Meanwhile, in the real world…
I usually start with something I drew in the real world. I just think more fluidly on paper. Sometimes I spend more time nailing things down that way; sometimes I don’t.
In this case I pretty obviously didn’t spend a lot of time on the drawing. It’s something I scribbled out in a small book, direct in pen. I scanned it at 300 dpi and played with the levels to get a crisp, workable image.
(In passing, one useful Photoshop trick for scanning is to scan in color, then either go to the ‘channels’ tab to drop out colored pencil underdrawing, or notebook lines, or play with the channel mixer. You can drop out colored pencil behind an ink drawing without bothering to erase. In practice I rarely use this, as I almost never formally ‘ink’ my drawings.)
Once I have an easy-to-read copy of my sketch, I turn it into a 1-bit image. Image->mode->bitmap; use a halftone screen. Why a halftone screen? I like the way it blurs out my lines; I will have to think about what’s going on when I’m making shapes in Illustrator. If I don’t think, I’m bored, and if I get bored I probably won’t finish the piece.
Illustrator.
The bulk of my process, of course, happens in Illustrator. I flip over to it (AI is pretty much always running on my machine, unless I’ve just rebooted) and open up a new document.
The page size usually doesn’t really matter beyond a vague suggestion of the overall composition I want. The color mode does. I prefer to work in CMYK, rather than RGB – in the event of printing something out, I won’t have to worry about losing the luminous colors something drawn with light is capable of. Unfortunately, this also means I can’t use most of the bitmap filters, as they only work in RGB documents. I find that most of them just end up wasting time in fooling with parameters for a half hour, then hitting ‘undo’ and just drawing something myself, so limiting the filters to a handful of blurs and noises is probably a good thing for me.
The very next thing I do is file->place and bring in my tiff. Note that ‘link’ is checked – this means that AI will just store a pointer to the tiff, rather than including the entire tiff in my file. This makes for smaller, faster-loading and -saving files. If the tiff is a part of the final image, I’d probably want to embed it in the file before moving it off of my machine, but that’s an incredibly rare thing for me.
I size the rough to fit the page, and give it a color that I probably won’t be using in the image. Then I set it to ‘difference’, lock off the layer it’s on, and name the layer ‘sketch’. Now I can keep this layer at the top of the stack, so my rough is always visible, without obscuring the shapes beneath too much. I’ll turn it off when I want to see things with no interference. Since it’s ‘difference’, I can see it no matter what’s beneath it – dark, light, it always contrasts.
Why a 1-bit tiff if I’m going to use opacity effects? Well, for one thing, I developed this habit back in AI8, before AI could do transparency. AI treats 1-bit tiffs differently from other included images: black pixels are drawn, while white ones are transparent. And the black pixels take on whatever color you assign the image. If you place a greyscale image, it will show up as an opaque rectangle. Nowadays? If I do this to a greyscale image, the white parts will still interact with stuff underneath if I use ‘difference’ mode. And I can’t tone it. I still usually use a 1-bit tiff. (If I’m being lazy and working from a jpeg I’d scanned for posting in my ‘sketchbook’ section, I’ll turn it into an opacity mask for a colored, difference-mode rectangle so I can still keep it at the top of the stack.)
To be continued…
There will be a few paragraphs about my basic ways of piling up paths here, once I write them. Complete with images. I got caught up in just finishing the piece I started here, instead of documenting its process!