Search Results

Search found 41789 results on 1672 pages for 'software development'.

Page 549/1672 | < Previous Page | 545 546 547 548 549 550 551 552 553 554 555 556  | Next Page >

• How do I get the compression on specific dynamic body

  - by Mike JM

  Sorry, I could not find any tag that would suit my question. Let me first show you the image and then write what I want to do: I'm using box2D. As you can see there are three dynamic bodies connected to each other (think of it as a table from front view).The LEG1 and LEG2 are connected to the static body. (it's the ground body). Another dynamic body is falling onto the table. I need to get the compression in the LEG1 and LEG2 separately. Joints have GetReactionForce() function which returns a b2Vec, which in turn has Length() and LengthSqd functions. This will give the total sum of the forces in any taken joint. But what I need is forces in individual bodies that are connected with joints. Once you connect several bodies with a single joint it again will show the sum of forces which is not useful.Here's the case iI'm talking about:

  Read the article

• Cocos2dx- Draw primitives(polygons) on Update

  - by Haider

  In my game I'm trying to draw polygons on on each step i.e. update method. I call draw() method to draw new polygon with dynamic vertices. Following is my code: void HelloWorld::draw(){glLineWidth(1);CCPoint filledVertices[] = {ccp(drawX1,drawY1),ccp(drawX2,drawY2), ccp(drawX3,drawY3), ccp(drawX4,drawY4)};ccDrawSolidPoly( filledVertices, 4, ccc4f(0.5f, 0.5f, 1, 1 ));} I call the draw() method from the update(float dt) method. The engine is behaving inconsistently i.e. sometimes it displays the polygons and on other occasions it does not. Is it the right approach to do such a task? If not what is the best way to display large number of primitives?

  Read the article

• Android device - C++ OpenGL 2: eglCreateWindowSurface invalid

  - by ThreaderSlash

  I am trying to debug and run OGLES on Native C++ in my Android device in order to implement a native 3D game for mobile smart phones. The point is that I got an error and see no reason for that. Here is the line from the code that the debugger complains: mSurface = eglCreateWindowSurface(mDisplay, lConfig, mApplication->window, NULL); And this is the error message: Invalid arguments ' Candidates are: void * eglCreateWindowSurface(void *, void *, unsigned long int, const int *) ' --x-- Here is the declaration: android_app* mApplication; EGLDisplay mDisplay; EGLint lFormat, lNumConfigs, lErrorResult; EGLConfig lConfig; // Defines display requirements. 16bits mode here. const EGLint lAttributes[] = { EGL_RENDERABLE_TYPE, EGL_OPENGL_ES2_BIT, EGL_BLUE_SIZE, 5, EGL_GREEN_SIZE, 6, EGL_RED_SIZE, 5, EGL_SURFACE_TYPE, EGL_WINDOW_BIT, EGL_RENDER_BUFFER, EGL_BACK_BUFFER, EGL_NONE }; // Retrieves a display connection and initializes it. packt_Log_debug("Connecting to the display."); mDisplay = eglGetDisplay(EGL_DEFAULT_DISPLAY); if (mDisplay == EGL_NO_DISPLAY) goto ERROR; if (!eglInitialize(mDisplay, NULL, NULL)) goto ERROR; // Selects the first OpenGL configuration found. packt_Log_debug("Selecting a display config."); if(!eglChooseConfig(mDisplay, lAttributes, &lConfig, 1, &lNumConfigs) || (lNumConfigs <= 0)) goto ERROR; // Reconfigures the Android window with the EGL format. packt_Log_debug("Configuring window format."); if (!eglGetConfigAttrib(mDisplay, lConfig, EGL_NATIVE_VISUAL_ID, &lFormat)) goto ERROR; ANativeWindow_setBuffersGeometry(mApplication->window, 0, 0, lFormat); // Creates the display surface. packt_Log_debug("Initializing the display."); mSurface = eglCreateWindowSurface(mDisplay, lConfig, mApplication->window, NULL); --x-- Hope someone here can shed some light on it.

