void VP_Cloud::process(const VertexProcessorInput* input, VertexProcessorOutput* output) const
{
PROCDATA_CLOUD* userOutput = (PROCDATA_CLOUD*)output->user;
const float* position = (const float*)input->data[0];
const float* normal = (const float*)input->data[3];
const float* texcoord = (const float*)input->data[4];
// get vertex position in world space
mcemaths_transform_m4v4(userOutput->worldPos, m_M, position);
// project world-position into shadow map's UV space, inp:userOutput->worldPos, outp:userOutput->shadowcoord
mcemaths_quatcpy(userOutput->shadowcoord, userOutput->worldPos);
// ALIGN16 float enlarge[4];
// mcemaths_quatcpy(enlarge, normal);
// mcemaths_mul_3_4(enlarge, 1.5f);
// mcemaths_add_3_4_ip(userOutput->shadowcoord, enlarge); // slightly enlarge the model for anti-acne
mcemaths_transform_m4v4_ip(userOutput->shadowcoord, m_shadowPV);
// project world-position into view space
mcemaths_quatcpy(output->position, userOutput->worldPos);
mcemaths_transform_m4v4_ip(output->position, m_PV);
// remove world-position's w (later we'll use it as an xyz coordinate)
userOutput->worldPos[3] = 0;
// transform normal
mcemaths_transform_m4v4(userOutput->normal, m_Mrot, normal);
// pass on texture coord
userOutput->texcoord[0] = texcoord[0];
userOutput->texcoord[1] = texcoord[1];
}
void IP_SingleColour::interpolateByContributes(void* interpolatedUserData, const void** vertexUserData, const float* correctedContributes) const
{
__declspec(align(16)) PROCDATA_SINGLECOLOUR temp[3];
copyUserDataSingleColour(temp, (PROCDATA_SINGLECOLOUR*)vertexUserData[0]);
copyUserDataSingleColour(temp + 1, (PROCDATA_SINGLECOLOUR*)vertexUserData[1]);
copyUserDataSingleColour(temp + 2, (PROCDATA_SINGLECOLOUR*)vertexUserData[2]);
mcemaths_mul_3_4(temp[0].normal, correctedContributes[0]);
mcemaths_mul_3_4(temp[0].worldPos, correctedContributes[0]);
mcemaths_mul_3_4(temp[0].shadowcoord, correctedContributes[0]);
mcemaths_mul_3_4(temp[1].normal, correctedContributes[1]);
mcemaths_mul_3_4(temp[1].worldPos, correctedContributes[1]);
mcemaths_mul_3_4(temp[1].shadowcoord, correctedContributes[1]);
mcemaths_mul_3_4(temp[2].normal, correctedContributes[2]);
mcemaths_mul_3_4(temp[2].worldPos, correctedContributes[2]);
mcemaths_mul_3_4(temp[2].shadowcoord, correctedContributes[2]);
PROCDATA_SINGLECOLOUR* output = (PROCDATA_SINGLECOLOUR*)interpolatedUserData;
mcemaths_quatcpy(output->normal, temp[0].normal);
mcemaths_quatcpy(output->worldPos, temp[0].worldPos);
mcemaths_quatcpy(output->shadowcoord, temp[0].shadowcoord);
mcemaths_add_3_4_ip(output->normal, temp[1].normal);
mcemaths_add_3_4_ip(output->worldPos, temp[1].worldPos);
mcemaths_add_3_4_ip(output->shadowcoord, temp[1].shadowcoord);
mcemaths_add_3_4_ip(output->normal, temp[2].normal);
mcemaths_add_3_4_ip(output->worldPos, temp[2].worldPos);
mcemaths_add_3_4_ip(output->shadowcoord, temp[2].shadowcoord);
}
void FP_SingleColour::process(const FragmentProcessorInput* input, FragmentProcessorOutput* output) const
{
const PROCDATA_SINGLECOLOUR* inData = (const PROCDATA_SINGLECOLOUR*)input->user;
