You can add location information to your Tweets, such as your city or precise location, from the web and via third-party applications. You always have the option to delete your Tweet location history. Learn more
I’ve spoken with four teams in the past three days that would benefit from what I’m terming “mobile real-time rendering infrastructure”: a source-available, approachable, extensible library/ecosystem for building 3D apps and visualizations without the overhead of a “game engine”
The question is, is this a viable niche? You’re using SceneKit, but you’ve hit performance cliffs. You want to use Unity/Unreal, but they’re massive and cumbersome. Is there a (cross-platform) middle road for small teams building interactive/entertainment/edu/AR/viz apps?
I haven’t done much research yet, but I see things like Cinder, OpenFrameworks, and any number of nascent projects and I have to think things could be simpler and easier. You shouldn’t have to go the full Godot to add 3D to your app.
SceneKit is decent. Can you give me an example of a performance cliff?
You need to render 10,000 instances of the same smallish geometry, dynamically updating each every frame. Bye.
Fair. Quite a few ‘easy to use’ engines would struggle with that without some performance tuning; including Unity out the box
Right. I’m looking 18 months down the road when the idea of adopting Unity is even more onerous to workaday coders who just want the simplicity of SceneKit without the blackbox lock-in.
Oh... I think that's still Scenekit in 18 months - there wasn't much in the way of new Secenkit features this year, no 'what's new in scenekit' I'd expect Scenekit and Metal will go together better next WWDC I doubt Apple are comfortable with just how many AR things are Unity
this is just my hypothesis - but the Scenekit, graphics and programming teams have only grown - and not shipped new features...
Isn't SceneKit performance fine for most "I don't need a full game engine"-scale projects? Are there particular areas of SceneKit where you've hit performance issues?
No. Yes.
Absolutely! Our team needs that for visual debugging of 3D algorithms. My teammate does prototyping in REGL (e.g. http://rreusser.github.io/iterative-closest-point/ …), which made for very fast prototyping and iteration, and misses that approach in iOS/Metal
REGL is a lightweight framework for functional WebGL: https://github.com/regl-project/regl … For all the faults of JS and WebGL, it's really great to be able to not care so much about types and iterate without a full recompile/quit/relaunch cycle
You can take a look to #raylib, it's a simple and easy-to-use C framework, easily tweakable and comprehensive. multiplatform. free and open source.https://www.raylib.com
If 'viable' means 'financially viable' I'm sceptical, because IME the decision what engine technology to use is hardly justified by 'hard facts', but irrational emotions and risk aversion. Hard to beat the established engines in that game ;)
Low barrier to entry, cross platform, vr support, basic contents... are you talking about a-frame & webvr? Who's the target audience?
Audience being who's making stuff (coders, artists?) and what's the output? Physical installations or apps?
Filament by Google may fit your needs. @romainguy is one of the people behind it.https://github.com/google/filament
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.
