Status Update, May 2025

Overview

• Shinmera: released PromptFont v1.9

• Shinmera: released random-noise

• Shinmera: organised the European Lisp Symposium in Zürich

• Shinmera: gave a talk on the SBCL port to the Nintendo Switch

• Shinmera: implemented fishing mechanics in Weiss

• Shinmera: attempt and fail (once again) to implement dual-quaternion skinning

• Shinmera: further work on the SF3 standard and related implementations

• Domolutz: plenty of work on the various animation involved

Some Details

This month was quite turbulent. Among continued health complications from the surgery in April, I also had a whooole bunch of work to do with organising the European Lisp Symposium at my office. I've also been under a whole heaping of stress from those things and the continual pressure resulting from trying to also develop Weiss, and having ever less funds to spend on that.

Anyway, I won't harp on about the more troubling and negative aspects of this month. Let's have a look at some of the actual developments instead! First, I think I'll talk a little bit about the symposium.

Obviously aside from the Monday and Tuesday on which it actually took place, there was a whole bunch of extra effort involved in getting the catering organised, the venue set up, drinks and badges gathered, and so on. Fortunately with some help from friends on the last two days ahead of the conference everything came together nicely without me also ending up dead from exhaustion.

During the actual conference I didn't get to actually chat much with anyone, since I was busy with organisational stuff, and just too drained to actually seek out conversations. But, from what little I did hear, people seemed to generally enjoy things, and especially the lunch and banquet catering was great.

While I don't regret offering to organise ELS, I also don't think I'm going to do it again any time soon. It is so much effort, and while that part isn't really much trouble to me, it is also a whole bunch of pressure and stress, which I cope with far less well.

If you're looking for recordings of the talks, please bear with us for a while longer. They need to be edited and uploaded, and we're all just volunteers doing this stuff in our free time when we have the energy to. The videos will be published in due time.

For what it's worth, the talk Charles and I gave on the Switch port went pretty well. Kandria now also runs at a... semi-stable 30 FPS, with random hitches. That means there's still work left to be done to optimise it, of course, but all the really hard work is done. It's also just that... I don't have much of any patience left for optimising this game, nor any money to spend on anyone else to do it for me. So, yeah, I don't know when I'll be making progress on this again.

Anyway, with all this other stuff going on Weiss took a little bit of a back seat, but there was still significant progress made. Most notably, you can now fish!

We're also still hammering away at the combat system, playing around with things such as hit stop, animation timing, and so on and so forth. It's a whole lotta experimenting, and honestly it still feels far from the way I would like it to be. It's also really hard to pinpoint where the problems lie, though, because there's just so many factors that contribute to the combat feel:

• enemy behaviour

• enemy hitboxes

• player movement

• player hitboxes

• animation timing

• level design

• visual effects

• sound effects

In particular what I am itching to try and tackle soon is the level design aspect. We've been stuck on the same test area for multiple years now and I'm feeling rather depressed about it. Trouble with that is that I still have no good idea on an editing workflow. Everything I've tried with Blender has been a chore in many respects, and every external tool I've tried has either also a chore workflow or does not meet the requirements I have.

Either way, it's a problem that needs to be tackled no matter what, might as well tackle it sooner. Another thing I'm worried about though is that our current level geometry testing method of using convex decomposition on the input geometry is still very brittle and prone to annoying edge cases.

If things don't work well once I have a somewhat significant level put together, I will have to investigate other solutions. The BSP work that I hired someone for early this year unfortunately also turned out to be unworkable for us, and I am very hesitant to try another approach, as it may also just be another dead end and I'd only know after spending weeks of hard work on it.

Anyhow, for now I need to first take care of my health. I've been far too stressed and moody recently, and if I don't figure out a way to correct that I'm concerned about burning out entirely.

So yes, things will be slowing down a bit in June. Maybe I should plan some actual holidays before my Psychology studies start in September, too.

I hope you're all doing well out there, and if you've attended ELS in person: thank you so much, I hope you had a good time.

Thank you all also once again and always for your continued support. Your donations have made a significant impact on my ability to keep working on these things, and for that I am extremely grateful, even if I don't know well how to show it.

— Love, Shinmera ❤️