Search Results

Search found 31839 results on 1274 pages for 'plugin development'.

Page 537/1274 | < Previous Page | 533 534 535 536 537 538 539 540 541 542 543 544  | Next Page >

  • Detecting collision between ball (circle) and brick(rectangle)?

    - by James Harrison
    Ok so this is for a small uni project. My lecturer provided me with a framework for a simple brickbreaker game. I am currently trying to overcome to problem of detecting a collision between the two game objects. One object is always the ball and the other objects can either be the bricks or the bat. public Collision hitBy( GameObject obj ) { //obj is the bat or the bricks //the current object is the ball // if ball hits top of object if(topX + width >= obj.topX && topX <= obj.topX + obj.width && topY + height >= obj.topY - 2 && topY + height <= obj.topY){ return Collision.HITY; } //if ball hits left hand side else if(topY + height >= obj.topY && topY <= obj.topY + obj.height && topX + width >= obj.topX -2 && topX + width <= obj.topX){ return Collision.HITX; } else return Collision.NO_HIT; } So far I have a method that is used to detect this collision. The the current obj is a ball and the obj passed into the method is the the bricks. At the moment I have only added statement to check for left and top collisions but do not want to continue as I have a few problems. The ball reacts perfectly if it hits the top of the bricks or bat but when it hits the ball often does not change directing. It seems that it is happening toward the top of the left hand edge but I cannot figure out why. I would like to know if there is another way of approaching this or if people know where I'm going wrong. Lastly the collision.HITX calls another method later on the changes the x direction likewise with y.

    Read the article

  • How to determine the end of list has been reached?

    - by Sweta Dwivedi
    I'm trying to animate my object according to a set of recorded values from kinect skeleton stream by saving the (x,y,z) stream from the skeletal data into a list and then set my objects x and y position from the x,y of the list. However, once the list end has been reached it starts to animate again from the start. I don't want that - I just want the model position to keep going in the positive X direction. Is there any way I can check if end of the list has been reached and to just update the model position in x direction? Or is there any other way to continue moving my sprite once the points in the list are over... i dont want it to start animating all the way again.. protected override void Update(GameTime gameTime) { //position += spriteSpeed * (float)gameTime.ElapsedGameTime.TotalSeconds; //// TODO: Add your update logic here using (StreamReader r = new StreamReader(f)) { string line; Viewport view = graphics.GraphicsDevice.Viewport; int maxWidth = view.Width; int maxHeight = view.Height; while((line = r.ReadLine()) != null) { string[] temp = line.Split(','); int x = (int) Math.Floor(((float.Parse(temp[0]) * 0.5f) + 0.5f) * maxWidth); int y = (int) Math.Floor(((float.Parse(temp[1]) * -0.5f) + 0.5f) * maxHeight); motion_2.Add(new Point(x, y)); } } position.X = motion_2[i].X; position.Y = motion_2[i].Y; i++; a_butterfly_up.Update(gameTime); a_butterfly_side.Update(gameTime); G_vidPlayer.Play(mossV); base.Update(gameTime); }

    Read the article

  • How can I determine if a cube is adjacent to another cube, and optimize its buffers if so?

    - by Christian Frantz
    I'm trying to optimize the rendering of a collection of cubes, (based on an answer I was given to another question I asked). I understand the logic behind occlusion culling, but I'm having trouble with the code. When I create a cube, I want to determine if that cube is touching another existing cube, and if so I don't want to generate the redundant data in my vertex or index buffers. I'm planning on making a method that I call from my cube constructor so that everytime I create a cube, these checks are made, and neither occluded face is ever drawn. How would I go about this?

    Read the article

  • How does a BSP tree work for Z sorting?

    - by Jenko
    I'm developing a 3D engine in software, and so I must compute Z sorting manually. I'm currently using the painters algorithm to sort triangles and then drawing them back-to-front. This causes artifacts that I'm trying to correct. Would using a dynamic BSP-tree ensure "correct Z sorting" of triangles? Why? Because the bounding volumes of triangles would be similar? Since I would have a single "world" BSP tree, would I have to remove and re-add any moved/scaled/rotated object into the tree? Is it possible to add triangles into a BSP tree without the expensive cutting process? Why do you need to cut triangles on the axis planes anyway? Is it faster to traverse a BSP tree from any angle, than to sort all tris each draw like the painters algorithm?

