On 6 May 2013 19:52, Stefan Krause <[email protected]> wrote: > A few days ago I found a nice article about Dart vs. Java performance: > http://dartogreniyorum.blogspot.co.uk/2013/05/performance-optimization-and-dart.html > I wanted to check how well V8 is doing with this benchmark, so I converted > to Java code to Javascript, but I ran into an issue I don't understand. > > If I execute the javascript code with node.js or with a V8 custom build (git > master) the code runs in less than two seconds (V8 32 bit: 1197 msecs, V8 64 > bit: 1549 msecs, node.js 64 bit: 1341 msecs). If I run it in the browser is > takes more than seven seconds (28.0.1499.0 canary : 7580 msecs, 27.0.1453.73 > beta: 7707 msecs) all on a MBP with OSX 10.8.3. > > The code stresses typed arrays (mostly Float64Array) quite heavily. What > could explain that large difference?
For historic reasons, Chrome currently uses a completely different implementation for typed arrays than the D8 shell (and presumably, Node). We are working on unifying these implementations right now, and that should provide a significant performance boost, especially over the current Chrome implementation. /Andreas -- -- v8-users mailing list [email protected] http://groups.google.com/group/v8-users --- You received this message because you are subscribed to the Google Groups "v8-users" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/groups/opt_out.
