I sent the poor boy on a trip to his local color print shop, to ask someone in production (I suggested someone over 40 years of age) how they used to cut Rubylith to mask for colors. Now, I've been in computer production since 1987, but I certainly remember cutting "ruby" for masks and figured the boy could find an old timer who knew his stuff. No such luck!
An internet search on my part also yielded nothing. Apparently Rubylith is still used in silk screening (good for them - we used to joke about the investor who held on to his Rubylith or Zipatone stock, hoping it would rise some day) but I could find nothing written about separations before computers. For the record, I never colored a comic this way, but I did plenty of overlays for newspapers and printers and can guess at the processes.
We at CAD like to be at the cutting edge of written communication, so, as a public service, here it is. The best article on the net on this subject because it is, apparently, the only article on the net on this subject.
Printing before the advent of computers used two processes for separating colors into CMYK (Cyan, Magenta, Yellow and Black). One process was to shoot separations from the photograph or artwork with a stat camera, ending up with four pieces of film that could be stripped together onto a larger project. The other process, used by comic book companies to color their black line drawings, was to use overlays.
The technique was to use three pieces of acetate lined up on top of each other over the artwork page, each representing C, M or Y. Usually this acetate was rubylith, a product still used in screen printing today (to print on material and other substrates). Where the reddish film was cut away from the acetate ink would not print. Where the film was left ON the acetate, the camera negative would leave a blank spot, and ink would print.
If that sounds complex wait, there's more. Separators would then use screens, dot patterns on sticky acetate, to create the screened back colors. I believe the only screens they used were 25 percent, 50 percent and 75 percent. These were then stuck on the three acetate overlays.
The process is why early comic books used a limited color palette. 100 percent cyan and 100 percent magenta is a deep blue; 25 percent cyan is a nice sky color; 100 percent magenta and 100 percent yellow is a bright red; 100 percent cyan and 100 percent yellow is a bright green; and so on.
I like to add 30 percent cyan to my reds to deepen the tone. Imagine the addition to the separator's workload to cut screens by hand for each red before the advent of computers.
As far as I know, there was no way to check their work. Comic companies on a tight budget were probably not willing to shell out the extra bucks for color proofs. The separators' work went right from rubylith overlays to the camera to shoot plates. That is why a plate is occassionally missing from a color in older comics.
Add to this that they were skilled craftsmen, usually working for printers, not comic companies (one notable exception being Marie Severin, who was on the staff at EC comics) and that they were working on deadline each month, churning out many pages a day. I have a great respect for these generally unknown people.
Some artists for independent companies throughout the early years became interested in doing their own overlays, since all kinds of dot patterns were available on acetate, including gradients. These movements were short lived, probably when the artists realized how hard it was.
Much of the coloring in the first 35 years of comic books was crude and rushed, but the same can be said for the writing, pencilling, inking and lettering. That is why, when an early comic in bright primaries is well executed using just the right colors to tell the story, with a limited palette, I am very impressed.
Coloring, Part Two, 2000
I first noticed the change in comic book coloring in the mid-1980s. First, common solid colors were replaced by more complicated color palettes. Then gradients - a soft blending of two or more colors, called "grads" in the industry - starting appearing everywhere. I started using Adobe Photoshop software for a printing company in 1987, so I knew what was happening. Companies were trying to color comics on computers.
Early on, it must have taken forever to color a page. Even coloring at 150 dots per inch and increasing the resolution later to 300 for printing was slow in those days. The patient artist who allowed his work to be pixellated in the early days must have been sweating bullets - it was all so new there were no systems and standards. The black line art was likely to be choppy.
By early 1990 coloring was getting out of hand. Every bump was formed with a grad, every belt buckle had a lens flare and every strand of hair was 3D. In my opinion, it took a while for the initial excitement of what COULD be done gave way to a real assessment of what SHOULD be done. I admire the painterly approach, as long as the basics of painting are applied. Much of that early 1990s stuff ignored (A) where the light was coming from, (B) shaping the figure instead of hiding it and (C) separating foreground from background.
Today (12/00) there are a variety of artistic approaches. In the new Crossgen Comics line, artists, inkers and colorists work side by side for a unified approach and, although complicated, the coloring helps the artwork instead of hiding it. Many Vertigo comics (DC) use a variety of greyed colors, difficult to code and proof, but interesting after a printing press lays down the inks. Solid colors without grads are making a comeback. I personally love to see just the right, bright solids working together on a page.
Comic book coloring has been an obscure, but interesting history to follow over the decades, combining technology with artistry and a hefty dose of comic fan opinion. Dave McKean has been a hit with his photo collage covers for over a decade and some artists have experimented with photo interiors. Can comic interiors using all these techniques together be far behind?
