Blog: How the comic is made

In this blog, I'll share the process of making a comic page for Leaving the Cradle!

I work with batches of pages, currently trying to do 10 at a time, and each of the described steps is done across the entire 10 pages sequentially. Working on several pages at once turned out to be a very good idea, since it somehow makes me finish all of them faster than making page by page in sequential order. I don't know why, really, but when I started doing batches, for the first time I started getting free time to do other stuff besides just the comic, though the feeling "oh, I have three months of free time before the buffer runs out! Woohoo!" is quite deceptive, since I usually need to spend ~2.5 months on making the next batch...

To track my progress, I use a Google Doc spreadsheet that I've downloaded from a fellow comic creator and adapted for my use.

...Though, I frequently forget to update it ^^"

The page starts with the script.

My scripts look like this (note that I don't write script in English, so this is translated) :

-----266---------------------------

ZANE AND THE PRESIDENTS ENTER THE CONFERENCE ROOM

ZANE

• Now that we won't be disturbed, I'd like to discuss something with all of you.

• It is in our mutual interest that the upcoming conversation not be mentioned outside this room.

ZANE BRINGS AN IMAGE OF A GRAY ON THE SCREEN

ZANE

• Have you ever come into contact with such creatures?

US PRESIDENT

• Is this some kind of joke?

• No.

ZANE DISPLAYS IMAGES OF RAHARRIAN AND OTHER GREYS ON THE SCREEN

US PRESIDENT

• What the...

ZANE

• Stories are emerging in the Alliance about abductions by strange beings for unclear purposes.

• These are the images of these creatures, recreated from eyewitness accounts.

• We've dubbed them "Ghosts".

-----267---------------------------

I tried to write at least a chapter ahead of the latest released page, but most of the time the script is only about one or two batches (or ~20 pages) ahead of the release.

When I have finished the script for the scene, I start thumbnailing approximate layouts and frame compositions for the page.

I printed comic templates on sketchbook pages. The small size of both the sketchbook and the individual templates in it means that I can't detail the content of the panels even if I try - which is perfect for thumbnailing.

I erase a lot during this phase, and usually the hardest part of this stage is to understand how to cut a section of the script into frames properly. Sometimes I'm unable to, and I need to rewrite a part of the script to thin it out, or the opposite, add more panels than I have anticipated.

After the thumbs for the entire batch of pages is done, I scan and bring them to Photoshop, where I then use them to make a clean panel layout.

If there are any new locations the action takes place in, I start making the necessary 3d models for that. Sometimes that takes quite a lot of time, as the locations can be big. On rare occasions, if the new location is a one-off background for just a couple of panels, I consider just drawing it, as it might be faster.

I try to reuse stuff I've already modeled whenever I can to save time - for instance, did you notice that the room Gharr was sentenced in is basically the same as Shuru's office, only elongated and with a higher ceiling?

I make two versions of the location, one for renders and one for dummy posing, that I dub "movie sets", from which I strip down anything except essential geometry and movable objects (like chairs and doors), and paint the whole thing white so that my drawing dummies would be visible on it.

Next, I import text from the script into the comic page, and start setting cameras and posing my drawing helper dummies on the set of the scene. I use screenshots from Photoshop as a background to remind myself of what approximate composition I had in mind in thumbs, and what's the current position of words in the panel. Comic does indeed use "set logic" when it comes to points of view, since very often for a good composition, cameras end up being out of the room, inside the walls - on a real movie set, there would be no wall.

At this phase, I usually tweak the wording of the text so that it would flow better as a bubble, or fit on the panel properly. I use a wide "stroke" effect on the text layers, which helps me to estimate approximately how big the resulting bubbles will be.

My drawing dummies are made by me, I did not use anything premade. Initially, they were just boxes for proportions, and I had to make two sets of sketches - first with raw anatomy, and then draw clothing over it. With the dummies getting more detailed and my skill increasing, that step has become redundant.

(On this particular page, I've forgot that China's and Protagonistan's presidents shoud've been switched places)

Cameras I've set up will be used to render out backgrounds. I've made a couple of scripts for Photoshop and for 3ds Max to help me with transferring panel properties from Photoshop (Size and position on the page), and for managing the rendering of the backgrounds themselves.

