• Announcements

    • khawk

      Download the Game Design and Indie Game Marketing Freebook   07/19/17

      GameDev.net and CRC Press have teamed up to bring a free ebook of content curated from top titles published by CRC Press. The freebook, Practices of Game Design & Indie Game Marketing, includes chapters from The Art of Game Design: A Book of Lenses, A Practical Guide to Indie Game Marketing, and An Architectural Approach to Level Design. The GameDev.net FreeBook is relevant to game designers, developers, and those interested in learning more about the challenges in game development. We know game development can be a tough discipline and business, so we picked several chapters from CRC Press titles that we thought would be of interest to you, the GameDev.net audience, in your journey to design, develop, and market your next game. The free ebook is available through CRC Press by clicking here. The Curated Books The Art of Game Design: A Book of Lenses, Second Edition, by Jesse Schell Presents 100+ sets of questions, or different lenses, for viewing a game’s design, encompassing diverse fields such as psychology, architecture, music, film, software engineering, theme park design, mathematics, anthropology, and more. Written by one of the world's top game designers, this book describes the deepest and most fundamental principles of game design, demonstrating how tactics used in board, card, and athletic games also work in video games. It provides practical instruction on creating world-class games that will be played again and again. View it here. A Practical Guide to Indie Game Marketing, by Joel Dreskin Marketing is an essential but too frequently overlooked or minimized component of the release plan for indie games. A Practical Guide to Indie Game Marketing provides you with the tools needed to build visibility and sell your indie games. With special focus on those developers with small budgets and limited staff and resources, this book is packed with tangible recommendations and techniques that you can put to use immediately. As a seasoned professional of the indie game arena, author Joel Dreskin gives you insight into practical, real-world experiences of marketing numerous successful games and also provides stories of the failures. View it here. An Architectural Approach to Level Design This is one of the first books to integrate architectural and spatial design theory with the field of level design. The book presents architectural techniques and theories for level designers to use in their own work. It connects architecture and level design in different ways that address the practical elements of how designers construct space and the experiential elements of how and why humans interact with this space. Throughout the text, readers learn skills for spatial layout, evoking emotion through gamespaces, and creating better levels through architectural theory. View it here. Learn more and download the ebook by clicking here. Did you know? GameDev.net and CRC Press also recently teamed up to bring GDNet+ Members up to a 20% discount on all CRC Press books. Learn more about this and other benefits here.
Sign in to follow this  
Followers 0
Juliean

Deferred renderer not working on second PC

10 posts in this topic

Hello,

 

I'm working parallell on my tower pc and my notebook. Yesterday I finally implemented my deferred rendering-system, which works fine on the tower. When cloning my repo on the notebook however, nothing shows up. Those are the steps I already took to debug (in PIX):

 

- I made sure that the cube actually gets rendered and transformed correctly.

 

- The g-buffer render-targets are set correctly.

 

- There is no output to the 3 g-buffer rts. "Debug this pixel" shows that the pixel is black, then it gets cleared to black (before the rendering), and then nothing.

 

- All rendertargets, camera etc... are correctly sized to the backbuffer (my app takes care of that automatically).

 

So uhm... since my notebook has an Ati graphics card I cannot use NVPerfStudio to debug it in detail like I normally would. GpuPerfStudio 1.2 seems to lack a d3d-server, so thats out of the question too. Let me give you a quick insight on my two effect files I use to render.

 

The effect used to render my cube:

/*

% Base shader for rendering 3D models
% Renders to a g-buffer

keywords: material classic

date: 120129

*/

float4x4 World : World;
shared float4x4 ViewProj;

texture2D Material;
texture2D Bump;

sampler MaterialSampler = sampler_state
{
	Texture = <Material>;
	
	MagFilter = Linear;
	MinFilter = Anisotropic;
	MipFilter = Linear;
	MaxAnisotropy = 16;
};

sampler BumpSampler = sampler_state
{
	Texture = <Bump>;
	
	MagFilter = Linear;
	MinFilter = Linear;
	MipFilter = Linear;
};


struct VS_INPUT
{
    float4  vPos            : POSITION0;
    float2  vTex0           : TEXCOORD0;
    float3  vNrm            : NORMAL0;
    float3  vTan            : TANGENT0;
    float3  vBin            : BINORMAL0;
};

struct VS_OUTPUT
{
	float4 vPos             : POSITION0;
	float4 vPos1            : POSITION1;
	float2 vTex0            : TEXCOORD0;
	float3x3 mTBN		: TEXCOORD1;
};

struct PS_OUTPUT
{
	float4 vMat       : COLOR0;
	float4 vPos       : COLOR1;
	float4 vNrm       :COLOR2;
};

