Joseph Dane wrote:
>> Interestingly enough, I use Eclipse 3.0.1 at work on Windows XP on a 2.7GHz
>> Celeron with 1GB of RAM, while at home I use Eclipse 3.0.1 on MacOS X 10.4.2
>> on a 12" PowerBook G4 with 1.25GB of RAM. I find the Mac Eclipse *MUCH*
>> snappier than the Windows Eclipse, especially startup.
>
>fascinating.
>
>I use eclipse exclusively on the mac, so I can't make any comparisons
>to the windows version. however, I'd rate the performance of eclipse
>on my 1GHz/1GB Powerbook as just barely acceptable. Startup, which I
>just timed for the first time, takes about a minute, and that was
>immediately after shutting down a previously running instance, for
>whatever difference that might make.
What other apps were you running? What Eclipse plugins did you have?
What hard drive characteristics (rpm, capacity, free)?
Maybe it's the Eclipse 3.0.1 vs. 3.1.1 difference.
Or maybe that extra 0.25 GB of RAM really matters.
Sometimes small details matter a lot.
Many things can reduce speed, and swapping is a very important one:
<http://lists.apple.com/archives/java-dev/2005/Aug/msg00053.html>
Swapping in Java has different consequences than other apps because Java's
heap is GC'ed, so triggering a GC can result in a lot of pages getting
swapped in, just to determine reachability. After heap rearrangement,
these may have to be swapped out just to get back to the original active
working context. Even garbage has to be swapped, unless the heap algorithm
is very clever indeed.
Without a carefully constructed and repeatable test case, it may be
fascinating to consider speed differences, but I would be reluctant to draw
any particular conclusions from them.
And that's not even considering the Quartz Debug trick.
-- GG
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Java-dev mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/java-dev/email@hidden
This email sent to email@hidden
