Search Results

Search found 19338 results on 774 pages for 'game loop'.

Page 402/774 | < Previous Page | 398 399 400 401 402 403 404 405 406 407 408 409  | Next Page >

• GLSL custom interpolation filter

  - by Cyan

  I'm currently building a fragment shader which is using several textures to render the final pixel color. The textures are not really textures, they are in fact "input data" to be used in the formula to generate the final color. The problem I've got is that the texture are getting bi-linear-filtered, and therefore the input data as well. This results in many unwanted side-effects, especially when final rendered texture is "zoomed" compared to original resolution. Removing the side effect is a complex task, and only result in "average" rendering. I was thinking : well, all my problems seems to come from the "default" bi-linear filtering on these input data. I can't move to GL_NEAREST either, since it would create "blocky" rendering. So i guess the better way to proceed is to be fully in charge of the interpolation. For this to work, i would need the input data at their "natural" resolution (so that means 4 samples), and a relative position between the sampled points. Is that possible, and if yes, how ? [EDIT] Since i started this question, i found this internet entry, which seems to (mostly) answer my needs. http://www.gamerendering.com/2008/10/05/bilinear-interpolation/ One aspect of the solution worry me though : the dimensions of the texture must be provided in an argument. It seems there is no way to "find this information transparently". Adding an argument into the rendering pipeline is unwelcomed though, since it's not under my responsibility, and translates into adding complexity for others.

  Read the article

• Matrix loading problems with jbullet and lwjgl

  - by Quintin

  The following code does not load the matrix correctly from jbullet. //box is a RigidBody Transform trans = new Transform(); trans = box.getMotionState().getWorldTransform(trans); float[] matrix = new float[16]; trans.getOpenGLMatrix(matrix); // pass that matrix to OpenGL and render the cube FloatBuffer buffer = ByteBuffer.allocateDirect(4*16).asFloatBuffer().put(matrix); buffer.rewind(); glPushMatrix(); glMultMatrix(buffer); glBegin(GL_POINTS); glVertex3f(0,0,0); glEnd(); glPopMatrix(); the jbullet is configured as so: CollisionConfiguration = new DefaultCollisionConfiguration(); dispatcher = new CollisionDispatcher(collisionConfiguration); Vector3f worldAabbMin = new Vector3f(-10000,-10000,-10000); Vector3f worldAabbMax = new Vector3f(10000,10000,10000); AxisSweep3 overlappingPairCache = new AxisSweep3(worldAabbMin, worldAabbMax); SequentialImpulseConstraintSolver solver = new SequentialImpulseConstraintSolver(); dynamicWorld = new DiscreteDynamicsWorld(dispatcher, overlappingPairCache, solver, collisionConfiguration); dynamicWorld.setGravity(new Vector3f(0,-10,0)); dynamicWorld.getDispatchInfo().allowedCcdPenetration = 0f; CollisionShape groundShape = new BoxShape(new Vector3f(1000.f, 50.f, 1000.f)); Transform groundTransform = new Transform(); groundTransform.setIdentity(); groundTransform.origin.set(new Vector3f(0.f, -60.f, 0.f)); float mass = 0f; Vector3f localInertia = new Vector3f(0, 0, 0); DefaultMotionState myMotionState = new DefaultMotionState(groundTransform); RigidBodyConstructionInfo rbInfo = new RigidBodyConstructionInfo(mass, myMotionState, groundShape, localInertia); RigidBody body = new RigidBody(rbInfo); dynamicWorld.addRigidBody(body); dynamicWorld.clearForces(); Nothing is rendered on the screen. What am I doing wrong?