  Read the article

• 2D non-tile based map editor

  - by Jonesy

  I am currently developing a relatively simple 2D, topdown oriented adventure game for the iPhone and was wondering what would be the easiest way to create the maps for my game. I figured I would need some kind of visual editor that would give me immediate feedback and would allow me to place all objects in the world exactly where I want them. I could then load the saved representation of the world I create in the editor in my game. So, I am looking for a simple map editor that allows me to do this. All the objects in my game are simply textured rectangles build up from two triangles. All I need to be able to do is position different rectangles/objects in the map, and give them a texture. I am using texture atlases, so it would be useful to be able to assign portions of textures to the objects. I then need to be able to extract all the objects from the saved representation of my maps, together with the name/identifier of the texture(atlas) they use, and the area of the texture atlas. I have looked at some tile-based map editors like Tiled and Ogmo, but they don't seem to be able to do what I want. Any suggestions? EDIT: a more concrete example: something like the GameMaker level editor, but then with added export functionality in a handy format.

  Read the article

• Using Ogre particle point billboards with shaders

  - by Jay

  I'm learning about using Ogre particles and had some questions about how the point type particles work. Q. I believe point type particles are implemented as a single position. Is one single vertex is passed to the vertex shader? Q. If one vertex is passed to the vertex shader then what gets sent to the fragment shader? Q. Can I pass the particle size to the shader? Perhaps with a custom parameter?

  Read the article

• Can and should a game design be patented?

  - by Christian

  I have an idea for a game that I want to develop and I feel is unique, and I'm wondering if I should patent it. I read on the web that games can be patented, but just because it can be done doesn't mean that it makes sense to do it. I actually don't really want patent it (it's expensive, a hassle and I don't believe in patenting of ideas... unless it's something truly revolutionary). However, I'm concerned a bigger company could come along, with more experienced game designers and developers and steal the idea.

  Read the article

• Flash framerate reliability

  - by Tim Cooper

  I am working in Flash and a few things have been brought to my attention. Below is some code I have some questions on: addEventListener(Event.ENTER_FRAME, function(e:Event):void { if (KEY_RIGHT) { // Move character right } // Etc. }); stage.addEventListener(KeyboardEvent.KEY_DOWN, function(e:KeyboardEvent):void { // Report key which is down }); stage.addEventListener(KeyboardEvent.KEY_UP, function(e:KeyboardEvent):void { // Report key which is up }); I have the project configured so that it has a framerate of 60 FPS. The two questions I have on this are: What happens when it is unable to call that function every 1/60 of a second? Is this a way of processing events that need to be limited by time (ex: a ball which needs to travel to the right of the screen from the left in X seconds)? Or should it be done a different way?

  Read the article

• Economic modelling - Resources for valuing goods

  - by Rushyo

  tl;dr: What economic/computer science books would you suggest for learning about economic valuation of goods and simulations thereof? I'm looking to create an economic model for a game based on goods created procedurally. Every natural resource and produced good would be procedurally generated, with certain goods being assigned certain uses. Fakesium might be used for the production of Weapon A and produced from Fakesium factories which use Dilithium and Widgets as reagents, where Widgets are also the product of Foo and Bar The problem is not creating the resources and their various production utlities - but getting the game's AI empires and merchants to correctly value the goods according to their scarcity, utility and production costs. I need to create a simulation of goods which allows the various game factions to assign a common value denominator (credits) to each resource, depending on how much its worth to that empire. I see the simulation being something like: "I have a high requirement for Weapon A. Since I don't have much of Fakesium, which is needed for Weapon A - I must have a high demand for Fakesium. If I can acquire Fakesium, devalue it. If not, increase its value - and also increase demand for Dilithium and Widgets too." This is very naive - because it may be much much cheaper for the empire to simply purchase Dilithium and Widgets directly rather than purchasing Fakesium, for example. Another example is two resources might allow the creation of Weapon A (Fakesium and Lieron), so we'd need to consider that. I've been scratching my head over the problem and it keeps growing. By the time the player joins the world, I'd expect enough iterations of this process to have occurred that prices would have largely normalised - and would then only trigger rarely to compensate for major changes (eg. if the player blows up the world's only Foo mine!) Could anyone suggest resources (books, largely) which outline this style of modelling, preferably in the context of simulations? Since this problem would never occur outside fantasy worlds, I figured this is probably the most likely place to find people who have encountered similar problems and I'm sure there's people who know of good places for Games Developers to start looking at less specific economic theory too. Additionally, does anyone know of any developers with blogs whose games or research applications perform similar modelling?

