Log in

No account? Create an account

Comic creation - John C. Kirk

Dec. 18th, 2006

03:11 am - Comic creation

Previous Entry Share Next Entry

Following up on Friday evening's post (the new version of my comic), here are some more details about the process involved.

As I mentioned ages ago, I now have an A4 Wacom drawing tablet. I actually bought this back in March, but I haven't had time to play with it until recently. The idea is that the tablet comes with a special pen, but it doesn't display any image directly; you have to look at the monitor for that (as you do with a mouse). The main difference between the pen and the mouse is the idea of absolute vs relative positions. With a mouse, it's all relative, so any movement on the mousemat gets reflected on the screen. If I find that my mouse is getting a bit too close to the edge of the mat, I can pick it up and put it back in the middle, and any movement through the air is ignored. By contrast, there's a 1-1 mapping between each point on the tablet and the screen, so if I pick up the pen and put it down somewhere else then the mouse pointer will hop across the screen to match this. I've been using it with the tablet on my lap, so if I find that the pen is getting too close to my body then I have to move the tablet rather than moving the pen.

There's a nice exercise in the tablet's manual: draw a grid (with the mouse), then use the pen to put a purple dot at each grid intersection, a blue cross over each dot, and a red circle around each cross. I enclose before and after images here:
Grid 1Grid 2
If you'd like to try it out, click on the first image to get a big version, save it on your computer, then try editing it with a mouse. I used the pencil tool in Microsoft Paint for this; there are of course more sophisticated options available, but the purpose of this exercise is to position the mouse pointer accurately, not to actually draw neat circles. It's amazing how much more accurate the pen is, even when I was just getting started.

It did take me a few days to get used to the pen, and it was frustrating at first, but it was worth it in the end, and there's no substitute for practice. It reminded me of a couple of similar transitions that I've made in the past. Back when I was at primary school, and I learnt to do joined up handwriting, it was slower than printing letters individually, and in one essay I swapped back to printing half-way through (possibly mid-sentence) which the teacher wasn't very impressed with. Similarly, when I started using the Dvorak keyboard layout ten years ago, my typing speed dropped dramatically; I'm not sure of the exact figures, but I could type at about 60 words per minute on QWERTY, as opposed to about 2 words per minute on Dvorak, and this was very frustrating. However, my top speed on Dvorak is now a lot better than it was on QWERTY: at least 120 wpm.

I've mainly been working from the book The DC Comics Guide to COLORING and LETTERING Comics, and the colorist highly recommended using a Wacom tablet. To quote: "It takes a good three or four days to get used to the process, but once you do, it feels as natural as drawing with a pencil. You'll find that going back to using a mouse will feel like trying to draw with a potato." I'm not intending to do away with my mouse for everyday work (partly because it's convenient to have it on the desk and tuck my tablet out of the way), but I definitely agree that the pen is necessary for precision work.

One other nice benefit of the tablet is that the pen has an eraser on the end. What this means in practice is that I can swap tools without using the toolbox, i.e. I can keep the pencil selected for the main end and just do a quick erase by flipping the pen around. As a related issue, this is an advantage of working digitally; if I do make a mistake, I haven't just ruined hours of work. As well as the book I mentioned, I've been referring to the QC Tutorial Of Doom, and one of his comments particularly stuck in my mind: "Remember- there's no such thing as 'cheating' in digital artwork- you do whatever you can think of to make your drawings look as nice as possible."

As an amusing sidenote, it does feel vaguely Dickensian to keep dipping the digital pen into its "inkwell" every so often. It doesn't need to be recharged or anything like that, but the idea is to avoid damaging the tip, and if you leave the pen lying on the tablet then it will remain "active", so I put it back in the inkwell if I need to type with both hands or look something up in a book.

