Hello hOORST1.
I'm sorry to hear that the LCD display daemon is giving you grief. I have just tested it again myself on my openSUSE 10.2 EM64T system and I do not have the problem you mention.
The amarok script has been made so that it minimizes the calls for refresh of the display... You are using the version from kde-apps.org, right?
If not you can find it here:
http://kde-apps.org/content/show.php/lcdproc_amarok?content=39109I am using lcdproc 5.1 which I have compiled myself. I haven't seen LCDd or lcdproc use much CPU at all...
I think the only way to avoid seeing the status display is to run a client with a reasonable large value. My Amarok script is currently pouncing everything else when music is playing and on startup and if you pause Amarok it's priority will be really low so you can see the other stuff the display is alternating through.
If for some reason this really bogs down your system you could fetch my shell library and run something like this:
. /usr/local/bin/lcdlib.sh
set_clock "$(date +%H:%M\ %d.%m.%Y)"
That will give you todays date and time which will auto-update -- no daemon required. If that seems too static you can follow up with:
clock_move
Strangely enough, the clock will then move...
Running in KDE or not really shouldn't be a factor, except for the fact that you will load some KDE libraries when Amarok starts. It should have no impact on the LCD display script at all...
I haven't tested any of this on Ubuntu, but my current guess is that since Feisty Fawn is still in beta there may be some issues they need to look at before this runs like it should...
All LCDd, lcdproc and liblcd.sh functions do are write to a serial port. That ought to be well supported in any distribution...
Let me know if I can help you further.

I have one issue with my MD8800 machine (well, two actually)
1) For some reason it started saying I may have a bad USB cable. This is related to an internal USB hub. It wasn't there in the beginning (at least the distro I used at the time didn't show it). This causes a fairly high load in itself. I luckily managed to get a kernel developer to make me a patch that would allow me to disable this port on this specific hub and after that all is well.
The only other way I could fight this would be to disable USB 2.0 either in the BIOS or my blacklisting ehci-hcd.
2) The card reader doesn't show up. Not sure if it's related to 1). Windows doesn't seem to have any problems so I wonder if the card-reader is implemented in a funky way.
I am montioning this mainly because you may or may not have a similar situation. If you have kernel messages about not being able to activate a specific USB port and probably the cable is bad, that might explain the load issue you mention.