For most of the 2000s, playing a game in a browser meant Flash. When Adobe ended support at the end of 2020 and every major browser blocked the plugin outright, tens of thousands of games stopped working overnight, and browser gaming was widely written off as a closed chapter.
What actually happened is that the web had already replaced every part of Flash, piece by piece, over the preceding decade, and the replacements were better. Most people simply never noticed, because the transition happened underneath games that kept working.
What the browser gained
| What a game needs | What replaced the plugin |
|---|---|
| Fast 2D drawing | Canvas and WebGL, running on the GPU |
| Real-time networking | WebSockets and WebRTC, with sub-100ms round trips |
| Sound | The Web Audio API, with sample-accurate scheduling |
| Smooth animation | requestAnimationFrame, synced to the display's refresh rate |
| Fast code | JIT-compiled JavaScript and WebAssembly |
| Working offline | Service workers and local storage |
| Touch and gamepad | Pointer Events and the Gamepad API |
Every one of those is a standard rather than a plugin, which means it works without permission from anybody, ships in every browser on every device, and cannot be switched off by a single vendor deciding to move on. The failure mode that killed Flash cannot happen to the web platform in the same way.
Why this matters most for classic games
There is a real ceiling here and it is worth naming: a browser tab is not going to run a hundred-gigabyte open world at the same fidelity as a native install. That is a genuine limit and it is not going away.
But classic games do not need any of that. Air hockey is two paddles, a puck and a score. Snake is a grid. Ludo is sixteen pieces and a die. These games were designed under constraints far tighter than a modern browser imposes. Snake originally ran on a Nokia 6110 with a 84×48 pixel display. Running them in a browser is not a compromise; it is more headroom than they were built for.
The friction argument
Consider what standing between a person and a game usually looks like: find the store listing, download a few hundred megabytes, wait, open it, create an account, verify an email, accept permissions, sit through a tutorial, find a friend by username. Each of those steps loses people. Multiply the losses and most of the group is gone before the game starts.
A link that opens and plays has one step. That is the entire pitch, and for casual and social play it beats a technically better game reliably, because the better game is the one nobody finished installing.
What no account really means
'No sign-up' is usually presented as convenience, and it is, but the more interesting half is what it means for your data. A game with no accounts has nothing to leak: no email, no password, no purchase history, no profile to sell or lose in a breach.
It also changes what analytics can even do. With no account there is no persistent identity to attach behaviour to, which is why every Tooston game stores its preferences in your own browser and asks before loading anything else at all. The details are in our privacy policy and cookie policy.
How to tell a good browser game from a bad one
- It starts instantly. If there is a loading bar for a game of Snake, something is wrong, usually an ad network rather than the game.
- It works on a phone. Not “has a mobile page”. Touch controls should be designed, not a mouse handler with taps bolted on.
- It does not ask for anything. No account, no notifications prompt, no email before you can play.
- It keeps up. Motion should be smooth on a mid-range phone, not just on a desktop.
- It tells you what it stores. A real cookie banner with a working reject button is a decent proxy for whether the rest of the site was built carefully.
Where this goes
The interesting shift is not technical any more. The platform question is settled. It is that a browser game's distribution is a URL, which means it can be shared into a group chat, linked from a search result, and opened by someone who has never heard of it, on a device you did not anticipate, without asking anyone's permission.
That is a genuinely better model for the kind of game you play for four minutes with a friend. It is the whole reason we build the way we do. You can see the result on the Tooston games page.