On the software side, I've been trying out some Adobe applications on a 30 day trial; I'll probably buy a copy of their Creative Suite in January. Specifically, I've been using Photoshop CS2 to do the colouring and Illustrator CS2 to do the lettering (including speech bubbles and sound effects), i.e. Photoshop for the raster stuff and Illustrator for vectors. Both of these applications produce single page documents, so I've also been looking at InDesign CS2 to combine them into a 4 page pdf. (This is partly based on advice from the Making A Comic Book Cover tutorial.) I'm not exactly an expert at any of these applications yet, but I did get an Amazon delivery today (Sunday!) with four of Adobe's "Classroom in a Book" volumes - a general introduction to the Creative Suite, and individual books for Photoshop, Illustrator, and InDesign.

When I drew the original comic pages (in February), I used A4 paper, mainly because I have reams of it for my printer. I've kept the same size for the digital files, although I have been trying to work out the best format for future comics (I have several in the pipeline). Scott McCloud is quite keen on the idea that web comics can be liberated from the restrictions of traditional printed comics, and one of his online Zot! strips takes advantage of this by having a very long panel for a falling scene. However, I'd quite like to see my comics in print one day; realistically, this would be from some form of "print on demand" company (aka "vanity press" in my case) such as Lulu or ComiXpress. I remember one of Piro's editorials at MegaTokyo a while back, where he said that he was having to reformat a lot of his artwork for publication (e.g. enlarging it to stretch to the new borders), although I can't find the details of that now. It would make sense for me to do my comics in the right format from the start, but hopefully A4 is good enough for now, and I can revise that if I come across any more info later on. British comics tend to be taller than US ones, so the comic format at Lulu refers to the standard US paperback size. For instance, the UK Transformers TPBs I have are slightly shorter/wider than A4, but roughly the same size. Then there's the manga "pocket sized" format, and I think that A4 should scale down to something vaguely suitable if necessary. The slight wrinkle here is that original US comic artwork is larger than the printed version, but that shouldn't be a problem if I work digitally and zoom in on it.

On a related note, there's a discussion here about how to create mini-comics (i.e. printing them out yourself and stapling the pages together). The amusing aspect is where one guy has a cunning technique for doing this without having a long arm stapler, and someone else says "but dude...you're telling people to bend staples with their fingers." Right now, that's not an issue for me ("The Gift" is only 4 pages long, so it would be printed out on one double-sided page of A3), but I would favour spending the relatively small amount of money on buying the right tool for the job. Planning ahead, I think I'll go to the UK Web & Mini Comix Thing 2007 on a fact-finding expedition.

I haven't really made any changes to the original artwork, although I did eliminate a few stray lines. The main improvement is that I was able to get rid of the big grey areas in the original version: I converted the scanned image from indexed to greyscale, then to a bitmap (50% threshold for determining whether a given pixel should be black or white), and saved it at 1-bit colour. I also separated the panels with gutters, and made those edges as clean/sharp as I could.

I just used "flat" colours, rather than any gradients/shadows. This is the same way that The Simpsons looks, rather than rendering a scene based on a specific light source. I've also been doing CMYK (Cyan/Magenta/Yellow/blacK) colours rather than RGB (Red/Green/Blue); I gather than CMYK is better for printed work, whereas RGB is better for the web. Then again, I generated the final gif files with a "Save for web..." option, so that may well be replacing them with the closest RGB equivalent.

The easy way to do colouring is to select/fill each region in one click, using something like the "Fill with colour" tool in Paint or the "Magic wand" tool in Photoshop. The problem is that this relies on each region being clearly delimited, i.e. all of the edges have to be closed. For instance, if you look at the bed in the very first panel, the line that separates the duvet from the sheets doesn't quite stretch all the way across, so any automated tool would fill up both regions together. In fact, I wound up with pretty much the entire page being coloured at once! Anyway, the colouring guy from the DC book recommends against the magic wand tool for professional work, and says that the lasso is the best tool to use. (This works much better in Photoshop than in Paint!)