float shadowFactor = PuresoftSamplerProjection::get(m_shadowTex, inData->shadowcoord);
ALIGN16 float L[4];
mcemaths_sub_3_4(L, m_lightPos, inData->worldPos);
mcemaths_norm_3_4(L);
ALIGN16 float E[4];
mcemaths_sub_3_4(E, m_cameraPos, inData->worldPos);
mcemaths_norm_3_4(E);
ALIGN16 float H[4];
mcemaths_add_3_4(H, E, L);
mcemaths_norm_3_4(H);
enum {LAMBERT, SPECULAR};
ALIGN16 float factors[4];
factors[LAMBERT] = mcemaths_dot_3_4(L, inData->normal);
float yawOfLight = (float)acos(mcemaths_dot_3_4(L, m_lightDir));
factors[LAMBERT] *= yawOfLight < fieldOfLight ? 1.0f : (float)opt_pow(cos(yawOfLight - fieldOfLight), 150);
factors[SPECULAR] = opt_pow(mcemaths_dot_3_4(H, inData->normal), (unsigned int)*m_specularExponent);
mcemaths_clamp_3_4(factors, 0, 1.0f);
ALIGN16 float outputColour[4];
mcemaths_quatcpy(outputColour, m_diffuse);
mcemaths_mul_3_4(outputColour, factors[LAMBERT]);
ALIGN16 float specularColour[4];
mcemaths_quatcpy(specularColour, m_diffuse);
mcemaths_mulvec_3_4(specularColour, m_specularColour);
mcemaths_mul_3_4(specularColour, factors[SPECULAR]);
ALIGN16 float ambientColour[4];
mcemaths_quatcpy(ambientColour, m_diffuse);
mcemaths_mulvec_3_4(ambientColour, m_ambient);
mcemaths_add_3_4_ip(outputColour, specularColour);
mcemaths_mul_3_4(outputColour, shadowFactor);
mcemaths_add_3_4_ip(outputColour, ambientColour);
mcemaths_clamp_3_4(outputColour, 0, 1.0f);
mcemaths_mul_3_4(outputColour, 255.0f);
PURESOFTBGRA bytesColour;
bytesColour.elems.r = (unsigned char)outputColour[2];
bytesColour.elems.g = (unsigned char)outputColour[1];
bytesColour.elems.b = (unsigned char)outputColour[0];
bytesColour.elems.a = 0;
output->write4(0, &bytesColour);
}
void VP_Shadow::process(const VertexProcessorInput* input, VertexProcessorOutput* output) const
{
ALIGN16 float shrink[4];
mcemaths_quatcpy(shrink, (const float*)input->data[0]);
mcemaths_mul_3(shrink, 0.9f);
mcemaths_transform_m4v4(output->position, m_PVM, shrink);
}
void VP_DiffuseOnly::process(const VertexProcessorInput* input, VertexProcessorOutput* output) const
{
PROCDATA_DIFFUSEONLY* userOutput = (PROCDATA_DIFFUSEONLY*)output->user;
const float* position = (const float*)input->data[0];
const float* normal = (const float*)input->data[3];
const float* texcoord = (const float*)input->data[4];
// get vertex position in world space
mcemaths_transform_m4v4(userOutput->worldPos, m_M, position);
// project world-position into shadow map's UV space, inp:userOutput->worldPos, outp:userOutput->shadowcoord
mcemaths_quatcpy(userOutput->shadowcoord, userOutput->worldPos);
mcemaths_transform_m4v4_ip(userOutput->shadowcoord, m_shadowPV);
// project world-position into view space
mcemaths_transform_m4v4(output->position, m_PVM, position);
// remove world-position's w (later we'll use it as an xyz coordinate)
userOutput->worldPos[3] = 0;
// transform normals
mcemaths_transform_m4v4(userOutput->normal, m_MR, normal);
// pass on texture coord
userOutput->texcoord[0] = texcoord[0];
userOutput->texcoord[1] = texcoord[1];
}
void FP_DiffuseOnly::process(const FragmentProcessorInput* input, FragmentProcessorOutput* output) const