    Read the article

  • Easiest turn-base games you can think of?

    - by Edgar Miranda
    I'm planning to get into the process of programming multiplayer turn-base games. I would like to start off by making some of the simplest (yet fun) multiplayer turn-base games out there. What are some that you can provide? For example... Tic-Tac-Toe Rock-Paper-Scissors Checkers Some not so easy games... 4 in a row chess poker In terms of "ease" of implementation I'm mainly looking at logic. For example, Rock-Paper-Scissors has very simple logic, while chess has logic that is more complicated. So far I have the following: Hexagon Heroes of Might and Magic Nine Men's Morris Connect 4 21 (card game) Pen the Pig (The Dot game) Memory Match

    Read the article

  • Cocos2d-x v3.1 for WinPhone 8 auto change texture after resumed from background

    - by Bình Nguyên
    I have some sprites in my game (for Windows Phone 8). These are my steps to reproduce the problem: Open game Play (this is an optional step) Press Windows button to send game to background Press Back button to resume game The problem is: After the game has resumed, some sprites exchange textures, some sprites go black (like there is no texture being bound). I'm using cocos2dx version 3.1.1. Can someone help me to solve this problem?

    Read the article

  • Worker roles in Windows Azure to host a multiplayer server

    - by MrWiggels
    I've been doing research on where to host a simple multi-player backend for a simple game I'm developing. So as a first choice I downloaded the Windows Azure SDK, which provides a nice and simple emulator environment where you can test out your application before uploading. I also download the Azure Social Game Toolkit (Visit), and followed as far as my understanding can take me. So, down to the main question. Is there anybody with experience developing Azure applications. I'm developing a Action RPG game, in a similar vein to Diablo III. I was thinking of putting up Matchmaking, Friends Lists, etc. Is there another way to connect to Azure services via something like UDP or TCP for sending packets or does everything have to go through HTTP requests? Is it even possible to use HTTP request/response for something like this? All game commands will be simple. Because the game server and the clients will be kept in-sync and will have deterministic actions, I'm just going to send actions like "Use Primary Skill" and "Use Secondary Skill". Any hints, ideas, light bulbs or a smack-in-the-face presentation will be much appreciated.

    Read the article

  • Moving in a diamond - enemy gets stuck

    - by Fibericon
    I have an enemy that I would like to move as follows: Start at (0, 200, 0) Move to (200, 0, 0) Move to (0, -200, 0) Move to (-200, 0, 0) Move to start point, repeat as long as it remains active. This is what I've done to achieve that: if (position.X < 200 && position.Y > 0) { Velocity = new Vector3(1, -1, 0) * speed; } else if (position.X >= 200 && position.Y <= 0 && position.Y > -200) { Velocity = new Vector3(-1, -1, 0) * speed; } else if (position.X <= 0 && position.Y <= -200) { Velocity = new Vector3(-1, 1, 0) * speed; } else { Velocity = new Vector3(1, 1, 0) * speed; } It moves to the second point, but then gets stuck and appears to vibrate in place. How should I be doing this?

    Read the article

  • What Type of Options should be on the Game Settings Menu?

    - by A13X
    I have seen a post about the main menu options here: UI: Main Menu options for mobile games. What options should be listed? What do users want to see? But I want to know what kind of options should/need to be available on the settings screen. I am making a rather simple 2D game for Android, but really I haven't found many aspects that warrant an options button or a check box besides turning the sound and music on/off. I was thinking graphics settings but then again, how many apps really need graphics settings besides immersive 3D ones?

    Read the article

  • isometric background that covers the viewport [on hold]

    - by Richard
    The background image should cover the viewport. The technique I use now is a loop with an innerloop that draws diamond shaped images on a canvas element, but it looks like a rotated square. This is a nice example: ,that covers the whole viewport. I have heard something about clickthrough maps, but what more ways are there that are most efficient with mobile devices and javascript? Any advice in grid design out there?.