  Read the article

• Resolution Independence in libGDX

  - by ashes999

  How do I make my libGDX game resolution/density independent? Is there a way to specify image sizes as "absolute" regardless of the underlying density? I'm making a very simple kids game; just a bunch of sprites displayed on-screen, and some text for menus (options menu primarily). What I want to know is: how do I make my sprites/fonts resolution independent? (I have wrapped them in my own classes to make things easier.) Since it's a simple kids game, I don't need to worry about the "playable area" of the game; I want to use as much of the screen space as possible. What I'm doing right now, which seems super incorrect, is to simply create images suitable for large resolutions, and then scale down (or rarely, up) to fit the screen size. This seems to work okay (in the desktop version), even with linear mapping on my textures, but the smaller resolutions look ugly. Also, this seems to fly in the face of Android's "device independent pixels" (DPs). Or maybe I'm missing something and libGDX already takes care of this somehow? What's the best way to tackle this? I found this link; is this a good way of solving the problem?: http://www.dandeliongamestudio.com/2011/09/12/android-fragmentation-density-independent-pixel-dip/ It mentions how to control the images, but it doesn't mention how to specify font/image sizes regardless of density.

  Read the article

• UDK game Prisoners/Guards

  - by RR_1990

  For school I need to make a little game with UDK, the concept of the game is: The player is the headguard, he will have some other guard (bots) who will follow him. Between the other guards and the player are some prisoners who need to evade the other guards. It needs to look like this My idea was to let the guard bots follow the player at a certain distance and let the prisoners bots in the middle try to evade the guard bots. Now is the problem i'm new to Unreal Script and the school doesn't support me that well. Untill now I have only was able to make the guard bots follow me. I hope you guys can help me or make me something that will make this game work. Here is the class i'm using to let te bots follow me: class ChaseControllerAI extends AIController; var Pawn player; var float minimalDistance; var float speed; var float distanceToPlayer; var vector selfToPlayer; auto state Idle { function BeginState(Name PreviousStateName) { Super.BeginState(PreviousStateName); } event SeePlayer(Pawn p) { player = p; GotoState('Chase'); } Begin: player = none; self.Pawn.Velocity.x = 0.0; self.Pawn.Velocity.Y = 0.0; self.Pawn.Velocity.Z = 0.0; } state Chase { function BeginState(Name PreviousStateName) { Super.BeginState(PreviousStateName); } event PlayerOutOfReach() { `Log("ChaseControllerAI CHASE Player out of reach."); GotoState('Idle'); } // class ChaseController extends AIController; CONTINUED // State Chase (continued) event Tick(float deltaTime) { `Log("ChaseControllerAI in Event Tick."); selfToPlayer = self.player.Location - self.Pawn.Location; distanceToPlayer = Abs(VSize(selfToPlayer)); if (distanceToPlayer > minimalDistance) { PlayerOutOfReach(); } else { self.Pawn.Velocity = Normal(selfToPlayer) * speed; //self.Pawn.Acceleration = Normal(selfToPlayer) * speed; self.Pawn.SetRotation(rotator(selfToPlayer)); self.Pawn.Move(self.Pawn.Velocity*0.001); // or *deltaTime } } Begin: `Log("Current state Chase:Begin: " @GetStateName()@""); } defaultproperties { bAdjustFromWalls=true; bIsPlayer= true; minimalDistance = 1024; //org 1024 speed = 500; }

  Read the article

• Resources for game networking in Java

  - by pudelhund

  I am currently working on a Java multiplayer game. The game itself (single player) already works perfectly fine and so does the chat. The only thing that is really missing is the multiplayer part. Sadly I am absolutely clueless on where to start with that. I roughly know that I will have to work with packages, and I also know many things about streaming etc (chat is already working). Oh and it should - according to this article - be a UDP server. My problem is that I can't find any resources on how to do this. A tutorial (book or website) would be perfect, alternatively a good example of an open source client/server (in Java of course) would be fine as well. If you feel like doing something helpful I'd also really appreciate someone "privately" teaching me via email or some chat program :) Thank you!