{
PURESOFTBGRA bytesColour;
ALIGN16 float outputColour[4], ambiemtColour[4];
const PROCDATA_DIFFUSEONLY* inData = (const PROCDATA_DIFFUSEONLY*)input->user;
float shadowFactor = PuresoftSamplerProjection::get(m_shadowTex, inData->shadowcoord);
PuresoftSampler2D::get4(m_diffuseTex, inData->texcoord[0], inData->texcoord[1], &bytesColour);
outputColour[0] = bytesColour.elems.b;
outputColour[1] = bytesColour.elems.g;
outputColour[2] = bytesColour.elems.r;
outputColour[3] = bytesColour.elems.a;
mcemaths_quatcpy(ambiemtColour, outputColour);
mcemaths_mulvec_3_4(ambiemtColour, m_ambient);
ALIGN16 float L[4];
mcemaths_sub_3_4(L, m_lightPos, inData->worldPos);
mcemaths_norm_3_4(L);
ALIGN16 float E[4];
mcemaths_sub_3_4(E, m_cameraPos, inData->worldPos);
mcemaths_norm_3_4(E);
ALIGN16 float H[4];
mcemaths_add_3_4(H, E, L);
mcemaths_norm_3_4(H);
enum {LAMBERT, SPECULAR};
ALIGN16 float factors[4];
factors[LAMBERT] = mcemaths_dot_3_4(L, inData->normal);
float yawOfLight = acos(mcemaths_dot_3_4(L, m_lightDir));
factors[LAMBERT] *= yawOfLight < fieldOfLight ? 1.0f : opt_pow(cos(yawOfLight - fieldOfLight), 150);
factors[SPECULAR] = opt_pow(mcemaths_dot_3_4(H, inData->normal), (unsigned int)*m_specularExponent);
mcemaths_clamp_3_4(factors, 0, 1.0f);
mcemaths_mul_3_4(outputColour, (factors[LAMBERT] + factors[SPECULAR]) * shadowFactor);
mcemaths_add_3_4_ip(outputColour, ambiemtColour);
mcemaths_clamp_3_4(outputColour, 0, 255.0f);
bytesColour.elems.r = (unsigned char)outputColour[2];
bytesColour.elems.g = (unsigned char)outputColour[1];
bytesColour.elems.b = (unsigned char)outputColour[0];
bytesColour.elems.a = 0;
output->write4(0, &bytesColour);
}
void IP_DiffuseOnly::interpolateByContributes(void* interpolatedUserData, const void** vertexUserData, const float* correctedContributes) const
{
__declspec(align(16)) PROCDATA_DIFFUSEONLY temp[3];
copyUserDataDiffuseOnly(temp, (PROCDATA_DIFFUSEONLY*)vertexUserData[0]);
copyUserDataDiffuseOnly(temp + 1, (PROCDATA_DIFFUSEONLY*)vertexUserData[1]);
copyUserDataDiffuseOnly(temp + 2, (PROCDATA_DIFFUSEONLY*)vertexUserData[2]);
mcemaths_mul_3_4(temp[0].normal, correctedContributes[0]);
mcemaths_mul_3_4(temp[0].worldPos, correctedContributes[0]);
mcemaths_mul_3_4(temp[0].texcoord, correctedContributes[0]);
mcemaths_mul_3_4(temp[0].shadowcoord, correctedContributes[0]);
mcemaths_mul_3_4(temp[1].normal, correctedContributes[1]);
mcemaths_mul_3_4(temp[1].worldPos, correctedContributes[1]);
mcemaths_mul_3_4(temp[1].texcoord, correctedContributes[1]);
mcemaths_mul_3_4(temp[1].shadowcoord, correctedContributes[1]);
mcemaths_mul_3_4(temp[2].normal, correctedContributes[2]);
mcemaths_mul_3_4(temp[2].worldPos, correctedContributes[2]);
mcemaths_mul_3_4(temp[2].texcoord, correctedContributes[2]);
mcemaths_mul_3_4(temp[2].shadowcoord, correctedContributes[2]);
PROCDATA_DIFFUSEONLY* output = (PROCDATA_DIFFUSEONLY*)interpolatedUserData;
mcemaths_quatcpy(output->normal, temp[0].normal);
mcemaths_quatcpy(output->worldPos, temp[0].worldPos);