  Read the article

• How to populate a form list with buttons using javascript

  - by StealingMana

  I made a script that, when you press one button(accessories) the selection(mylist) populates with one array(accessoryData), and when you hit the other button(weapons) the other array(weaponData) populates the selection. However, in the current state of the code the second button press is not re-populating the selection. What is wrong here? Also if there is a more efficient way to do this, that might be helpful. Full code function runList(form, test) { var html = ""; var x; dataType(test); while (x < dataType.length) { html += "<option>" + dataType[x]; x++; } document.getElementById("mylist").innerHTML = html; }

  Read the article

• Cross-platform builds with OGRE3D via CMake. Any tips?

  - by frarees

  I've been trying to compile a simple project for both OSX and Windows platforms, using OGRE3D, but I've got some problems on the way. I'm using CMake to create my platform specific project files (VS solution & Xcode project). Some problems I found are: OGRE3D source is distributed in 2 flavors, Windows sources and UNIX/OSX sources. In OSX, compiling dependencies (freetype, FreeImage and specially OIS) is such a pain. I don't know how to handle precompiled dependencies (they exist for both Win & Mac). May sound like a noob question, but I would appreciate some tips on this. Resources, forum posts, anything. There exists any "cross-platform base project for OGRE3D" on the net? Would be really helpful if someone who already managed to do this can bring some light. Btw, I'm not basing the project on OGRE3D, it's just that is the biggest library I'm probably using, so I depend a lot on it. Thanks in advantage!

  Read the article

• How do I rotate a sprite with ccbezierTo in cocos2d-x?

  - by user1609578

  In cocos2d-x, I move a sprite with ccbezierTo like this: // use for ccbezierTo bezier.controlPoint_1 = ccp(m_fish->getPositionX() + 200, visibleSize.height/2 + 300); bezier.controlPoint_2 = ccp(m_fish->getPositionX() + 400, visibleSize.height/2 - 300); bezier.endPosition = ccp(m_fish->getPositionX() + 600,visibleSize.height/2); bezier1.controlPoint_1 = ccp(m_fish->getPositionX() + 800, visibleSize.height/2 + 300); bezier1.controlPoint_2 = ccp(m_fish->getPositionX() + 1000, visibleSize.height/2 - 300); bezier1.endPosition = ccp(m_fish->getPositionX() + 1200,visibleSize.height/2); bezierForward = CCBezierTo::create(6, bezier); nextBezier = CCBezierTo::create (6,bezier1); m_fish->runAction(CCSequence::create( bezierForward, nextBezier, NULL)); How can I make my sprite rotate while moving it with CCBezierTo?

  Read the article

• How do I swap two objects in a GC language without triggering GC?

  - by TenFour04

  I have two array lists. that I want to swap each frame. My question is, does the variable 'temp' need to be a member variable to avoid triggering GC, assuming this method is called on dozens of objects each frame? I'm not creating a new object, just a new reference to an object. public void LateUpdate(){ ArrayList<int> temp = previousFrameCollisions; previousFrameCollisions = currentFrameCollisions; currentFrameCollisions = temp; currentFrameCollisions.clear(); } I've been told there's no reason to make a primitive into a member variable just to avoid GC, so my best guess is that this also applies to object references.