    Read the article

  • How do I separate codes with classes?

    - by Trycon
    I have this main class: package javagame; import org.newdawn.slick.GameContainer; import org.newdawn.slick.Graphics; import org.newdawn.slick.SlickException; import org.newdawn.slick.state.BasicGameState; import org.newdawn.slick.state.StateBasedGame; public class tests extends BasicGameState{ public boolean render=false; tests1 test = new tests1(); public tests(int test) { // TODO Auto-generated constructor stub } @Override public void init(GameContainer arg0, StateBasedGame arg1) throws SlickException { // TODO Auto-generated method stub } @Override public void render(GameContainer arg0, StateBasedGame arg1, Graphics g) throws SlickException { // TODO Auto-generated method stub if(render==true) { g.drawString("Hello",100,100); } } @Override public void update(GameContainer gc, StateBasedGame s, int delta) throws SlickException { // TODO Auto-generated method stub test.render=render; test.update(gc, s, delta); } @Override public int getID() { // TODO Auto-generated method stub return 1000; } } and its sub-class: package javagame; import org.newdawn.slick.GameContainer; import org.newdawn.slick.Input; import org.newdawn.slick.state.StateBasedGame; public class tests1 { public boolean render; public void update(GameContainer gc, StateBasedGame s, int delta) { Input input = gc.getInput(); if(input.isKeyPressed(Input.KEY_X)) { render=true; } } } I was finding a way to prevent many codes in one class. I'm new to java. When I try running my game, then when I press X, it does not work. How am I suppose to fix that?

    Read the article

  • Understanding math used to determine if vector is clockwise / counterclockwise from your vector

    - by MTLPhil
    I'm reading Programming Game AI by Example by Mat Buckland. In the Math & Physics primer chapter there's a listing of the declaration of a class used to represent 2D vectors. This class contains a method called Sign. It's implementation is as follows //------------------------ Sign ------------------------------------------ // // returns positive if v2 is clockwise of this vector, // minus if anticlockwise (Y axis pointing down, X axis to right) //------------------------------------------------------------------------ enum {clockwise = 1, anticlockwise = -1}; inline int Vector2D::Sign(const Vector2D& v2)const { if (y*v2.x > x*v2.y) { return anticlockwise; } else { return clockwise; } } Can someone explain the vector rules that make this hold true? What do the values of y*v2.x and x*v2.y that are being compared actually represent? I'd like to have a solid understanding of why this works rather than just accepting that it does without figuring it out. I feel like it's something really obvious that I'm just not catching on to. Thanks for your help.

    Read the article

  • How do I create a third Person View using DXUTCamera in DX10?

    - by David
    I am creating a 3d flying game and using DXUTCamera for my view. I can get the camera to take on the characters position, But I would like to view my character in the 3rd person. Here is my code for first person view: //Put the camera on the object. D3DXVECTOR3 viewerPos; D3DXVECTOR3 lookAtThis; D3DXVECTOR3 up ( 5.0f, 1.0f, 0.0f ); D3DXVECTOR3 newUp; D3DXMATRIX matView; //Set the viewer's position to the position of the thing. viewerPos.x = character->x; viewerPos.y = character->y; viewerPos.z = character->z; // Create a new vector for the direction for the viewer to look character->setUpWorldMatrix(); D3DXVECTOR3 newDir, lookAtPoint; D3DXVec3TransformCoord(&newDir, &character->initVecDir, &character->matAllRotations); // set lookatpoint D3DXVec3Normalize(&lookAtPoint, &newDir); lookAtPoint.x += viewerPos.x; lookAtPoint.y += viewerPos.y; lookAtPoint.z += viewerPos.z; g_Camera.SetViewParams(&viewerPos, &lookAtPoint); So does anyone have an ideas how I can move the camera to the third person view? preferably timed so there is a smooth action in the camera movement. (I'm hoping I can just edit this code instead of bringing in another camera class)

