Optimization! Plus: what's next?
Arcannon v0.1.0: An unexpected launch!
Pictured: Arcannon developer absolutelyivy, moments before launch.
Hey everyone! It's been so exciting to read all of your feedback over the last week. I didn't expect to get so many players and comments on such an early build of Arcannon. I'm so grateful to everyone that's given it a shot!
One major throughline in the reviews has been performance, so I've been hard at work the last few days putting out patches 0.1.5 through 0.1.15 addressing a variety of performance drawbacks. The game should run a lot better now on most machines, but there's still some pesky issues the Godot profiler hasn't helped me identify yet. For whatever reason, the browser version still runs significantly worse in full screen, so I'd recommend playing it in the embed window or downloading it for now.
Patch Notes
- Changed arc prediction to pre-calculate all angles at round start, instead of re-calculating every time the mouse moves.
- Per carllacan's suggestion on incrementaldb.com, replaced Mote sprites with a simple shader. I couldn't get the boss shader to look right, so it's still a .png for now.
- Simplified enemy data and logic. Since there will be many enemies on screen, I wanted to make sure each node was as simple as possible. There's still room for improvement here, and I may implement object pooling in the future to reduce memory usage further.
- Read Noise values from a .png texture instead of generating a FastNoiseLite instance at runtime.
- Prerender effects/shaders during the loading screen. This is meant to fix the stuttering that happens on first shot / first hit, but it doesn't quite work consistently, maybe because it's rendered off screen.
- Added a sick new background shader. This doesn't make anything faster. It's just awesome!
Road to v0.2.0
So, what's next for Arcannon? We're still at the proof-of-concept stage, so I want to iron out a solid gameplay loop before I put together a Steam page. Here's the problem space I'll be working in, gathered from player feedback:
- The game needs more variety. More enemy types and more skills, especially skills that do more than increase a stat number.
- The player needs a little more agency. Though the game has minimal interaction, players still want to make decisions and understand their consequences, even if those consequences are small.
- Better skill progression. As many players have pointed out, certain skills feel like they are in the wrong place, and can even halt progression if chosen or ignored. Ideally, the tree should branch out in a way that naturally allows a variety of viable builds. A new skill should never make the player's build worse, but some can be more powerful than others.
- Faster pace. Each round should feel noticeably different than the last. This game isn't about difficulty or grinding, it's about a satisfying progression loop, and currently that loop is way too slow.
Thanks for sticking with me, and I'll make sure to post more announcements as I experiment with new features!
Leave a comment
Log in with itch.io to leave a comment.