  Read the article

• How do I make camera move at same speed when rotating and moving forward

  - by dez

  I made a camera in DX9. To move forward I press the Up arrow. To rotate on the Y axis I use the mouse. When I perform these movements on their own the camera moves at the speed I want. However, if I hold down Up and move the mouse at the same time then the camera moves a lot faster than it should. I want it to move at the same speed as it does when only the Up arrow is pressed. I think I need to normalize something somewhere but not sure what and not sure where. Have tried various combinations without success so if anyone can point me in the right direction that would be great. Thanks. I've post code below. #define KEY_DOWN(vk_code) ((GetAsyncKeyState(vk_code) & 0x8000) ? 1 : 0) LRESULT WINAPI MsgProc( HWND hWnd, UINT msg, WPARAM wParam, LPARAM lParam ) { if( KEY_DOWN(VK_UP)) MovePlayer(D3DXVECTOR3(0, 0, -1.0f)); if( KEY_DOWN(VK_DOWN)) MovePlayer(D3DXVECTOR3(0, 0, 1.0f)); switch( msg ) { case WM_MOUSEMOVE: ProcessMouseInput(); } } void MovePlayer( D3DXVECTOR3 in_vec ) { D3DXMATRIX CameraRot; D3DXMatrixRotationY(&CameraRot,D3DXToRadian(AngleY)); D3DXVECTOR3 CameraRotTarget; D3DXVec3TransformNormal(&CameraRotTarget,&in_vec,&CameraRot); CameraPos += (m_timeElapsed * CameraRotTarget); } void ProcessMouseInput() { GetCursorPos( &CurrentMouseState ); if ((CurrentMouseState.x != GameMouseState.x) || (CurrentMouseState.y != GameMouseState.y)) { int dx = CurrentMouseState.x - GameMouseState.x; int dy = CurrentMouseState.y - GameMouseState.y; AngleY+=m_timeElapsed*dx*7.0f; } GameMouseState = CurrentMouseState; // Set back to window center in Render function } VOID UpdateCamera() { D3DXVECTOR3 CameraOrigTarget(0, 0, -1); D3DXVECTOR3 CameraOrigUp(0, 1, 0); D3DXMATRIX CameraRot; D3DXMATRIX CameraRotX; D3DXMatrixRotationX(&CameraRotX,D3DXToRadian(AngleX)); D3DXMATRIX CameraRotY; D3DXMatrixRotationY(&CameraRotY,D3DXToRadian(AngleY)); CameraRot = CameraRotX * CameraRotY; D3DXVECTOR3 CameraRotTarget; D3DXVec3TransformNormal(&CameraRotTarget,&CameraOrigTarget,&CameraRot); D3DXVECTOR3 CameraTarget; CameraTarget = CameraPos + CameraRotTarget; D3DXVECTOR3 vUpVec( 0.0f, 1.0f, 0.0f ); D3DXMatrixLookAtLH( &matView, &CameraPos, &CameraTarget, &vUpVec ); g_pd3dDevice->SetTransform( D3DTS_VIEW, &matView ); D3DXMatrixPerspectiveFovLH( &matProj, D3DX_PI / 4, 1.0f, 1.0f, 100.0f ); g_pd3dDevice->SetTransform( D3DTS_PROJECTION, &matProj ); }

  Read the article

• Determining if something is on the right or left side of an object?

  - by meds

  I have a character in a 3D world which is facing an arbitrary direction on a flat plane, the player can click on the left or right side of the character and based on which side is clicked on a different action happens. How can I determine which side the click occured on? Obviously for straight on ahead (0,0,1) I can simply use the x coordinate of the click point to determine if it's the left or right hand side, but what about other cases?

  Read the article

• sprite group doesn't support indexing

  - by user3956

  I have a sprite group created with pygame.sprite.Group() (and add sprites to it with the add method) How would I retrieve the nth sprite in this group? Code like this does not work: mygroup = pygame.sprite.Group(mysprite01) print mygroup[n].rect It returns the error: group object does not support indexing. For the moment I'm using the following function: def getSpriteByPosition(position,group): for index,spr in enumerate(group): if (index == position): return spr return False Although working, it just doesn't seem right... Is there a cleaner way to do this? EDIT: I mention a "position" and index but it's not important actually, returning any single sprite from the group is enough

  Read the article

• Dropping multiple objects using an array in Actionscript?