mcemaths_quatcpy(output->texcoord, temp[0].texcoord);
mcemaths_quatcpy(output->shadowcoord, temp[0].shadowcoord);
mcemaths_add_3_4_ip(output->normal, temp[1].normal);
mcemaths_add_3_4_ip(output->worldPos, temp[1].worldPos);
mcemaths_add_3_4_ip(output->texcoord, temp[1].texcoord);
mcemaths_add_3_4_ip(output->shadowcoord, temp[1].shadowcoord);
mcemaths_add_3_4_ip(output->normal, temp[2].normal);
mcemaths_add_3_4_ip(output->worldPos, temp[2].worldPos);
mcemaths_add_3_4_ip(output->texcoord, temp[2].texcoord);
mcemaths_add_3_4_ip(output->shadowcoord, temp[2].shadowcoord);
}
void VP_CloudShadow::process(const VertexProcessorInput* input, VertexProcessorOutput* output) const
{
PROCDATA_CLOUDSHADOW* userOutput = (PROCDATA_CLOUDSHADOW*)output->user;
const float* position = (const float*)input->data[0];
const float* texcoord = (const float*)input->data[4];
ALIGN16 float position_adjusted[4], pvm[16];
mcemaths_quatcpy(position_adjusted, position);
mcemaths_mul_3(position_adjusted, 0.95f);
mcemaths_transform_m4m4(pvm, m_PV, m_M);
mcemaths_transform_m4v4(output->position, pvm, position_adjusted);
userOutput->texcoord[0] = texcoord[0];
userOutput->texcoord[1] = texcoord[1];
}
void FP_SingleColourNoLighting::process(const FragmentProcessorInput* input, FragmentProcessorOutput* output) const
{
const PROCDATA_SINGLECOLOUR* inData = (const PROCDATA_SINGLECOLOUR*)input->user;
ALIGN16 float outputColour[4];
mcemaths_quatcpy(outputColour, m_diffuse);
mcemaths_mul_3_4(outputColour, 255.0f);
PURESOFTBGRA bytesColour;
bytesColour.elems.r = (unsigned char)outputColour[2];
bytesColour.elems.g = (unsigned char)outputColour[1];
bytesColour.elems.b = (unsigned char)outputColour[0];
bytesColour.elems.a = 0;
output->write4(0, &bytesColour);
}
void IP_CloudShadow::interpolateByContributes(void* interpolatedUserData, const void** vertexUserData, const float* correctedContributes) const
{
__declspec(align(16)) PROCDATA_CLOUDSHADOW temp[3];
copyUserData(temp, (PROCDATA_CLOUDSHADOW*)vertexUserData[0]);
copyUserData(temp + 1, (PROCDATA_CLOUDSHADOW*)vertexUserData[1]);
copyUserData(temp + 2, (PROCDATA_CLOUDSHADOW*)vertexUserData[2]);
mcemaths_mul_3_4(temp[0].texcoord, correctedContributes[0]);
mcemaths_mul_3_4(temp[1].texcoord, correctedContributes[1]);
mcemaths_mul_3_4(temp[2].texcoord, correctedContributes[2]);
PROCDATA_CLOUDSHADOW* output = (PROCDATA_CLOUDSHADOW*)interpolatedUserData;
mcemaths_quatcpy(output->texcoord, temp[0].texcoord);
mcemaths_add_3_4_ip(output->texcoord, temp[1].texcoord);
mcemaths_add_3_4_ip(output->texcoord, temp[2].texcoord);
}
void IP_Planet::interpolateByContributes(void* interpolatedUserData, const void** vertexUserData, const float* correctedContributes) const
{
__declspec(align(16)) PROCDATA_PLANET temp[3];
copyUserData(temp, (PROCDATA_PLANET*)vertexUserData[0]);
copyUserData(temp + 1, (PROCDATA_PLANET*)vertexUserData[1]);
copyUserData(temp + 2, (PROCDATA_PLANET*)vertexUserData[2]);
mcemaths_mul_3_4(temp[0].tangent, correctedContributes[0]);
mcemaths_mul_3_4(temp[0].binormal, correctedContributes[0]);
mcemaths_mul_3_4(temp[0].normal, correctedContributes[0]);
