matthewmatosis
patreon
Various Improvements (and an update on art posts)
🕑 Added 2022-04-30 10:28:22 +0000 UTCIt's been a great month for progress, most of which is showcased in this video. The biggest win is a rewrite of the collision system to be 4 times faster, easier to understand/modify and more robust than it ever has been. Other improvements are covered in the video if you'd like to know more.
The only thing I'd like to add which isn't in the video relates to art posts. Last month we revealed the protagonist and a few other tidbits. The reception was great and I'm really looking forward to showing more but it's a bit of a catch-22. As I mentioned before, Betelgeuse differs from the other projects because it was primarily inspired by a world I envisioned rather than some specific gameplay idea I wanted to pursue. As you now know, it's a world of alien insects so I hope whatever story we convey can be done without dialogue. There are many aspects of the world/story which I'd like to reveal naturally through the audiovisual elements rather than just explaining them outright. This makes me hesitant to reveal more art since I'd rather not have to explain all the details before players have a chance to go through the game. All of this is contrasted against my desire to provide value to patrons in the here and now since a fully finished game is still far from guaranteed.
Here's the compromise I've arrived at. When it makes sense to do so, I will occasionally share concept art but I won't explain in detail what any of the art depicts. This way, even if the game doesn't release, you can experience some of that discovery for yourself as we reveal more details about the world just by showing you more stuff. I suppose this means you might form some misconceptions about things but maybe that's fun in its own way. My hope is that this approach will leave enough gaps for there to be some mystery left should you actually get to play the game.
If this is a dealbreaker for you, let me know in the comments. Unless I see a lot of pushback, I'll use this as the art posting policy going forward, it seems like the most balanced approach to me.
Comments
Matthewmatosis
Sorry for the late reply to this. I have a few reasons for preferring not to talk about the collision system at length. Some of them selfish (I only just decided to move on from it and am reluctant to share that work immediately) some of them pragmatic (it would be better to push further with the game so as to give it time to settle before potentially polluting the world with a poor collision system). Maybe I'll feel more comfortable doing it eventually but at the very least I'm convinced it's not a good idea right now. Sorry if that's a disappointment.
Kyrith
Would you consider sharing your precise tile collision code? I'm doing pixel art development in GMS2 as well (top down in my case) and would adore an opportunity to see how you dealt with this famous problem. Even if you don't feel like your solution is 'perfect,' I'd just really like to see more methods for handling this
DominickKimbley
So I know I'm a week behind so you're probably not reading the comments on this post anymore but I'll offer this suggestion anyway since this is how I'm handling the AI predictive system in my game. Keep in mind, mine is built in Unreal Engine 4 and is a 3D game so it's not 100% transferable but it should still work out somewhat. Instead of calculating the predictive system based on 15 separate actors that are all moving more or less in the correct direction, it might be good to just have one actor that is only location data, that is just 15 frames ahead of where you are. You may incorporate one or two additional actors for arcing movements but other than that, just maintain location data and alter it to affect where the game knows you're going to be. This'll be harder for my game which has more fluid movement in terms of momentum but for a Gob-like this should be easy enough for you to clamp down into a series of predictable values. Try it and see what you get.
