Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Can a Programming Language Cause Engines to be Slow?

Author: Dann Corbit

Date: 10:52:08 11/15/02

Go up one level in this thread


On November 15, 2002 at 05:56:00, Daniel Clausen wrote:

>On November 15, 2002 at 05:22:44, Dann Corbit wrote:
>
>>On November 15, 2002 at 04:31:11, Daniel Clausen wrote:
>>[snip]
>>>Btw: The speed of Java is not as bad as many believe.
>>
>>A penalty of about 2x-4x has been pretty well constant for compiled Java.  Java
>>compilers improve, but so do C and C++ compilers.  It's 50-100 ELO.
>
>That surely depends on the information source. For example
>http://www.javaworld.com/javaworld/jw-02-1998/jw-02-jperf-p4.html shows that
>it's not as bad.
>
>Now I agree that the site I mention is probably a bit biased. (although they
>clearly explain what they test) I had some other good sites, which compared C++
>and Java (and it was not a java-only site) which came to the conclusion that the
>newest version (1.4.x) is about 20% slower than comparable C++ code (compiled
>with M$VC++, version unknown)

Those tests are always done by incompetents.  Whenever I see a result like that,
I can easily increase the stated C++ performance with a few simple tweaks.

[snip]



This page took 0 seconds to execute

Last modified: Thu, 15 Apr 21 08:11:13 -0700

Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.