VS_OUTPUT mainVS(VS_INPUT i)
{
	VS_OUTPUT o;

	float4x4 WorldViewProj = mul(World,ViewProj);
	o.vPos = mul(i.vPos, WorldViewProj);
	o.vPos1 = mul(i.vPos, World);

	o.vTex0 = i.vTex0;

	float3 N = normalize(mul(i.vNrm, World));
    float3 T = normalize(mul(i.vTan, World));
    float3 B = normalize(mul(i.vBin, World));
    o.mTBN = float3x3(T, B, N);

	return o;
}

PS_OUTPUT mainPS(VS_OUTPUT i)
{
	PS_OUTPUT o;

	o.vMat = tex2D(MaterialSampler,i.vTex0);

	o.vPos = i.vPos1;

	o.vNrm = tex2D(BumpSampler, i.vTex0);
	o.vNrm.rgb = 2.0* o.vNrm.rgb - 1.0;
    o.vNrm.rgb = mul(o.vNrm.rgb, i.mTBN);
	o.vNrm.rgb = normalize(o.vNrm.rgb);
	o.vNrm.rgb = 0.5 * o.vNrm.rgb + 0.5;
	o.vNrm.a = 1.0f;

	return o;
}

technique technique0 {
	pass p0 {
		CullMode = Ccw;
		VertexShader = compile vs_3_0 mainVS();
		PixelShader = compile ps_3_0 mainPS();
	}
}

The effect for the lights:

/*

% Shader for performing deferred lighting
% Renders: ambient

keywords: light deferred

date: 120202

*/

/*

Ambient Paramters

*/

float3 cAmbientColor;
shared texture2D cMaterial;

sampler cMaterialSampler = sampler_state
{
	Texture = <cMaterial>;
	
	MagFilter = Point;
	MinFilter = Point;
	MipFilter = Point;
};

/*

Directional Paramters

*/

float3 cLightDirection;
float3 cCameraPos;
float3 cDiffuseColor;
float3 cSpecularColor;
shared texture2D cNormal;

sampler cNormalSampler = sampler_state
{
	Texture = <cNormal>;
	
	MagFilter = Point;
	MinFilter = Point;
	MipFilter = Point;
};

shared texture2D cPosition;

sampler cPositionSampler = sampler_state
{
	Texture = <cPosition>;
	
	MagFilter = Point;
	MinFilter = Point;
	MipFilter = Point;
};

/*

Primitive Vertex Shader

*/

struct VS_INPUT
{
    float3  vPos            : POSITION0;
    float2  vTex0           : TEXCOORD0;
};

struct VS_OUTPUT
{
	float4 vPos             : POSITION0;
	float2 vTex0            : TEXCOORD0;
};

struct PS_OUTPUT
{
	float4 vMat       : COLOR0;
	float4 vPos       : COLOR1;
	float4 vNrm       :COLOR2;
};

VS_OUTPUT vsQuad(VS_INPUT i)
{
    VS_OUTPUT o;

	o.vPos = float4(i.vPos,1.0);
	
	o.vTex0 = i.vTex0;

	return o;
}

/*

Ambient Lighting

*/

float4 psAmbient(VS_OUTPUT i) : COLOR0
{
	float4 vDiffuseMaterial;
	vDiffuseMaterial.rgb = tex2D(cMaterialSampler, i.vTex0).rgb;
	vDiffuseMaterial.rgb *= cAmbientColor;
	
	vDiffuseMaterial.a = 1.0;
	
	return vDiffuseMaterial;
};

technique Ambient {
	pass p0 {

		CullMode = None;
		ZEnable = false;
		AlphaBlendEnable = true;
		Srcblend = One;
		Destblend = One;
		VertexShader = compile vs_3_0 vsQuad();
		PixelShader = compile ps_3_0 psAmbient();
	}
}

