marko10_000
/
openclonk
forked from Mirrors/openclonk
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

mesh: Implement texturing 

We really need a proper way to handle lighting. The current way is way too
random and just chosen so that the objects I'm testing with look good.
stable-5.2
Armin Burgmeier 2009-09-09 22:54:12 -04:00
parent 95b4738d66
commit fe34708fd6
1 changed files with 51 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

55
src/platform/StdGL.cpp
View File

@ -34,6 +34,10 @@
#include <math.h>
#include <limits.h>
// TODO: Are the GL constants guaranteed to be continuous, so we can just
// use GL_TEXTURE0 + i instead of TEXNUMS[i]?
static const int TEXNUMS[] = { GL_TEXTURE0, GL_TEXTURE1, GL_TEXTURE2, GL_TEXTURE3, GL_TEXTURE4, GL_TEXTURE5, GL_TEXTURE6, GL_TEXTURE7, GL_TEXTURE8, GL_TEXTURE9, GL_TEXTURE10, GL_TEXTURE11, GL_TEXTURE12, GL_TEXTURE13, GL_TEXTURE14, GL_TEXTURE15, GL_TEXTURE16, GL_TEXTURE17, GL_TEXTURE18, GL_TEXTURE19, GL_TEXTURE20, GL_TEXTURE21, GL_TEXTURE22, GL_TEXTURE23, GL_TEXTURE24, GL_TEXTURE25, GL_TEXTURE26, GL_TEXTURE27, GL_TEXTURE28, GL_TEXTURE29, GL_TEXTURE30, GL_TEXTURE31 };
static void glColorDw(DWORD dwClr)
{
glColor4ub(GLubyte(dwClr>>16), GLubyte(dwClr>>8), GLubyte(dwClr), GLubyte(dwClr>>24));
@ -125,7 +129,11 @@ bool CStdGL::UpdateClipper()
glViewport(iX, RenderTarget->Hgt-iY-iHgt, iWdt, iHgt);
glMatrixMode(GL_PROJECTION);
glLoadIdentity();
gluOrtho2D((GLdouble) iX, (GLdouble) (iX+iWdt), (GLdouble) (iY+iHgt), (GLdouble) iY);
// Set clipping plane to -1000 and 1000 so that large meshes are not
// clipped away.
glOrtho((GLdouble) iX, (GLdouble) (iX+iWdt), (GLdouble) (iY+iHgt), (GLdouble) iY, -1000.0f, 1000.0f);
//gluOrtho2D((GLdouble) iX, (GLdouble) (iX+iWdt), (GLdouble) (iY+iHgt), (GLdouble) iY);
//gluOrtho2D((GLdouble) 0, (GLdouble) xRes, (GLdouble) yRes, (GLdouble) yRes-iHgt);
return true;
}
@ -379,14 +387,35 @@ void CStdGL::PerformMesh(StdMeshInstance &instance, float tx, float ty, float tw
const float dy = ty + ry*thgt;
const float scx = twdt/(box.x2 - box.x1);
const float scy = thgt/(box.y2 - box.y1);
#if 0
// Scale so that Z coordinate is between -1 and 1, otherwise parts of
// the mesh could be clipped away by the near or far clipping plane.
// This technique might also enable us not to clear the depth buffer
// after every mesh rendering - we could simply scale the first mesh
// of the scene so that it's Z coordinate is between 0 and 1, scale
// the second mesh that it is between 1 and 2, and so on.
// This of course requires an orthogonal projection so that the
// meshes don't look distorted - if we should every decide to use
// a perspective projection we need to think of something different.
// TODO: This is currently commented out because it gets the lighting
// wrong (objects are too dark). Instead we changed the clipping plane
// to be large enough. I think this is still a got idea, though,
// when we fix the lighting. Maybe we need to adapt the normals
// somehow...
const float scz = 1.0/(box.z2 - box.z1);
#else
const float scz = 1.0;
#endif
// Keep aspect ratio:
//if(scx < scy) scy = scx;
//else scx = scy;
glTranslatef(dx, dy, 0.0f);
glScalef(scx, scy, 1.0f);
glScalef(scx, scy, scz);
// Put a light source in front of the object
const GLfloat light_position[] = { 0.0f, 0.0f, 15.0f*Zoom, 1.0f };
const GLfloat light_position[] = { 0.0f, 0.0f, box.z2 + 15.0f*Zoom, 1.0f };
glLightfv(GL_LIGHT0, GL_POSITION, light_position);
glEnable(GL_LIGHT0);
@ -416,6 +445,13 @@ void CStdGL::PerformMesh(StdMeshInstance &instance, float tx, float ty, float tw
{
const StdMeshMaterialPass& pass = technique.Passes[i];
// TODO: Test this when the material is loaded, and mark the
// technique as non-working if it has passes with more
// textures than this number.
GLint max_texture_units;
glGetIntegerv(GL_MAX_TEXTURE_UNITS, &max_texture_units);
assert(pass.TextureUnits.size() <= max_texture_units);
// Set up material
#if 0
glMaterialfv(GL_FRONT, GL_AMBIENT, pass.Ambient);
@ -425,7 +461,12 @@ void CStdGL::PerformMesh(StdMeshInstance &instance, float tx, float ty, float tw
glMaterialfv(GL_FRONT, GL_EMISSION, pass.Emissive);
glMaterialf(GL_FRONT, GL_SHININESS, pass.Shininess);
// TODO: Set up texture units
for(unsigned int j = 0; j < pass.TextureUnits.size(); ++j)
{
glActiveTexture(TEXNUMS[j]);
glEnable(GL_TEXTURE_2D);
glBindTexture(GL_TEXTURE_2D, pass.TextureUnits[j].GetTexture().texName);
}
// Render mesh
// TODO: Use glInterleavedArrays? Hm, might be impossible as
@ -448,6 +489,12 @@ void CStdGL::PerformMesh(StdMeshInstance &instance, float tx, float ty, float tw
RenderMeshVertex(vtx3, pass, &Transform, ClrModMap, dwModClr);
}
glEnd(); // GL_TRIANGLES
for(unsigned int j = 0; j < pass.TextureUnits.size(); ++j)
{
glActiveTexture(TEXNUMS[j]);
glDisable(GL_TEXTURE_2D);
}
}
glDisable(GL_LIGHT0);