Tuesday 22 April 2014

Tech Feature: HPSL Shading Language


Overview

HPL3 is our first engine to support both PC and consoles. To make it easy to support multiple platforms and multiple shading languages we have decided to use our own shading language called HPSL. Shader code written in HPSL goes through a shader parser to translate it to the language used by the hardware.
The shader written in HPSL is loaded into the engine at runtime, the code is then run through a preprocess parser that strips away any code that is not needed by the effect or material. After that the stripped code is translated to the language used by the hardware (GLSL #330 on PC and PSSL on the PS4) and then compiled.
HPSL uses the same syntax as the scripting or engine code. HPSL is based on GLSL #330 but some of the declarations are closer to HLSL.

// Example code
@ifdef UseTexture
uniform cTexture2D aColorMap : 0;
@endif

void main(in cVector4f px_vPosition,
                in cVector4f px_vColor,
                in cVector4f px_vTexCoord0,
                out cVector4f out_vColor : 0)
{
          cVector4f vColor = px_vColor;


@ifdef UseTexture
                    vColor *= sample(aColorMap, px_vTexCoord0.xy);
          @endif

          out_vColor = vColor;
}

//Preproccess step
void main(in cVector4f px_vPosition,
                in cVector4f px_vColor,
                in cVector4f px_vTexCoord0,
                out cVector4f out_vColor : 0)
{
          cVector4f vColor = px_vColor;

          out_vColor = vColor;
}

// Translation step
#version 330
#extension GL_ARB_explicit_attrib_location : enable

in vec4 px_vColor;
in vec4 px_vTexCoord0;
layout(location = 0) out vec4 out_vColor;

void main()
{
          vec4 px_vPosition = gl_FragCoord;
          bool px_bFrontFacing = gl_FrontFacing;
          int px_lPrimitiveID = gl_PrimitiveID;

          vec4 vColor = px_vColor;

          out_vColor = vColor;
}

Preprocessing

All the shader code used in SOMA is handwritten. In order to keep all the relevant code at the same place and to be able to quickly optimize shaders HPL3 uses a preprocessing step. This has been used for our previous games as well. A preprocessor goes thorugh the code and removes large chunks that are not needed or used by the effect or material. The lighting shader used in SOMA contains code used by all the different light types. Changing a preprocess variable can change a light from a point light to a spotlight or can be used to enable shadow mapping. The preprocessor strips blocks of code that are not used, this increases performance since code that has no visual effects is removed completely. Another feature of the preprocess parser is the ability to change the value of a constant variable, this can be used to change the quality of an effect.

// SSAO code
for(float d = 0.0; d < $kNumSamples; d+=4.0)
{
          // perform SSAO…
}
The preprocessor makes it easy to do complex materials with multiple textures and shading properties while only performing the heavy computations for the materials that need it.

Translation

After the preprocess strips the code it is ready to get translated. In the first step all the variable types and special functions are converted to the new language. Then the main entry function is created and all the input and output is bound to the correct semantics. In the last step the translated code is scanned for texture and buffers that get bound to the correct slot. 

Compilation

The translated code is then compiled. If a compilation error occurred the translated code is printed to the log file along with the error message and corresponding row for easy debugging.

Summary

In order to deliver the same visual experience to all platforms and to make development faster we decided on using our own shading language. The code is translated to the language used by the hardware and compiled at runtime. Supporting other shading languages in the future will be very easy since we only need to add another converter. 
HPSL translates to GLSL #330 which requires OpenGL 3.3 (DirectX 10 feature set). This means that SOMA will require a DirectX 10 or newer graphic card.
Modders will still be able to write shader code directly in GLSL if they chose to.

HPSL Reference

Syntax

HPSL uses the same syntax used by the scripting language.
 
Variable Type
Description
int
32 bit signed integer
uint
32 bit unsigned integer
bool
Stores true or false
float
32 bit float
double
64 bit float
cVectorXf
Vector of floats
cVectorXl
Vector of signed integers
cVectorXu
Vector of unsigned intergers
cMatrixXf
Square float matrix
cMatrixXxXf
Non-square matrix (Ex cMatrix2x4f)
cBuffer
Container of multiple variables that get set by the CPU


Texture Type
Description
cTexture1D
Single dimension texture
cTexture2D
Standard 2D texture
cTexture3D
Volume texture
cTextureCube
Cubemap texture
cTextureBuffer
A large single dimension texture used to store variables
cTexture2DMS
A 2D render target with MSAA support
cTextureXCmp
A shadow map texture used for comparison operations
cTextureXArray
Array of cTextureX textures

A texture contains both the image and information about what happens when it is sampled. If you are used to OpenGL/GLSL then this is nothing new. DirectX uses a different system for storing this information. It uses a texture for storing the data and a separate sampler_state that controls filtering and clamping. Using the combined format makes it easy to convert to either GLSL or HLSL.
Textures need to be bound to a slot at compilation time. Binding is done by using the “:” semantic after the texture name.

//bind diffuse map to slot 0
uniform cTexture2D aDiffuseMap : 0;

Variable Type Modifier
Description
uniform
A variable or texture that is set by the CPU
in
Read only input to a function
out
Output of a function
inout
Read and write input and output to a function
const
A constant value that must be initialized in the declaration and can’t be changed

Entry Point and Semantics

The entry point of a shader program is the “void main” function. Input and output of the shader is defined as arguments to this function. The input to the vertex shader comes from the mesh that is rendered. This might be information like the position, color and uv mapping of a vertex. What the vertex shader outputs is user defined, it can be any kind of information that the pixel shader needs. The output of the vertex shader is what gets sent to the pixel shader as input. The variables are interpolated between the vertices of the triangle. The input of the pixel shader and the output of the vertex shader must be the same or else the shaders won’t work together. Finally the output of the pixel shader is what is shown on the screen. The pixel shader can output to a of maximum 4 different render targets at the same time.
Some of the input and output are System defined semantics. System Semantics are set or used by the hardware. 

System Semantic
Description
Type
Shader Type
px_vPosition
Vertex position output. Pixel shader input as screen position. This is required by all shaders
cVector4f
Vertex (out), Pixel (in)
: X
Output color slot, where X must be in the range 0-3
cVector4
Pixel (out)
vtx_lVertexID
Index of the current vertex
int
Vertex (in)
vtx_lInstanceID
Index of the current instance
int
Vertex (in)
px_lPrimitiveID
Index of the triangle this pixel belongs to
int
Pixel (in)
px_bFrontFacing
Indicates if the pixel belongs to the front or back of the primitive
bool
Pixel (in)

Input to the vertex shader is user defined. HPL3 has a few user defined semantics that work with our mesh format.

Mesh Semantic
Description
Type
vtx_vPosition
Position of the vertex
cVector4f
vtx_vTexCoord0
Primary UV coord
cVector4f
vtx_vTexCoord1
Secondary UV coord
cVector4f
vtx_vNormal
World space normal
cVector3f
vtx_vTangent
World space tangent, w contains binormal direction
cVector4f
vtx_vColor
Color
cVector4f
vtx_vBoneIndices
Index of the bones used to modify this vertex
cVector4l
vtx_vBoneWeight
Weight to multiply the bones with
cVector4f

It is possible to add more user defined semantics if needed
//vertex shader
uniform cMatrixf a_mtxModelViewProjection;

void main(in cVector4f vtx_vPosition,
               in cVector4f vtx_vColor,
               in cVector4f vtx_vTexCoord0,
               out cVector4f px_vColor,
               out cVector4f px_vTexCoord0,
              out cVector4f px_vPosition)
{                          
          px_vPosition = mul(a_mtxModelViewProjection, vtx_vPosition);
          px_vColor = vtx_vColor;
                             px_vTexCoord0 = vtx_vTexCoord0;
}

//pixel shader
uniform cTexture2D aColorMap : 0;

void main(in cVector4f px_vPosition,
               in cVector4f px_vColor,
               in cVector4f px_vTexCoord0,
               out cVector4f out_vColor : 0)
{
         out_vColor = px_vColor * sample(aColorMap, px_vTexCoord0.xy);
}

Functions

HPSL is based on OpenGL 3.3 and GLSL version 330 and supports almost all of the GLSL arithmetic functions.
There are some functions that are different from GLSL. This is to make it easier to support HLSL and PSSL.

Arithmetic Function
Description
mul(x, y)
Multiplies two matrices together (multiplying by using * not supported for matrices)
lerp(x, y, t)
Interpolates between two values

Texture sampling use functions specific to the HPSL language.

Texture Function
Description
sample(texture, uv)
sample(texture, uv, offset)
Samples a texture at the specified uv coordinate. Can be used with an integet offset
sampleGather(texture, uv)
sampleGather(texture, uv, offset)
Samples a texture but returns only the red component of each texel corner
sampleGrad(texture, uv, dx, dy)
sampleGrad(texture, uv, dx, dy, offset)
Performs texture lookup with explicit gradients
sampleLod(texture, uv, lod)
sampleLod(texture, uv, lod, offset)
Samples the texture at a specific mipmap level
sampleCmp(texture, uv, comp_value)
sampleCmp(texture, uv, comp_value, offset)
Performs texture lookup and compares it with the comparison value and returns result
load(texture, position)
Gets the value of a texel at the integer position
getTextureSize(texture, lod)
Returns the width and height of the texture lod
getTextureLod(texture, uv)
Gets the lod that would get sampled if that uv coord is used
getTextureLevelCount
Gets the number of MipMap levels

It is also possible to use language specific code directly. Some languages and graphic cards might have functions that are more optimized for those systems and then it might be a good idea to write code specific for that language.

@ifdef Lang_GLSL
                  vec4 vModifier = vec4(lessThan(vValue, vLimit));
@else
                  cVector4f vModifier = step(vValue, vLimit);
@endif


5 comments:

  1. Nice idea to write a higher level shader language which translates to different graphics API shader languages for portability. Did you write the parser from scratch or is it built on top of another parser engine (and only define grammar)?

    I like your tech articles, very informative - and your games of course (played all three Penumbra games the last easter days, next to be played is AAMFP). Keep up the good work! :D I slowly become a Frictional Games fan. ^^

    Btw could you also write a tech article about Occlusion Culling / Visibility Determination (and maybe Content Streaming) in HPL3? As I've read, you now support indoor and large outdoor scenes and I am interested in which techniques you use for both scene types and how you manage seamless transition between indoor and outdoor.

    HPL1 uses portal culling and HPL2 CHC, right (Wiki says that ;D)? Do you use any preprocessing in your engines like generating a PVS or anything like that?

    ReplyDelete
    Replies
    1. Thanks.
      The parser is built from scratch. The preprocessor part has been in the engine since HPL2 but the translation part is new for HPL3.

      HPL3 uses CHC culling just as HPL2. There have been some improvements to it and Im going to add some more improvements later, so i might make a blog post about that. The only thing we bake is static geometry and physics, but that is baked when you start a map the first time (it has been prebaked when the game is released)

      Delete
  2. OMG, I didn't expect this - this is so awesome from the perspective of the modding community. Great job!

    ReplyDelete
  3. I read on Thomas' twitter that you're thinking about supporting DirectX or running on Xbox One. Is that the reason you're mentioning HLSL in the article? Would DirectX be solely for Xbox or would that be an option on PC, too?

    ReplyDelete
  4. I read on Thomas' twitter that you're thinking about supporting DirectX

    ReplyDelete

Note: only a member of this blog may post a comment.