The background render script started out very humble; it was basically just setting render size based on the camera's name, and preparing the scene for render by hiding layers and toggling render options depending on what pass I was rendering (initially, there were just two - colors and lineart). It was 200 lines of code long. However rudimentary it was, it helped me speed up background renders significantly.

The version I use in 2025 is a tiny bit more complex now, with a total of 2600 lines of code across 3 different scripts...

It manages dummy posing minutia, exports properly sized screenshots back to Photoshop (before that, I had to screencap the program and resize it in Photoshop manually, trying to match frame proportions), records animation on the scene objects so that they would be in proper places in a proper frame (Before that, I could keep animations of a set only for a specific page, and had to erase them when I needed a new one), I can pick which cameras I want to render, and the complexity of switching the render modes have greatly increased as well, so much that I now actually need a memo to remember what the script is looking for.

...After the render is done, I use another script that almost automates importing the heap of renders (6 to 14 images for each panel) in Photoshop, setting them up as panel background folders, and distributing the panels on the page. That one script is 1800 lines long now, and originated as a cannibalization of Photoshop's stock script for making panoramas (since it was one that had a nice UI window that allowed for selecting several files for processing).

I cannot stress far enough, just HOW MUCH time the scripts save me (I have a couple more that are less interesting and much more technical in nature, like one that slices text copypasted from the comic script into separate phrase layers). Doing background renders used to be a big task that could take an entire week; nowadays, I almost don't think about it. Knowing how to code (Like, how it operates, not any specific language) really is an essential part of using a computer.

However, there are downsides, too. I became so reliant on these scripts that I can't change my pipeline now, not even the software version, out of fear that the update will break one of my scripts. I've thought about moving to Blender and Krita away from Max and Photoshop, but that would basically destroy my ability to make pages on the current schedule, and return me to the dreaded times of 2018, when the "schedule" was "~2 pages a month, if you're lucky".

So I think I'll wait until starting a fresh project before changing my workflow.

After the backgrounds are taken care of, I proceed with sketching.

I once heard somewhere that blue color helps with sketches, since this way it is harder to see line imperfections and random smudges, so you have less urge to refine your sketches. I think it helps, but for me - just a little, my sketches still end up relatively defined and crisp.

Other than making new models for background, this is the most time-consuming part of the process; I consider myself lucky if I manage to sketch one page in under 6-8 hours.

While drawing, I keep a PureRef reference board with all my characters and other useful things on a separate screen, to ensure I won't veer off-model or forget a detail that should be there.

After the sketch is done, I proceed with drawing clean lineart over it (also known as "inking").

I'll shill for a little program called "Lazy Nezumi Pro". It's a tool that can stabilize pen movement and pressure, and it has loads of useful modes, scripts and rulers, like a perspective tool. Plus, it works with any app, so I can have unified stroke processing anywhere I draw. It's not free, but it is the best $30 I've ever spent. It turned making lineart from a PITA to a relaxed and fast process where I don't need to redraw every single line 50 times before I get it right, and I use it basically almost always when I need to use tablet, no matter if for comic or for work.

After the lineart is done, I create a color folder and start filling flats, using lineart layer as borders. I import a palette file of the characters and keep those palettes close to where I will need them, to speed up the process.

After the flats are done, it's a home stretch. I paint the shading in four layers (harsh shadows, soft shadows, skin rimlight, and highlights), create a mask for where characters should be occluded by the background, and add final bells and whistles like effects, if the page has any.

As a final touch, I make bubbles for the phrases I laid out in the very beginning and add sound effects. Usually, at this point nothing else needs to be done, but sometimes I have to make quick last time edits because bubble covers more than I expected. I use same stroke effect as before to ensure bubble edges do not get too close to the words.

After this, I make a translated version of the texts, make a thumbnail for the update notification icons in RSS and on social media, re-examine the pages several times to see if I have missed anything, and it's fully done, upload it to the buffer!

Time to sit back and relax start writing script for the next portion of the comic...