 |
 | Home | Contact Us | FAQ | Search & Site Map | Link to Us |
 | Sign In | Join | Other 45 Sites in Network |
Java Forum / Virtual Machine / June 2004Java's performance far better that optimized C++
http://www.sys-con.com/story/print.cfm?storyid=45250 Any comments? Thanks Gaurav >http://www.sys-con.com/story/print.cfm?storyid=45250 the key is obviously the improved method calls. This is done though inlining. The big advantage hotspotting has is knowing what to inline. Inlining everything just gobbles up RAM with no payback.  SignatureCanadian Mind Products, Roedy Green. Coaching, problem solving, economical contract programming. See http://mindprod.com/jgloss/jgloss.html for The Java Glossary. >>http://www.sys-con.com/story/print.cfm?storyid=45250 > > the key is obviously the improved method calls. This is done though > inlining. The big advantage hotspotting has is knowing what to > inline. Inlining everything just gobbles up RAM with no payback. Although note that gcc was run with -O2, which does not inline at all. gcc -O3 would have been a more reasonable comparison. A more interesting question: why does anyone pay any attention to microbenchmarks? David Hopwood <david.nospam.hopwood@blueyonder.co.uk > http://www.sys-con.com/story/print.cfm?storyid=45250 > > Any comments? > > Thanks > Gaurav They forgot gcj. >http://www.sys-con.com/story/print.cfm?storyid=45250 > >Any comments? Yeah, yahoos that don't proofread their subject line probably don't have a clue. You also misspelled far. I followed up to comp.lang.c++ so you can get properly flamed for being a troll. > http://www.sys-con.com/story/print.cfm?storyid=45250 > > Any comments? > > Thanks > Gaurav Does 'apples and oranges' ring a bell? Big cycle suckers can be better - unless you need small and efficient. The answer is: it depends. Benchmarks are management fodder, and often the source of bad decisions. >Benchmarks are management fodder, and often the source of bad decisions. Focusing on benchmarks is often being penny wise and pound foolish. What counts is getting the product out the door on time, bug free on budget, and doing that release after release. Whether it runs 5% faster nobody will notice. In fact for many programs it has to run 100% faster for it to be even psychologically noticeable. If you have a bottleneck you will get far more return on your time and money by profiling and tuning the critical section, especially choosing a smarter algorithm, than you ever will by backing off to a more manual language that introduces far more problems than it solves. SignatureCanadian Mind Products, Roedy Green. Coaching, problem solving, economical contract programming. See http://mindprod.com/jgloss/jgloss.html for The Java Glossary. Not sure if GCC is the best optimized compiler. May be these tests should be rerun with MS C++ compiler on Windows and Intel C++ 8.0 on Linux. Sandeep -- http://www.EventHelix.com/EventStudio EventStudio 2.0 - Generate Sequence Diagrams (no drawing required) > http://www.sys-con.com/story/print.cfm?storyid=45250 > > Any comments? Just one: BAAAAAAAAAAAAAAAAAAAAAAAAAHAHAHHAHAHAHAHHAAAAAAAAAAAAA!!!!! Jesus von Neumann CHRIST you Javapologists, GIVE IT THE HELL UP ALREADY! Consistently Stupid, thank you for another meaningless comment. > > http://www.sys-con.com/story/print.cfm?storyid=45250 > > [quoted text clipped - 5 lines] > > Jesus von Neumann CHRIST you Javapologists, GIVE IT THE HELL UP ALREADY! Free MagazinesGet these publications absolutely FREE for up to 12 months. There are no hidden fees and no obligation. Simply choose a title, complete the application form and submit it. Read more ...
|
Reply to this Message
|
 Sign In
 Join
 My Latest Posts My Monitored Threads My Blog My Photo Gallery My Profile My Homepage Start New Thread Enable EMail Alerts Rate this Thread
|
©2008 Advenet LLC Privacy Policy - Terms of Use
This website includes both content owned or controlled by Advenet as well as content owned or controlled by third parties.