Posted by
ctrueden on
Jun 12, 2009; 7:34pm
URL: http://imagej.273.s1.nabble.com/A-note-on-JVM-performance-tp3692131p3692134.html
Hi Grant,
Studies disagree on exactly how much slower Java is than C++ and in what
ways. People like to broadly generalize about these technologies when
unfortunately it can't really be done objectively.
I wrote up a short bit on Java vs. C++ on the LOCI software web site:
http://www.loci.wisc.edu/ome/faq.html#javaPeople like to cite the Shootout article but it's already five years old. A
lot changes in that time when it comes to software, so such articles are
often immediately out of date.
Still, I agree that Java's performance is quite reasonable for most
applications and people should think carefully about their application's
needs before resorting to C++ for performance reasons.
-Curtis
On Fri, Jun 12, 2009 at 12:10 PM, Grant Harris <
[hidden email]> wrote:
> This article, Java vs C++ "Shootout" Revisited
> (<a href="http://java.sys-con.com/node/45250##">http://java.sys-con.com/node/45250##), makes a strong case to refute the
> still prevalent myth that Java is slow. "[E]ven with JVM startup time,
> Java
> is still faster than C++ in many of these tests."
>
> It also points out that "no one should ever run the client JVM when given
> the choice... Unfortunately, Java applications and applets run by default
> in
> the client VM," Lea observes. "The Server VM is much faster than the Client
> VM, but it has the downside of taking around 10% longer to start up, and it
> uses more memory."
>
> All you need to do is add -server to the java.exe arguments. I'd like to
> hear if anyone finds a marked improvement in using ImageJ.
> Cheers.
> -- Grant Harris, MBL
>