Jump to content

  • Log In with Google      Sign In   
  • Create Account

Banner advertising on our site currently available from just $5!


1. Learn about the promo. 2. Sign up for GDNet+. 3. Set up your advert!


Adding Primitives to Shadow Mapping


Old topic!
Guest, the last post of this topic is over 60 days old and at this point you may not reply in this topic. If you wish to continue this conversation start a new topic.

  • You cannot reply to this topic
2 replies to this topic

#1 KurtO   Members   -  Reputation: 247

Like
0Likes
Like

Posted 18 September 2012 - 01:08 PM

Hi!

I am using this tutorial to try to understand how to use shadow-mapping
http://www.gamedev.net/page/resources/_/technical/graphics-programming-and-theory/soft-edged-shadows-r2193

I have downloaded the source code and got it to work. I understand in a small way how the shadowmap is rendered and using the effect file to calculate the shadows.

Now i am trying to add Primitives like triangles/cubes and sprites to my scene but can´t get it to work, i get the Triangle to show, but not in World-Space with the other models. I am completely stuck and have no ide where that code would be?

Where should i render my primitives in this code and how can i get them in world-space?

I have made lots of demo with just triangles / cubes and sprites with blending etc, it is just that the shadow-mapping code confuses me.

Any help is greatfully accepted. =)

My UpdateFrame function:
[source lang="cpp"]HRESULT FrameMove(){ // // Camera space matrices // // Computee the world matrix D3DXMATRIX matWorld; D3DXMatrixIdentity( &matWorld ); // Compute the view matrix D3DXMATRIX matView; D3DXMatrixLookAtLH( &matView, &g_vEyePos, &g_vEyeAim, &g_vUp ); // Compute the projection matrix D3DXMATRIX matProj; D3DXMatrixPerspectiveFovLH( &matProj, D3DXToRadian(60.0f), (FLOAT)SCREEN_WIDTH / (FLOAT)SCREEN_HEIGHT, 1.0f, 1024.0f ); // Compute the world-view-projection matrix D3DXMATRIX matWorldViewProj = matWorld * matView * matProj; // World inverse matrix D3DXMATRIX matWorldIT; D3DXMatrixInverse( &matWorldIT, NULL, &matWorld ); D3DXMatrixTranspose( &matWorldIT, &matWorldIT ); // // Light space matrices // // View matrix D3DXVECTOR3 vLightPos = D3DXVECTOR3( 40.0f, 10.0f, -50.0f ); D3DXVECTOR3 vLightAim = D3DXVECTOR3( 0.0f, 10.0f, 0.0f ); D3DXMatrixLookAtLH( &matView, &vLightPos, &vLightAim, &g_vUp ); // Compute the projection matrix D3DXMatrixOrthoLH( &matProj, 45.0f, 45.0f, 1.0f, 1024.0f ); // Compute the light-view-projection matrix D3DXMATRIX matLightViewProj = matWorld * matView * matProj; // Compute the texture matrix float fTexOffs = 0.5 + (0.5 / (float)SHADOW_MAP_SIZE); D3DXMATRIX matTexAdj( 0.5f, 0.0f, 0.0f, 0.0f, 0.0f, -0.5f, 0.0f, 0.0f, 0.0f, 0.0f, 1.0f, 0.0f, fTexOffs, fTexOffs, 0.0f, 1.0f ); D3DXMATRIX matTexture = matLightViewProj * matTexAdj; // Setup the effect variables g_pEffect->SetMatrix( "g_matWorldViewProj", &matWorldViewProj ); g_pEffect->SetMatrix( "g_matLightViewProj", &matLightViewProj ); g_pEffect->SetMatrix( "g_matWorld", &matWorld ); g_pEffect->SetMatrix( "g_matWorldIT", &matWorldIT ); g_pEffect->SetMatrix( "g_matTexture", &matTexture ); g_pEffect->SetVector( "g_vLightPos", (D3DXVECTOR4*)&vLightPos ); g_pEffect->SetVector( "g_vEyePos", (D3DXVECTOR4*)&g_vEyePos ); g_pEffect->SetVector( "g_vLightColor", &D3DXVECTOR4( 1.0f, 1.0f, 1.0f, 0.5f ) ); g_pEffect->SetBool("g_bFiltered", g_bFiltered); return S_OK;}[/source]
I think the scene is set, i can rotate and walk around when manipulatin the View but my primitives are not placing correctly.