    Read the article

  • Drawing a random x,y grid of objects within a prespective

    - by T Reddy
    I'm wrapping my head around OpenGL ES 2.0 and I think I'm trying to do something very simple, but I think the math may be eluding me. I created a simple, flat-ish cylinder in Blender that is 2 units in diameter. I want to create an arbitrary grid of these edge to edge (think of a checker board). I'm using a 3D perspective with GLKit: CGSize size = [[self view] bounds].size; _projectionMatrix = GLKMatrix4MakePerspective(GLKMathDegreesToRadians(45.0f), size.width/size.height, 0.1f, 100.0f); So, I managed to manually get all of these cylinders drawn on the screen just fine. However, I would like to understand how I can programmatically "fit" all of these cylinders on the screen at the same time given the camera location, screen size, cylinder diameter, and the number of rows/columns. So the net effect is that for small grids (i.e., 5x5) the objects are closer to the camera, but for large grids (i.e., 30x30) the objects are farther away. In either case, all of the cylinders are visible.

    Read the article

  • How to create an Orthographic display in OpenGL (ES) that handles different screen sizes and orientations?

    - by Piku
    I'm trying to create an iPad/iPhone game using GLES2.0 that contains a 3D scene with a heads-up-display/GUI overlaid on the top. However, this problem would also apply if I were to port my game to a computer and run the game in a resizable window, or allow the user to change screen resolutions... When trying to make the 2D GUI/HUD work I've made the assumption that all I'm really doing is drawing a load of 2D textured 'quads' on the screen and am trying to treat the orthographic projection as an old-style 2D display with 0,0 in the upper left and screenWidth,ScreenHeight in the lower right. This causes me all sorts of confusion when I rotate my ipad into Landscape mode since I can't work out what to put into my projection and modelview matrices to turn everything around the right way. It also gets messy if I want to support the iPad's large screen, an iPhone or a Retina display since I have to then draw three sets of textures for everything and work out which ones to use. Should I be trying to map the 2D OpenGL co-ords 1:1 with the screen? While typing out this question it occurs to me that I could keep my origin in the centre, still running -1/+1 along the axes. This would let me scale my 2D content appropriately on the different screen sizes, but wouldn't I end up with the textures being scaled and possibly losing quality? I'm using OpenGLES 2.0 and have a matrix library that has equivalents to the GLES1.1 glOrthof() and glFrustrum() calls.

    Read the article

  • What happened to .fx files in D3D11?

    - by bobobobo
    It seems they completely ruined .fx file loading / parsing in D3D11. In D3D9, loading an entire effect file was D3DXCreateEffectFromFile( .. ), and you got a ID3DXEffect9, which had great methods like SetTechnique and BeginPass, making it easy to load and execute a shader with multiple techniques. Is this completely manual now in D3D11? The highest level functionality I can find is loading a SINGLE shader from an FX file using D3DX11CompileFromFile. Does anyone know if there's an easier way to load FX files and choose a technique? With the level of functionality provided in D3D11 now, it seems like you're better off just writing .hlsl files and forgetting about the whole idea of Techniques.

    Read the article

  • Isometric algorithm producing tiles in wrong draw order

    - by David
    I've been toying with isometric and I just cant get the tiles to be in the right order. I'm probably missing something obvious and I just can't see it. Even at the risk of looking stupid, here's my code: for (int i = 0; i < Tile.MapSize; i++) { for (int j = 0; j < Tile.MapSize; j++) { spriteBatch.Draw( Tile.TileSetTexture, new Rectangle( (-j * Tile.TileWidth / 2) + (i * Tile.TileWidth / 2), (i * (Tile.TileHeight - 9) / 2) - (-j * (Tile.TileHeight - 9) / 2), Tile.TileWidth, Tile.TileHeight), Tile.GetSourceRectangle(tileID), Color.White, 0.0f, new Vector2(-350, -60), SpriteEffects.None, 1.0f); } } And here's what I end up with: messed up map Yep, bit of an issue. If anyone could help, I'd appreciate it.