  - by Eratosthenes

  I'm trying to get these fireBalls to drop more often, I'm not sure if I'm using Math.random correctly. Also, for some reason I'm getting a null reference because I think the fireBalls array waits for one to leave the stage before dropping another one? This is the relevant code: var sun:Sun=new Sun var fireBalls:Array=new Array() var left:Boolean; function onEnterFrame(event:Event){ if (left) { sun.x = sun.x - 15; }else{ sun.x = sun.x + 15; } if (fireBalls.length>0&&fireBalls[0].y>stage.stageHeight){ // Fireballs exit stage removeChild(fireBalls[0]); fireBalls.shift(); } for (var j:int=0; j<fireBalls.length; j++){ fireBalls[j].y=fireBalls[j].y+15; if (fireBalls[j].y>stage.stageHeight-fireBall.width/2){ } } if (Math.random()<.2){ // Fireballs shooting from Sun var fireBall:FireBall=new FireBall; fireBall.x=sun.x; addChild(fireBall); fireBalls.push(fireBall); } }

  Read the article

• Handling commands or events that wait for an action to be completed afterwards

  - by virulent

  Say you have two events: Action1 and Action2. When you receive Action1, you want to store some arbitrary data to be used the next time Action2 rolls around. Optimally, Action1 is normally a command however it can also be other events. The idea is still the same. The current way I am implementing this is by storing state and then simply checking when Action2 is called if that specific state is there. This is obviously a bit messy and leads to a lot of redundant code. Here is an example of how I am doing that, in pseudocode form (and broken down quite a bit, obviously): void onAction1(event) { Player = event.getPlayer() Player.addState("my_action_to_do") } void onAction2(event) { Player = event.getPlayer() if not Player.hasState("my_action_to_do") { return } // Do something } When doing this for a lot of other actions it gets somewhat ugly and I wanted to know if there is something I can do to improve upon it. I was thinking of something like this, which wouldn't require passing data around, but is this also not the right direction? void onAction1(event) { Player = event.getPlayer() Player.onAction2(new Runnable() { public void run() { // Do something } }) } If one wanted to take it even further, could you not simply do this? void onPlayerEnter(event) { // When they join the server Player = event.getPlayer() Player.onAction1(new Runnable() { public void run() { // Now wait for action 2 Player.onAction2(new Runnable() { // Do something }) } }, true) // TRUE would be to repeat the event, // not remove it after it is called. } Any input would be wonderful.

  Read the article

• Techniques for lighting a texture (no shadows)

  - by Paul Manta

  I'm trying to learn about dynamic shadows for 2D graphics. While I understand the basic ideas behind determining what areas should be lit and which should be in shadow, I don't know how I would "lighten" a texture in the first place. Could you go over various popular techniques for lighting a texture and what (dis)advantages each one has? Also, how is lighting a texture with colored light different from using white light?

  Read the article

• Y Axis inverted on vertex output

  - by Yonathan Klijnsma

  I've got my project running and somehow it seems my vertex y components are inverted. 10 in the positive on Y goes down and 10 negative on the Y axis goes up. I can't find anything with the initialization and I am not doing any negative scaling in the view matrix. I've never had something like this happen before, does anyone have some tips or things to look for ? How I am sending verteces to the GPU ( Currently intermediate mode ) glVertex3f( x_pos_n, 10, z_pos ); I am using CG in the project but even without shaders the Y axis seems to be inverted.

  Read the article

• 16-bit PNGs in Slick2D

  - by Neglected

  I'm working on a project and I'm using some 3rd party sprites just to get it off the ground; recently I've come into a hitch. Slick2D doesn't seem to want to load my images. That is, it will warn me that images are the wrong bit-depth. All the images are in 16-bit PNG form (PNG is required for transparency). Is there any way I can disable the warning (being the bad guy programmer (the console print for each individual load REALLY SLOWS DOWN the image)) or is there another solution? I was thinking about converting all images (using imagemagick) to .gif (with an alpha channel). Would there be any loss in quality between formats? EDIT: I tried using imagemagick but some of the sprites use pure black so I can't do that without wrecking the image. EDIT2: using "identify" on any of the images show them as being 8-bit.. but Slick2D won't load them. What the hell? D: EDIT3: Issue solved (ish). If you are googling this then just disable the java png loader from slick by sticking this somewhere in your code (like the main method): System.setProperty("org.newdawn.slick.pngloader", "false");

  Read the article

• how can I change object look point?

