返回资讯中心
外部精选
软件工程
#开发者实时流

DragonRuby – I never wanted to create a game engine

I never wanted to build a game engine. I’m an artist; for me, an engine was just a means to an end. But after four commercial releases, I realized I was losing the only edge I had as an indie: my time and my love of the…

Hacker News 最新amirrajan2026/10/17 · 2026/10/179 分钟阅读

以下正文同步自 Hacker News 最新,版权归原站所有,已转换为易读排版。

I never wanted to build a game engine. I’m an artist; for me, an engine was just a means to an end. But after four commercial releases, I realized I was losing the only edge I had as an indie: my time and my love of the craft. I got approval for the Nintendo Switch, and instead of celebrating, I felt dread. My initial mistake was using the “prescribed” tools for each target platform (the native frameworks the platform holders tell you to use). It’s vendor lock-in. To reach a new audience, I had to do a rewrite, every time. I wasn’t making art anymore; I was a full-time janitor for my own code.

I knew that the success of my first title was synonymous to winning the lottery. I was honest enough with myself to admit it probably wouldn’t happen again. In an industry governed by hits, the only path to sustainability was to maximize exposure by casting the widest net possible from Day One: release to every available platform right out the gate.

I felt like I was going insane talking to other indies about this. They’d just shrug and say, “That’s not possible. This is just how it is.” They were trapped in the Windows + Steam Only Delusion, the idea that you can just “start on PC” and worry about the rest later. They were willfully leaving 60% of their potential revenue on the table because their tools made anything else feel impossible. They were sleepwalking into a future where, if they ever did get lucky, they’d spend the next two years in “porting hell,” trading their best years for platform maintenance.

I spent six months vetting the giants (and at the cost of my sanity, continue to keep a pulse on them). They all had the “cross-platform” checkmark on their feature grid. It was a lie.

To the industry giants, “cross-platform” just means they can spit out a binary for a target. To me, that’s worthless. If I have to spend weeks fixing skewed aspect ratios, broken input modalities, performance issues, and “leaky” OS abstractions, that isn’t a cross-platform engine, it’s a fucking liability.

The big engines have successfully convinced developers that this friction is a “user problem.” They don’t have to innovate because it’s cheaper to invest in marketing that spreads a specific narrative: “Cross-platform is just hard.” By making you believe the struggle is an inherent law of nature, they absolve themselves of the responsibility to build better tools.

I didn’t build DragonRuby to “rescue” the industry. I built it for me. I needed a way to survive as a solo creator without drowning in technical debt every time a new console launched, a new iOS/Android version dropped, or a new chipset architecture emerged.

To make this happen, I partnered with Ryan C. Gordon (a core maintainer of libSDL). We are two sides of the same coin: Ryan knows platform unification better than anyone on the planet, and I know the commercial indie pipeline (from the first line of code to the final storefront link). Together, we built a toolkit that actually respects our time (and by extension, the time of every indie dev, solo dev, and small team that can’t simply throw money at a problem).

It took three years of my life I’ll never get back to solidify the “bones.” DragonRuby exists because I refused to trade my career for a fight with a build pipeline and platform fragmentation. It is a warning to every other indie out there: this is the path that lies ahead of you. Stop and consider the possibility that it doesn’t need to be like this.

正文由 探How 从来源站点 RSS 同步,内容未经改写;遇到排版缺失或需要图片、视频时请以原文为准。

打开原文
去社区讨论 →