7
votes
What programming/technical projects have you been working on?
This is a recurring post to discuss programming or other technical projects that we've been working on. Tell us about one of your recent projects, either at work or personal projects. What's interesting about it? Are you having trouble with anything?
I've been working on my so far unnamed golf game and I'd love it if you could look at a screenshot or two and criticize my graphics!
First take a look and then below I say what I'm struggling with: https://imgur.com/a/EiLzcuI
This was originally supposed to be variation on Links LS 2000 or something like that. Not a copy, but something like "the way old golf games from that era look in our memory". So I wanted to use some modern updates to the same concept - simple textures, 2D billboarded plants.
But now that I have implemented pseudo-3D shading to the 2D trees I feel like I've accidentally made the plants almost realistic, while the ground materials still look like they're from cca 2000. And aside from changing the grass colors, which will be done, I don't really know what to do, because making grass materials semi-realistic is a pretty difficult task that usually involves techniques too advanced for an in-browser javascript game that should run on a phone.
I could just go the other way and make all the textures (including the plants) pixelated, but I'm not sure I want to do that. I really like how the damn trees look now and they're so cheap to render, plus I'm not sure I want to follow the retro trend so literally.
The skybox is a placeholder and so is the HUD and partially the map (it has ugly colors/textures now). Feel free to spit on those, but they will likely not last.
to me it looks good, i see what you mean about the contrast but it also didnt stand out to me when i first saw the screenshots. i dont think the realistic trees are distracting since golf has very clear play areas / obstacles, and the cca 2000 stuff almost serves as a way to keep that separation clear.
like, people already kinda know internally that if something is higher resolution then its prolly a skybox or something scenery, and not a playable object or whatever. and maybe keeping the grass simpler might help add realism elsewhere (like landing impacts / chunks taken out) without needing those features to match the now higher realism
Thanks, yeah, that makes sense. I'm currently exploring how more modern golf games do it, and it seems like they too mostly use simple textures, though sometimes supplemented with things like grass particles.
one thing is you might be able to render the bushier grass slightly above where the actual collider is? so the ball looks a bit visually sunken in/ enveloped.
Yeah, the collision mesh is different from the rendered mesh, so this should be easy and I was already considering it. I'll test it.
I see imgur seems to be broken at the moment, so I reuploaded the screens here: https://postimg.cc/gallery/LJ8hbFg
Hit a big milestone on the main project I'm involved with at work. Four years ago, all the devs were fed up with various aspects of our brittle and hard-to-maintain graphql-based API, so I proposed we create a new REST-based one built on a modern stack and put all new functionality into that one. This week we finally removed the final remaining calls to the old API and waved goodbye to that whole deployment for good. I've never seen so many emojis fly by as during the slack huddle where we announced that.
Recreated a browser game called Empire Attack. Named it Tessera, https://tessera.qot.nz.
https://jason.schwarzenberger.co.nz/posts/2026/qot-games-and-recreating-empire-attack/
I'm also in the middle of trying to make an election simulation game, where you can 'play' as either candidates/parties/election officials.
https://election.qot.nz
Currently it's basically just a results simulator without much of a game to it.
It was not obvious to me that you had to actually have 800 in the bank to build a mine, and you couldn't simply increase the population to 800.
My mother is staying with us again for the winter and I was a bit embarrassed that I couldn't really explain how to use the TV so she would understand it. We inherited a Bose surround sound system from the previous owners and added a Google TV streamer, so there are three remotes. It's complicated. But my wife and I rarely watch TV, so we didn't do anything about it before.
So, I'm on a quest to build a simple, old-fashioned TV user interface that only requires a single remote. I'm calling it MomTV and you can try out the web version here. The UI is inspired by Channel Surfer and cable TV user interfaces in general. On the upside, there are no ads, thanks to YouTube premium. The downside is that it can only play YouTube videos and live feeds that allow embedding. But I think I can find enough content. (Some channels are placeholders and I'll be improving them.)
It turns out that building an Android TV app is not difficult with AI assistance. There is a Tailscale app for Android TV that can be used to set up a private network connection to a Linux VM in the cloud. (In this case, running on exe.dev.) The AI installed the Android command-line tools and walked me through setting it up. I can ask for features and it will build them and reinstall the app on Google TV.
Also, I haven't tried this yet, but it looks like Google TV will support a kiosk mode so I can replace the entire UI with my own app. I was disappointed with Google TV when I first tried it, because the default UI is an ad-ridden monstrosity, unlike the original Chromecast device. But it does seem to be a decent platform to build your own TV user interface.
I did a thing in my old place where I mounted a Raspberry Pi to the wall, it had an IR emitter and it would send the commands to the devices. Rigged a few macros to make some parts easy. Ages ago, wonder if I still have that code....
I'm building out https://d50x.com which is tools to help angel investment fund managers.