  Read the article

• Loading a new instance of a class through XML not working quite right

  - by Thegluestickman

  I'm having trouble with XML and XNA. I want to be able to load weapon settings through XML to make my weapons easier to make and to have less code in the actual project file. So I started out making a basic XML document, something to just assign variables with. But no matter what I changed it gave me a new error every time. The code below gives me a "XML element 'Tag' not found", I added and it started to say the variables weren't found. What I wanted to do in the XML file as well, was load a texture for the file too. So I created a static class to hold my texture values, then in the Texture tag of my XML document I would set it to that instance too. I think that's were the problems are occuring because that's where the "XML element 'Tag' not found" error is pointing me too. My XML document: <XnaContent> <Asset Type="ConversationEngine.Weapon"> <weaponStrength>0</weaponStrength> <damageModifiers>0</damageModifiers> <speed>0</speed> <magicDefense>0</magicDefense> <description>0</description> <identifier>0</identifier> <weaponTexture>LoadWeaponTextures.ironSword</weaponTexture> </Asset> </XnaContent> My Class to load the weapon XML: public static class LoadWeaponXML { static Weapon Weapons; public static Weapon WeaponLoad(ContentManager content, int id) { Weapons = content.Load<Weapon>(@"Weapons/" + id); return Weapons; } } public static class LoadWeaponTextures { public static Texture2D ironSword; public static void TextureLoad(ContentManager content) { ironSword = content.Load<Texture2D>("Sword"); } } I'm not entirely sure if you can load textures through XML, but any help would be greatly appreciated.

  Read the article

• Alchemy like game for the web, open source. Any ideas for element combinations?

  - by JohnDel

  I created a web game like the Android game Alchemy. It's open source and in the back-end you can create your own elements / your own game. I was wondering what elements - ideas would be good to implement as a prototype / demo? Some ideas are: Colors Programming languages Chemical Compounds Same as the original alchemy Evolution of biological organisms What do you think? Any specific combination ideas?

  Read the article

• jMonkey Quest Database

  - by theJollySin

  I am building a game in jMonkey (Java) and I have so far only used default quest text. But now I need to start populating a lot of quests with text. My design requires A LOT of quests texts. What is the best way to build a database of quest texts in jMonkey? I don't have a lot of real experience with databases. Is there a database that integrates well with jMonkey? Here are the ideal properties I want in my database, in order of priority: Reasonably light learning curve Easy portability (in Java) to Windows, Linux, and Mac OSX Good interface with Java Good interface with jMonkey The ability to add properties to the quests: ID, level, gender, quest chain ID, etc. Or am I wrong in thinking I need to use some giant monster like SQL? I haven't been able to find much information on this, so are people using some non-database methods for storing things like quest text in jMonkey?

  Read the article

• Poor mobile performance when running from Eclipse

  - by Yajirobe_LOL

  So after weeks of thinking my rendering code was bad, I accidentally discovered the following: Running my game on a Nexus S From Eclipse (Debug as - Android application): 12fps From the device while still attached to USB (getting log info in Eclipse still): 24fps From the device while not attached via USB: 56fps I was wondering if anyone else has issues like this? I mean, the problem really isn't a problem since the final release build will likely have good performance, but for the time being I don't want to have to keep (un)plugging my device in and out when testing code all day long. Is there some remedy for this or does anyone have any input/advice? Thanks.

  Read the article

• Using heavyweight ORM implementation for light based games

  - by Holland

  I'm just about to engulf myself in an MVC-based/Component architecture in C#, using MySQL's connector/Net for the data storage, and probably some NHibernate/FluentNHibernate Object-relational-mapping to map out the data structure. The goal is to build a scalable 2D RPG. Then I think about it...and I can't help but think this seems a little "heavy weight" for a 2D RPG, especially one which, while I plan to incorporate a lot of functionality and entertaining gameplay, may be ported to something like Windows Phone or Android in the future. Yet, on the other hand even a 2-Dimensional RPG can become very complicated, and therefore must incorporate a lot of functionality. While this can be accomplished with text/XML/JSON for data storage, is there a better way? Is something such as Object-Relational-Mapping useful in such an application? So, what do you think? Would you say that there is a place for such technologies? I don't know what to think...

