Live data from Hacker News

Ambient: The Multiplayer Game Engine

ambient.run

101–110 of 148 posts

Re: Ambient: The Multiplayer Game Engine

#101
post #97

Biggest issue I have with my current game engine is compile times are about a minute for C++. I don’t see it anywhere but would love to find that Ambient can go from code change to running multiplayer clients in < 10 seconds

Yeah that's the idea; as a user you build multiple independent modules, which each are compiled in isolation, and which each produce a WASM module. The runtime then loads that module. All of this is very fast on the Rust toolchain with incremental builds, though of course it depends on the user breaking their code up into reasonably sized modules. The modules can independently be hot-reloaded (we had this before but it went away in a refactor, but we're planning to re-add again).

Re: Ambient: The Multiplayer Game Engine

#102
post #20

A couple of questions... Is it using Vulkan at the back or WebGL? I'm surprised that it's not mentioned in large bold text, typically that is the case. How are you financing this project? You're hiring people, so clearly there is a budget, but the engine is license under MIT, is it based on subscription for support or royalties?

Hi, Ambient CPTO here; We're using wgpu (a rust crate), which supports a number of different backends, including Vulcan and WebGL (and Directx, WebGPU, Metal).

We're VC backed, which is how we finance the project right now. Long term we plan to monetize additional (optional) services around Ambient, such as game servers.

Re: Ambient: The Multiplayer Game Engine

#105

I wonder how it can scale? Even with the battle royale craze, it seems like nothing has really replicated Planetside 2's black magic netcode (or at least amassed a playerbase big enough to show it off).

Planetside 2 netcode is terrible, people forget that it runs at 5hz, at that sim speed any game can put 100's of people on screen.

Also client side hit detection, lol. Full of cheat and completely inacurate gameplay.

Re: Ambient: The Multiplayer Game Engine

#106

"Why not Rust?" is a question I ask about game development a lot these days. I'm an engineer with a decade of experience as a web services engineer, and 5 years of game development engineering. Rust seems like the right direction for game developers -- safer code, faster compile times, more portable binaries. Ambient looks very cool, but I had some questions -- 1) WebAssembly, but no native web player? One of the mos…

"Why not Rust?" Well, if you have got access to a language designed for making games like Jai, then that would presumably be better than Rust.

Is there a road map for when Jai might be available.

Re: Ambient: The Multiplayer Game Engine

#107
post #15

I have to ask: Why not contribute to Bevy instead? Edit: Ah... I see it's a SaaS startup that wants to offer the servers as a service for game developers. Building on someone else's engine would make the hosting brand weaker.

Engines need to be high-performance across machines and networks, a challenge in itself, and their maintenance is more complicated than an operating system's. This is an observation based on the current engine and OS market. Why is it bad that they're a SaaS business? What other business models are ok for making game engines?

Re: Ambient: The Multiplayer Game Engine

#108

Engines that reached escape velocity often did so by having usable artist tools and [...] no, that's pretty much the defining factor. Artists used them. Seeing people in the age of 2 commercial engines (and only one you can reliably ship with...) attempt to challenge architectures the incumbents instilled, knowing the challenge of getting games made with a new engine, gives my cynicism a run for its' money.

Yep, good tools was the minimum a decade ago, these days you also need an asset store. I suspect the next decade you will need to have integrated AI tools. The only opensource game engine that looks interesting is Godot, they are working hard on making their tools better and growing their asset store.

"Integrated", imho, is exactly how large engines will win. Modularity is a huge challenge (ABI, Data Pipeline, where compute happens), but without modularity, only engines with teams of hundreds will be serious contestants.

Re: Ambient: The Multiplayer Game Engine

#109
post #64

Earlier quoted context omitted.

Besides what the others answered, not only it showed off what was capable on iPhone, it showed what OpenGL ES 3.0 (the WebGL 2.0 foundation) could do. Most stuff on the Web thus far, with exception of Google Earth and a few ecommerce sites, is closer to Flash 2D games than anything else. Hence when a game engine advertises itself as WebGPU, I expect a browser demo at the same level as Infinity Blade, Unreal Citadel,…

WebGPU doesn't necessarily mean "in the browser" though Rust game engines are using it as a library for native games

It does, anywhere else is better served with middleware that has full exposure to native 3D APIs without the constraints of an API designed to be called in a browser sandbox.

Re: Ambient: The Multiplayer Game Engine

#110

Earlier quoted context omitted.

Ambient co-founder here. You're right, thanks for the feedback, we'll update the website with one soon.

Regarding that video, stuttering on a travelling content made me tick. Video has a lot of duplicate frames, not in a perfect ratio of the encoded 60Hz. • Check https://global-uploads.webflow.com/63e51fd138280b42988eb5e6/... in VLC, pause the video then press the "e" key to advance frame by frame. • Open https://www.youtube.com/watch?v=f_BaT3LDwo8 , advance with the "." (dot) key.

I was wondering what was going on. I'm sensitive to low refresh rates (my screen is 144Hz, had 2 concussions) but watching that video was nauseating.
Post reply on HN