This is the mail archive of the
cygwin-xfree@cygwin.com
mailing list for the Cygwin XFree86 project.
Re: Fwd: RE: [FEATURE REQUEST] LBX
- To: Brian Genisio <briang at OasisAdvancedEngineering dot com>
- Subject: Re: Fwd: RE: [FEATURE REQUEST] LBX
- From: Alexander Gottwald <alexander dot gottwald at informatik dot tu-chemnitz dot de>
- Date: Tue, 26 Jun 2001 15:58:11 +0200 (CEST)
- Cc: huntharo at msu dot edu, cygwin-xfree at sources dot redhat dot com
As long as LBX does not cause any problems, we should leave it in the server.
On Tue, 26 Jun 2001, Brian Genisio wrote:
> It may be true that LBX is not necessarily the best extension to include in an X
> server. It removes the dynamics of an application to use an extension, since it
> talks to a proxy as a server. The proxy then may or may not be able to connect
> to the real server, making it much harder for the app to determine if it should
> use LBX or not.
>
> Still, LBX does speed up X communication through slower connections. Also, over
> a LAN, if there are a lot of images, it speeds things up. I have done some
> comparisons by with a bandwith/latency limiting router with an X app in the
> following circumstances :
> 1. No compression
> 2. LBX compression
> 3. DXPC compression (similar to LBX, but is not a server extension, and requires
> 2 proxys)
>
> I found that compression definately made a difference. LBX was overall, a better
> solution as for speed. DXPC did much better at crunching large images (XPutImage
> calls), but was slower with normal operation.
>
> With the nature of what a Cygwin XFree Server is commonly used for, I believe LBX
> is more helpful than in other servers. Many people use the X server in windows
> to display remote applications, thus making LBX useful to have. Plus, the
> current state of cygwin and the XFree server is pretty slow, so it can use any
> help it can get....
>
> Unless, of course, LBX does not speed things up under the XFree server in
> Cygwin. My tests were only done with Linux XFree, and Hummingbird eXceed for
> Windows. I assume the same results are true with cygwin, and my assumption may
> be false.
>
> Ok, I have done enough rambling on. Have a good one !!!
> Brian
>
>
> >
> > >I found the article that gave me the impression that getting rid of LBX
> > >would be a good thing, it is the article that Keith Packard and Jim Gettys
> > >wrote about the RandR extension:
> > >http://www.xfree86.org/~keithp/talks/randr/randr/randr.html#tex2html1
> > >
> > >"more importantly, the extension framework has isolated ``bad'' ideas from
> > >the core X functionality allowing their eventual atrophy into irrelevance."
> > >
> > >The footnote for that statement says:
> > >http://www.xfree86.org/~keithp/talks/randr/randr/randr.html#foot74
> > >"E.g. PEX, XIE, LBX, along with wide lines and arcs in the core protocol ...
> > >"
> > >
> > >Now, Jim Gettys (original writer of the X Window System) and Keith Packard
> > >(RandR, Shadow, FB, XFree86) are arguably the two primary heavyweights in
> > >the X world today. When they talk, I march. They said LBX sucks, so I
> > >removed LBX.
> > >
> > >Of course, LBX is back in now, but I didn't want anyone to think that I was
> > >making a rash decision in disabling LBX. On a side note, Keith Packard had
> > >recently turned off building PEX and XIE for XFree86 and encouraged other
> > >developers to do the same thing for other platforms, so it seemed perfectly
> > >reasonable to disable LBX as well.
> > >
> > >Harold
>
>
--
Alexander.Gottwald@informatik.tu-chemnitz.de
http://www.gotti.org
phone: +49 3725 349 80 80 mobile: +49 172 7854017