    Read the article

  • Texture errors in CubeMap

    - by shade4159
    I am trying to apply this texture as a cubemap. This is my result: Clearly I am doing something with my texture coordinates, but I cannot for the life of me figure out what. I don't even see a pattern to the texture fragments. They just seem like a jumble of different faces. Can anyone shed some light on this? Vertex shader: #version 400 in vec4 vPosition; in vec3 inTexCoord; smooth out vec3 texCoord; uniform mat4 projMatrix; void main() { texCoord = inTexCoord; gl_Position = projMatrix * vPosition; } My fragment shader: #version 400 smooth in vec3 texCoord; out vec4 fColor; uniform samplerCube textures void main() { fColor = texture(textures,texCoord); } Vertices of cube: point4 worldVerts[8] = { vec4( 15, 15, 15, 1 ), vec4( -15, 15, 15, 1 ), vec4( -15, 15, -15, 1 ), vec4( 15, 15, -15, 1 ), vec4( -15, -15, 15, 1 ), vec4( 15, -15, 15, 1 ), vec4( 15, -15, -15, 1 ), vec4( -15, -15, -15, 1 ) }; Cube rendering: void worldCube(point4* verts, int& Index, point4* points, vec3* texVerts) { quadInv( verts[0], verts[1], verts[2], verts[3], 1, Index, points, texVerts); quadInv( verts[6], verts[3], verts[2], verts[7], 2, Index, points, texVerts); quadInv( verts[4], verts[5], verts[6], verts[7], 3, Index, points, texVerts); quadInv( verts[4], verts[1], verts[0], verts[5], 4, Index, points, texVerts); quadInv( verts[5], verts[0], verts[3], verts[6], 5, Index, points, texVerts); quadInv( verts[4], verts[7], verts[2], verts[1], 6, Index, points, texVerts); } Backface function (since this is the inside of the cube): void quadInv( const point4& a, const point4& b, const point4& c, const point4& d , int& Index, point4* points, vec3* texVerts) { quad( a, d, c, b, Index, points, texVerts, a.to_3(), b.to_3(), c.to_3(), d.to_3()); } And the quad drawing function: void quad( const point4& a, const point4& b, const point4& c, const point4& d, int& Index, point4* points, vec3* texVerts, const vec3& tex_a, const vec3& tex_b, const vec3& tex_c, const vec3& tex_d) { texVerts[Index] = tex_a.normalized(); points[Index] = a; Index++; texVerts[Index] = tex_b.normalized(); points[Index] = b; Index++; texVerts[Index] = tex_c.normalized(); points[Index] = c; Index++; texVerts[Index] = tex_a.normalized(); points[Index] = a; Index++; texVerts[Index] = tex_c.normalized(); points[Index] = c; Index++; texVerts[Index] = tex_d.normalized(); points[Index] = d; Index++; } Edit: I forgot to mention, in the image, the camera is pointed directly at the back face of the cube. You can kind of see the diagonals leading out of the corners, if you squint.

    Read the article

  • Logic in Entity Components Systems

    - by aaron
    I'm making a game that uses an Entity/Component architecture basically a port of Artemis's framework to c++,the problem arises when I try to make a PlayerControllerComponent, my original idea was this. class PlayerControllerComponent: Component { public: virtual void update() = 0; }; class FpsPlayerControllerComponent: PlayerControllerComponent { public: void update() { //handle input } }; and have a system that updates PlayerControllerComponents, but I found out that the artemis framework does not look at sub-classes the way I thought it would. So all in all my question here is should I make the framework aware of subclasses or should I add a new Component like object that is used for logic.

    Read the article

  • How to automatically render all opaque meshes with a specific shader?