/*

Directional Lighting

*/

float4 psDirectional(VS_OUTPUT i) : COLOR0
{
	float3 vDiffuseMaterial;
	vDiffuseMaterial.rgb = tex2D(cMaterialSampler, i.vTex0).rgb;
	
	float4 vWorldNrm = tex2D(cNormalSampler, i.vTex0);
	vWorldNrm.rgb = 2.0 * vWorldNrm.rgb - 1.0;

	float3 vWorldPos = tex2D(cPositionSampler, i.vTex0).rgb;
	float3 vCameraDir = cCameraPos - vWorldPos;
	float3 vLightDir = normalize(cLightDirection);
	float3 vDiffuseIntensity = max(0, dot(vLightDir, vWorldNrm));
    float3 Reflect = normalize(2 * -vDiffuseIntensity * vWorldNrm - vLightDir);
	float shadow = saturate(4* vDiffuseIntensity);
    float4 specular = max(pow(dot(Reflect, normalize(-vCameraDir)), 32), 0); // R.V^n

	float4 color;
	color.rgb = shadow * (vDiffuseIntensity * cDiffuseColor.xyz * vDiffuseMaterial.rgb + specular * cSpecularColor.xyz);
	
	color.a = 1.0;
	
	return color;
};

technique Directional {
	pass p0 {
		
		CullMode = None;
		ZEnable = false;
		AlphaBlendEnable = true;
		Srcblend = One;
		Destblend = One;
		VertexShader = compile vs_3_0 vsQuad();
		PixelShader = compile ps_3_0 psDirectional();
	}
}


Then, let me show you my (temporary) code for rendering the cube and my lights:

 

#pragma once
#include "SystemManager.h"
#include "Components.h"

class DeferredRenderSystem : public EntitySystem<DeferredRenderSystem>
{
public:
	DeferredRenderSystem(Gfx3D& gfx3D): m_pGfx3D(&gfx3D) {};
	void Update(EntityManager& entityManager)
	{
		//todo: HUGE cleanup
		EntityManager::entityVector* vEntities = entityManager.EntitiesWithComponents<ModelComponent, PositionComponent>();
		Camera camera;
		D3DXMATRIX matrix;
		D3DXMatrixIdentity(&matrix);
		Effect* effect;
		Model* model;
		Material* material;
		// set g-buffer
		//todo: implement "ClearAllBound"
		m_pGfx3D->Clear3DRenderTarget(RenderTargets::DEFERRED_POSITION);
		m_pGfx3D->Clear3DRenderTarget(RenderTargets::DEFERRED_MATERIAL);
		m_pGfx3D->Clear3DRenderTarget(RenderTargets::DEFERRED_NORMAL);
		m_pGfx3D->Set3DRenderTarget(0, RenderTargets::DEFERRED_MATERIAL);
		m_pGfx3D->Set3DRenderTarget(1, RenderTargets::DEFERRED_POSITION);
		m_pGfx3D->Set3DRenderTarget(2, RenderTargets::DEFERRED_NORMAL);
		for(auto entity : *vEntities)
		{
			PositionComponent* pos = entity->GetComponent<PositionComponent>();
			ModelComponent* mod = entity->GetComponent<ModelComponent>();
			effect = m_pGfx3D->GetEffect(mod->m_effect);
			model = m_pGfx3D->GetModel(mod->m_model);
			material = m_pGfx3D->GetMaterial(mod->m_material);
			effect->SetMatrix("ViewProj", camera.GetViewProjectionMatrix());
			effect->SetTexture("Material", material->GetTexture());
			effect->SetTexture("Bump", material->GetBumpMap());
			D3DXMatrixTranslation(&matrix, pos->m_x, pos->m_y, pos->m_z);
			effect->SetMatrix("World", matrix);
			effect->Begin();
			model->Render();
			effect->End();
		}
		delete vEntities;
		//unbind g-buffer
		m_pGfx3D->Set3DRenderTarget(0, RenderTargets::BACKBUFFER);
		m_pGfx3D->Set3DRenderTarget(1, RenderTargets::BACKBUFFER);
		m_pGfx3D->Set3DRenderTarget(2, RenderTargets::BACKBUFFER);
	}
private:
	Gfx3D* m_pGfx3D;
};

