4 ms·

I was wondering what the end goal of asm.js was - and the ability to run games in your browser at near-desktop speeds is _very_ impressive. Now, how long until

I was wondering what the end goal of asm.js was - and the ability to run games in your browser at near-desktop speeds is _very_ impressive. Now, how long until Chrome supports it...

GotNothing 14y ago
As soon as they find a way to put ads in it...
- joshrice 14y ago
  on the loading screens!
- notatoad 14y ago
  yes, because chrome is so full of ads...
- lucian1900 14y ago
  Chrome has supported NaCl for a long time, which is quite a bit saner than asm.js
  - hosay123 14y ago
    You show up on every NaCl-related thread claiming it is awesome and asm.js is crap, but that's as far as it goes - you never explain why. You're basically wasting time because what you say is never actionable. Please defend your opinions - inform us!
    - lucian1900 14y ago
      (P)NaCl exposes pretty much the same environment and tool chain devs get writing native apps. Predictable performance, decent debugging, etc. One can even decide to use platform-specific features if strictly necessary. The problem it has is that it is largely opaque to the JITs in browser engines (even PNaCl, since JS JITs have nothing to do with llvm). One way to fix that would be to define a simple byte code as a compilation target and have people compile to that, in addition to compilation to JS with emacripten. The end result would be, in my opinion, much cleaner than the ugly hacks in asm.js Furthermore, by design, asm.js will always need guards in the generated code, thus some overhead that can never be fully removed (although JITs will help with this). In particular, one can write "x = +x" in JS, but then change the type of x, thus invalidating the invariant. A guard will be needed to guarantee correctness, since asm.js is a subset of JS, but the outer set still exists. It seems to me that web technologies are getting more and more crazy. To me it appeared obvious that asm.js is not a nice approach, but I can see many people disagree.
      - BrendanEich 14y ago
        You are mistaken, there are no type guards within asm.js-verified/compiled/linked code. Read more at http://asmjs.org/ http://asmjs.org/. /be
  - gcp 14y ago
    You're too early for April Fools, but you almost got me.
AndrewDucker 14y ago
It occurs to me that this also makes the web much more plausible as an application-delivery system. And who cares what OS you're running - if you can compile your application to asm.js, which runs at only half the speed of C[1], then you can make it available to everyone. [1] See: the image here: http://www.extremetech.com/computing/151403-firefox-sticks-it-to-google-with-odinmonkey-which-can-boost-javascript-performance-by-1000-or-more http://www.extremetech.com/computing/151403-firefox-sticks-i...
- pjmlp 14y ago
  Assuming the browser vendor cares to support it...
  - drawkbox 14y ago
    ... and at comparable speeds... Sadly this is one area where plugins still rule and advance innovation. That has made it into the web quite a bit in html5 but still a bit to go before it is stable for 3D without a plugin. Three.js, fallbacks to canvas, etc are awesome and work today (hardware limitations on mobile, browser limitations like IE no webgl). But for intense 3d games some hardware acceleration is needed that is consistent. I am working on a bunch of canvas mobile apps right now and the speeds are problematic with many sprites, would love more hardware access. If Mozilla can truly get 3d mobile browser gaming going with this (which currently there is no solution other than hardware accelerated canvas) then that opens up interesting areas. Also opens up new security issues. In terms of business/product use that isn't highlighting new tech, it is still a bit away.
mratzloff 14y ago
I just played the demo game in Chrome and it ran at 30-60 fps on "high resolution" on a 2-year-old MacBook Pro, so... maybe it doesn't matter that much in the short term.

I was wondering what the end goal of asm.js was - and the ability to run games in your browser at near-desktop speeds is _very_ impressive. Now, how long until Chrome supports it...

GotNothing 14y ago

As soon as they find a way to put ads in it...

joshrice 14y ago
on the loading screens!
notatoad 14y ago
yes, because chrome is so full of ads...
lucian1900 14y ago
Chrome has supported NaCl for a long time, which is quite a bit saner than asm.js
- hosay123 14y ago
  You show up on every NaCl-related thread claiming it is awesome and asm.js is crap, but that's as far as it goes - you never explain why. You're basically wasting time because what you say is never actionable. Please defend your opinions - inform us!
  - lucian1900 14y ago
    (P)NaCl exposes pretty much the same environment and tool chain devs get writing native apps. Predictable performance, decent debugging, etc. One can even decide to use platform-specific features if strictly necessary. The problem it has is that it is largely opaque to the JITs in browser engines (even PNaCl, since JS JITs have nothing to do with llvm). One way to fix that would be to define a simple byte code as a compilation target and have people compile to that, in addition to compilation to JS with emacripten. The end result would be, in my opinion, much cleaner than the ugly hacks in asm.js Furthermore, by design, asm.js will always need guards in the generated code, thus some overhead that can never be fully removed (although JITs will help with this). In particular, one can write "x = +x" in JS, but then change the type of x, thus invalidating the invariant. A guard will be needed to guarantee correctness, since asm.js is a subset of JS, but the outer set still exists. It seems to me that web technologies are getting more and more crazy. To me it appeared obvious that asm.js is not a nice approach, but I can see many people disagree.
    - BrendanEich 14y ago
      You are mistaken, there are no type guards within asm.js-verified/compiled/linked code. Read more at http://asmjs.org/ http://asmjs.org/. /be
- gcp 14y ago
  You're too early for April Fools, but you almost got me.

AndrewDucker 14y ago

It occurs to me that this also makes the web much more plausible as an application-delivery system. And who cares what OS you're running - if you can compile your application to asm.js, which runs at only half the speed of C[1], then you can make it available to everyone. [1] See: the image here: http://www.extremetech.com/computing/151403-firefox-sticks-it-to-google-with-odinmonkey-which-can-boost-javascript-performance-by-1000-or-more http://www.extremetech.com/computing/151403-firefox-sticks-i...

pjmlp 14y ago
Assuming the browser vendor cares to support it...
- drawkbox 14y ago
  ... and at comparable speeds... Sadly this is one area where plugins still rule and advance innovation. That has made it into the web quite a bit in html5 but still a bit to go before it is stable for 3D without a plugin. Three.js, fallbacks to canvas, etc are awesome and work today (hardware limitations on mobile, browser limitations like IE no webgl). But for intense 3d games some hardware acceleration is needed that is consistent. I am working on a bunch of canvas mobile apps right now and the speeds are problematic with many sprites, would love more hardware access. If Mozilla can truly get 3d mobile browser gaming going with this (which currently there is no solution other than hardware accelerated canvas) then that opens up interesting areas. Also opens up new security issues. In terms of business/product use that isn't highlighting new tech, it is still a bit away.

mratzloff 14y ago

I just played the demo game in Chrome and it ran at 30-60 fps on "high resolution" on a 2-year-old MacBook Pro, so... maybe it doesn't matter that much in the short term.