    - by dsilva.vinicius
    I have a specular outline shader that I want to be used on all opaque meshes of the scene whenever a specific camera renders. The shader is working properly when it is manually applied to some material. The shader is as follows: Shader "Custom/Outline" { Properties { _Color ("Main Color", Color) = (.5,.5,.5,1) _OutlineColor ("Outline Color", Color) = (1,0.5,0,1) _Outline ("Outline width", Range (0.0, 0.1)) = .05 _SpecColor ("Specular Color", Color) = (0.5, 0.5, 0.5, 1) _Shininess ("Shininess", Range (0.03, 1)) = 0.078125 _MainTex ("Base (RGB) Gloss (A)", 2D) = "white" {} } SubShader { Tags { "Queue"="Overlay" "RenderType"="Opaque" } Pass { Name "OUTLINE" Tags { "LightMode" = "Always" } Cull Off ZWrite Off // Uncomment to show outline always. //ZTest Always CGPROGRAM #pragma target 3.0 #pragma vertex vert #pragma fragment frag #include "UnityCG.cginc" struct appdata { float4 vertex : POSITION; float3 normal : NORMAL; }; struct v2f { float4 pos : POSITION; float4 color : COLOR; }; float _Outline; float4 _OutlineColor; v2f vert(appdata v) { // just make a copy of incoming vertex data but scaled according to normal direction v2f o; o.pos = mul(UNITY_MATRIX_MVP, v.vertex); float3 norm = mul ((float3x3)UNITY_MATRIX_IT_MV, v.normal); float2 offset = TransformViewToProjection(norm.xy); o.pos.xy += offset * o.pos.z * _Outline; o.color = _OutlineColor; return o; } float4 frag(v2f fromVert) : COLOR { return fromVert.color; } ENDCG } UsePass "Specular/FORWARD" } FallBack "Specular" } The camera used fot the effect has just a script component which setups the shader replacement: using UnityEngine; using System.Collections; public class DetectiveEffect : MonoBehaviour { public Shader EffectShader; // Use this for initialization void Start () { this.camera.SetReplacementShader(EffectShader, "RenderType=Opaque"); } // Update is called once per frame void Update () { } } Unfortunately, whenever I use this camera I just see the background color. Any ideas?

    Read the article

  • Explicit resource loading in Ogre (Mogre)

    - by sebf
    I am just starting to learn Mogre and what I would like to do is to be able to load resources 'explicitly' (i.e. I just provide an absolute path instead of using a resource group tied to a directory). This is very different to manually loading resources, which I believe in Ogre has a very specific meaning, to build up the object using Ogres methods. I want to use Ogres resource management system/resource loading code, but to have finer control over which files are loaded and in what groups they are. I remember reading how to do this but cannot find the page again; I think its possible to do something like: Declare a resource group Declare the resource(s) (this is when the actual resource file name is provided) Initialise the resource group to actually load the resource(s) Is this the correct procedure? If so, is there any example code showing how to do this?

    Read the article

  • Moving AI in a multiplayer game

    - by Smallbro
    I've been programming a multiplayer game and its coming together very nicely. It uses both TCP and UDP (UDP for movement and TCP for just about everything else). What I was wondering was how I would go about sending multiple moving AI without much lag. At first I used TCP for everything and it was very slow when people moved. I'm currently using a butchered version of this http://corvstudios.com/tutorials/udpMultiplayer.php for my movement system and I'm wondering what the best method of sending AI movements is. By movements I mean the AI chooses left/right/up/down and the player can see this happening. Thanks.

    Read the article

  • Simple collision detection in Unity 2D

    - by N1ghtshade3
    I realise other posts exist with this topic yet none have gone into enough detail for me. I am attempting to create a 2D game in Unity using C# as my scripting language. Basically I have two objects, player and bomb. Both were created simply by dragging the respective PNG to the stage. I have set up touch controls to move player left and right; gravity of any kind is not needed as I only require it to move x units when I tap either the left or right side of the screen. This movement is stored in a script called playerController.cs and works just fine. I also have a variable health = 3 for player, which is stored in healthScript.cs. I am now at a point where I am stuck. I would like it so that when player collides with bomb, health decreases by one and the bomb object is destroyed. So what I tried doing is using a new script called playerPhysics.cs, I added the following: void OnCollisionEnter2D(Collision2D coll){ if(coll.gameObject.name=="bomb") GameObject.Destroy("bomb"); healthScript.health -= 1; } While I'm fairly sure I don't know the proper way to reference a variable in another script and that's why the health didn't decrease when I collided, bomb never disappeared from the stage so I'm thinking there's also a problem with my collision. Initially, I had simply attached playerPhysics.cs to player. After searching around though, it appeared as though player also needed a rigidBody attached to it, so I did that. Still no luck. I tried using a circleCollider (player is a circle), using a rigidBody2D, and using all manner of colliders on one and/or both of the objects. If you could please explain what colliders (if any) should be attached to which objects and whether I need to change my script(s), that would be much more helpful than pointing me to one of the generic documentation examples I've already read. Also, if it would be simple to fix the health thing not working that would be an added bonus but not exactly the focus of this question. Bear in mind that this game is 2D; I'm not sure if that changes anything. Thanks!

