Wednesday, 28 December 2016

Cerebus Volume One: the Original Artwork: part 3 of 3

Sean Michael Robinson:


Greetings friends,

I hope all of you reading this have had a fulfilling and relaxing holiday season.

As you might be able to suss out from the title of this post, I've  spent the past two weeks posting highlights or (for various reasons) interesting images we're received as part of the Cerebus Art Dragnet, which managed to net more than sixty original art pages for the newly-restored (and currently at the printers) Cerebus Volume One. (Here's more information about the mammoth volume one restoration, and why the original artwork is almost always the best way forward, given enough time to restore the tone.)

This week will be the last look at those pages, after which, I'm hoping, you'll get a chance to see the new Cerebus Volume One for yourselves and see how it looks in person. But for today, we'll focus on the color scans of the original artwork, rather than the restoration itself.

And what better way to start than by taking a look at one of the only original art pages I've seen from issue 12, the one and only Cerebus issue completed with duotone board?

Dave tackles the issue in the notes of Cerebus Archive Number One (now available from your local comic shop). But the short version is, the entire issue was inked on a special paper that could produce two shades of "gray" when exposed to different chemicals that could be applied with a brush. Dave was inspired to take up the technique by Bernie Wrightson's spectacular use of it in some of his horror comics of the seventies. (A similar type of board, with a much more mechanical pattern, was used by Eastman and Laird on their run of Teenage Mutant Ninja Turtles).

Issue 12 didn't reproduce quite as well as Dave had hoped, however, a problem exacerbated by the original negative being replaced by a dupe shot from a printed issue, sometime after the first printing. So the issue itself has had some fascination for me from a reproduction standpoint, and every time I get a chance to see a page of original artwork from it, I'm very curious.

This page came to us courtesy of Scott. Like every page I've seen from this issue, the duotone shading has faded in a pretty extreme way, now having the appearance of a walnut ink wash than a tone that might reproduce in black and white.



You can see that the darker tone still has its intended shape, but the lighter tint has all but disappeared. I did some playing around with the color channels of the scan to see if it was stronger in one channel over another, and sure enough, the yellow channel looked pretty reasonable after some selective sharpening. But still not good enough to reproduce how it was intended to look, the light washes splotchy rather than a continuous tone. (In the end I ended up combining the original art with a print scan of the original page, to great effect. See the book when it's available in January!)







Also of note — the very funny dialogue appears to have been wholly composed in the margins of the page itself, supporting Young Dave's assertion (in the first Swords of Cerebus introduction) that he composes each page one page at a time. 



(Edit: Dave pointed out in the comments below that one could theoretically eliminate the duotone entirely now that it's changed color so much. And he's right! Below is an example of what that looks like, using Photoshop's "Black and White" dialogue controls)




And then there are the rarest pages of all — the pages where a restored print scan looks close to indistinguishable from the original artwork. These are mostly pages where the photography was exceptional, and there was very little tiny detail to pick up — no fleck tone, no spatter, no fine pen or crayon shading or anything else out of the ordinary.

Here's a scan of the splash page from Cerebus #7, which marks the first time in Cerebus that Dave Sim used a particular rendering technique (as noted in the dedication to the original purchaser.) Also of note — the use of white to mark off the foreground from the middle ground, and to clean up the border somewhat (and help finish the "rising sun by implication" negative space inside the logo, a neat bit of design, bringing a third spiral shape to the page).




Here's another original from James G's collection, this time from issue 10. This issue overall has always looked nice in print, probably a lot due to the spaciousness (owing to the snow) and the nice High Society-esque reliance on lots of white, lots of black, and only one or two mid-tones. There's also a lot of nice applied texture in the large flecks of snow (the actual toothbrush work being reserved for the two very nice splash pages) You can see, however, that even on a page like this, largely printer-proof, benefits from going back to he original artwork in the texture of the hair and outfits of the figures, places where some tighter-rendered textures crept in.





Other issue ten pages (like this one scanned by Jason C. at ComicLink — thanks Jason!) benefit in much more obvious ways. Ever seen any of that white hatching before on this panel? Well, you will now!