  - by jques

  I have tried to load image but system does not give permission. Please look image http://www.rps.net/gunslinger/scrnshot/gunslinger33.jpg I have two arm with two gun, I want to rotate these arm with mouse. For example, if I move mouse position to the left, arms with guns should be move also. Since this is hoby project, I am a bit confort to ask below question ; What should I do to achieve my wish ? some explanation: perspective view gun in viewing direction left click = trigger left gun perspective Feel free to change the title Thanks

  Read the article

• How to disguise a serverside mob as another?

  - by Shaun Wild

  I've been working a Minecraft sever mod and i want to be able to add a new entity to the server, but then make the server send the packets to the client, imitating another mob, for example.. Lets say say i have EntityPlayerNPC.class, what i want to do is have all of the packets that get sent to the client look like they are from that of another player which is on the player, therefore allowing me to add custom NPC's... Thinking about the theory i'm sure this can be done. I've tried looking around for where the packets are being sent from and whatnot, can anyone think up a solution? edit: i tried adding a new constructor to the Packet20NamedEntitySpawn class like so: public Packet20NamedEntitySpawn(String username, EntityLiving e){ this.entityId = 0; this.name = username; this.xPosition = MathHelper.floor_double(e.posX * 32.0D); this.yPosition = MathHelper.floor_double(e.posY * 32.0D); this.zPosition = MathHelper.floor_double(e.posZ * 32.0D); this.rotation = (byte)((int)(e.rotationYaw * 256.0F / 360.0F)); this.pitch = (byte)((int)(e.rotationPitch * 256.0F / 360.0F)); this.metadata = e.getDataWatcher(); } unfortunatley, that didn't work :(

  Read the article

• How can I track a falling ball with a camera?

  - by Jason

  I have been trying to get my camera to follow a falling ball but with no success. here is the code float cameraY = (FrustumHeight / 2)+((ball.getPosition().y) /2) - (FrustumHeight /2); if (cameraY < FrustumHeight/2 ) cameraY = FrustumHeight/2; camera.position.set(0f,cameraY, 0f); Gdx.app.log("test",camera.position.toString()); camera.update(); camera.apply(Gdx.gl10); batch.setProjectionMatrix(camera.combined); batch.begin(); batch.draw(backgroundRegion, camera.position.x - FrustumWidth / 2, -cameraY - (FrustumHeight/2) , 320, 480); batch.draw(ballTexture, (camera.position.x - FrustumWidth / 2) + ball.getPosition().x,-cameraY + ball.getPosition().y - (FrustumHeight/2) , 32, 32); I'm sure I am doing this completely wrong - what is the correct way to do this?

  Read the article

• How can I perform a masked erase in SDL2?

  - by Kvisle

  I'm trying to implement some shadow/lighting effects in my 2D-project, and I've concluded that if there is an easy way to perform a masked erase on an SDL_Texture, it would make the drawing operations quite cheap. Let's say I have a texture of the part of the level where light is not meant to be rendered. I also have a texture with my "light map"; I want to use this to just draw omni lights from my light sources. Then I want to use the first image to 'subtract' the portions of the light map that are not to be rendered on the final scene. Then I draw my "light map" texture on top of my scene, with additive blending enabled. This sounds like a good theory in my head, but I can't see any functions in the SDL2 API that let me do masked erase from a texture. Am I overlooking something? Does anything like this exist?

  Read the article

• Particles are not moving correctly [closed]

  - by cr33p

  I want to make a particle explosion, after something gets destroyed, but somehow only one line of mixed colors show up on the screen. Here's the header: http://pastebin.com/JW5bPLj2 Here's the source: http://pastebin.com/KHmFqytD I don't get what's wrong, as it's nearly the same as in "Programming Linux Games" Can somebody help me fix that? PS: "Uint32 delta" is needed to update the pixels based on time. PSS: Maybe I should add that it's programmed in C and includes SDL. EDIT: Found the problem. It was the "drawParticles" function. The problem was, that I passed a double to "offset" (as particles[i].x, etc are all doubles). So I ended up with values like ~MAX_INT because I didn't cast the doubles properly to ints.

