Re: compilation of java for 64 bit
Tom Anderson wrote:
On Sun, 27 Dec 2009, Mike Schilling wrote:
zigzagdna wrote:
Looks like safest way is to use 64bit jvm when on 64 bit OS, however
it runs slightly slower
Read it more carefully: *may* run slightly slower.
Without compressed OOPs, *will probably* run moderately slower.
Probability based on what factors, and asserted with what evidence?
With compressed OOPs, *will probably not* run noticeably slower.
As M. Pornin posted, there are large categories of Java programs that run
appreciably faster with a 64-bit JVM. The probability of any given program
running faster or slower is directly related to the probability that one will
write a program whose profile fits the "faster" or "slower" category. Thomas
Pornin also asserted that for the "overwhelming majority" of programs, "little
to no performance difference is to be observed between a 32-bit and a 64-bit
JVM." I don't know what his evidence is, either, but his claim contradicts
yours somewhat.
The only claim I trust here is the one made by every writer on the subject of
software performance engineering, that it takes measurement to know.
The OP should concentrate on good algorithms and clean, maintainable code.
The performance will likely take care of itself then.
As Peter Duniho remarked, in response to the OP's statement that, "[In the
e]nvironment I work in, there is not any time to do any performance comparison":
Then by definition, performance doesn't matter. Don't worry about it.
--
Lew