Segfault on running FreeOrion in linux

Problems and solutions for installing or running FreeOrion, including discussion of bugs if needed before posting a bug report on GitHub. For problems building from source, post in Compile.

Moderator: Oberlus

Message
Author
User avatar
kroddn
Static Linker
Posts: 347
Joined: Thu Jun 28, 2007 10:28 am

Re: Segfault on running FreeOrion in linux

#16 Post by kroddn »

Geoff the Medio wrote:It was always "mostly statically linked" as far as I was aware... (although this does make it significantly less so...)..
Of course, but the only dependencies where /usr/lib/libGL and similar libs, which naturally cannot be linked statically, because with different graphic card drivers come differenct libGLs.

I am looking forward to having OGRE plugins linked in statically. The download size increased from 129MB to 163MB now, having many shared libs shipped with it.

Does the windows version run on computers using ATI graphic cards?

User avatar
Geoff the Medio
Programming, Design, Admin
Posts: 12802
Joined: Wed Oct 08, 2003 1:33 am
Location: Munich

Re: Segfault on running FreeOrion in linux

#17 Post by Geoff the Medio »

I don't recall seeing any obviously ATI-specific bug reports, but haven't tested it on an ATI machine myself. It (mostly) runs on my laptop onboard intel GMA card, though, which should be less powerful than almost anything made by ATI in the last few years.

User avatar
OndrejR
Space Dragon
Posts: 339
Joined: Thu Oct 02, 2008 11:00 pm
Location: Slovakia

Re: Segfault on running FreeOrion in linux

#18 Post by OndrejR »

According this Ogre manual from 2006 ATI doesn't support vertex fetch in their cards (Radeon X1n00). nVidia do support it in both their 6600 and above range. ATI support an alternative called 'Render to Vertex Buffer', but this is not standardized(in 2006) and is very much different in its implementation, so cannot be considered to be a drop-in replacement. This is the case even though the Radeon X1n00 cards claim to support shader model 3 (which requires vertex texture fetch).

And according Vertex and Fragment Programs Manual hardware support for vertex texture fetching is not ubiquitous and says to use the uses_vertex_texture_fetch directive when declaring vertex programs that use vertex textures, so that if the client hardware does not support it, technique fallback can be enabled, as described in manual.

It is not required to use this attribute for DirectX-targeted shaders, since vertex texture fetching is only supported in OpenGL shader model 3 and above, and can state this as a required syntax in the shader definition.

So the question is: If it is still true that current ATI drivers and newest graphics says that it support shader model 3 and in fact it doesn't is not FreeOrion fault that FreeOrion doesn't run on ATI cards. But maybe switching to fragment directive via using above directive for vertex texture fetching is solution to this problem. Another solution is to try open source ATI drivers.

I can't test this because I don't have ATI.

And another question is if nvidia-cg-toolkit is required on ATI or there is some alternative.

User avatar
kroddn
Static Linker
Posts: 347
Joined: Thu Jun 28, 2007 10:28 am

Re: Segfault on running FreeOrion in linux

#19 Post by kroddn »

Maybe we should open another discussion thread for those topics instead of posting them here?

For me the question should be: how can we have FreeOrion run on more platforms - currently, the limitation to nvidia chipsets seems not to be a good way.

User avatar
francys
Space Squid
Posts: 54
Joined: Sat Oct 25, 2008 8:53 pm
Location: Victoria, BC

Re: Segfault on running FreeOrion in linux

#20 Post by francys »

kroddn wrote:For me the question should be: how can we have FreeOrion run on more platforms - currently, the limitation to nvidia chipsets seems not to be a good way.
Having the same issue, I cannot but agree with you.

User avatar
kroddn
Static Linker
Posts: 347
Joined: Thu Jun 28, 2007 10:28 am

Re: Segfault on running FreeOrion in linux

#21 Post by kroddn »

I just testet FreeOrion on a windows machine with an ATI card. The game crashes, just like the linux version, after the main window appears (keeps black).

oreg log added.
Attachments
ogre_log.zip
(3.13 KiB) Downloaded 59 times

User avatar
OndrejR
Space Dragon
Posts: 339
Joined: Thu Oct 02, 2008 11:00 pm
Location: Slovakia

Re: Segfault on running FreeOrion in linux

#22 Post by OndrejR »

I think that this problem should be answered on Ogre forum where are a lot of Ogre programmers. On this forum it is only tzlaine.