  Read the article

• Keyboard is not detecting as3

  - by Kaoru

  i got this problem, when i press 1 in keyboard, the function does not run, even i already declared it. How do i solve this? Here is the code: public function Character() { Constant.loaderMario.load(new URLRequest("mario.swf")); addEventListener(KeyboardEvent.KEY_DOWN, selectingPlayer); } private function selectingPlayer(event:KeyboardEvent):void { isoTextMario = new Word(); isoTextChocobo = new Word(); if (event.keyCode == Keyboard.NUMBER_1|| event.keyCode == 49) { Constant.marioSelected = true; if (Constant.marioSelected == true) { isoTextMario.marioCharacter(); addChild(isoTextMario); } } else if (event.keyCode == Keyboard.NUMBER_2 || event.keyCode == 50) { Constant.chocoboSelected = true; if (Constant.chocoboSelected == true) { isoTextChocobo.chocoboCharacter(); addChild(isoTextChocobo); } } }

  Read the article

• Artificial Intelligence ... how to make an object roam freely/avoid other objects, and model consciousness? [on hold]

  - by help bonafide pigeons

  Say a simple free roam battle scene in which a player runs around freely and engages in battle with other enemies/objects, as shown below: The dragon/dinosaur (or whatever that thing I drew appears to be) will, by some measure, try and avoid attacks so it is modeled to appear to have a conscious desire to avoid pain. My question is ... since this is very complex, many possible strategies for solving this, algorithms, etc., what is the basic idea behind how this would be accomplished in any sort? Like, we can assume the enemy in the picture is not just going to aimlessly hop around and avoid, but freely be modeled to behave as if it were really exploring/fighting. For the best example I can give, witness the behavior of the enemies in Final Fantasy 12 in this video: http://www.youtube.com/watch?v=mO0TkmhiQ6w How do the pros, or how would anyone attempt solve/implement this? PS: I have tried several times to give an image the "illusion" that is has a conciousness, but aside from emulating a real animal's consciousness in complete, I fall short and get choppy moving images that follow predictable patterns, error-prone movements, and the worst imaginable scenario of a battle engagement.

  Read the article

• Fast, accurate 2d collision

  - by Neophyte

  I'm working on a 2d topdown shooter, and now need to go beyond my basic rectangle bounding box collision system. I have large levels with many different sprites, all of which are different shapes and sizes. The textures for the sprites are all square png files with transparent backgrounds, so I also need a way to only have a collision when the player walks into the coloured part of the texture, and not the transparent background. I plan to handle collision as follows: Check if any sprites are in range of the player Do a rect bounding box collision test Do an accurate collision (Where I need help) I don't mind advanced techniques, as I want to get this right with all my requirements in mind, but I'm not sure how to approach this. What techniques or even libraries to try. I know that I will probably need to create and store some kind of shape that accurately represents each sprite minus the transparent background. I've read that per pixel is slow, so given my large levels and number of objects I don't think that would be suitable. I've also looked at Box2d, but haven't been able to find much documentation, or any examples of how to get it up and running with SFML.

  Read the article

• How do I retain previously drawn graphics?

  - by Cromanium

  I've created a simple program that draws lines from a fixed point to a random point each frame. I wanted to keep each line on the screen. However, it always seems to be cleared each time it draws on the spriteBatch even without GraphicsDevice.Clear(color) being called. What seems to be the problem? protected override void Draw(GameTime gameTime) { spriteBatch.Begin(); DrawLine(spriteBatch); spriteBatch.End(); base.Draw(gameTime); } private void DrawLine(SpriteBatch spriteBatch) { Random r = new Random(); Vector2 a = new Vector2(50, 100); Vector2 b = new Vector2(r.Next(0, 640), r.Next(0,480)); Texture2D filler= new Texture2D(GraphicsDevice, 1, 1, false, SurfaceFormat.Color); filler.SetData(new[] { Color.Black }); float length = Vector2.Distance(a, b); float angle = (float)Math.Atan2(b.Y - a.Y, b.X - a.X); spriteBatch.Draw(filler, a, null, Color.Black, angle, Vector2.Zero, new Vector2(length,10.0f), SpriteEffects.None, 0f); } What am I doing wrong?

