More Steady Progress - Clickable Art

Hey all, I'm back with another update on work for v0.33!

For the last few days I've been focused on clickable art, particularly the new bedroom clickables that I'm using for the upcoming night time events. Since drawing characters takes me a good chunk of time (even with the rough style I'm currently using) I decided I should improve their bed art as well, rather than have to redraw everything in the future. That involved fixing their perspective, getting the scale right, and giving them some rough shading. I also cleaned up some other clickable objects, like the Nerd's computer, while I was looking at the files for them.

I've just finished the last bit of sleeping art (the Cheerleader; who had the most pose detail AND the most awkward perspective), which means I've now got 100% of the art for v0.33 finished! I've got all of the basic triggers set up for the night time events too, so all that remains is to write the actual dialogue and narration. That always takes longer than I expect, so rather than tempt fate I'm just going to say it'll be done Soon-ish™.

Besides my direct work on AVFH I've spent some time working on art-related stuff. I've long been frustrated with the amount of time I need to spend going from a mostly-finished sketch to clean lineart that I can colour and shade. I went searching for a mechanical solution to produce clear vector lines - which would save me hours of tedious retracing time and let me polish up a lot of my practice drawings so I can share them. I actually experimented with a machine learning method first (more on that another day, when I've got more time to write and make some example images), but the approach that seems to be working actually goes back to my old favourite open source drawing program: GIMP!

The G'MIC library of plugins had some very useful filters that pretty reliably (and faithfully) converted my rough sketch into a cleaner raster image. From there I could take it into Inkscape (also open source!) and vectorize it, but I think the clean raster version will be plenty for what I need it for. Here's an eample:

The final result would still take some light cleanup, but it's much easier to work with than rough lineart from my sketch. The bits that need the most cleanup - the hair, in this example - are also the bits that are the least clear in my original sketch, which feels correct. Hopefully I'll be able to put this all to work in the near future!

Alright, that's enough writing from me, I'm going to go get started with some writing!