts gaining density when it really shouldn't. I've see it myself on Stouffer step wedges, but I haven't printed this HSL wedge enough to see it there yet. I think Katherine's test is a big piece of the puzzle. I think what she has shown (or proven) is that gum needs some light to react in a proper photographic (read: logarithmically exposed) way. She has a properly exposed print here. The fuzzy shadow area at the bottom is where the black ink cart has kicked in and is strongly blocking light. She's going to choose a colour for blocking in her neg' in the 1.05 logD to 1.35'ish UV density range. Normal for gum. But waaaay down on the bottom those black densities are in the 1.8 to 2.7 or higher range. Useless for gum printing. Almost no light is getting through, but maybe just a little to cause the effect we're seeing. Staining, fogging, inversion. We can debate the word, but the way it has affected the paper is undeniable. Personally, I think staining or fogging is best because inversion would denote that the tones progress in a linear fashion which they don't really appear to do. The dMax never really shows up in end of that area. Like Katharine I have no interest in debating it either. Not many gum printers would want this in a print (although, being gum, someone might ). I think it's footnote material myself. An interesting footnote.

~m

 

[mkochsch]

Nailed it

I think you've nailed it or are very close. My next post was going to be something to the effect that maybe we're seeing the interaction of two variables. The question is, if it is a pigment load issue, what's the actual mechanism that's causing the stain? Exposure or lack thereof seems to be part of it. Why isn't the gum wiped clean off? Does a longer soak get rid of the stain eventually? Would we, or do we, see the same stain in an unexposed paper treated with that emulsion?
~m