Wednesday, 19 August 2015

Church & State I: A Pox On Dotville

Sean Michael Robinson:

Howdy everyone!

This will be a brief one, as I'm still recovering from last week's marathon work session to finish the newly restored Church & State I. Rest assured, plans are proceeding for Church & State II as I write this, and we should have a schedule for that book firmed up by a week from now. I'll be using the next few weeks of posts to demonstrate the work we did on the book and give you an idea of how things turned out.

In the meanwhile, this would be a great time to pay a visit to your local comic store and place an order for Church & State I! 

I thought this week I'd give you all a peek into how I dealt with one of the most challenging and aggravating problems facing C&S I. Issues 66, 67, and 70-73 all seem to have been photographed with a poorly calibrated stat camera, causing all manner of grit and pencil schmutz to be picked up in the photography along with the ink itself. In the pages where we had access to the original art, this wasn't a problem at all, but the majority of these pages were sourced from the negatives. 

The top tier of page 429, as it appears in my tenth printing of Church & State. Notice the "grimy" looking tone, which in closeup looks like areas with swollen, bulging dots. 

This is a problem where I probably should have just accepted this is how it is. This is how the pages were photographed, and short of retoning all the pages there was nothing to do about it, so how about I just relax and accept it?

Fortunately for my neurotic nature, I didn't have to accept it the way it was-- there was a solution available to me.

Here's a screenshot of a scan of the original negative. As you can see, although there's no dot gain unlike the printed example above, the problem is still there.

So I open up the Levels command and take a look.

By moving the midpoint very far to the left, and moving the black point up just a bit, we've now driven a visible wedge between the noise (non-desireable elements) and the signal (the actual ink on paper.) This is enough to then use sharpening that targets only the signal, and then use further levels adjustments to "knock out" the remainder of the noise.

Of course, it also leaves the rest of the page overexposed, too light. Look at how far away those halftone dots are from each other! Surely there would be a way to return the page to its original value?

There was indeed ... but I'm afraid that'll have to wait for another day.

Perfect? No, but it's a big improvement, and short of scans of the original art being sent to us, it's the best way forward I've found yet for these problem pages.

Of course, now that I've mostly solved it, this is a problem that disappears for a good decade or so of the book. Flipping forward, I think it affects a good 60 page chunk of Minds, and then that might be it for the remainder of the series. I'd like to think there was a purpose to it anyway, though, even if it only affects a hundred or two hundred pages or so total, even if only to indulge my need for constant improvement.

Next week-- more Mara! Less me!! Less tech talk!!!


Jeff Seiler said...

Thank you, thank you, thank uo, Sean and Mara!

Barry Deutsch said...

Wow! That looks MUCH better! It's really impressive.