User avatar
OndrejR
Space Dragon
Posts: 339
Joined: Thu Oct 02, 2008 11:00 pm
Location: Slovakia

Re: Segfault on running FreeOrion in linux

#23 Post by OndrejR »

I found some ogre log for ATI HD3870 on Ogre forum on Windows platform and in this log vertex texture fetch is supported.

Code: Select all

20:51:51: RenderSystem Name: OpenGL Rendering Subsystem
20:51:51: GPU Vendor: ati
20:51:51: Device Name: ATI Radeon HD 3870
20:51:51: Driver Version: 2.1.7769.0
20:51:51: * Fixed function pipeline: yes
20:51:51: * Hardware generation of mipmaps: yes
20:51:51: * Texture blending: yes
20:51:51: * Anisotropic texture filtering: yes
20:51:51: * Dot product texture operation: yes
20:51:51: * Cube mapping: yes
20:51:51: * Hardware stencil buffer: yes
20:51:51: - Stencil depth: 8
20:51:51: - Two sided stencil support: yes
20:51:51: - Wrap stencil values: yes
20:51:51: * Hardware vertex / index buffers: yes
20:51:51: * Vertex programs: yes
20:51:51: * Fragment programs: yes
20:51:51: * Geometry programs: no
20:51:51: * Supported Shader Profiles: arbfp1 arbvp1 glsl ps_1_1 ps_1_2 ps_1_3 ps_1_4
20:51:51: * Texture Compression: yes
20:51:51: - DXT: yes
20:51:51: - VTC: no
20:51:51: * Scissor Rectangle: yes
20:51:51: * Hardware Occlusion Query: yes
20:51:51: * User clip planes: yes
20:51:51: * VET_UBYTE4 vertex element type: yes
20:51:51: * Infinite far plane projection: yes
20:51:51: * Hardware render-to-texture: yes
20:51:51: * Floating point textures: yes
20:51:51: * Non-power-of-two textures: yes
20:51:51: * Volume textures: yes
20:51:51: * Multiple Render Targets: 4
20:51:51: - With different bit depths: yes
20:51:51: * Point Sprites: yes
20:51:51: * Extended point parameters: yes
20:51:51: * Max Point Size: 8192
20:51:51: * Vertex texture fetch: yes
20:51:51: - Max vertex textures: 16
20:51:51: - Vertex textures shared: yes
20:51:51: * Render to Vertex Buffer : no
20:51:51: * GL 1.5 without VBO workaround: no
20:51:51: * Frame Buffer objects: yes
20:51:51: * Frame Buffer objects (ARB extension): no
20:51:51: * Frame Buffer objects (ATI extension): no
20:51:51: * PBuffer suppport: no
20:51:51: * GL 1.5 without HW-occlusion workaround: no
20:51:51: Registering ResourceManager for type Texture
20:51:51: ResourceBackgroundQueue - threading disabled
20:51:51: Particle Renderer Type 'billboard' registered
20:51:51: SceneManagerFactory for type 'BspSceneManager' registered.
20:51:51: Registering ResourceManager for type BspLevel
20:51:51: SceneManagerFactory for type 'OctreeSceneManager' registered.
20:51:51: SceneManagerFactory for type 'TerrainSceneManager' registered.
20:51:51: Creating viewport on target 'TEmpire', rendering from camera 'camera', relative dimensions 8192L: 0 T: 0 W: 1 H: 1 ZOrder: 0
20:51:51: Viewport for camera 'camera', actual dimensions 8192L: 0 T: 0 W: 640 H: 480
20:51:51: Parsing scripts for resource group Autodetect
20:51:51: Finished parsing scripts for resource group Autodetect
20:51:51: Parsing scripts for resource group General
20:51:51: Parsing script terrain.material
20:51:51: Parsing script tools.material
20:51:51: Parsing script trees.material
20:51:51: Parsing script gui.material
20:51:51: Parsing script te.fontdef
20:51:51: Finished parsing scripts for resource group General
20:51:51: Parsing scripts for resource group Internal
20:51:51: Finished parsing scripts for resource group Internal
This means that your graphic card is outdated.

User avatar
kroddn
Static Linker
Posts: 347
Joined: Thu Jun 28, 2007 10:28 am

Re: Segfault on running FreeOrion in linux

#24 Post by kroddn »

OndrejR wrote:This means that your graphic card is outdated.
Puh. Thats hard to hear. And not really an satisfying answer, as this graphic chip is even a never one than my nvidia chip.

Post Reply