What I think S&box is missing

I recently watched a video discussing S&box and why it still seems to struggle with player interest. I started writing a comment under it, but it became way too long for YouTube, so I figured I'd put the thoughts here instead.

I bought Garry's Mod in 2013 and it was basically the game I grew up playing. I spent most of my pre-teen and teenage years on it, not just playing but also creating servers, configuring them, making addons and generally messing around with the game. I've also followed S&box since it was first announced around 2017.

Some of this might be obvious, already implemented, or heavily discussed before. This is mostly just a brain dump of what came to mind while watching the video and thinking about what made Garry's Mod work so well for me.

I don't think S&box's biggest problem is simply a lack of good games. I think part of the issue is that it's missing some of the structure that made Garry's Mod so easy for normal players to build on top of.

In Garry's Mod, most players were concentrated around a relatively small number of gamemodes like Sandbox, DarkRP, TTT, Deathrun and Prop Hunt. What made it interesting was that servers could take the same gamemode and turn it into completely different experiences using different weapons, jobs, vehicles, progression, maps, rules, HUDs and other addons.

The server owner didn't need to build all of that themselves. You could install existing content, configure it, mix it together and put your own twist on the gamemode.

Those experiences also don't need to be anywhere near standalone-game quality. Some of the best GMod servers were pretty rough technically. They were fun because of the rules, addons, progression, objectives and communities around them.

I think S&box needs more of that. There should be strong, reusable systems for things like weapons, vehicles, inventories, HUDs and admin tools that are easy to combine. Creating a completely new gamemode should still be possible, but it probably shouldn't be the default starting point. Most people will likely create better experiences by starting with something solid and building on top of it.

I also think creating content for those systems should be much easier. If someone wants to make a HUD, weapon pack, inventory UI or similar addon, there should be clear and standardized ways for it to integrate with existing gamemodes and systems. Ideally, you make a good addon once and it can be used across lots of different experiences, rather than having to build it specifically for one game. Whether that comes from standardizing more of the underlying engine interfaces, or from encouraging creators to build on top of standardized foundations like Sandbox, the important part is that addons should be able to plug into lots of different experiences without being rewritten for each one.

One thing Garry's Mod did really well was keeping the gap between having an idea and actually trying it very small. You could follow a YouTube tutorial, download some addons, edit a few files and immediately see the result in-game.

The S&box editor is much more powerful, but that can also make it intimidating. Ideally, someone should be able to take an existing foundation, add Workshop content, configure some rules and objectives and make something fun without touching code at all.

We're also in a different time now because of AI. Maybe people today are more willing to create things from scratch because AI makes programming easier, but that can also contribute to the low-quality AI slop people complain about.

I think AI could actually help solve that if the editor used it more as a guide. An integrated coding agent could default to encouraging people to build on existing gamemodes, systems and content instead of immediately reinventing everything. Maybe when starting a project, the editor asks a few simple questions about what you're trying to make and adapts itself around that.

Someone making their own server could get help finding and combining existing content, while someone specifically trying to create a new gamemode could choose a more open and explorative setup.

Whenever I made a Garry's Mod server, I wanted people to actually play it. So I advertised it, invited friends and tried to build a community. Those friends could then bring other friends.

If thousands of people can easily create their own TTT, DarkRP or Deathrun server, you also have thousands of people with a reason to bring others into the game. Lowering the barrier to creation doesn't just create more content. It creates more people actively trying to attract players.

Garry's Mod had an enormous amount of content, including loads of bad content, but you weren't expected to browse all of it at once.

You usually chose a gamemode first. If I wanted TTT, I clicked TTT and browsed TTT servers. If I wanted DarkRP, I did the same there.

Then you could scan the top 10 or 20 servers, look at the player count and read a few tags in the server name like custom weapons, cars or leveling. Within seconds you had a rough idea of what made each server different. Low-quality servers existed too, but they usually ended up further down the list.

S&box feels much more overwhelming by comparison. You're presented with lots of completely different experiences of wildly varying quality and expected to browse banners until something looks interesting. There might be something great buried in there, but most players probably aren't going to spend that much time looking.

I think there could be a clearer distinction between the underlying gamemode and the server or experience built on top of it.

TTT could be the actual gamemode, while many communities run their own versions with different maps, weapons, progression and rules. Those servers could still be prominently discoverable and feel completely different without all presenting themselves as unrelated standalone games.

That would also make reusable content more valuable, since creators could spend more time on what actually makes their experience different instead of rebuilding the same basic systems.

I understand why letting S&box games become standalone releases is attractive, but I'm not sure it's good for S&box itself long-term.

A platform like this becomes much more valuable when everyone shares the same playerbase. If someone installs S&box because their friend invited them to one game, they're now also a potential player for every other experience on the platform.

I'd rather see S&box make staying attractive through things like the Play Fund, discoverability, Workshop integration, infrastructure and access to an existing playerbase. It's similar to releasing something on the App Store instead of putting it on your own random website. The platform itself has value because the users are already there.

I don't think S&box mainly needs more people trying to create polished games from scratch.

I think it needs to make it extremely easy for people to take strong foundations, combine good existing content, change some rules, create an objective, host their own version and invite their friends.

Let experienced developers create completely new gamemodes when they actually need to. For everyone else, make building something fun on top of existing content the easiest and most obvious path.

A huge part of what made Garry's Mod special was that almost anyone could take something that already existed, turn it into their own thing and get other people playing it.

I think S&box could benefit a lot from recreating more of that with the tools we have today.