kscreensaver shows garbage in dual screen mode in the area which is not 3d (opengl) capable.

Bug #63565 reported by Moreno Bartalucci
2
Affects Status Importance Assigned to Milestone
KDE Artwork
New
Undecided
Unassigned
kdeartwork (Ubuntu)
Invalid
Low
Unassigned

Bug Description

I use a dual monitor configuration (with merged framebuffer) with an ATI adapter (ATI Technologies Inc RV350 [Mobility Radeon 9600 M10]).

I use the open source radeon drivers (with R300 acceleration).
glxinfo gives this (among the other information): "OpenGL renderer string: Mesa DRI R300 20040924 AGP 8x TCL"

I'm using kubuntu dapper (6.06.1).

The first screen is set at 1600x1200, while the second one is at 1280x1024, thus the total horizontal resolution is 1600+1280=2880.

I know the 3d acceleration on this card do not support the whole area, part of the second screen is not working with 3d graphics.

When an opengl screen saver starts (kscreensaver), I see garbage on the uncovered area.
With Ubuntu Breezy (gnome), however, that same area was blanked (=kept black).

It happens with almost all the opengl screen savers, currently, however, I'm using keuphoria.

Revision history for this message
Sebastian Rode (sebastian-ro-deactivatedaccount) wrote :

Thank you for taking the time to report this bug and helping to make Ubuntu better. You reported this bug a while ago and there hasn't been any activity in it recently. We were wondering is this still an issue for you? Can you try with latest Ubuntu release? Thanks in advance.

Changed in kdeartwork:
status: New → Incomplete
Revision history for this message
Moreno Bartalucci (moreno-bartalucci) wrote : Re: [Bug 63565] Re: kscreensaver shows garbage in dual screen mode in the area which is not 3d (opengl) capable.

Yes, the bug is still fully there.
Tried this morning with Kubuntu Hardy updated recently (a few days ago).

I tried it on a laptop Compaq/HP nw8000 (the same one I was using at the bug
report time).

Now the configuration of the dual head is changed (MergedFB is no more active,
I had to figure out how to use xrandr).

Furthermore, the screensaver GL i used for the test is almost unusable because
it's extremely slow (if I remember well it wasn't so slow with Dapper), but
this behaviour could be due to some other settings.
The screen saver I tried is Euphoria.

In the meantime I changed notebook but I still have the old one for testing.

Alle sabato 23 febbraio 2008, Sebastian Rode ha scritto:
> Thank you for taking the time to report this bug and helping to make
> Ubuntu better. You reported this bug a while ago and there hasn't been
> any activity in it recently. We were wondering is this still an issue
> for you? Can you try with latest Ubuntu release? Thanks in advance.
>
> ** Changed in: kdeartwork (Ubuntu)
> Status: New => Incomplete

Revision history for this message
Harald Sitter (apachelogger) wrote :

Technically said, there is no reason you should have one head with 3d and one without. I.e. I think this should be fixed in the Driver.

Anyway, please report a bug on http://bugs.kde.org and paste the bug URL here... this issue should really be fixed in KDE directly.

Changed in kdeartwork:
importance: Undecided → Low
status: Incomplete → New
Revision history for this message
Moreno Bartalucci (moreno-bartalucci) wrote :

As I wrote in the decription of the bug, it is NOT a problem of "one head yes, one head no", the problem is on the total horizontal resolution.
In my configuration, in fact, it is 2880 pixels (1600 on the first head + 1280 on the second head).
The 3d hardware acceleration on this card, however, supports only until a certain number of pixels which is below my number of pixels (i don't remember in this moment how much it is).
The result in my case is that the first head and part of the second one can be drawn using the 3d hardware acceleration, the remaining part of the second head cannot.
This is not a bug: it is how this video card is designed for.
The bug is: in case of this card, any attempt to draw anything beyond the available 3d hardware specification of the card results in almost-random garbage drawn on that part of the second screen.
This is not beautiful, nor useful being a screen saver used to avoid a permanent pattern drawn on the same area of the screen.
I don't know if the "responsible" for this is to be considered the radeon driver or the application drawing the data.
In any case, if there is a better place where i can post this bug I'm ready to do it.

Revision history for this message
Harald Sitter (apachelogger) wrote :

I still think this should be implemented in the driver or at least X11 rather than any specific application, implementing a protection in every application using 3D is 100% unproductive assuming it can be down on a lower level for all the applications at once.

As said, please report the bug at http://bugs.kde.org and paste the bug URL here.

By the way, the screen saver use case of avoiding a permanent pattern drawn on the same area of screen isn't existing for at least 8 years. ;-)

Revision history for this message
Moreno Bartalucci (moreno-bartalucci) wrote :

I fully agree with you in both the subjects :-)

Anyway I was not sure where to post this bug, so I started posting it for the package which showed the problem to me.

It's a little bit frustrating, however, knowing, after 1 and half year that I posted 'my' bug in the wrong place.

In any case I can only say thanks to everybody who made ubuntu a reality.

Ubuntu made me enter the linux world and, even if I can't say I'll never leave it, I'm sure I'll never go back.

Revision history for this message
Harald Sitter (apachelogger) wrote :

Your bug seems to be a problem with the KDE program itself, and not with our KDE packages. While we appreciate your issue, it would be better if it was tracked at https://bugs.kde.org, so that the KDE developers can deal with this speedily and have direct communication with you as the reporter for more effective debugging.

 Thanks!

Changed in kdeartwork (Ubuntu):
status: New → Invalid
Revision history for this message
Moreno Bartalucci (moreno-bartalucci) wrote :

As I don't own the computer showing this problem anymore, I can't go on with this.

To me this bug report can be erased: it's been completely useless.

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.