class DeferredLightSystem : public EntitySystem<DeferredLightSystem>
{
public:
	DeferredLightSystem(Gfx3D& gfx3D): m_pGfx3D(&gfx3D) {};
	void Update(EntityManager& entityManager)
	{
		m_pGfx3D->Clear3DRenderTarget(RenderTargets::FINAL_SCENE);
		m_pGfx3D->Set3DRenderTarget(0, RenderTargets::FINAL_SCENE);
		//ambient
		EntityManager::entityVector* vEntities = entityManager.EntitiesWithComponents<AmbientComponent>();
		Effect* pEffect = m_pGfx3D->GetEffect(L"Light.fx");
		if(vEntities->size() != 0)
		{
			pEffect->SetTechnique("Ambient");
			for(auto entity: *vEntities)
			{
				AmbientComponent* pComponent = entity->GetComponent<AmbientComponent>();
				FColor color = pComponent->m_ambient;
				pEffect->SetFloat3("cAmbientColor", color.r, color.g, color.b);
				pEffect->SetTexture("cMaterial", m_pGfx3D->Get3DRenderTarget(RenderTargets::DEFERRED_MATERIAL));
				pEffect->Begin();
				m_pGfx3D->RenderScreen();
				pEffect->End();
			}
		}
		delete vEntities;
		//directional
		vEntities = entityManager.EntitiesWithComponents<DirectionComponent, DiffuseComponent, SpecularComponent>();
		pEffect = m_pGfx3D->GetEffect(L"Light.fx");
		if(vEntities->size() != 0)
		{
			pEffect->SetTechnique("Directional");
			//todo: real camera
			pEffect->SetFloat3("cCameraPos", 10.0f, 0.0f, 0.0f);
			for(auto entity: *vEntities)
			{
				DirectionComponent* pDirection = entity->GetComponent<DirectionComponent>();
				pEffect->SetFloat3("cLightDirection", pDirection->m_x, pDirection->m_y, pDirection->m_z);
				//diffuse component
				FColor color = entity->GetComponent<DiffuseComponent>()->m_diffuse;
				pEffect->SetFloat3("cDiffuseColor", color.r, color.g, color.b);
				//specular component
				FColor specularColor = entity->GetComponent<SpecularComponent>()->m_specular;
				pEffect->SetFloat3("cSpecularColor", specularColor.r, specularColor.g, specularColor.b);
				//set g-buffer
				pEffect->SetTexture("cMaterial", m_pGfx3D->Get3DRenderTarget(RenderTargets::DEFERRED_MATERIAL));
				pEffect->SetTexture("cPosition", m_pGfx3D->Get3DRenderTarget(RenderTargets::DEFERRED_POSITION));
				pEffect->SetTexture("cNormal", m_pGfx3D->Get3DRenderTarget(RenderTargets::DEFERRED_NORMAL));
				//render
				pEffect->Begin();
				m_pGfx3D->RenderScreen();
				pEffect->End();
			}
		}
		delete vEntities;
		//unbind g-buffer
		m_pGfx3D->Set3DRenderTarget(0, RenderTargets::BACKBUFFER);
		pEffect->SetTexture("cMaterial", NULL);
		pEffect->SetTexture("cNormal", NULL);
		pEffect->SetTexture("cPosition", NULL);
	};

private: 

	Gfx3D* m_pGfx3D;
};

 

Other than that, no device render states etc... are set. The backbuffer and z-buffer is cleared afterwards, begin() and end() is applied correctly, as my gui using the LPD3DXSPRITE is working properly (see attachment).

 

 

So, I know this is probably even harder to tell than for me, but do you have any other idea what could be failing here? Using PIX I determined that almost everything is correct, there just is no output to my g-buffers... I'm pretty desperate after 3 hours of successless debugging, so any idea would be highly appreciated!

 

 

 

0

Share this post


Link to post
Share on other sites

Does you laptop support enough render targets? Older GPUs had limit, so depending what you have it might cause issues. Also make sure laptop actually uses GPU, not integrated.

0

Share this post


Link to post
Share on other sites

Yes, render targets are supported, my laptop is not two weeks old with an AMD Radeon 7550/7650 GPU, no integrated. Plus, PIX is showing me it is capable of at least 4 rendertargets, and since I'm checking all the HRESULTs and throw an exception on fail I should get one if too little rendertargets where supported, I quess. Even if too little rendertargets were supported, shouldn't at least the first (material, in my case) buffer be filled? All of them are empty, tested via PIX as well as rendering each of them to the backbuffer.

 

Could it be a fault of Windows 8? I can't enable the debug runtime (options greyed out) to check if there are any debug outputs aside from the methods returning a fail message... my tower PC is running Win7, could this be a reason?

Edited by The King2
0

Share this post


Link to post
Share on other sites

I know we have had some issues with this and found a resolution. let me get telanor to take a look and see what he can find, if anything.

0

Share this post


Link to post
Share on other sites

My guess is that you have errors/warnings in the D3D runtime (for example binding a texture that is still bound as RenderTarget).

Not all drivers/GPUs behave the same when there are warnings/errors ...

