Hi! I installed my system a few weeks ago, when i did xine was working fine. When I installed xgl on my system all of a sudden my remote on Xine became distorted,(like so http://i68.photobucket.com/albums/i1...6/93bbd2d5.jpg )

I tried to remove the xine program, and reinstalling it, but still same problem. I also did an upgrade on my system, and one of the packages updated was xine, i though it was a fix for my problem but nope same thing still happening. It's just this minor annoyance that is keeping me from enjoying my Xubuntu box :-p.
If you need any specs or outpu from a .conf file i will be more that happy to post it.

Almost forgot its Xubuntu edgy beta.

Thanks for you help in advanced.

Hi there. I was having the same problem with Xine and XGL on my Mandriva 2007. A workaround that worked for me is to edit /etc/imrc. You will find a part of the file that reads as:

"# This will turn shared pixmaps on or off (off forces off, on lets imlib
# work it out). This is yet another speedup. leave it on unless it doesn't
# work.. then turn it off.
SharedPixmaps on"

Change "on" to "off" and save the file. (as root, of course)

Hope it works for you too.
Cheers

I'm also having this issue in Xubuntu, though it doesn't seem as if I have that file. Do you know what is responsible for? maybe ubuntu calls it something different.

Unfortunately i do not know what Ubuntu or its variants would call it or use in its stead. Looking at the file, all i can tell you is that it says it "contains palette entries for a global palette for all Imlib based programs." So perhaps look for anything related to Imlib on your system?
You could try to create the file, but i have no idea how that would affect the overall performance of your OS. If you want to try, here is the full text of my config:

################################
# Config file for Imlib #
################################

# The file that contains palette entries for a global palette for all Imlib
# based programs.
# options: full path to palette file
PaletteFile /etc/im_palette.pal
# This defines if when the display is greater than 8 bit, that it still remaps
# the images to the palette defined, rather than using "perfect" rendering
# options: yes/no
PaletteOverride no
# If remapping to the palette, whether to use Floyd-Steinberg dithering. Saying
# yes will slow things down though.
# options: yes/no
Dither yes
# when remapping to the palette, saying fast will reduce accuracy, but improve
# speed quite considerably
# options: fast/slow
Remap fast
# This turns on dithering for 15/16 bpp. This makes smooth gradients look much
# smoother - in fact almost perfect. You will find it nigh impossible to tell
# the difference between 15/16bpp dithered and 24bpp. Unless you have extra
# CPU to burn, its not recommended, unless you are a image quality freak, and
# you insist on maximum quality in 15/16bpp. It does slow things down. It
# would be best to leave it off and let the applications themselves allow
# you to select it for certain purposes only.
HighQuality off
# This option if specified off will force MIT-SHM off, otherwise will allow
# Imlib to work it out itself.
Mit-Shm on
# This will turn shared pixmaps on or off (off forces off, on lets imlib
# work it out). This is yet another speedup. leave it on unless it doesn't
# work.. then turn it off.
SharedPixmaps off
# This speeds up rendering considerably, but may not work on your hardware
# due to it bypassing a few layers and byte-twiddling the rendered image data
# manually, and due to endianess, bit-ordering or RGB ordering it may screw up
# and not work, so try it.. if things work great!, if not, wait until a
# renderer for your situation is written, or write one yourself and donate
# it. It's easy to do, just look at rend.c
FastRender on
# This is in fact a workaround due to Solaris's shared memory theories.
# This specifies the maximum size of a shared memory chunk in bytes. If an
# image is larger that this in bytes for the video mode you're in, imlib will
# not use MIT-SHM. if you comment this out, imlib will use as much memory as
# necessary to render the image.
# Shm_Max_Size 1000000
# This turns Image loading (24) bit caching on or off. HIGHLY suggested to be
# turned ON!
Image_Cache on
# Image cache size in bytes. As with any cache, the more, the better. If you
# load the same image more than once. Imlib will used a previously loaded
# copy, and if its freed, the Image_Cache_Size amount of bytes of image data
# are kept even after being freed, in case the same image is loaded again soon
# afterwards. Neat eh?
Image_Cache_Size 524288
# This turns the pixmap caching system on or off. If on, only well-behaved
# programs that conform to the specs for using Imlib will exhibit the
# behavior as expected. It is suggested to leave this on, as it will boost
# performance considerably, speed-wise and memory-wise. The reason apps need
# to be well-behaved is so that they don't go drawing on, and XFreePixmap'ing
# these pixmaps themselves, because this will trample all over the cache
# and give very horrid effects, or even make the apps crash with segfaults or
# Xlib errors.
Pixmap_Cache on
# Pixmap cache is in **-> BITS <-**... the end result is APPROXIMATELY
# 10000000 bits of pixmap make your Xserver grow by 1Mb of RAM (VERY rough).
# As with any cache, the more, the better. The more you have, the less likely
# it is that you will get cache misses and so performance on scaling the same
# image to commonly used sizes (ie if 3 or 4 sizes of the same image are used)
# will be lightning fast, in fact in some tests I did, in 16bpp up to 38 times
# as fast, and in 8bpp (with dithering on) up to 105 times faster!!! (these
# are nominal figures obtained on my machine. these are MAXIMUM speedup
# results. Results may vary on other machines and according to the way
# programs are written and use Imlib)
Pixmap_Cache_Size 5242880
# This FORCES Imlib to use the hexadecimal visual id stated here if it is
# defined in the imrc. This bypasses Imlib's routines that hunt for the best
# visual. You can obtain a list of visual ID's using the xdpyinfo command.
# You should only need this if Imlib doesn't pick the correct visual or you
# have strange hardware/Xserver combinations.
#ForceVisualID 22
# This allows Imlib to fall back on Imagemagick and/or NETPBM
# utilities if it can't load the file.
Fallback on
# Default Gamma, Brightness and Contrast stuff....
Gamma 1.0
Brightness 1.0
Contrast 1.0
Red_Gamma 1.0
Red_Brightness 1.0
Red_Contrast 1.0
Green_Gamma 1.0
Green_Brightness 1.0
Green_Contrast 1.0
Blue_Gamma 1.0
Blue_Brightness 1.0
Blue_Contrast 1.0
Ordered_Dither on

I couldn't find a similar file already existing, but I did find a thread on the ubuntuforums explaining saying that it does work to create a file to do that, by typing this command:

Presumably if there's nothing else in the file it just keeps all the defaults. This works fine but just sets if for your user. I tried making /etc/imrc with the same thing to see if it would affect things globally, but didn't seem to work right away. I'll post back if it changes in the morning after a reboot, maybe needs to read the file at that time.

Edit: Yeah, the only way I get it to work is for the file .imrc to be in the home folder for each user. Not much of a problem for me since I only have two user accounts on my computer though.

Hi, all! Thanks for your replies. Sorry I didnt have a chance to answer for i was gone for the long weekend. Anyways down to business. Ok I found the file you were referring too Blackhawkckc
I edited it, but it did not catch. So i tried what mcmillan said and make a file in the home folder called .imrc
and i copy and pasted the info from the other file located in
/etc/imlib/imrc Now it works like charm But I lost my beryl (not due to this) so now i need to get it back lol.

Thanks all for your help!