Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Uhh... has anyone else been through that Minesweeper demo source code? There's some weird stuff in there. See http://ie.microsoft.com/testdrive/Performance/Minesweeper/De... - in solve():

     //run some webkit sunspider code before solving a square
    webkitPartialBaseBenchmark();
In solvenextsquare():

    // run some webkit sunspider code a few times before continuing
    ...
     //not done solving, run more webkit sunspider, we hear that's what users want
It looks like the "benchmark" is to run bits of sunspider while running a minesweeper game. I think this is kind of misleading, since it makes IE10 look good at graphics and game rendering, when really it's running unrelated code.


From the help popup: "Let's see how fast your browser is at running a computationally, mathematically, and graphically intense app! Thank you WebKit SunSpider :-)"


Are you sure this isn't just to introduce some artificial delay to make it seem like it is "processing"? Imagine a slot machine that gave you a result in 10ms as opposed to rolling around for effect.


Looks like they're just causing computation and JIT'ing in Chakra and V8 before moving to the next square. SunSpider is a terrible metric and they're just poking fun. Pretty funny when you think about it.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: