With D3D11 and GL3, support for fixed pipeline functionality was removed. Meaning you can only render objects using shaders.
While GPU Program Scripts offer you maximal control and flexibility over how your objects are rendered, writing and maintaining them is also a very time consuming task.
The Run Time Shader System or RTSS for short is the Ogre way of managing Shaders and their variations. Initially it was created as a drop-in-replacement to the Fixed-Function Pipeline (FFP) for RenderSystems that lacked it. However, since then it grew to a general way to express shader functionality in Material Scripts without having to manually write shaders.
While the resulting shaders are less optimized, they offer the following advantages:
Save development time e.g. when your target scene has dynamic lights and the number changes, fog changes and the number of material attributes increases the total count of needed shaders dramatically. It can easily cross 100 and it becomes a time consuming development task.
Reusable code - once you've written the shader extension you can use it anywhere due to its independent nature.
Custom shaders extension library - enjoy the shared library of effects created by the community. Unlike hand written shader code, which may require many adjustments to be plugged into your own shader code, using the extensions library requires minimum changes.
For fixed function function properties, the RTSS will read the standard pass and texture_unit definitions, so no changes are required. To enable features that go beyond the possibilities of the FFP, you have to define an additional rtshader_system block with the respective properties.
For instance, the FFP only allows per-vertex lighting. To request per-pixel lighting, you would add the following block to a pass:
// RT Shader system section.
rtshader_system
{
// Override lighting stage with per pixel lighting.
lighting_stage per_pixel
}
To modify the default lighting stage see below. For more examples see Media/RTShaderLib/materials/RTShaderSystem.material.
RTSS Pass properties
Here are the attributes you can use in a rtshader_system block of a pass {}:
with gbuffer, this specifies the data to be written into one or two MRT targets. Possible values are depth, normal, viewpos, normal_viewdepth and diffuse_specular
Normal map contains normal data in tangent space. This is the default normal mapping behavior and it requires that the target mesh will have valid tangents within its vertex data.
object_space
Normal map contains normal data in object local space. This normal mapping technique has the advantages of better visualization results, lack of artifacts that comes from texture mirroring usage, it doesn't requires tangent and it also saves some instruction in the vertex shader stage. The main drawback of using this kind of normal map is that the target object must be static in terms of local space rotations and translations.
parallax
Normal map contains normal data in parallax corrected tangent space The restrictions of tangent_space apply. Additionally the alpha channel of the normal texture is expected to contain height displacement data. This is used for parallax corrected rendering.
parallax_occlusion
An extension of parallax, which samples the texture multiple times to allow using a larger displacement value without getting artifacts.
height_scale
displacement scale factor, when using parallax or parallax_occlusion
texcoord_index
the texcoord attribute index to read the uv coordinates from
layered_blend
Apply photoshop-like blend effects to texture layers.
The RTSS manages a set of opaque isolated components (SubRenderStates) where each implements a specific effect. These "effects" include Fixed Function transformation and lighting. At the core these components are plain shader files providing a set of functions; e.g. SGX_Light_Directional_Diffuse(), SGX_Light_Point_Diffuse().
Correctly ordering these functions, providing them with the right input values and interconnecting them is the main purpose of the RTSS.
As you can see the RTSS also resolved the required parameters and routed them into the correct functions. See Creating custom shader extensions for details about parameter resolution.
Now that you know what the RTSS does, you are probably wondering how to change which functions are emitted per stage. Let's say, change the lighting from the FFP style per-vertex lighting to per-pixel lighting.
The RTSS is flexible enough to "just" move the according calculations from the vertex shader to the pixel shader.
Core features of the system
Runtime shader generation synchronized with scene state. Each time scene settings change, a new set of shaders is generated.
Full Fixed Function Pipeline (FFP) emulation. This feature is most useful combined with render systems that don't provide any FFP functionality (OpenGL ES 2.0, D3D11 etc).
Shader language independent interface: the logic representation of the shader programs is completely independent from the target shader language. You can generate code for different shader languages from the same program.
Pluggable interface for different shader languages.
Pluggable interface for shader based functions in a seamless way. Each function will be automatically combined with the rest of the shader code.
Smart program caching: each shader program is created only once and may be used by multiple passes.
Automatic vertex shader output register compacting: no more compacting variables by hand. In case the amount of used vertex shader outputs exceeds the maximum allowed (12 to 32, depending on D3DPSHADERCAPS2_0.NumTemps), a compacting algorithm packs the vertex shader outputs and adds unpack code in the fragment shader side.
Material script support, for both export and import.
Controlling shader re-generation
By default the RTSS synchronizes with the active SceneManager regarding the fog settings and the number of active lights. This can result in frame-drops when new lights are added, as all managed Materials are updated for the new light-count. On the other hand, the generated shaders might include too many lights for the targeted budget.
to set a fixed number of lights the materials should consider.
The RTSS in Depth
When the user asks the system to generate shaders for a given technique he has to provide a name for the target technique scheme. The system then creates a new technique based on the source technique but with a different scheme name.
The idea behind this concept is to use Ogre's built-in mechanism of material schemes, so all the user has to do in order to use the new technique is to call Ogre::Viewport::setMaterialScheme.
Before each viewport update, the system performs a validation step of all associated shader based techniques it created. This step includes automatic synchronization with the scene lights and fog states. When the system detects that a scheme is out of date it generates the appropriate shaders for each technique new.
The following steps are executed in order to generate shaders for a given technique:
For each pass in the technique the system builds a set of sub render states that describe the logic process of the rendering pipeline from the draw call submission until the final pixel color.
Each render state is translated into a set of logic shader programs (currently only pixel and vertex shader). The logic programs are then sent to specific shader language writers that produce source code for the respective shader language. The source code is used to create the GPU programs that are applied to the destination pass. Before rendering of an object that uses generated shaders the system allows each sub render state to update the GPU constants associated with it.
Main components
The following is a partial list of components within the RTSS. These components are listed as they have great importance in understanding controlling and later extending the RTSS system.
ShaderGenerator
The ShaderGenerator is the main interface to the RTSS system. Through it you can request to generate and destroy the shaders, influence from what parts to create the shaders, and control general system settings such as the shading language and shader caching.
RenderState classes
The RenderState is the core component of the system. It aggregates the stages that the final shader will be created from. These stages are referred to as SubRenderStates. It is possible to bypass the rest of the RTSS and use RenderStates directly to manually generate shaders for arbitrary passes.
RenderStates exist on two levels:
SGScheme RenderStates describe the SubRenderStates that will be used when creating a shader for a given material scheme.
SGPass TargetRenderState describe the SubRenderStates that will be used when creating a specific pass of a specific material.
When a shader is generated for a given material the system combines the SubRenderStates from both RenderStates to create a shader specific for a material pass in a specific scheme.
SubRenderState classes
Sub-render states (SRS) are components designed to generate the code of the RTSS shaders. Each SRS usually has a specific role to fill within the shader's construction. These components can be combined in different combinations to create shaders with different capabilities.
By default, Ogre adds the following 5 SRSs to every scheme RenderState to recreate the functionality provided by the fixed pipeline
Ogre::RTShader::SRS_VERTEX_COLOUR - Calculate the base diffuse and specular color of the object regardless of lights or textures. The color is calculated based on the ambient, diffuse, specular and emissive properties of the object and scene and the specified color tracking.
Ogre::RTShader::SRS_FOG - Modulates the color of a pixel based on the scene or object fog parameters.
SubRenderStateFactory
As the name suggests, sub render state factories are factories that produce sub render states. Each factory generates a specific SRS.
These type of components are noteworthy for two reasons. The first and obvious one is that they allow the system to generate new SRSs for the materials it is asked to generate. The second reason is that they perform as script readers and writers allowing the system to create specific or specialized SRSs per material.
Default implementation of a Listener to use with the Ogre::RTShader system.
Definition OgreSGTechniqueResolverListener.h:46
Customizing the default RenderState
Let's say, you wanted to globally change the default per-pixel lighting mode of the RTSS back to the FFP style per-vertex lighting. For this you have to grab the global RenderState associated with the active material scheme, as
Next, you have to create the FFPLighting SRS that should be used for shader generation and add to the set.
Note
adding a SRS will automatically override the default SRS for the same stage. In the example we override the Ogre::RTShader::FFP_LIGHTING stage.
Creating the shader based technique
This step will create a new, shader-based, technique based on the given technique. Calling Ogre::RTShader::ShaderGenerator::createShaderBasedTechnique() will cause the system to generate SubRenderStates based on the source technique and add a new technique using the given scheme name to the material. The passes of this new technique will receive shaders generated and updated by the system during as described in the Shader generation at runtime section below.
To use the generated technique, change the material scheme of your viewport(s) to the scheme name you passed as argument to this method.
// Create shader based technique from the default technique of the given material.
you can automate the shader generation process for all materials. First set the viewport scheme to the destination scheme of the RTSS shaders. Second register to the Ogre::MaterialManager::Listener implementing handleSchemeNotFound() - e.g. OgreBites::SGTechniqueResolverListener
Shader generation at runtime
During the application runtime the ShaderGenerator instance receives notifications on a per frame basis from its target SceneManager. At this point it checks the material scheme in use. In case the current scheme has representations in the manager, it executes its validate method. The SGScheme validation includes synchronization with scene light and fog settings. In case it is out of date it will rebuild all shader generated techniques.
The first step is to loop over every SGTechnique associated with this SGScheme and build its RenderStates - one for each pass.
The second step is to loop again on every SGTechnique and acquire a program set for each SGPass.
Note
The shaders are only automatically updated for lights and fog changes. If you change the source pass after initial shader creation, you must call Ogre::RTShader::ShaderGenerator::invalidateMaterial manually.
