Re: BUG: gnome terminal horrible performance



On Sat, Jul 31, 2004 at 09:45:47AM +0200 or thereabouts, Carsten Weinberg wrote:
> Ryan McDougall wrote:
> 
> >This is not really fair at all. 
> >
> Well, this kind of ignorance is not only not fair its also not helpfull.
> 

I came late to this thread. But a few comments. 

Ryan is absolutely right about gnome-terminal having to deal
with some things xterm doesn't. As well as Pango, I don't think 
xterm talks to the accessibility layer at all, for example.
We can all come up with examples where gnome-terminal behaves
slowly. I have a copious bug report of my own at 

http://bugzilla.gnome.org/show_bug.cgi?id=118657
(Drastic slowdown (particularly in terminals) when 'accessibility'
key in gconf on)

Note that comment number 14 in that bug has concrete suggestions
on how to measure the amount of information gnome-terminal is
sending to at-spi-registryd (accessibility-related). I am 
rebuilding Gnome at the moment and intend to try this, but
you don't need to rebuild Gnome as a whole, just ORBit-2,
if you want to add your own findings to it.

We can't just remove the accessibility code to speed things
up. Or the Pango stuff because "who wants their Korean spam
rendered correctly in a gnome-terminal anyway?" Or tabs in
a gnome-terminal because we can all use 'screen' anyway.

> Ryan, as it is gnome-terminal is unusable. Even on my very fast machine
> its much too slow.

"very fast machine" is not clear enough to give accurate 
information. Unfortunately, I don't have the background to 
suggest the things you need to include to generate numbers
that help someone fix the problem. If I knew, I would not
be sending bug reports of "Bill, help, what do I need to 
type to debug this?" :(

If there are other gnome-terminal or vte hackers lurking
on this list, they are welcome to suggest other things to 
investigate as well as ORBit debugging spew. 

Telsa




[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]