Posted March 12, 2023 by Anime Garlicoin
Game development is an evening and weekend hobby for me. As such, my progress is not only non-linear, but it is extremely inconsistent. Typical sessions will be along the lines of one of the following, in order of likelihood:
I am still driven by an unceasing desire to make something cool and interesting, and I have several pages of typed notes and ideas that I want to implement. I still daydream about new mechanics/NPCs during my commute and at work, and I still research new ways to improve things or approach development. Like anyone with minor ADHD, though, sometimes ideas get lost or I completely underestimate the complexity required to achieve my goal.
I am still stuck trying to keep this game at under 1-GB in order to release on this platform. Every few weeks, I will package the game to test outside of the Editor or on a less-capable machine, and discover that I have to yet again trim back texture sizes or simplify/combine meshes in order to reduce the size of the released game package. Luckily, removing a particular tree variety and replacing all of its instances with another one is a very easy trick I have the ability to keep doing for a while. Unfortunately, I keep setting more and more meshes to use Nanite, and forget to remember that the slight performance increase it gives comes at the cost of higher VRAM and disk
Performance is another important aspect, and is something I have been specifically working on improving this weekend. It may be that users of 1000-series NVidia GPUs will not be able to expect more than 30-FPS; I don't know for sure this point. But I'm doing my best to keep the VRAM requirements and processing power at a minimum. With my current machine, I am definitely CPU-bottlenecked and have to spend the rest of today trying to figure out what assets are in particular causing the problem. I have some good ideas as to what the culprit(s) are, but I have convinced myself to figure out how to use Unreal Insights to tell me EXACTLY what it is first, and my first attempts at this have failed so far. Always something new to learn, so enough writing and time to go get that bread!