In article 
,
 apple2freak{at}gmail.com wrote:
[...]
> I responded with specific benchmarks using the Java 6 client because 
> I think that is much more representative of what the average person 
> will use in order to run Java programs.

I see that both -client and -server JITC options are included in the JRE 
that comes with the SDK:

http://java.sun.com/docs/hotspot/HotSpotFAQ.html#compiler_types>

I imagine the average person will use whichever optimizes performance, 
unless (s)he determines performance to be irrelevant.

On Mar 7, 3:11 pm, mdj  wrote:
> > > Conversely, you'll be hard-pressed to find a benchmark where Java 
> > > comes out ahead by more than a few percentage points.

Indeed, I often choose Java to leverage the rich, cross-platform 
features of the SDK; or I might choose Ada to leverage strong type 
checking; or I might choose some other language for some other reason. 
I don't really worry about performance until the profiler tells me 
where I've made a design flaw. Then, I'm more likely to consider an 
algorithmic variation than a new language, although nothing precludes 
invoking an external routine.

On the Apple II, the Java runtime isn't quite as rich, but the native 
interface is good old 6502, source included.

[...]

I like the comparison sites like http://shootout.alioth.debian.org/> 
and http://www.rosettacode.org/> to study optimized code, but I'm not 
sure the results are dispositive of language choice.

-- 
John B. Matthews
trashgod at gmail dot com
http://sites.google.com/site/drjohnbmatthews>
--- SBBSecho 2.12-Win32