These performance tests are dodgy at the best of times, simply because the way you approach problems are different in different languages. For example, Java is all about "create lots of objects" where C++ might take a more conservative approach (but at the expense of more convoluted code). C++ might suit video codecs, where you set up your buffers and the code just blisters over them. Java is probably going to suit anything with a complex, evolving object graph (well, the garbage collection certainly does).
Who was it (Gosling?) who said "C people think malloc is fast". But I do think it rocks how far the JVM has come. And it's all down to Java being an abstract sandbox (the thing which caused it to be slow in the first case). So an accurate benchmark would have to do something non-trivial and the code would have to be written in a very similar way, otherwise you're not really comparing the same thing.
-- You received this message because you are subscribed to the Google Groups "The Java Posse" group. To post to this group, send email to javapo...@googlegroups.com. To unsubscribe from this group, send email to javaposse+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/javaposse?hl=en.