GPU Shading and Rendering: OpenGL Shading Language

1
(No Transcript)
2
GPU Shading and RenderingOpenGL Shading Language

• Marc Olano
• UMBC

3
OpenGL Shading

• High level language
• OpenGL Shading Language GLslang GLSL
• Integrated into OpenGL API (no extra run-time)

4
Organization

• API
• Vertex Shading
• Fragment Shading
• Lots of demos.
• 2-year old Apple PowerBook G4/1.5GHz
• ATI Mobility Radeon 9700

5
API-integrated

• Compiler built into driver
• Presumably they know your card best
• IHVs must produce (good) compilers
• Use built-in parameters (glColor, glNormal, )
• Add your own
• Other options can still produce low-level code
• Cg, ASHLI, RapidMind,
• With loss of integration

6
Using High-level Code

• Create shader objectS glCreateShader(GL_VERTEX_
  SHADER)S glCreateShaderObjectARB(GL_VERTEX_SHAD
  ER_ARB)
• Vertex or Fragment
• Load shader into objectglShaderSource(S, n,
  shaderArray, lenArray)glShaderSourceARB(S, n,
  shaderArray, lenArray)
• Array of strings
• Compile objectglCompileShader(S)glCompileShaderA
  RB(S)

7
Loading Shaders

• glShaderSource(S, n, shaderArray, lenArray)
• One string containing entire mmapd file
• Strings as includes
• Varying variables between vertex and fragment
• Strings as lines
• Null-terminated if lenArray is Null or length-1

8
Using High-level Code (2)

• Create program objectP glCreateProgram()P
  glCreateProgramObjectARB()
• Attach all shader objectsglAttachShader(P,
  S)glAttachObjectARB(P, S)
• Vertex, Fragment or both
• Link togetherglLinkProgram(P)glLinkProgramARB(P)
  
• UseglUseProgramObject(P)glUseProgramObjectARB(P)

9
Using High-level Code (3)

• Where is my attributes/uniforms
  parameter?iglGetAttribLocation(P,myAttrib)ig
  lGetUniformLocation(P,myAttrib)
• Set themglVertexAttrib1f(i,value)glVertexAttribP
  ointer(i,)glUniform1f(i,value)

10
Using Low-level Code

• Load shaderglProgramStringARB(GL_VERTEX_PROGRAM_A
  RB, GL_PROGRAM_FORMAT_ASCII_ARB, length, shader)
• Vertex or fragment
• Single string (vs. array)
• EnableglEnable(GL_VERTEX_PROGRAM_ARB)

11
Useful Tools

• Shader debugger
• Immediate updates
• Choose model/texture
• Tweak parameters
• Examine/dump frames
• Several available
• Not hard to build

• OpenGL debugger
• Trace of calls made
• Examine resources
• Breakpoints/actions
• Graph performance
• A couple of choices

12
gDEBugger A Professional OpenGL Debugger and
Profiler

• Provides graphic pipeline information needed to
  find bugs and to optimize application
  performance
• Shortens debugging and profiling time
• Improves application quality
• Optimizes application performance

13
Free gDEBugger License for Academic Users!

• OpenGL ARB and Graphic Remedy Academic Program
• Annual program for all OpenGL Academic users
• License of the full feature version for one year
• Includes all software updates
• A limited number of free licenses available
  fornon-commercial developers who are not in
  academia
• More details http//academic.gremedy.com

14
Non-windows OS

• Linux
• gDEBugger in progress
• Apple OpenGL Profiler and Driver Monitor
• Free part of OS / Developer tools

15
Vertex DemoBlend Positions
16
High-level Code
void main() float Kin gl_Color.r
// key input // screen position from vertex
and texture vec4 Vp ftransform() vec4
Tp vec4(gl_MultiTexCoord0.xy1.8-.9, 0.,1.)
// interpolate between Vp and Tp
gl_Position mix(Tp,Vp,pow(1.-Kin,8.)) //
copy to output gl_TexCoord0
gl_MultiTexCoord0 gl_TexCoord1 Vp
gl_TexCoord3 vec4(Kin)
17
Main Function
void main() float Kin gl_Color.r
// key input // screen position from vertex
and texture vec4 Vp ftransform() vec4
Tp vec4(gl_MultiTexCoord0.xy1.8-.9, 0.,1.)
// interpolate between Vp and Tp
gl_Position mix(Tp,Vp,pow(1.-Kin,8.)) //
copy to output gl_TexCoord0
gl_MultiTexCoord0 gl_TexCoord1 Vp
gl_TexCoord3 vec4(Kin)
18
Use Standard OpenGL State
void main() float Kin gl_Color.r
// key input // screen position from vertex
and texture vec4 Vp ftransform() vec4
Tp vec4(gl_MultiTexCoord0.xy1.8-.9, 0.,1.)
// interpolate between Vp and Tp
gl_Position mix(Tp,Vp,pow(1.-Kin,8.)) //
copy to output gl_TexCoord0
gl_MultiTexCoord0 gl_TexCoord1 Vp
gl_TexCoord3 vec4(Kin)
19
Built-in Types
void main() float Kin gl_Color.r
// key input // screen position from vertex
and texture vec4 Vp ftransform() vec4
Tp vec4(gl_MultiTexCoord0.xy1.8-.9, 0.,1.)
// interpolate between Vp and Tp
gl_Position mix(Tp,Vp,pow(1.-Kin,8.)) //
copy to output gl_TexCoord0
gl_MultiTexCoord0 gl_TexCoord1 Vp
gl_TexCoord3 vec4(Kin)
20
Swizzle / Channel Selection
void main() float Kin gl_Color.r
// key input // screen position from vertex
and texture vec4 Vp ftransform() vec4
Tp vec4(gl_MultiTexCoord0.xy1.8-.9, 0.,1.)
// interpolate between Vp and Tp
gl_Position mix(Tp,Vp,pow(1.-Kin,8.)) //
copy to output gl_TexCoord0
gl_MultiTexCoord0 gl_TexCoord1 Vp
gl_TexCoord3 vec4(Kin)
21
Vector Construction
void main() float Kin gl_Color.r
// key input // screen position from vertex
and texture vec4 Vp ftransform() vec4
Tp vec4(gl_MultiTexCoord0.xy1.8-.9, 0.,1.)
// interpolate between Vp and Tp
gl_Position mix(Tp,Vp,pow(1.-Kin,8.)) //
copy to output gl_TexCoord0
gl_MultiTexCoord0 gl_TexCoord1 Vp
gl_TexCoord3 vec4(Kin)
22
Built-in Functions
void main() float Kin gl_Color.r
// key input // screen position from vertex
and texture vec4 Vp ftransform() vec4
Tp vec4(gl_MultiTexCoord0.xy1.8-.9, 0.,1.)
// interpolate between Vp and Tp
gl_Position mix(Tp,Vp,pow(1.-Kin,8.)) //
copy to output gl_TexCoord0
gl_MultiTexCoord0 gl_TexCoord1 Vp
gl_TexCoord3 vec4(Kin)
23
Vertex Fragment DemoFresnel Environment Map
24
Trick 1 Where is the Eye
ObjectSpace
ModelViewMatrix
EyeSpace
Projection Matrix
ClipSpace