    Read the article

  • Formula for three competing heroes, each has one they can beat and one they're beaten by

    - by Georgiadis Abraam
    I am trying to design a game for a project I have, The main idea is: 3 Types of heroes 3 Stats per hero There are no levels involved so the differences must be located on stats. Fight logic - The logic of fight is that type1hero has good chances winning type2hero, type2hero has good chances type3hero and type3hero has good chances winning type1hero. For over a week I am trying to find a stats based formula that will allow me to fix this but I can't, I was meddling with numbers yesterday and it was decent but I can't extract the formula out of it. Could you please guide me or give me hints on how should I start creating formulas on a Non lvl game that fulfills the fight logic?

    Read the article

  • Efficiently separating Read/Compute/Write steps for concurrent processing of entities in Entity/Component systems

    - by TravisG
    Setup I have an entity-component architecture where Entities can have a set of attributes (which are pure data with no behavior) and there exist systems that run the entity logic which act on that data. Essentially, in somewhat pseudo-code: Entity { id; map<id_type, Attribute> attributes; } System { update(); vector<Entity> entities; } A system that just moves along all entities at a constant rate might be MovementSystem extends System { update() { for each entity in entities position = entity.attributes["position"]; position += vec3(1,1,1); } } Essentially, I'm trying to parallelise update() as efficiently as possible. This can be done by running entire systems in parallel, or by giving each update() of one system a couple of components so different threads can execute the update of the same system, but for a different subset of entities registered with that system. Problem In reality, these systems sometimes require that entities interact(/read/write data from/to) each other, sometimes within the same system (e.g. an AI system that reads state from other entities surrounding the current processed entity), but sometimes between different systems that depend on each other (i.e. a movement system that requires data from a system that processes user input). Now, when trying to parallelize the update phases of entity/component systems, the phases in which data (components/attributes) from Entities are read and used to compute something, and the phase where the modified data is written back to entities need to be separated in order to avoid data races. Otherwise the only way (not taking into account just "critical section"ing everything) to avoid them is to serialize parts of the update process that depend on other parts. This seems ugly. To me it would seem more elegant to be able to (ideally) have all processing running in parallel, where a system may read data from all entities as it wishes, but doesn't write modifications to that data back until some later point. The fact that this is even possible is based on the assumption that modification write-backs are usually very small in complexity, and don't require much performance, whereas computations are very expensive (relatively). So the overhead added by a delayed-write phase might be evened out by more efficient updating of entities (by having threads work more % of the time instead of waiting). A concrete example of this might be a system that updates physics. The system needs to both read and write a lot of data to and from entities. Optimally, there would be a system in place where all available threads update a subset of all entities registered with the physics system. In the case of the physics system this isn't trivially possible because of race conditions. So without a workaround, we would have to find other systems to run in parallel (which don't modify the same data as the physics system), other wise the remaining threads are waiting and wasting time. However, that has disadvantages Practically, the L3 cache is pretty much always better utilized when updating a large system with multiple threads, as opposed to multiple systems at once, which all act on different sets of data. Finding and assembling other systems to run in parallel can be extremely time consuming to design well enough to optimize performance. Sometimes, it might even not be possible at all because a system just depends on data that is touched by all other systems. Solution? In my thinking, a possible solution would be a system where reading/updating and writing of data is separated, so that in one expensive phase, systems only read data and compute what they need to compute, and then in a separate, performance-wise cheap, write phase, attributes of entities that needed to be modified are finally written back to the entities. The Question How might such a system be implemented to achieve optimal performance, as well as making programmer life easier? What are the implementation details of such a system and what might have to be changed in the existing EC-architecture to accommodate this solution?

    Read the article

< Previous Page | 533 534 535 536 537 538 539 540 541 542 543 544  | Next Page >