  Read the article

• Blending textures together, texture fade over / fade in

  - by Deukalion

  What is the best way to render a texture overlapping effect? Like in this example: I want either the grass to fade in to the snow texture, or the other way around. No rough edges. Somehow make them blend over. So the grass has a bit of snow or the snow has a bit of grass How is this possible during runtime? If that's possible. I don't render this by using the SpriteBatch, since the ground isn't rectangles (they can be moved). This is the way I render each shape (each one of those squares): // LoadTexture // Apply EffectPass device.DrawUserIndexedPrimitives<VertexPositionNormalTexture> ( PrimitiveType.TriangleList, render.Item.Points, // Array of VertexPositionNormalTexture 0, render.Item.Points.Length, render.Item.Indexes, // Array of int indexes (triangulation) 0, render.Item.Indexes.Length / 3, VertexPositionNormalTexture.VertexDeclaration );

  Read the article

• xna orbit camera troubles

  - by user17753

  I have a Model named cube to which I load in LoadContent(): cube = Content.Load<Model>("untitled");. In the Draw Method I call DrawModel: private void DrawModel(Model m, Matrix world) { foreach (ModelMesh mesh in m.Meshes) { foreach (BasicEffect effect in mesh.Effects) { effect.EnableDefaultLighting(); effect.View = camera.View; effect.Projection = camera.Projection; effect.World = world; } mesh.Draw(); } } camera is of the Camera type, a class I've setup. Right now it is instantiated in the initialization section with the graphics aspect ratio and the translation (world) vector of the model, and the Draw loop calls the camera.UpdateCamera(); before drawing the models. class Camera { #region Fields private Matrix view; // View Matrix for Camera private Matrix projection; // Projection Matrix for Camera private Vector3 position; // Position of Camera private Vector3 target; // Point camera is "aimed" at private float aspectRatio; //Aspect Ratio for projection private float speed; //Speed of camera private Vector3 camup = Vector3.Up; #endregion #region Accessors /// <summary> /// View Matrix of the Camera -- Read Only /// </summary> public Matrix View { get { return view; } } /// <summary> /// Projection Matrix of the Camera -- Read Only /// </summary> public Matrix Projection { get { return projection; } } #endregion /// <summary> /// Creates a new Camera. /// </summary> /// <param name="AspectRatio">Aspect Ratio to use for the projection.</param> /// <param name="Position">Target coord to aim camera at.</param> public Camera(float AspectRatio, Vector3 Target) { target = Target; aspectRatio = AspectRatio; ResetCamera(); } private void Rotate(Vector3 Axis, float Amount) { position = Vector3.Transform(position - target, Matrix.CreateFromAxisAngle(Axis, Amount)) + target; } /// <summary> /// Resets Default Values of the Camera /// </summary> private void ResetCamera() { speed = 0.05f; position = target + new Vector3(0f, 20f, 20f); projection = Matrix.CreatePerspectiveFieldOfView(MathHelper.PiOver4, aspectRatio, 0.5f, 100f); CalculateViewMatrix(); } /// <summary> /// Updates the Camera. Should be first thing done in Draw loop /// </summary> public void UpdateCamera() { Rotate(Vector3.Right, speed); CalculateViewMatrix(); } /// <summary> /// Calculates the View Matrix for the camera /// </summary> private void CalculateViewMatrix() { view = Matrix.CreateLookAt(position,target, camup); } I'm trying to create the camera so that it can orbit the center of the model. For a test I am calling Rotate(Vector3.Right, speed); but it rotates almost right but gets to a point where it "flips." If I rotate along a different axis Rotate(Vector3.Up, speed); everything seems OK in that direction. So I guess, can someone tell me what I'm not accounting for in the above code I wrote? Or point me to an example of an orbiting camera that can be fixed on an arbitrary point?