• Where is the Eye in Eye Space?
• (0,0,0)? Not necessarily!
• Know where it is in Clip Space
• (0,0,-1,0), looking in the (0,0,1,0) direction
• Assuming GL_LESS depth test
• Invert projection to find the eye!
• Works for any eye position, or even parallel
  projection.

25
Trick 2 Subtract Homogeneous Points

• Homogeneous point vec4(V.xyz, V.w)
• 3D equivalent V.xyz/V.w
• Defers division, makes perspective, translation,
  and many things happy
• Vector subtraction VE
• V.xyz/V.w E.xyz/E.w
• (V.xyzE.w E.xyzV.w)/(V.wE.w)

26
Trick 3 Skip Division for Normalize

• normalize(V.xyz/V.w) normalize(V.xyz)
• If V.w isnt negative
• Put it all together
• normalize(V-E)
• normalize(V.xyzE.w - E.xyzV.w)

27
OpenGL State DemoVertex Lighting
28
Lighting Vectors in Eye Space
void main() // convert shading-related
vectors to eye space vec4 P
gl_ModelViewMatrixgl_Vertex vec4 E
gl_ProjectionMatrixInversevec4(0,0,-1,0) vec3
V normalize(E.xyzP.w-P.xyzE.w) vec3 N
normalize(gl_NormalMatrixgl_Normal)
29
Accumulate Each Light
// accumulate contribution from each light
gl_FrontColor vec4(0) for(int i0
iltgl_MaxLights i) vec3 L
normalize(gl_LightSourcei.position.xyzP.w
- P.xyzgl_LightSourcei.position.w)
vec3 H normalize(LV) float diff
dot(N,L) gl_FrontColor
gl_LightSourcei.ambient if (diff gt 0.)
gl_FrontColor
gl_LightSourcei.diffuse diff
gl_FrontColor gl_LightSourcei.specul
ar max(pow(dot(N,H), gl_FrontMaterialShinines
s),0.)
30
Standard Vertex Shader Stuff
// standard texture coordinate and position
stuff gl_TexCoord0 gl_TextureMatrix0gl_Mu
ltiTexCoord0 gl_Position ftransform()

31
Noise

• Controlled, repeatable randomness
• Still spotty implementation
• Can use texture or compute

32
Noise Characteristics

• Repeatable
• Locally continuous but distant points
  uncorrolated
• values -1,1, average 0
• 1/2  1 cycle per unit
• Versions for n-D input

33
Noise Subtleties

• Many noise functions based on a lattice
• Like a spline between integer coordinates
• Hash of integer coordinates ? control points
• Interpolating values easy but poor
• Even with higher-order interpolation
• Perlins noise
• Passes through 0 at each integer
• Hash gives gradient

34
Modified Noise Olano 2005

• Three relatively independent modifications
• New computable hash
• Change gradient computation
• Reorder computation
• Variety of computation/texture options
• Can just store in a texture
• Can compute with some texture accesses
• Can compute with no texture accesses

35
Computable Hash

• Normal hash chains access to permutation texture
• Want totally computable hash
• mod(kx2, m)
• Still chain for higher-D
• hash(floor(P.x) hash(floor(P.y)))
• Not quite as good, but cheap computable
• Noise usually not used alone

36
Gradient

• 3D Gradient (fract(P.x), fract(P.y))
• Each sign from one bit of hash
• Made slightly more difficult without bitwise ops
• Allows noise(x) noise(x,0)
• If 2D noise is stored in a texture
• Can share the same texture for 1D noise as well
• Not normally true!

37
Reordered Computation

• Refactor to be able to build n-D noise from two
  shifted calls to n-1 D noise
• If 2D noise is stored in a texture
• Can build 3D noise from 2 texture accesses
• Can build 4D noise from 4 texture accesses

38
Shader Design Strategies

• Learn and adapt from RenderMan
• Noise
• Layers
• Multiple Passes
• Baked computation

39
(No Transcript)