Normally you can see the debug output inside pix (or inside visual studio 2012).

Edited by Yourself
0

Share this post


Link to post
Share on other sites

@riuthamus:

 

Thanks, that would be great!

 

@Yourself:

 

Unfortunately (really, I'd be glad if such a thing was going on) there are no errors/warning (except a lot of "ignoring redundand sampler state set"), three days ago I encountered the bug that you described with binding a texture that is still a rendertarget, no such thing anymore... So unless GPUs behave different in the errors or warnings they throw, this is out of the questions.

Edited by The King2
0

Share this post


Link to post
Share on other sites

struct VS_INPUT
{
float4 vPos : POSITION0;                  <---- should this be a float3? 
float2 vTex0 : TEXCOORD0;
float3 vNrm : NORMAL0;
float3 vTan : TANGENT0;
float3 vBin : BINORMAL0;
};

 

That maybe not be your problem. I know I had some problems with using a float4 instead of a float3. The rest of the code look fine. 

Edited by mgoss
0

Share this post


Link to post
Share on other sites

That maybe not be your problem. I know I had some problems with using a float4 instead of a float3. The rest of the code look fine.

 

No, that isn't the problem here, indeed it was needed to expand the input parameter to float4 and adjusting the pixel shader accordingly to get the world position rendered out to the g-buffer correctly. The transformation is shown to be correct in the PIX, too. Pre-VS, Post-VS and Viewport all show the cube at the position where it is meant to be. It just isn't rendered out. Thanks for the suggestion anyway.

 

Some more suggestions? I'm pretty desperate now, I tried to disable everything - z-buffer, scissor test, etc... . I even used a direct renderer instead of my deferred one, which produces exactly the same result. Any help would be highly apprechiated!

0

Share this post


Link to post
Share on other sites

So thanks to Git I was finally able to find out the problem - I'm posting the solution in case someone else encounters that too.

 

Take a look at these lines from my shaders:

 

 

struct VS_OUTPUT
{
    float4 vPos             : POSITION0;
    float4 vPos1            : POSITION1;
    float2 vTex0            : TEXCOORD0;
    float3x3 mTBN        : TEXCOORD1;
}

VS_OUTPUT mainVS(VS_INPUT i)
{
    VS_OUTPUT o;

    float4x4 WorldViewProj = mul(World,ViewProj);
    o.vPos = mul(i.vPos, WorldViewProj);
    o.vPos1 = mul(i.vPos, World);

    o.vTex0 = i.vTex0;

    float3 N = normalize(mul(i.vNrm, World));
    float3 T = normalize(mul(i.vTan, World));
    float3 B = normalize(mul(i.vBin, World));
    o.mTBN = float3x3(T, B, N);

    return o;
}

Don't see anything suspicious? Neigther did I, but let me tell you, if I change the following line accordingly:

 

 

    o.vPos1 = mul(i.vPos, World);
 //change this line to:
    o.vPos1 = o.vPos;

Then it will work. Of course now I have the position in Screen-Space, which I don't want. Here's the solution:

 

struct VS_OUTPUT
{
    float4 vPos             : POSITION0;
    float2 vTex0            : TEXCOORD0;
    float4 vPos1            : TEXCOORD1;
    float3x3 mTBN        : TEXCOORD2;
};


Its so simple. One f***ing line of code, and all those weeks of trying and trying again without getting it to work.
Moral of the story, never declare your g-buffer attributes as POSITIONX, even if they represent, well, a position. May someone explain to me, why? Especially: Why does one PC accept that, and another one doesn't?

Edited by The King2
0

Share this post


Link to post
Share on other sites

I thought POSITIONn semantics weren't passed to the pixel shader (only used for the rasterizer). I'm surprised it worked on the first computer.

 

FWIW, I get compile errors about POSITION if I compile that under the ps_2_0 profile, but not ps_3_0. So, at least the fxc compiler thinks it's ok to use POSITION in the pixel shader (for ps_3_0).

0

Share this post


Link to post
Share on other sites

I thought POSITIONn semantics weren't passed to the pixel shader (only used for the rasterizer). I'm surprised it worked on the first computer.

 

POSITION0 isn't passed to the pixel shader, POSITION1 however is, though it still seems to being used by the rasterizer too. This would somehow explain why no pixels where drawn on my notebook. Still I'd be curious what was really going on, and why it worked on my tower. Well, I quess I'll never find out though...

0

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now
Sign in to follow this  
Followers 0