And with that, I'm out of time for this week! Last week's post had a special request from great cartoonish Scott Yoshinaga, who asked for a more technical explanation of the restoration work. I.e. how it's done — technical details, scanners, the works. 

Well, careful what you wish for Scott! And keep your eye on this space in January :)



11 comments:

Dave Sim said...

Sean - Thanks for this!

It would be interesting to see if you could find any example of this particular duo-shade pattern(s) either in reproduction or original art form and "fly it in" where that's possible to do.

Long term, I mean.

It seems to me that what you're doing with the colour channels is only going to get more sophisticated in the long term. We'll have some means, mechanically, at some future date to say "Anything with this density of black, retain, anything that isn't this black, delete". Of course black is a relative term with the varying density of the India ink but that seems to me part of the "recognition evolution" required.

Once all of the mottled duo-shade has been deleted, you could then create (I'd say RE-create but that would incorrectly imply that it was done right in the first place) the originally intended areas from the light and dark patterns you've found elsewhere.

It's funny to remember how JAZZED I had been about going to New York City in 1979 for Phil Seuling's Comic Art Con largely because I could go to an artstore that had duoshade board and buy enough to do an issue of CEREBUS (and show off my prize to Gene Day -- most of #12 was drawn at his place in Gananoque).

I imagine Gene was thinking, "He's doing it wrong. I don't know the right way to do it, but I'm pretty sure he's doing it wrong." But was too polite to say anything.

Sean R said...

Hey Dave,

If we got the rest of the pages in through the Dragnet, that would indeed be possible! What I did this time was replace segments of the lighter tone with a print scan of the page which had been "de-dot-gained" to open up the tones. But yeah, the process you're describing is doable using Photoshop's "Black and White" dialogue box, where you can grab various color bandwidths and change the amount represented in the resulting grayscale image until the color you're looking to eliminate is mostly gone. I used this on a few originals that had been hand-colored by Gerhard after the fact...

I've added an example of this to the post. :)

Dave Sim said...

Hi Sean! Actually what I was suggesting was deleting the duoshade and then replacing it. As I say, that would require finding an example of what this duoshade tone was supposed to look like (either in a printed version or -- unlikely -- non-faded artwork) and then importing that and putting on top of the de-duoshaded artwork.

I'm suggesting it because we only have the two pages right now, so I think it would be worth financing you reconstructing what the pages were supposed to look like, breaking that up into multiple tasks:

1) Areas you're pretty sure this is what it was supposed to look like

2) Areas you're guessing this is what it was supposed to look like

3) Areas you're suggesting this is what it should look like

and then let me have a look here at AMOC.

Dave Sim said...

Is there enough clear duoshade for you to clone it? That would seem to me to be step one. And then do three different versions of the page according to the guidelines above.

It ain't going to be easy. You're going to need to duplicate the brush effects I got from brushing the developer on and (in many instances) guess which effects you're looking at are intentional and which are just Really Badly Done Duoshade.

As a believer in "signs" and Signs, it seems that the fact that only two of these pages have turned up so far is telling us to figure out what we're (you) are going to do with them in microcosmic detail so we can apply those labour intensive lessons to the pages that come in. And provide a guideline for whoever might be remastering the pages long after we're both dead.

I'll try and stay true to what I THINK my original intentions were. And then, at the end of the day, we'll say "Okay -- original intentions aside -- what would make it a nicer looking duoshade page?"

Sean R said...

Hey Dave,

Sounds good! I was just pointing out that the first step (removing the old shading) is doable now fairly easily. Yeah, the second step would be something else, especially figuring out how the transition areas between the shades should look. I wonder if there's some New Old Stock duoshade of the same brand hanging out there in the ether somewhere, and whether it might be possible to use the chemicals on it now to generate the effect? or if even the unexposed paper is ruined by time and light. That's my first inclination. I can look around and see if I can get an idea of what the effect is supposed to look like, specifically, what the densities might be intended. Can you find the brand name of the duoshade you used on the page that you still have and let me know? Otherwise maybe I can find some really well-produced Wrightson stuff that would give me an idea-- maybe something of his that was printed on slicker stock than typical...