  Read the article

• Level selection view - similiar to Angry Bird's

  - by Piotr

  I am making game and need to prepare view for level selection. Could you recommend me some opensource library which could I use? I need icons to vibrate after long pressing one of them, some callbacks after choosing them, possibility to prepare custom icon's view, page control and horizontal scrolling. I was trying to use OpenSpringBoard but weirdly couldn't see scrollview and pagecontrol working in this project - it seems that there's possibility to use only one page. On the other hand, myLauncher(https://github.com/dlinsin/myLauncher) isn't so easy to include in project, as I need a seperate view with some delegate methods. I need to be compatible with iOS 4.2

  Read the article

• Cannot convert parameter 1 from 'short *' to 'int *' [closed]

  - by Torben Carrington

  I'm trying to learn pointers and since I recently learned that short int takes up less memory [2 bytes as apposed to the long int's memory usage of 4 which is the default for int] I wanted to create a pointer that uses the memory address of a short integer. I'm following a tutorial in my book about Pointers and it's using the Swap function. The problem is I receive this error the moment I change everything from int to short int: error C2664: 'Swap' : cannot convert parameter 1 from 'short *' to 'int *' 1 Types pointed to are unrelated; conversion requires reinterpret_cast, C-style cast or function-style cast Since my code is so small here is the whole thing: void Swap(short int *sipX, short int *sipY) { short int siTemp = *sipX; *sipX = *sipY; *sipY = siTemp; } int main() { short int siBig = 100; short int siSmall = 1; std::cout << "Pre-Swap: " << siBig << " " << siSmall << std::endl; Swap(&siBig, &siSmall); std::cout << "Post-Swap: " << siBig << " " << siSmall << std::endl; return 0; }

  Read the article

• Setting up cube map texture parameters in OpenGL

  - by KaiserJohaan

  I see alot of tutorials and sources use the following code snippet when defining each face of a cube map: for (i = 0; i < 6; i++) glTexImage2D(GL_TEXTURE_CUBE_MAP_POSITIVE_X + i, 0, InternalFormat, size, size, 0, Format, Type, NULL); Is it safe to assume GL_TEXTURE_CUBE_MAP_POSITIVE_X + i will properly iterate the following cube map targets, GL_TEXTURE_CUBE_MAP_NEGATIVE_X, GL_TEXTURE_CUBE_MAP_POSITIVE_Y, GL_TEXTURE_CUBE_MAP_NEGATIVE_Y etc?

  Read the article

• Which algorithm used in Advance Wars type turn based games

  - by Jan de Lange

  Has anyone tried to develop, or know of an algorithm such as used in a typical turn based game like Advance Wars, where the number of objects and the number of moves per object may be too large to search through up to a reasonable depth like one would do in a game with a smaller search base like chess? There is some path-finding needed to to engage into combat, harvest, or move to an object, so that in the next move such actions are possible. With this you can build a search tree for each item, resulting in a large tree for all items. With a cost function one can determine the best moves. Then the board flips over to the player role (min/max) and the computer searches the best player move, and flips back etc. upto a number of cycles deep. Finally it has found the best move and now it's the players turn. But he may be asleep by now... So how is this done in practice? I have found several good sources on A*, DFS, BFS, evaluation / cost functions etc. But as of yet I do not see how I can put it all together.

  Read the article

• Blender DirectX exporter to Panda3D

  - by jakebird451

  I have been experimenting with Panda3D lately. I have a character made in Blender with various bones and currently with one animation that I wish to export to a *.x format for Panda3D. My current attempt was to export the model was to first export with bones [Armatures] by checking the "Export Armatures" button in the export menu (file name: char.x). Thanks to the *.x file format, I read the file and it seems to have the same bone structure format as the model (with parenting and matrix positional data). The second export was selecting Animations - Full Animation to provide just the animation (file name: char_idle.x). The models exported just fine. I am not sure about the animation yet, but the file seems to be just fine. This is my code for loading the model into python & Panda3D: self.model = Actor("char.x",{"char_idle.x"}) When I run the program the command line provides a couple of errors, the main errors of interest are: :Actor(warning): char.x is not a character! and ... File "C:\Panda3D-1.8.0\direct\actor\Actor.py", line 284, in __init__ if (type(anims[anims.keys()[0]])==type({})): AttributeError: 'set' object has no attribute 'keys' The first error is the most interesting to me. The model works if I leave the animation dictionary blank. With no animations loaded the character appears in its un-animated T position, however the actor warning still shows up. The character should include the various bones when I exported the model right? I am not that experienced with blender, I'm just a programmer. So if the problem lies in blender please try to keep that in mind when posting a reply. I'll try my best to keep up. I also tried to print out the bone structure without any animations loaded and it provides a similar error with the line print self.model.listJoints(): File "C:\Panda3D-1.8.0\direct\actor\Actor.py", line 410, in listJoints Actor.notify.error("no part named: %s" % (partName)) File "C:\Panda3D-1.8.0\direct\directnotify\Notifier.py", line 132, in error raise exception(errorString) StandardError: no part named: modelRoot I really hope it is a simple exporting fix.