One tip for the lasso: you don't have to return to your starting point. If you finish the line elsewhere, the program will automatically generate a straight line to fill in the gap. I started out by doing a big selection for the easy shape in the middle of each region (e.g. a big rectangle for the duvet) and then doing several small regions to cover the fiddly bits on the edges. However, after practicing for a while, I can now select some regions all in one go. I zoom in a lot for this, normally at 200% view, so that I can make it pixel perfect. I've also been using the pencil to fill in any bits that I missed with the lasso, but that is a lot slower.

For the lettering, I've been using some free fonts from Blambot. Specifically, I've used "Letter-O-Matic" for the speech bubbles, "Blam Blam" for the slamming door sound fx, "Whoopass" for the alarm clock ringing sound fx, "Fat Stack" for the title/ending, and "Mothership" for the creator name/copyright info and the date/time caption at the start. Unfortunately, the "Mothership" font doesn't include the actual copyright symbol ©, and I don't think that (C) is actually legally binding, but it will do for now until I can find something better. Even though it's now computer lettered, everything is still in upper case. That's due to the fonts I'm using, but I believe that the underlying reason for that is to reduce the amount of space required between lines. In lower case, there are letters like "g" and "p" which go further down than other characters, whereas in upper case everything is level.

Interestingly, the file size for the new versions is about the same as the original versions; some are a few kilobytes bigger and others are a few kb smaller, but the extra colour info hasn't bulked out the files, which is nice.

Following up on something I said in February, I like to think that there are some subtleties with the body language here. If you look at the end of page 1, the couple are holding hands, because she is excited about her surprise. When he tells her about the brick on page 2, she pulls back one of her hands, and by panel 3 she's withdrawn both arms.

I think that this comic really does benefit from the colouring, since it's able to elevate the poor artwork. (There is the crude aphorism that "you can't polish a turd", but I'd like to think that doesn't apply here!) There are a few places where the colours are odd, but that's basically due to the underlying drawings. For instance, the two characters have shark/zombie eyes (plain white), but I didn't draw any pupils/irises to give colours to. Similarly, I'd like to colour the backgrounds, rather than having them floating in a void, but at the moment there's no clear distinction between the walls and the floor. I'd also like to have the guy wearing a Crystal Palace replica shirt, rather than a plain red T-shirt, but that means that I need to draw the appropriate lines to separate it into stripes.

On page 1, the thought bubble images (panel 3) came out slightly odd. The flowers were supposed to be roses, although they actually look more like poppies, with the black centre. I think the way to correct that is to redraw them with multiple layers of petals, but that's then an art change rather than a colouring change. So, I hereby declare that poppies are an equally romantic plant, thus solving the problem! As shuripentu mentioned, I missed out the "Milk Tray" logo from the box of chocolates; that was deliberate, because I was hoping that the colour made the illustration clearer, and it's a bit of a cheat to have the product name on the inside of the box, but I will now see about adding that back in. The diamond and pearls should really be clear/translucent rather than white/grey, but that requires more advanced colouring techniques, so hopefully the current method does convey what they're supposed to be.

For page 3, I followed sammoore's advice about the speech bubbles, by moving the first one up to the top/centre, and hopefully that makes it easier to follow the sequence. More generally, all of the tails are now pointing to the right place, which should help. Also, the lettering should now be more legible than my handwritten version (partly because I've bypassed the scanning process).

I intend to redo the entire thing in due course, but I need to improve my fundamental drawing skills first, which will take me a while. I've been reading various books about storytelling methods, so the next ones on my list will focus on the actual art skills. I have The DC Comics Guide to PENCILLING Comics and How to Draw Comics the Marvel Way sitting in my pile, and I might also benefit from a life drawing class. I didn't get on very well with art lessons at school, mainly because my teachers held the view that pictures shouldn't look exactly like the object you're drawing/painting; if you want realism, take a photo, and save art for the impression you get from the object. I can see their point, in that portraits/sculptures are mainly relics from a time before photos. On the other hand, I think that comics are an exception - if you're telling a story that hasn't actually happened then you can't take a photo, so it's good to make the pictures realistic.

