Hacker News Reader

Hacker News

Comment by pwnna | Hacker News Reader

Parent Full thread

pwnna

How is this made?

It's pretty slow on my computer. Would ASM.js work better?

View on HN

I'm getting 15 FPS in Chrome on my Surface Pro.

It uses OpenGL ES it seems, doesn't work in IE 11.

Edit: It would be interesting to know if this would perform better using WebGL and running it in IE 11, anyone has any thoughts on that?

60fps (cap I think) on Firefox 25 on i5 desktop.

It's GPU-, not CPU bound.

Yeah I assumed this when I noticed even the Retina laptops with the i7s weren't doing as well. I have an Nvidia GTX 560 Ti which must be doing most of the work.

32 FPS - Chrome - Windows 7 - Thinkpad X1 Carbon i7

22 FPS on Firefox..

Huh. A bug should be filed..

45 FPS: Chrome on 2012 Macbook Pro. Looks fantastic.

51 FPS on new Macbook Pro Retina

It's gorgeous... reminds me of the good old "visualization" days of Winamp and MusicMatch

I could be wrong, but my understanding is that WebGL uses the exact same API as ES, including the same stripped down shader language.

25 FPS on Chrome/2013 Macbook Pro

Weird! I get 60 on Chrome 31.0.1650/2013 Macbook Pro.

60fps on chrome

20 FPS - Chrome - MacBook Air/2011

I'm getting 38 FPS on a 2012 Macbook Air with Chrome. Didn't expect the HD4000/HD3000 difference to be so big.

It looks like most of the heavy lifting is done in shaders, so asm.js probably wouldn't improve it much.

It looks like there's some sort of gravity involved. So then how do you get around the O(n^2) problem of comparing the contribution of each particle to every other particle? Did they implement an quadtree of some sort, or is it something else?

It looks like it's using a fluid solver. So, it's not that every particle is interacting with every other particle, but that there's a flow vector being established at each point on the screen and that's being added to each particle's position.

Disclaimer: I haven't actually analyzed the source.

I'm wondering what optimization you can to that do with a quadtree. The only way I can see is if you sacrifice accuracy on distant particles by quantizing and counting. You could use geometric hashing to do the same thing.

Anyway, I don't think this demo is exhibiting inter-particle attraction, but I'm not positive.

Exactly right. The most common scheme uses a quadtree and runs around O(n log n).

http://en.wikipedia.org/wiki/Barnes%E2%80%93Hut_simulation

Reply on news.ycombinator.com