Just When You Thought You Knew It All Department - I'm consistently amazed by the little details I can forget to address at my advanced age. By email, Curtis points out that coloring with rubylith is even more complicated than I detail below, due to SCREEN ANGLES.
Whenever one screen of a different percentage is printed on top of another, the screens must be angled to avoid moire patterns, which are psychedelic looking swirls. Computers figure these angles automatically (did you ever notice, in your printer's preferences, choices for percentages in your CMYK output?) but someone cutting by hand would have to have some other way to angle the screens, either by eyeing it (seems impossible to me) or by shooting the solid ruby in a stat camera to create screens (seems inordinately expensive and complex - figure CMYK plus 25% CMYK plus 50% CMYK plus 75% CMYK and you have 16 overlays).
"I've actually dropped colors in by hand like this (I use amberlith)," wrote Curtis. "and it gets very complicated. I really can't see how anyone ever even thought of doing it like that. I'm not an old timer, just a young graphics afficionado who has been taught the old fashioned ways.
"On a side note I believe different approached were tried in the seventies with separate overlays done for the original art and using the process camera to shoot them individually and then burning a contact neg from the separate percentages and grads - but I'm not sure on that one," wrote Curtis. "I know things like that have been used in printing in general though."
Thanks for your input, Curtis. This will give me one more thing to track down with the upcoming comic convention season on the way. Anyone who has seen an old timer doing a complex skill by hand knows they can move pretty fast. Maybe it WAS all done by hand! I'd end up cutting off a finger or two with the Exacto blade.
Coloring, Part Three, 2005
Khouri wrote in to clarify the coloring processes discussed in "Coloring Comic Books Before Computer," parts one and two of which have been in our article section (above) for some time.
His father was, and still is, in the prepress business, which explains him being involved with the Kaleidescope software.
In the late 80's I wrote software to run on a PC for doing vector based spot color. The primary reason was for Sunday newspaper inserts like for your local supermarket. Some comic illustrators got wind of it and immediately became our most prolific users.
It was a bit like Illustrator, except Illustrator wasn't around yet for me to copy. The output was primarily to PostScript film output on a Linotype machine although I also wrote a rasterizer for color thermal printer proofs. It was called Kaleidescope, but we sold the software and I lost track of what happened to it after a few years. I know that Image comics made huge use of it.
Anyway, I can add some more history to your article. The coloring process was explained to me by people at Marvel when we were recoloring some classic comics for hard cover books they were producing.
The 11 x 17 inked and lettered final art board was photocopied 66 percent to final size. The colorist specs out what colors go where by penciling in codes like R50Y50 for 50 percent yellow and 50 percent magenta. They used "red" and "blue" instead of "magenta" and "cyan."
These marked up photocopies were sent out to work-at-home moms with a stack of pin-register punched acetate (not the more expensive rubylith) and pot of "opaque" (a water soluble paint that comes as a solid like those water color kit books I get for my kids now).
These people were paid by the page, so they could go as fast or slow as they wanted, as long as the work was correct. Most were probably paid per hour below what would be minimum wage.
They would use the opaque to paint in the areas on the acetate that were to get the color. They made one acetate each for Y25, Y50, Y, R25, R50, R, B25, B50 and B. Back at the shop, they make contact negatives in a vacuum frame and at the same time put together 16 pages for a full printing plate. Now they have three masks per color, one each for 25 percent, 50 percent and 100 percent. Then they make the plates, exposing each color plate three times, using the 25 percent with a 25 percent screen, etc.
Khouri: Thank you very much for writing, that clears up a lot. I've cut Ruby, but never went through any process this elaborate.
This also explains why some of the coloring was obviously made with a brush, instead of the crisp lines of a knife. You can tell who was on a crushing deadline by how many brush strokes (and brush stroke mistakes) are visible in certain '60s comics.
It's amazing that this complicated process was less expensive than simply painting color on a single overlay and shooting the color plates from that! Labor, it appears, was cheaper than using the available technology.
And, now that you mention it, I think I remember hearing about Kaleidescope.
Khouri wrote back:
Also, the opaqued acetate process is a lot more tolerant of variations in the process like how old the developer chemicals are. At that time, to make a color separation of a painting required three exposures (a fourth for a black, but we'll ignore that) through a halftone screen. The exposure time and processing of that film is critical. The halftone screen is a large film with fuzzy continuous tone dots on it and is meant to be in contact with the exposed film. They are expensive and if one gets a scratch, that scratch will show up in everything that is halftoned with it, so the halftone screen then becomes garbage.
All that analog stuff makes me shudder.
So, with the process they were using, even the skilled labor could be less skilled and cheaper. They could change their film processing chemicals less frequently, reducing costs there. All this stuff adds up.
Thanks for writing in, Khouri! Call me a techy geek, but I love this stuff. I look on it as an unsung skill in the history of comics.