Page 4 of 6
Posted: Fri Aug 21, 2009 3:16 pm
by Murloc992
Valmond wrote:Murloc992, could you tell me what your ping is (the first ping is marked in the output window, "Got server time 14346843, ping=123)?
and also, can I send yoy another shader to see if it's the lightmap or the triplanar mapping that is causing all the black parts?
Lonesome Ducky, I have added debug output to a new version if you want to test it (which would be very cool if you do
)
[edit] just FYI the game runs on Open GL at the moment, haven't been able to recompile Nadros excellent
Cg binding with DirectX (well, when I recompile the Cg binding it doesn't work so for the moment it's OpenGL only).
Ping is 100+- always.
Well i am on linux at the moment (switched) and can't test the game, sorry
Posted: Fri Aug 21, 2009 4:00 pm
by Lonesome Ducky
Oh ok, it's having trouble compiling the shader. My video card is an ATI Radeon x1650 pro.
Posted: Fri Aug 21, 2009 7:11 pm
by Nadro
Hi, I try run Your game, but I had 2 problems:
1. Error on compiling shader (Radeon HD4850)
2. Game crash after: " * Start Main Loop *"
First problem I solved by repair "atlas.frag" shader (You have to replace "gl_TexCoord[0]" by "vec2(gl_TexCoord[0])", but app crash still.
I try release improved version of IrrCg soon. With latest Irrlicht SVN and new IrrCg version You can will optimize Your game.
Posted: Fri Aug 21, 2009 7:19 pm
by Lonesome Ducky
Nadro wrote:Hi, I try run Your game, but I had 2 problems:
1. Error on compiling shader (Radeon HD4850)
2. Game crash after: " * Start Main Loop *"
First problem I solved by repair "atlas.frag" shader (You have to replace "gl_TexCoord[0]" by "vec2(gl_TexCoord[0])", but app crash still.
I try release improved version of IrrCg soon. With latest Irrlicht SVN and new IrrCg version You can will optimize Your game.
Same with me. If you check the top of the console it gives errors on compiling the shader.
Posted: Fri Aug 21, 2009 9:23 pm
by Murloc992
It seems that ATI cards won't 'eat' trashy shaders, while NVidia takes em' with pleasure
NVidia > ATI.
Proof: NVidia can survive till the actual gameplay.
/jokeform
Posted: Fri Aug 21, 2009 10:31 pm
by Nadro
Murloc992 wrote:It seems that ATI cards won't 'eat' trashy shaders, while NVidia takes em' with pleasure
NVidia > ATI.
Proof: NVidia can survive till the actual gameplay.
/jokeform
No in this case ATI > Nvidia, because NV drivers don't restrict GLSL specification. ATI drivers are better compatible with GLSL specification, so in my opinion ATI cards are better for programming applications, which usage OpenGL.
Posted: Sat Aug 22, 2009 12:01 am
by Valmond
@Murloc992
No problem
@Lonesome Ducky
That's weird, normally your card has these specs:
Shader Model 3.0
Complete feature set also supported in OpenGL® 2.0
What error do you get?
Do you have recent drivers (if they aren't really old that shouldn't be the problem)?
[edit]Ho I had forgotten to post my message (so I just posted it without checking if someone else had posted), Nice to see you Nadro!
[edit2] at 2 Ã clock in the morning I just can't make a good post, have to wait til the morning... thanks for the tests though and I'll continue tomorrow!
Posted: Sat Aug 22, 2009 2:24 am
by Lonesome Ducky
My drivers are at most a month old. I'm not on my regular computer right now, so I'll post the errors soon as I can get on it. But I'm really looking forward to testing the game
Posted: Sat Aug 22, 2009 12:28 pm
by Nadro
Lonesome Ducky wrote:My drivers are at most a month old. I'm not on my regular computer right now, so I'll post the errors soon as I can get on it. But I'm really looking forward to testing the game
You have the same error log that me and other AMD-ATI users, so:
Code: Select all
Fragment shader failed to compile with the following errors:
ERROR: 0:47: 'alphamap' : no matching overloaded function found
ERROR: 0:47: 'texture2D' : no matching overloaded function found
ERROR: 0:47: =' : cannot convert from 'const float' to '4-component vector of float'
...
ERROR: compilation errors. No code generated.
As I said this cause by use "gl_TexCoord[0]" instead of "vec2(gl_TexCoord[0])".
Posted: Sat Aug 22, 2009 12:30 pm
by Valmond
*ashamed* I actually loaded my old shaders (compiling them but without using them)...
I removed them and I also added quite a lot of debug output for the first run
in main so if someone want to test I will know a bit better where the app crashes.
@Lonesome Ducky
Working on it, working on it
@Nardo
I see forward to a new release
Posted: Sat Aug 22, 2009 12:40 pm
by Valmond
Posted at same time :p
and sorry again for the old shaders not working still being there
(Fixed).
Posted: Sat Aug 22, 2009 12:40 pm
by Nadro
Hi Valmond,
Application crash after:
"first run: old GUI"
In this version shaders are ok (You probably exclude atlas.frag from compilation process?).
Posted: Sat Aug 22, 2009 3:20 pm
by Valmond
Thanks this was exactly what I needed (even if I'm quite puzzled by the fact a GUI is causing the crash... it doesn't do anything at startup
)
and yes, the old shaders was compiled but not used (hence the problem) so I removed them.
I have not found the problem as it seems to be no one (
) so I opted for the idea the socket thread does
something it shouldn't or some other sync. problem and made the game load up some data from the server
before entering the main loop (ie. waiting for the clients playerdata to come from the server).
I also added some printf to the GUI in case there actually Is a bug in the GUI.
Posted: Sat Aug 22, 2009 3:26 pm
by Lonesome Ducky
Works great for me now! The gui is a little big, and the first window doesn't have a close button. Is there a way to attack?
Posted: Sat Aug 22, 2009 3:34 pm
by Valmond
Yes!
just right click on the outside of the window (this is the old GUI, not very good).
To attack, right click on a monster (the NPC:s are your 'friends', you can't attack them
) and when you are close enough you'll hit (repeatedly) automatically.
Does the landscape seems okay (no big black blobs everywhere like it did for Murloc992)?