By the way, I saw a bunch of TMNT artwork in person a year or two ago, including some of the pages from the crossover, and they have suffered the exact same fate. I also saw how they were able to put out those FIrst editions with the duoshade eliminated and painted color instead--they'd been keeping at-size photocopies of the artwork prior to putting in the duoshade chemicals! And prior to a few other rendering bits, it looks like.

Sean R said...

The issue with cloning it is finding a large enough piece that I can either use as-is, or "tile" it together. When I needed big chunks of tone for the C + S restoration, I just used the unused tone you sent me, which was perfect — large sheets made it easy. But if there's not a lot of it exposed it makes it a bit harder.

Scott Yoshinaga said...

Yay! Thanks Sean! Can't wait to see some of your process steps in January. I look forward to your weekly posts. It's amazing what you are able to do with all of the originals. You could probably write a whole book about the process and sell it with the amount of experience you have!

Have a safe New Year!

Dave Sim said...

I've got some old Letraset catalogues but, as I say, this was art paper you literally had to go to NYC to find. I'm sure I had a catalogue -- which would have had sampled reproductions of the two shades and what they looked like overlapped -- from that store decades ago but it's long gone. Letraset never had a duo shade analogue. The last time I talked to Kevin Eastman he had bought what he was pretty sure was the last of the existent supply -- and was down to the last few sheets of that.

I think the best bet would be Warren magazines (CREEPY, EERIE and VAMPIRELLA) from the time. Duo-shade was really pretty rare even through the 60s and 70s and 80s. I'll take a look and see if I can find a) a close approximation or the exact pattern b) a story with enough large areas that cloning and tiling it is possible. It's possible that Wrightson's THE LAST MEN from BADTIME STORIES was done with it. It's either that or something very close to it. Wrightson's "Cool Air" was duoshade but I think it was dot duoshade and I'm not sure I have the story. Possibly there are some frontispieces he did in the Warren magazines that would be done in those textures, but most of the ones I'm picturing were duo-shade cross-hatching.

Dave Sim said...

I think you're right that -- whatever condition the developer and art paper were in thirty years ago -- sitting around for 30 years on a shelf somewhere wouldn't do either any favours.

It's really amazing that none of us thought of that: don't use this @#$%: it's got temporary look/drastic downhill erosion written all over it! NO, NO! LOOK AT ME, MA! I'M RUSS HEATH! I'M BERNI WRIGHTSON!
I'M ROY CRANE! What a bunch of maroons! What a bunch of ultra-maroons! :)

Sean R said...

I'll dig around and see what I can find. Maybe check the back of the board for the page you have and see if there's a watermark. What was the name of the store? If I'm lucky I might be able to find something similar to this —

http://www.ebay.com/itm/ca1970-Hammermill-Paper-Company-sample-book-on-PRINTING-with-Duotone-Duo-Tone-/311608998738?hash=item488d57fb52:g:SJAAAOSw3mpXJ1RR

Will keep you abreast of any developments!

Dave Sim said...

Hi Sean - Checked my original printing of Wrightson's BADTIME STORIES under the jeweller's loupe. Bingo! That's the duoshade pattern I used on #12. First story in the book. The story is called "The Last HUNTERS" not "The Last MEN". I recalled that Heritage Auctions had auctioned the originals in recent memory. Which turns out to be May 2013.

You can see even Wrightson was having trouble with the stuff on the first two or three pages although not nearly as much trouble as I was going to have with it. And, of course, Wrightson being a genius, by page 5 he had it nailed.

The first panel on page 5 has a broad area of ground with the dark tone and the last panel on page 5 has a sky done in the light tone.

BADTIME STORIES was printed on glossy stock, so Sandeep should be able to get a good clone-able copy of pg 5 through scanning, but... I WAS wondering what the Heritage scans look like. I'm not "online adept" enough to get past the splash page on their site, let alone enlarge anything. The splash page is a very fuzzy scan.

Happy New Year to you and Rachel and Jasper!