|
public Draw2DImage ( |
||
| image | ||
| destPos | IrrlichtNETCP.Position2D | |
| 리턴 | void | |
static void Main(string[] args)
{
//We choosed OpenGL because it is cross-platform and we only have the openGL shader
//So please do not change this unless you change the shader !
IrrlichtDevice device = new IrrlichtDevice(DriverType.Direct3D9,
new Dimension2D(640, 480),
32, false, true, true, false);
//We set a new working directory
device.FileSystem.WorkingDirectory = "../../medias";
device.OnEvent += new OnEventDelegate(device_OnEvent);
//We set a basic caption
string caption = "Irrlicht .NET CP Shaders and Particles Example";
//We set our handlers
Driver = device.VideoDriver;
Scene = device.SceneManager;
//We have already seen that... The only special thing is the pointless emitter (what a funny name isn't it ?)
//Which is detailed just down.
Texture fire = Driver.GetTexture("fire.bmp");
ParticleSystemSceneNode particles = Scene.AddParticleSystemSceneNode(false, null, -1);
particles.SetEmitter(new PointlessEmitter());
//particles.AddAffector(new PointlessAffector());
particles.SetMaterialTexture(0, fire);
particles.SetMaterialType(MaterialType.TransparentAddColor);
particles.SetMaterialFlag(MaterialFlag.Lighting, false);
particles.ParticleSize = new Dimension2Df(50, 50);
particles.ParticlesAreGlobal = false;
particles = Scene.AddParticleSystemSceneNode(false, null, -1);
particles.SetEmitter(new PointlessEmitter());
//particles.AddAffector(new PointlessAffector());
particles.SetMaterialTexture(0, fire);
particles.SetMaterialType(MaterialType.TransparentAddColor);
particles.SetMaterialFlag(MaterialFlag.Lighting, false);
particles.ParticleSize = new Dimension2Df(50, 50);
particles.Position = new Vector3D(0, 400, 0);
particles.ParticlesAreGlobal = false;
particles = Scene.AddParticleSystemSceneNode(false, null, -1);
particles.SetEmitter(new PointlessEmitter());
//particles.AddAffector(new PointlessAffector());
particles.SetMaterialTexture(0, fire);
particles.SetMaterialType(MaterialType.TransparentAddColor);
particles.SetMaterialFlag(MaterialFlag.Lighting, false);
particles.ParticleSize = new Dimension2Df(50, 50);
particles.Position = new Vector3D(0, -400, 0);
particles.ParticlesAreGlobal = false;
//Here we only create 3 cubes and add a texture... Nothing exciting
SceneNode cube1, cube2, cube3;
cube1 = Scene.AddCubeSceneNode(40f, null, -1);
cube2 = Scene.AddCubeSceneNode(40f, null, -1);
cube3 = Scene.AddCubeSceneNode(40f, null, -1);
cube1.SetMaterialTexture(0, Driver.GetTexture("rockwall.bmp"));
cube2.SetMaterialTexture(0, Driver.GetTexture("rockwall.bmp"));
cube3.SetMaterialTexture(0, Driver.GetTexture("rockwall.bmp"));
//Here comes the fun... We create two low level shaders (taken from Irrlich's shader Example) and we set the base material as
//Solid for the first one and transparent for the second
int mat = Driver.GPUProgrammingServices.AddShaderMaterialFromFiles("opengl.vsh", "opengl.psh", OnShaderSet, MaterialType.Solid, 0);
int mat2 = Driver.GPUProgrammingServices.AddShaderMaterialFromFiles("opengl.vsh", "opengl.psh", OnShaderSet, MaterialType.TransparentAddColor, 0);
//And now we add both materials... Notice that no cast is needed because SetMaterialType
//Has an overload especially made for shaders !
cube2.SetMaterialType(mat);
cube3.SetMaterialType(mat2);
cube2.Position = new Vector3D(0, -40, 0);
cube3.Position = new Vector3D(0, 40, 0);
//We create a fixed cam for our render target scene
CameraSceneNode fixedcam = Scene.AddCameraSceneNode(null);
fixedcam.Position = new Vector3D(50, 0, -1000);
fixedcam.FarValue = 10000f;
fixedcam.AddAnimator(Scene.CreateFlyCircleAnimator(fixedcam.Target, 5000f, 0.001f));
//And a fps cam for our main scene
CameraSceneNode fpscam = Scene.AddCameraSceneNodeFPS(null, 100f, 400f, false);
Scene.ActiveCamera.Position = new Vector3D(50, 0, -1000);
Scene.ActiveCamera.FarValue = 10000f;
Texture mask = Driver.AddTexture(new Dimension2D(128, 128), "", ColorFormat.A8R8G8B8);
//Here is another cool feature from .NET CP
//Direct access to textures via Texture.Lock/Unlock and Texture.SetPixel/GetPixel
//Uncomment these lines to see the effect :)
/*int w = mask.OriginalSize.Width;
int h = mask.OriginalSize.Height;
double maxdistance = Math.Sqrt(Math.Pow(w, 2) + Math.Pow(h, 2));
mask.Lock();
for (int x = 0; x < w; x++)
for (int y = 0; y < h; y++)
{
double distance = Math.Sqrt(Math.Pow((w - x), 2) + Math.Pow((h - y), 2));
double coeff = distance / maxdistance;
int color = (int)(255 - (255 * coeff));
Color pixel = new Color((int)(color / 1.5), (int)(color / 1.5), (int)(color / 2), (int)(color / 2));
mask.SetPixel(x, y, pixel);
}
mask.Unlock();
//Again a useful feature... Direct texture saving to common formats such as png,jpg,bmp or gif...
//Without coding anything, your texture is directly saved in Irrlicht's working directory !
mask.Save("image.png");*/
//You may have noticed that GetPixel/SetPixel is very slow even if you lock the texture.
//Another insecure but quite faster way to proceed is to convert our lock to a pointer.
//It needs unsafe code but works very fast...
//However it is very unsecure and you must know perfectly what you are doing...
//That's why Modify/Retrieve exists (look down)
/*int w = mask.OriginalSize.Width;
int h = mask.OriginalSize.Height;
double maxdistance = Math.Sqrt(Math.Pow(w, 2) + Math.Pow(h, 2));
IntPtr lockresult = mask.Lock();
unsafe
{
int* directacces = (int*)(void*)lockresult;
int pitch = mask.Pitch / 4;
for (int x = 0; x < w; ++x)
for (int y = 0; y < h; ++y)
{
double distance = Math.Sqrt(Math.Pow((w - x), 2) + Math.Pow((h - y), 2));
double coeff = distance / maxdistance;
int color = (int)(255 - (255 * coeff));
Color pixel = new Color((int)(color / 1.5), (int)(color / 1.5), (int)(color / 2), (int)(color / 2));
directacces[x + y * pitch] = pixel.NativeColor;
}
}
mask.Unlock();*/
//Here is the SAFEST AND FASTEST way to modify our texture.
//A simple delegate which takes the coords of the pixel and returns the color
//It is called on each pixel and you can even return false if you don't want to change the pixel.
//We can now create our mask without speed or compatibility issue !
int w = mask.OriginalSize.Width / 2;
int h = mask.OriginalSize.Height / 2;
double maxdistance = Math.Sqrt(Math.Pow(w, 2) + Math.Pow(h, 2));
//We create a delegate to modify each pixel
ModifyPixel del = delegate(int x, int y, out Color col)
{
//Here is our formula... You can modify it as you wish, this is just an example !
double distance = Math.Sqrt(Math.Pow((w - x), 2) + Math.Pow((h - y), 2));
double coeff = distance / maxdistance;
int color = (int)(255 - (255 * coeff));
//We now set the out color
col = new Color((int)(color / 1.5), (int)(color / 1.5), (int)(color / 2), (int)(color / 2));
//And we return true since we created a color.
//If you return false, the current pixel won't be modified !
return true;
};
//And we modify our texture with our delegate
mask.Modify(del);
//Uncomment this line to save our mask on a portable network graphic file !
//You do not need to know jpg, bmp, gif, png... formats since they are converted automatically
//mask.Save("mask.png");
Console.WriteLine();
Console.WriteLine("============================================");
Console.WriteLine("Features List :");
Console.WriteLine();
for (VideoDriverFeature feat = 0; feat < VideoDriverFeature.Count; feat++)
Console.WriteLine(feat + " = " + Driver.QueryFeature(feat));
Console.WriteLine("============================================");
//Here is our logo
Texture logo = Driver.GetTexture("NETCPlogo.png");
//Here is another feature, RenderTarget
Texture renderTarget = Driver.CreateRenderTargetTexture(new Dimension2D(320, 240));
int lastfps = -1, fps = 0;
while (device.Run() && !Exit)
{
Driver.BeginScene(true, true, Color.Gray);
//Driver.SetRenderTarget(renderTarget, true, true, Color.TransparentBlue);
//Scene.ActiveCamera = fixedcam;
//Scene.DrawAll();
//Driver.SetRenderTarget(null, true, true, Color.Gray);
Scene.ActiveCamera = fpscam;
Scene.DrawAll();
//Ok, seems like Irrlicht 1.3.1 doesn't like drawing render target on OpenGL
//I guess we will disable this for now...
//Driver.Draw2DImage(renderTarget, new Position2D(0, 0), false);
Driver.Draw2DImage(mask, new Rect(new Position2D(), Driver.ScreenSize), new Rect(new Position2D(), mask.OriginalSize), Color.White, true);
//And finally our logo is painted
Driver.Draw2DImage(logo, new Position2D(0, Driver.ScreenSize.Height - logo.OriginalSize.Height),
new Rect(new Position2D(0, 0),
logo.OriginalSize), Color.White, true);
Driver.EndScene();
fps = Driver.FPS;
if(fps != lastfps)
{
device.WindowCaption = caption + " - FPS : " + fps;
lastfps = fps;
}
}
device.Close();
}
public static void Main(string[] args)
{
//Shall we fullscreen ? (new verb I just invented)
bool fullscreen = false;
//We check the optimal video mode (resolution and screen depth)
//As a default we choose 800x600x32
VideoMode optimalmode;
optimalmode.Resolution = new Dimension2D(800, 600);
optimalmode.Depth = 32;
//OpenGL is platform independent whereas DirectX isn't
//As a default we choose OpenGL
DriverType drivertype = DriverType.OpenGL;
//We set a cool caption for our example
string caption = "Irrlicht.NET CP Example on ";
//Is the application started on a Windows platform ?
//You may wonder why I did not wrote "Environment.OSVersion.Platform == PlatformID.Unix"
//The reason is simple : We do not need to know if we are on an unix platform,
//We need to know if we ARE NOT ON A WINDOWS platform, that's different indeed.
if (Environment.OSVersion.Platform != PlatformID.Win32Windows &&
Environment.OSVersion.Platform != PlatformID.Win32NT &&
Environment.OSVersion.Platform != PlatformID.Win32S)
{
caption += "Linux/Unix System";
//If the script is compiled on debug mode, we set 1024x768x32 with fullscreen.
//I guess most of users will have at least a 15' screen...
#if !DEBUG
fullscreen = true;
optimalmode.Resolution = new Dimension2D(1024, 768);
optimalmode.Depth = 32;
#endif
}
else //We ARE on a Windows Platform
{
caption += "Windows";
//Uncomment this line and the engine will use DirectX 9 on a Windows system
//I don't like Direct3D 9 but well the choice is yours ;)
//drivertype = DriverType.Direct3D9;
//Here we are, if the engine is compiled with release settings,
//We can play with a funny toy : fakedevice.
//It does not seem to work on Linux (perhaps one of those bugs that only appear on some cursed computers like mine)
//With this device, we will determine the ideal video mode by forcing
//The application to launch on the desktop video mode
//If a bug occurs (meaning video mode is null for instance)
//We set the same video mode as on Linux : 1024x768x32 which is the most common video mode.
#if !DEBUG
IrrlichtDevice fakedevice = new IrrlichtDevice(DriverType.Null, new Dimension2D(), 16, false, false, false, false);
optimalmode = fakedevice.DesktopVideoMode;
if(optimalmode.Resolution.Width == 0 || optimalmode.Resolution.Height == 0)
{
optimalmode.Resolution = new Dimension2D(1024, 768);
optimalmode.Depth = 32;
}
fakedevice.Dispose();
fullscreen = true;
#endif
}
//We add a cool caption which says which renderer (OpenGL or Direct3D9 for instance) we use
caption += " With " + drivertype + " - ";
//Here we are, we create the device with settings we determined before
IrrlichtDevice device = new IrrlichtDevice(drivertype,
optimalmode.Resolution,
optimalmode.Depth, fullscreen,
true, //Stencil Buffer (for shadow)
false, //Vertical Synchronisation (use it if you want your application not to go over 70 FPS)
false); //Anti Aliasing
scene = device.SceneManager; //We get some object such as the scene manager
driver = device.VideoDriver;
guienv = device.GUIEnvironment;
device.FileSystem.WorkingDirectory = "../../medias"; //We set Irrlicht's current directory to %application directory%/media
device.CursorControl.Visible = false; //Let's hide the cursor
device.OnEvent += new OnEventDelegate(device_OnEvent); //We had a simple delegate that will handle every event
device.WindowCaption = caption; //And we set a basic caption
//We create a floor mesh which is actually an hill plane mesh without height
AnimatedMesh floormesh = scene.AddHillPlaneMesh("_MyHill_", new Dimension2Df(250, 250),
new Dimension2D(5, 5),
0f, new Dimension2Df(0, 0),
new Dimension2Df(10, 10));
//We make the planar texture mapping to set the texture resolution
scene.MeshManipulator.MakePlanarTextureMapping(floormesh.GetMesh(0), 0.006f);
//We create a new mesh with tangents. It is needed for Parallax mapping
Mesh tFloor = scene.MeshManipulator.CreateMeshWithTangents(floormesh.GetMesh(0));
Texture heightmap = driver.GetTexture("rockwall_height.bmp"); //Our normal map
driver.MakeNormalMapTexture(heightmap, 10f); //We make a quite exagerated normal map (just to show off)
//And here we are, we finally create the scene node
SceneNode floor = scene.AddMeshSceneNode(tFloor, null, -1);
//We get the material
Material mat = floor.GetMaterial(0);
mat.Texture1 = driver.GetTexture("rockwall.bmp"); //Diffuse texture
mat.Texture2 = heightmap; //Normal map
mat.MaterialType = MaterialType.ParallaxMapSolid; //The beautiful Parallax Mapping
mat.MaterialTypeParam = 0.035f; //Parameter for the height of parallax mapping
//We create a FPS camera which is a basic camera controlled by
//Arrow keys and mouse such as the camera in Quake or Doom
CameraSceneNode cam = scene.AddCameraSceneNodeFPS(null, 50f, 100f, true);
cam.Position = new Vector3D(0, 100, -100);
cam.FarValue = 10000f; //We want to see all the scene
//We add the sword (don't tell me that it is not a katana, I already know it !)
SceneNode katana = scene.AddMeshSceneNode(scene.GetMesh("katana.x").GetMesh(0), cam, -1);
Vector3D KInitialRotation = new Vector3D(0, 20, -90);
Vector3D KInitialPosition = new Vector3D(23, -3, 40);
//We set our materials (the n 1 is the sword and the 0 the stick)
katana.Scale = new Vector3D(100f, 100f, 100f);
mat = katana.GetMaterial(1);
mat.DiffuseColor = new Color(255, 90, 90, 90);
mat.AmbientColor = new Color(255, 90, 90, 90);
mat.EmissiveColor = new Color(255, 90, 90, 90);
mat.SpecularColor = new Color(255, 90, 90, 90);
Material mat2 = katana.GetMaterial(0);
mat2.DiffuseColor = new Color(255, 10, 10, 10);
mat2.EmissiveColor = new Color(255, 120, 80, 0);
//We add a little shining effect with particles on the sword
ParticleSystemSceneNode particles = scene.AddParticleSystemSceneNode(false, katana, -1);
particles.SetEmitter(particles.CreateBoxEmitter(scene.GetMesh("katana.x").GetMesh(0).GetMeshBuffer(1).BoundingBox,
new Vector3D(0, 0.002f, 0),
1000, 1000,
new Color(0, 255, 255, 255),
new Color(0, 255, 255, 255),
200, 200, 0));
particles.ParticleSize = new Dimension2Df(1f, 3f); //A funny size for our funny effect
particles.AddAffector(particles.CreateFadeOutParticleAffector(new Color(0, 0, 0, 0), 100));
particles.SetMaterialTexture(0, driver.GetTexture("fire.bmp"));
particles.SetMaterialType(MaterialType.TransparentAddColor);
particles.SetMaterialFlag(MaterialFlag.Lighting, false);
//We create three dwarves with shadows
AnimatedMeshSceneNode dwarf = scene.AddAnimatedMeshSceneNode(scene.GetMesh("dwarf.x"));
dwarf.AnimationSpeed = 15;
dwarf.Position = new Vector3D(0, 0, 100);
dwarf.AddShadowVolumeSceneNode(-1, true, 10000f); //Wow... It was really hard to create it !
dwarf.Scale = new Vector3D(1.5f, 1.5f, 1.5f);
//Here we have our light that will simply rotate around the dwarf
LightSceneNode dwarflight = scene.AddLightSceneNode(dwarf, new Vector3D(0, 0, 0), Colorf.White, 10000f, -1);
dwarflight.AddAnimator(scene.CreateFlyCircleAnimator(new Vector3D(0, 100, 0), 100f, 0.001f));
dwarf = scene.AddAnimatedMeshSceneNode(scene.GetMesh("dwarf.x"));
dwarf.AnimationSpeed = 10;
dwarf.Position = new Vector3D(-100, 0, 100);
dwarf.AddShadowVolumeSceneNode(-1, true, 10000f);
dwarf.Scale = new Vector3D(1.5f, 1.5f, 1.5f);
dwarf = scene.AddAnimatedMeshSceneNode(scene.GetMesh("dwarf.x"));
dwarf.AnimationSpeed = 20;
dwarf.Position = new Vector3D(100, 0, 100);
dwarf.AddShadowVolumeSceneNode(-1, true, 10000f);
dwarf.Scale = new Vector3D(1.5f, 1.5f, 1.5f);
//We set the shadow color.
//I reduced a lot the opacity since our light is not supposed to be the sun
//And the scene is at night... Thus a shadow is not something very visible.
scene.ShadowColor = new Color(100, 0, 0, 0);
//We had a simple billboard to represent physically the light
BillboardSceneNode lightbill = scene.AddBillboardSceneNode(dwarflight, new Dimension2Df(20, 20), -1);
Texture fire = driver.GetTexture("fire.bmp");
lightbill.SetMaterialTexture(0, fire);
lightbill.SetMaterialType(MaterialType.TransparentAddColor);
lightbill.SetMaterialFlag(MaterialFlag.Lighting, false);
//We add a simple skybox
scene.AddSkyBoxSceneNode(null, new Texture[] { driver.GetTexture("irrlicht2_up.jpg"),
driver.GetTexture("irrlicht2_dn.jpg"),
driver.GetTexture("irrlicht2_lf.jpg"),
driver.GetTexture("irrlicht2_rt.jpg"),
driver.GetTexture("irrlicht2_ft.jpg"),
driver.GetTexture("irrlicht2_bk.jpg") },
-1);
//We had a simple texture that will be rendered each time
Texture matthias = driver.GetTexture("matthias.png");
//Here is our logo
Texture logo = driver.GetTexture("NETCPlogo.png");
//Just a simple demonstration of a feature which is not on Irrlicht .NET :
//Listing children of the node as a simple Array.