  Read the article

• Lwjgl camera causing movement to be mirrored

  - by pangaea

  I'm having a problem in that everything is rendered and the movement is fine. However, everything seems to be mirrored. In the sense that the TriangleMob should move towards me, but it doesn't instead it mirrors my action. I move forward the TriangleMob moves backwards. I move left, it moves right. I move backwards, it moves forward. The code works if I do this glPushMatrix(); glTranslatef(-position.x, -position.y, -position.z); glCallList(objectDisplayList); glPopMatrix(); However, I'm scared this will cause a problem later on. I suppose the code works. However, shouldn't the call be glPushMatrix(); glTranslatef(position.x, position.y, position.z); glCallList(objectDisplayList); glPopMatrix(); I think the problem could be caused by how I'm doing the camera, which is this glLoadIdentity(); glRotatef(player.getRotation().x, 1.0f, 0.0f, 0.0f); glRotatef(player.getRotation().y, 0.0f, 1.0f, 0.0f); glRotatef(player.getRotation().z, 0.0f, 0.0f, 1.0f); glTranslatef(player.getPosition().x, player.getPosition().y, player.getPosition().z);

  Read the article

• Physics Engine [Collision Response, 2-dimensional] experts, help!! My stack is unstable!

  - by Register Sole

  Previously, I struggle with the sequential impulse-based method I developed. Thanks to jedediah referring me to this paper, I managed to rebuild the codes and implement the simultaneous impulse based method with Projected-Gauss-Seidel (PGS) iterative solver as described by Erin Catto (mentioned in the reference of the paper as [Catt05]). So here's how it currently is: The simulation handles 2-dimensional rotating convex polygons. Detection is using separating-axis test, with a SKIN, meaning closest points between two polygons is detected and determined if their distance is less than SKIN. To resolve collision, simultaneous impulse-based method is used. It is solved using iterative solver (PGS-solver) as in Erin Catto's paper. Error-correction is implemented using Baumgarte's stabilization (you can refer to either paper for this) using J V = beta/dt*overlap, J is the Jacobian for the constraints, V the matrix containing the velocities of the bodies, beta an error-correction parameter that is better be < 1, dt the time-step taken by the engine, and overlap, the overlap between the bodies (true overlap, so SKIN is ignored). However, it is still less stable than I expected :s I tried to stack hexagons (or squares, doesn't really matter), and even with only 4 to 5 of them, they hardly stand still! Also note that I am not looking for a sleeping scheme. But I would settle if you have any explicit scheme to handle resting contacts. That said, I would be more than happy if you have a way of treating it generally (as continuous collision, instead of explicitly as a special state). Ideas I have: I would try adding a damping term (proportional to velocity) to the Baumgarte. Is this a good idea in general? If not I would not want to waste my time trying to tune the parameter hoping it magically works. Ideas I have tried: Using simultaneous position based error correction as described in the paper in section 5.3.2, turned out to be worse than the current scheme. If you want to know the parameters I used: Hexagons, side 50 (pixels) gravity 2400 (pixels/sec^2) time-step 1/60 (sec) beta 0.1 restitution 0 to 0.2 coeff. of friction 0.2 PGS iteration 10 initial separation 10 (pixels) mass 1 (unit is irrelevant for now, i modified velocity directly<-impulse method) inertia 1/1000 Thanks in advance! I really appreciate any help from you guys!! :)

  Read the article

< Previous Page | 398 399 400 401 402 403 404 405 406 407 408 409  | Next Page >