mcemaths_mul_3_4(temp[0].worldPos, correctedContributes[0]);
mcemaths_mul_3_4(temp[0].texcoord, correctedContributes[0]);
mcemaths_mul_3_4(temp[0].shadowcoord, correctedContributes[0]);
mcemaths_mul_3_4(temp[1].tangent, correctedContributes[1]);
mcemaths_mul_3_4(temp[1].binormal, correctedContributes[1]);
mcemaths_mul_3_4(temp[1].normal, correctedContributes[1]);
mcemaths_mul_3_4(temp[1].worldPos, correctedContributes[1]);
mcemaths_mul_3_4(temp[1].texcoord, correctedContributes[1]);
mcemaths_mul_3_4(temp[1].shadowcoord, correctedContributes[1]);
mcemaths_mul_3_4(temp[2].tangent, correctedContributes[2]);
mcemaths_mul_3_4(temp[2].binormal, correctedContributes[2]);
mcemaths_mul_3_4(temp[2].normal, correctedContributes[2]);
mcemaths_mul_3_4(temp[2].worldPos, correctedContributes[2]);
mcemaths_mul_3_4(temp[2].texcoord, correctedContributes[2]);
mcemaths_mul_3_4(temp[2].shadowcoord, correctedContributes[2]);
PROCDATA_PLANET* output = (PROCDATA_PLANET*)interpolatedUserData;
mcemaths_quatcpy(output->tangent, temp[0].tangent);
mcemaths_quatcpy(output->binormal, temp[0].binormal);
mcemaths_quatcpy(output->normal, temp[0].normal);
mcemaths_quatcpy(output->worldPos, temp[0].worldPos);
mcemaths_quatcpy(output->texcoord, temp[0].texcoord);
mcemaths_quatcpy(output->shadowcoord, temp[0].shadowcoord);
mcemaths_add_3_4_ip(output->tangent, temp[1].tangent);
mcemaths_add_3_4_ip(output->binormal, temp[1].binormal);
mcemaths_add_3_4_ip(output->normal, temp[1].normal);
mcemaths_add_3_4_ip(output->worldPos, temp[1].worldPos);
mcemaths_add_3_4_ip(output->texcoord, temp[1].texcoord);
mcemaths_add_3_4_ip(output->shadowcoord, temp[1].shadowcoord);
mcemaths_add_3_4_ip(output->tangent, temp[2].tangent);
mcemaths_add_3_4_ip(output->binormal, temp[2].binormal);
mcemaths_add_3_4_ip(output->normal, temp[2].normal);
mcemaths_add_3_4_ip(output->worldPos, temp[2].worldPos);
mcemaths_add_3_4_ip(output->texcoord, temp[2].texcoord);
mcemaths_add_3_4_ip(output->shadowcoord, temp[2].shadowcoord);
}
void VP_SingleColour::process(const VertexProcessorInput* input, VertexProcessorOutput* output) const
{
PROCDATA_SINGLECOLOUR* userOutput = (PROCDATA_SINGLECOLOUR*)output->user;
const float* position = (const float*)input->data[0];
const float* normal = (const float*)input->data[3];
const float* texcoord = (const float*)input->data[4];
// get vertex position in world space
mcemaths_transform_m4v4(userOutput->worldPos, m_M, position);
// project world-position into shadow map's UV space, inp:userOutput->worldPos, outp:userOutput->shadowcoord
mcemaths_quatcpy(userOutput->shadowcoord, userOutput->worldPos);
mcemaths_transform_m4v4_ip(userOutput->shadowcoord, m_shadowPV);
// remove world-position's w (later we'll use it as an xyz coordinate)
userOutput->worldPos[3] = 0;
// project model-position into view space
mcemaths_transform_m4v4(output->position, m_PVM, position);
// transform normal
mcemaths_transform_m4v4(userOutput->normal, m_MR, normal);
}
const vec4& vec4::operator= (const vec4& src)
{
mcemaths_quatcpy(&x, src);
return *this;
}