More generally, this does remind me of the trailers for the South Park film about 10 years ago. They had the typical dramatic voice-over, showing wireframe 3D models, and talking about using the latest cutting edge technology to create the film, then showed a scene with the low quality animation that everyone knows and loves. The new teaser for the Simpsons movie takes a similar approach (link via sulkyblue). Realistically, I'm never going to be an artist at the same level as Steve Epting (Captain America) or Brent Anderson (Astro City). On the other hand, there are almost 5000 people following the LJ feed for xkcd (a comic drawn with stick figures), plus however many people visit the main website for that comic directly, so I think that a significant number of people will overlook poor art for good writing.

I remember something that Peter David wrote in his "But I Digress" column (from Aug 17, 1990): "writing is something you do because it's impossible not to". I mainly write this blog for myself, and I'm glad to have it as an outlet. Similarly, I have various comic stories that I'm planning to tell, but I'm not intending to change career, and I like the idea of being able to dump the stories out of my brain into permanent storage. One reason that I've been playing with this particular story is that I want to hone my skills before I move onto others. In particular, I'm delaying "A monkey's tale" until I can draw sufficiently detailed hands. (Did you notice the subtle teaser I slipped in there?)

Digressing slightly, back when I chose my A levels at school I was intending to do a Law degree (hah!), so it didn't really matter which subjects I took. I chose to do Maths, Latin, and English Literature; Maths because I was good at it (straight As all the way through) and Latin/English because I enjoyed them. Beyond that, I also wanted to maintain some kind of balance between science and art; nowadays I'd say that I like the ideal of the "Renaissance Man", although I hardly qualify as such. Still, I am trying to maintain those other interests, and in particular I'd like to get back into piano playing. I used to do that regularly at secondary school, and I bought a MIDI keyboard a few years ago, but I haven't used it much; that's mainly because I haven't sorted out the software issues (I think I need some kind of sequencer). I've bought some new books of sheet music recently ("Ultimate TV Themes" and "Avenue Q"), and I'd like to play them. This would just be for my own benefit, or possibly a few friends, rather than any kind of public recitals, but I think it would still be worthwhile.

Anyway, not all of my goals will necessarily reach fruition, but at least I don't feel that I'm stuck in a rut, so all in all life is good.


[User Picture]
Date:December 19th, 2006 12:56 am (UTC)
"I don't think that (C) is actually legally binding" - in the UK you don't need to assert your copyright at all in order for it to be valid - the convention of doing so is merely a reminder and a convenience for the viewer. Besides, it's probably a moot point until you're sufficiently famous for people to be scouring your old livejournal posts for printable souvenirs. ;-)
(Reply) (Thread)
[User Picture]
Date:January 21st, 2007 10:42 pm (UTC)



> Back when I was at primary school, and I learnt to do joined up handwriting, it was >slower than printing letters individually, and in one essay I swapped back to printing >half-way through (possibly mid-sentence) which the teacher wasn't very impressed >with.

If your teacher still lives, find him/her and share this info:

The woes and failures of handwriting instruction come very largely from teachers damnation-bent on equating "good handwriting" with "doing it all joined-up" ... when actually, according to a 1998 study in the JOURNAL OF EDUCATIONAL RESEARCH (citation below) the fastest and most legible handwriters break about half the rules of joined-up writing.
It turns out that the fastest handwriters (and especially the fastest LEGIBLE handwriters) /a/ join only some letters, not all of them — using only the easiest joins, skipping the rest — and /b/ use some joined-up and some printed letter-shapes (where printed and cursive letters seriously "disagree" in shape, highest-speed highest-legibility handwriters tend to go for the printed shape).

Graham, S., Berninger, V., & Weintraub, N. (1998). The relationship between handwriting style and speed and quality. Journal of Educational Research, volume 91, issue number 5, (May/June 1998), pages 290-297.

In other words — joined-up writing comes in, at best, second-best. For more information on the matter (and what to do about it), visit the Handwriting Repair [tm] web-site at http://learn.to/handwrite or http://www.global2000.net/handwritingrepair
(Reply) (Thread)