So (I''m not trying to start an irrationaly argument here, so don''t think that I am), are you saying that if I wrote a C/C++ program that looked a lot like a Java VM (That is why I was refering to when I said "Java" above) and had a "Script" file that happened to look a lot like Java code, and I compiled and ran it, that somehow it is going to slower than Java?

BTW: Don''t take this stuff too seriously, it''s better just to think of this as a casual argument, not religious persecution.

"Finger to spiritual emptiness underlying everything." -- How a C manual referred to a "pointer to void." --Things People Said

http://druidgames.cjb.net/

I''m not saying anything about that case. All I''m saying is that there is no absolute limit that stipulates that Java cannot possibly execute faster than what the corresponding code would have in the language the JVM was written in.

If we''re talking strict interpretation - sure. But otherwise, no (i.e. JIT:ing, etc).

Oh, and sorry about posting anonymously all the time. I''m not a regular (not a fan of web forums), so I didn''t register.

I''m Peter Schuller (peter.schuller@infidyne.com) - there, I''m not anonymous anymore

In any case; I''m not trying to claim that Java is faster than C. Like I said before, the object of the test we ran wasn''t to show that Java was as fast / faster than C, but that is was comparable.

A more accurate test of real world performance would involve much function calling in the C version and more method invokations in the Java verison. Because in a language like Java, method invokations are what''s going to be the potential bottleneck.