OpenOffice.org Performance

Marc Lijour (Professeur d'Informatique) marc-bbkyySd1vPWsTnJN9+BGXg at public.gmane.org
Sun Nov 2 03:09:30 UTC 2003


Le 1 Novembre 2003 17:48, cbbrowne-HInyCGIudOg at public.gmane.org a écrit :
> > Small nitpick: OO is written in Java afaik, and most of the time is spent
> > by the Java hulk stepping out of its package(s). If OO would be written
> > in C++ and statically linked (shudder) it should come up at about disk
> > speed, assuming the right optimizations are used, just like Excel on M$.

Thanksfully OOo is getting better and faster with this new release (OOo-1.1 
and StarOffice 7).

> WRONG.
>
> OpenOffice.org depends on:
>  a) libc6 (meaning some of it is in C)
>  b) libstdc++ (meaning some of it is in C++)
>
> There are NO inherent dependancies on Java, except in certain
> extensions.
>
> OpenOffice.org IS written in C++.
>
> If the issues with its performance have to do with the language that it
> is implemented in, then it is C++ that is to blame, not Java.
>
> Reimplementing things in C++ is _NOT_ a universal panacea; the fact that
> it imposes considerable complexity of managing memory allocation and
> deallocation on programmers makes it quite easy for a move to C++ to
> worsen problems rather than to make them better.
>
> My installations of OpenOffice.org contain none of the Java components,
> and are certainly not affected by inefficiencies of that nature.
>
> OO _is_ quite slow, and the only language that can be pointed at for
> blame for that is C++.
> --
> wm(X,Y):-write(X),write('@'),write(Y). wm('cbbrowne','ntlug.org').
> http://www3.sympatico.ca/cbbrowne/finances.html
> Health is merely the slowest possible rate at which one can die.
> --
> The Toronto Linux Users Group.      Meetings: http://tlug.ss.org
> TLUG requests: Linux topics, No HTML, wrap text below 80 columns
> How to UNSUBSCRIBE: http://tlug.ss.org/subscribe.shtml

--
The Toronto Linux Users Group.      Meetings: http://tlug.ss.org
TLUG requests: Linux topics, No HTML, wrap text below 80 columns
How to UNSUBSCRIBE: http://tlug.ss.org/subscribe.shtml





More information about the Legacy mailing list