This is the mail archive of the java-discuss@sourceware.cygnus.com mailing list for the Java project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

Re: AWT


Thomas Reilly wrote:
> The question is whether to do a lightweight GLIB or heavyweight GTK+.
> I spoke briefly with Peter Mehlitz (sp?) at JavaOne (he did Kaffe's
> AWT) and he said their lightweight implementation made swing apps run
> faster than on your typical Sun JDK.  I can't remember exactly why, I
> don't see why you couldn't have a heavyweight peer model along side a
> fast lightweight model.  Maybe their lightweight implementation is
> just more efficient.

I'm inclined to agree with your last sentence - Motif is not the world's
most efficient toolkit.

I'm very suspicious of the performance claims of all lightweight
toolkits, even without the overwhelming counterexample provided by
Swing. They do the same work on the X client side that would otherwise
be performed by the X server - an advantage only if the client is
running on a more powerful machine and/or manages a more efficient
implementation of window control and event propagation than the X server
does (unlikely, given the server's years of tuning). They also cause
significantly more network traffic between X client and X server than do
heavyweight toolkits. I've been looking for some documentary evidence of
any advantages from lightweight toolkits... I'd most appreciate it if
anyone could point me to some.

Nathan Meyers
nmeyers@teleport.com

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]