foreach (SceneNode node in cam.Children)
Console.WriteLine("Child : " + node.ToString() + " Parent : " + node.Parent.ToString());
//Another cool new feature :
//FileSystem class with features like listing every file that Irrlicht can load
//(including files on zip archives for instance)
//foreach (FileListItem item in device.FileSystem.FileList)
// Console.WriteLine(item.ToString());
double random = 0;
Random chance = new Random();
int lastfps = -1;
while (device.Run() && !Exit)
{
//We get the FPS to synchronize all movements
//Notice that I advice NOT TO USE THE FPS to synchronize
//Because it is updated every second whereas it can change at every time
//Use the time between each loop instead
//(We use it here because it is not very important but be very careful)
int FPS = driver.FPS;
if(FPS != lastfps)
{
//And here we are, we set the caption of the main window
device.WindowCaption = caption + "FPS : " + FPS;
lastfps = FPS;
}
if (FPS < 10)
FPS = 70;
random += ((2000f / FPS) + chance.Next(-10, 10)) / 1000f;
//If someone has played the demo about thirty years...
if (random >= double.MaxValue)
random = 0;
//Sinuses and Cosines are very useful because we all know that
//they are between -1 and 1. Thus we don't need to add some "if"
//And we can control the domain easilly
//We had a realstic move of the sword because try to hold one during 10 minutes
//You won't be static... Tested !
//A nice thing would be to generate it using the camera movement speed or things like that...
katana.Position = KInitialPosition +
new Vector3D((float)(Math.Sin(random) / 6.0f),
(float)(Math.Sin(random) / 10f),
(float)(Math.Cos(random) / 6.0f));
katana.Rotation = KInitialRotation +
new Vector3D((float)(Math.Cos(random) / 2f),
(float)(Math.Cos(random)),
(float)(Math.Sin(random) / 2f));
//We clear the back buffer and begin the scene
driver.BeginScene(true, true, Color.From(255, 50, 50, 50));
//First we draw all 3D Objects such as our sword or the dwarves
//Notice that the order is very important... Try to move this line just before the
//driver.EndScene and you will see the difference
scene.DrawAll();
//3 rectangles that could for instance represent Health, Mana and Endurance points...
driver.Draw2DRectangle(new Rect(new Position2D(10, 5), new Dimension2D(300, 15)), Color.Red);
driver.Draw2DRectangle(new Rect(new Position2D(10, 25), new Dimension2D(300, 15)), Color.Blue);
driver.Draw2DRectangle(new Rect(new Position2D(10, 45), new Dimension2D(300, 15)), Color.Green);
//A little image loaded before that could represent for instance our character
driver.Draw2DImage(matthias, new Position2D(driver.ScreenSize.Width - matthias.OriginalSize.Width, 0),
new Rect(new Position2D(0, 0),
matthias.OriginalSize), Color.White, true);
//And finally our logo is painted
driver.Draw2DImage(logo, new Position2D(0, driver.ScreenSize.Height - logo.OriginalSize.Height),
new Rect(new Position2D(0, 0),
logo.OriginalSize), Color.White, true);
//Finally we draw everything in the GUI Environment... CF the GUI example
guienv.DrawAll();
//End of the scene, the back buffer is displayed
driver.EndScene();
}
//ALWAYS DISPOSE THE DEVICE AT THE END
//It is REQUIRED on Linux because if you don't, XWindow will stay at the old video mode
//And you can't imagine how ugly it is not to get back to the original video mode
device.Close();
}