The actual acquiring process is done by the TargetRenderState that generates CPU program representation, send them to a matching ProgramWriter that is chosen by the active target language, the writer generates source code that is the basis for the GPU programs. The result of this entire process is that each technique associated with the SGScheme has vertex and pixel shaders applied to all its passes. These shaders are synchronized with scene lights and fog settings.
Creating custom shader extensions
Although the system implements some common shader based effects such as per pixel lighting, normal map, etc., you may find it useful to write your own shader extensions.
In order to extend the system with your own shader effects you'll have to follow these steps:
Implement the SubRenderState interface - This is the main class that is responsible for the actual effect processing such as preparing the destination pass, updating the CPU shader programs, updating the GPU shader parameters etc.
Implement the SubRenderStateFactory interface: This class will allow the RTSS to create instances of the previous class via code or script as well as export it to material script file.
Add shader files that will supply all the actual shader functions your SubRenderState needs. In order to support multiple shader languages, Cross-platform macros are provided. These shaders should be placed in a resource location known to the resource manager.
Implementing the SubRenderState requires overriding the pure methods of the base class.
Ogre::RTShader::SubRenderState::getType() should return a unique string that identifies the sub class implementation. That value is shared among all instances and can be stored in a static string variable. It is used by the system to match between the SubRenderState instance and the factory that should destroy it.
Ogre::RTShader::SubRenderState::getExecutionOrder() should return integer value that will use the system to sort all SubRenderState instances of the same render state before each one of them will create its part in the CPU shader programs. Note that:
The execution order does not imply the order of the parameter definitions and function calls within the generated shader.
If an execution number is set to be the same as one of the basic fixed pipeline SRSs, then that SRS will be built instead of the fixed pipeline SRS.
Ogre::RTShader::SubRenderState::copyFrom() a simple copy method that uses the system when copying one instance to another. Note: Only configuration data attributes should be copied here.
The SubRenderState supplies default implementation for this method which breaks down this method into three stages:
Resolving parameters
this stage should grab all the needed parameters for this SubRenderState. Typically there are several SubRenderStates working on a common set of Parameters - either to cooperate or because they use the same inputs. Therefore parameters are not resolved by name (except for local variables), but rather by symbolic constants. These can either be of Ogre::GpuProgramParameters::AutoConstantType, which should already be familiar to you or of Ogre::RTShader::Parameter::Content.
You can think of the latter as an extension of the Cg/ HLSL Semantics to the actual content of the parameter.
In case of the Ogre::RTShader::FFPTransform we need the world view projection matrix and vertex shader input and output position parameters.
this stage should provide the name of the external shader library files that contains the actual shader code needed by this SubRenderState. In case of the Ogre::RTShader::SRS_TEXTURING it will add the common and texturing library for both vertex and pixel shader program.
this stage creates the function calls within this SubRenderState requires. To add function invocations, you first need to obtain a Ogre::RTShader::FunctionStageRef for the respective stage. In case of the Ogre::RTShader::SRS_FOG it will add vertex depth calculation to the vertex shader program.
The arguments to the function are the ones you resolved in the first step and the function name must be available in one of the libraries you provided in the second step. You can add call as many functions as you need. The calls will appear in the same order in the generates shader source code.
Make sure the parameter semantic (in/out) in the SubRenderState code matches to your shader code implementation you supplied in the library file. GLSL will fail to link to library functions if it won't be able to find a perfect function declaration match.
Ogre::RTShader::SubRenderState::preAddToRenderState(): this method called before adding this SubRenderState to a parent RenderState instances. It allows this SubRenderState to exclude itself from the list in case the source pass is not matching. I.E in case of SubRenderState that perform lighting calculations it can return false when the given source pass specifies that lighting calculations disabled for it.
This method also let the SubRenderState to opportunity to modify the destination pass. I.E the Ogre::RTShader::NormalMapLighting instance adds the normal map texture unit in this context.
Ogre::RTShader::SubRenderStateFactory::createInstanceImpl(): This method should return instance for the SubRenderState sub class.
Ogre::RTShader::SubRenderStateFactory::createInstance(): This method should return instance for the SubRenderState sub class using the given script compiler parameters. Implement this method if you want to be able to create your custom shader extension from material script.
Ogre::RTShader::SubRenderStateFactory::writeInstance(): This method should write down the parameters of a given SubRenderState instance to material script file. Implement this method if you want to be able to export a material that contains your custom shader extension.
Tips for debugging shaders
A couple of notes on debugging shaders coming from the RTSS:
It will make compilation problems easier to detect.
Once a shader is written to the disk, as long as you don't change the code behind it, the same shader will be picked up in the next application run even if its content has changed. If you have compilation or visual problems with the shader you can try to manually tinker with it without compiling the code again and again.
Other common problems with creating shaders in RTSS usually occur from defining vertex shader parameters and using them in the pixel shader and vice versa. So watch out for those.
1.9.8