This is the Render Scene:

[source lang="cpp"]HRESULT Render(){ // Clear the viewport g_pd3dDevice->Clear( 0, NULL, D3DCLEAR_TARGET | D3DCLEAR_ZBUFFER, 0x00000000, 1.0f, 0 ); // Begin rendering the scene if( SUCCEEDED( g_pd3dDevice->BeginScene() ) ) { g_pd3dDevice->SetFVF(CUSTOMFVF); g_pd3dDevice->SetStreamSource(0, v_buffer, 0, sizeof(CUSTOMVERTEX)); g_pd3dDevice->DrawPrimitive(D3DPT_TRIANGLELIST, 0, 1); // Grab the old render target and depth buffer g_pd3dDevice->GetRenderTarget( 0, &g_pOldColorRT ); g_pd3dDevice->GetDepthStencilSurface( &g_pOldDepthRT ); // Render the scene depth to the shadow map g_pd3dDevice->SetRenderTarget( 0, g_pShadowSurf ); g_pd3dDevice->SetDepthStencilSurface( g_pShadowDepth ); // Clear the viewport g_pd3dDevice->Clear( 0, NULL, D3DCLEAR_TARGET | D3DCLEAR_ZBUFFER, 0xFFFFFFFF, 1.0f, 0 ); // Set the technique g_pEffect->SetTechnique( "techShadow" ); // Render the effect UINT uPasses = 0; g_pEffect->Begin( &uPasses, 0 ); for( UINT uPass = 0; uPass < uPasses; uPass++ ) { // Set the current pass g_pEffect->BeginPass( uPass ); // Draw the floor g_pScene->DrawSubset(0); // Draw the statue g_pScene->DrawSubset(1); g_pScene->DrawSubset(2); // End the current pass g_pEffect->EndPass(); } // End the effect g_pEffect->End(); if( g_bSoftShadows ) { // Render the shadowed scene into the screen map g_pd3dDevice->SetRenderTarget( 0, g_pScreenSurf ); g_pd3dDevice->SetDepthStencilSurface( g_pNewDepthRT ); // Clear the viewport g_pd3dDevice->Clear( 0, NULL, D3DCLEAR_TARGET | D3DCLEAR_ZBUFFER, 0x00000000, 1.0f, 0 ); // Set the technique g_pEffect->SetTechnique( "techUnlit" ); // Set the textures g_pEffect->SetTexture( "tShadowMap", g_pShadowMap ); // Render the effect uPasses = 0; g_pEffect->Begin( &uPasses, 0 ); for(int uPass = 0; uPass < uPasses; uPass++ ) { // Set the current pass g_pEffect->BeginPass( uPass ); // Draw the floor g_pScene->DrawSubset(0); // Draw the statue g_pScene->DrawSubset(1); g_pScene->DrawSubset(2); // End the current pass g_pEffect->EndPass(); } // End the effect g_pEffect->End(); // Blur the screen map horizontally g_pd3dDevice->SetRenderTarget( 0, g_pBlurSurf[0] ); g_pd3dDevice->SetDepthStencilSurface( g_pNewDepthRT ); // Clear the viewport g_pd3dDevice->Clear( 0, NULL, D3DCLEAR_TARGET | D3DCLEAR_ZBUFFER, 0x00000000, 1.0f, 0 ); // Set the technique g_pEffect->SetTechnique( "techBlurH" ); // Compute the Gaussian offsets GetGaussianOffsets(TRUE, D3DXVECTOR2(1.0f / (FLOAT)SHADOW_MAP_SIZE, 1.0f / (FLOAT)SHADOW_MAP_SIZE), g_vSampleOffsets, g_fSampleWeights); g_pEffect->SetValue("g_vSampleOffsets", g_vSampleOffsets, 15 * sizeof(D3DXVECTOR2)); g_pEffect->SetValue("g_fSampleWeights", g_fSampleWeights, 15 * sizeof(FLOAT)); // Set the textures g_pEffect->SetTexture( "tScreenMap", g_pScreenMap ); // Render the effect uPasses = 0; g_pEffect->Begin( &uPasses, 0 ); for(int uPass = 0; uPass < uPasses; uPass++ ) { // Set the current pass g_pEffect->BeginPass( uPass ); // Draw the quad g_pd3dDevice->SetStreamSource( 0, g_pQuadVB, 0, sizeof(QuadVertex) ); g_pd3dDevice->SetFVF( FVF_QUADVERTEX ); g_pd3dDevice->DrawPrimitive( D3DPT_TRIANGLESTRIP, 0, 2 ); // End the current pass g_pEffect->EndPass(); } // End the effect g_pEffect->End(); // Blur the screen map vertically g_pd3dDevice->SetRenderTarget( 0, g_pBlurSurf[1] ); g_pd3dDevice->SetDepthStencilSurface( g_pNewDepthRT ); // Clear the viewport g_pd3dDevice->Clear( 0, NULL, D3DCLEAR_TARGET | D3DCLEAR_ZBUFFER, 0x00000000, 1.0f, 0 ); // Set the technique g_pEffect->SetTechnique( "techBlurV" ); // Compute the Gaussian offsets GetGaussianOffsets(FALSE, D3DXVECTOR2(1.0f / (FLOAT)SHADOW_MAP_SIZE, 1.0f / (FLOAT)SHADOW_MAP_SIZE), g_vSampleOffsets, g_fSampleWeights); g_pEffect->SetValue("g_vSampleOffsets", g_vSampleOffsets, 15 * sizeof(D3DXVECTOR2)); g_pEffect->SetValue("g_fSampleWeights", g_fSampleWeights, 15 * sizeof(FLOAT)); // Set the textures g_pEffect->SetTexture( "tBlurHMap", g_pBlurMap[0] ); // Render the effect uPasses = 0; g_pEffect->Begin( &uPasses, 0 ); for(int uPass = 0; uPass < uPasses; uPass++ ) { // Set the current pass g_pEffect->BeginPass( uPass ); // Draw the quad g_pd3dDevice->SetStreamSource( 0, g_pQuadVB, 0, sizeof(QuadVertex) ); g_pd3dDevice->SetFVF( FVF_QUADVERTEX ); g_pd3dDevice->DrawPrimitive( D3DPT_TRIANGLESTRIP, 0, 2 ); // End the current pass g_pEffect->EndPass(); } // End the effect g_pEffect->End(); // Restore the render target and depth buffer g_pd3dDevice->SetRenderTarget( 0, g_pOldColorRT ); g_pd3dDevice->SetDepthStencilSurface( g_pOldDepthRT ); SAFE_RELEASE( g_pOldColorRT ); SAFE_RELEASE( g_pOldDepthRT ); // Finally, render the shadowed scene g_pEffect->SetTechnique( "techScene" ); // Set the textures g_pEffect->SetTexture( "tBlurVMap", g_pBlurMap[1] ); g_pEffect->SetTexture( "tSpotMap", g_pSpotMap ); // Render the effect uPasses = 0; g_pEffect->Begin( &uPasses, 0 ); // Do we need to render the scene in wireframe mode if( g_bWireframe ) g_pd3dDevice->SetRenderState( D3DRS_FILLMODE, D3DFILL_WIREFRAME ); for( int uPass = 0; uPass < uPasses; uPass++ ) { // Set the current pass g_pEffect->BeginPass( uPass ); // Draw the floor g_pEffect->SetTexture( "tColorMap", g_pColorMap_Floor ); g_pEffect->CommitChanges(); g_pScene->DrawSubset(0); // Draw the statue g_pEffect->SetTexture( "tColorMap", g_pColorMap_Statue ); g_pEffect->CommitChanges(); g_pScene->DrawSubset(1); g_pScene->DrawSubset(2); // End the current pass g_pEffect->EndPass(); } // End the effect g_pEffect->End(); // Restore the render state g_pd3dDevice->SetRenderState( D3DRS_FILLMODE, D3DFILL_SOLID ); } else { // Restore the render target and depth buffer g_pd3dDevice->SetRenderTarget( 0, g_pOldColorRT ); g_pd3dDevice->SetDepthStencilSurface( g_pOldDepthRT ); SAFE_RELEASE( g_pOldColorRT ); SAFE_RELEASE( g_pOldDepthRT ); // Otherwise, render the scene with hard shadows g_pEffect->SetTechnique( "techSceneHard" ); // Set the textures g_pEffect->SetTexture( "tShadowMap", g_pShadowMap ); g_pEffect->SetTexture( "tSpotMap", g_pSpotMap ); // Render the effect uPasses = 0; g_pEffect->Begin( &uPasses, 0 ); // Do we need to render the scene in wireframe mode if( g_bWireframe ) g_pd3dDevice->SetRenderState( D3DRS_FILLMODE, D3DFILL_WIREFRAME ); for(int uPass = 0; uPass < uPasses; uPass++ ) { // Set the current pass g_pEffect->BeginPass( uPass ); // Draw the floor g_pEffect->SetTexture( "tColorMap", g_pColorMap_Floor ); g_pEffect->CommitChanges(); g_pScene->DrawSubset(0); // Draw the statue g_pEffect->SetTexture( "tColorMap", g_pColorMap_Statue ); g_pEffect->CommitChanges(); g_pScene->DrawSubset(1); g_pScene->DrawSubset(2); // End the current pass g_pEffect->EndPass(); } // End the effect g_pEffect->End(); // Restore the render state g_pd3dDevice->SetRenderState( D3DRS_FILLMODE, D3DFILL_SOLID ); } // End rendering the scene and present it g_pd3dDevice->EndScene(); g_pd3dDevice->Present( NULL, NULL, NULL, NULL ); } return S_OK;}[/source]

Sponsor:

#2 Seabolt   Members   -  Reputation: 751

Like
1Likes
Like

Posted 20 September 2012 - 02:20 PM

So shadow mapping is done generally in such a way that it is decoupled from your geometry. Meaning, it doesn't matter how you draw your geometry, as long as it's drawn into the shadow map.

So your flow normally goes like this:
BeginFrame

Bind your shadow camera
Bind your shadow map

Draw your occluders, (scene)

Bind your normal camera
Bind your render target

Draw your scene
Compare your object's depth in shadow camera view space vs the depth in the shadow map
Shade if your depth is greater than the depth in the map
Draw your object with applied shading
End Frame

This is a gross over simplification, but that's the goal.
Perception is when one imagination clashes with another

#3 KurtO   Members   -  Reputation: 247

Like
0Likes
Like

Posted 21 September 2012 - 07:42 AM

thanks, this makes it easier to know what to look for in all sample codes.




Old topic!
Guest, the last post of this topic is over 60 days old and at this point you may not reply in this topic. If you wish to continue this conversation start a new topic.



PARTNERS