  Read the article

• Low-level GPU code and Shader Compilation

  - by ktodisco

  Bear with me, because I will raise several questions at once. I still feel, though, that overall this can be treated as one question that may be answered succinctly. I recently dove into solidifying my understanding of the assembly language, low-level memory operations, CPU structure, and program optimizations. This also sparked my interest in how higher-level shading languages, GLSL and HLSL in particular, are compiled and optimized, as well as what formats they are reduced to before machine code is generated (assuming they are not converted directly into machine code). After a bit of research into this, the best resource I've found is this presentation from ATI about the compilation of and optimizations for HLSL. I also found sample ARB assembly code. This sort of addressed my original curiosity, but it raised several other questions. The assembler code in the ATI presentation seems like it contains instructions specifically targeted for the GPU, but is this merely a hypothetical example created for the purpose of conceptual understanding, or is this code really generated during shader compilation? If so, is it possible to inspect it, or even write it in place of the higher-level syntax? My initial searches for an answer to the last question tell me that this may be disallowed, but I have not dug too deep yet. Also, along the same lines, are GLSL shader programs compiled into ARB assembly code before machine code is generated, and is it possible to write direct ARB assembly? Lastly, and perhaps what I am most interested in finding out: are there comprehensive resources on shader compilation and low-level GPU code? I have been unable to find any thus far. I ask simply because I am curious :)

  Read the article

• Taking fixed direction on hemisphere and project to normal (openGL)

  - by Maik Xhani

  I am trying to perform sampling using hemisphere around a surface normal. I want to experiment with fixed directions (and maybe jitter slightly between frames). So I have those directions: vec3 sampleDirections[6] = {vec3(0.0f, 1.0f, 0.0f), vec3(0.0f, 0.5f, 0.866025f), vec3(0.823639f, 0.5f, 0.267617f), vec3(0.509037f, 0.5f, -0.700629f), vec3(-0.509037f, 0.5f, -0.700629), vec3(-0.823639f, 0.5f, 0.267617f)}; now I want the first direction to be projected on the normal and the others accordingly. I tried these 2 codes, both failing. This is what I used for random sampling (it doesn't seem to work well, the samples seem to be biased towards a certain direction) and I just used one of the fixed directions instead of s (here is the code of the random sample, when i used it with the fixed direction i didn't use theta and phi). vec3 CosWeightedRandomHemisphereDirection( vec3 n, float rand1, float rand2 ) float theta = acos(sqrt(1.0f-rand1)); float phi = 6.283185f * rand2; vec3 s = vec3(sin(theta) * cos(phi), sin(theta) * sin(phi), cos(theta)); vec3 v = normalize(cross(n,vec3(0.0072, 1.0, 0.0034))); vec3 u = cross(v, n); u = s.x*u; v = s.y*v; vec3 w = s.z*n; vec3 direction = u+v+w; return normalize(direction); } ** EDIT ** This is the new code vec3 FixedHemisphereDirection( vec3 n, vec3 sampleDir) { vec3 x; vec3 z; if(abs(n.x) < abs(n.y)){ if(abs(n.x) < abs(n.z)){ x = vec3(1.0f,0.0f,0.0f); }else{ x = vec3(0.0f,0.0f,1.0f); } }else{ if(abs(n.y) < abs(n.z)){ x = vec3(0.0f,1.0f,0.0f); }else{ x = vec3(0.0f,0.0f,1.0f); } } z = normalize(cross(x,n)); x = cross(n,z); mat3 M = mat3( x.x, n.x, z.x, x.y, n.y, z.y, x.z, n.z, z.z); return M*sampleDir; } So if my n = (0,0,1); and my sampleDir = (0,1,0); shouldn't the M*sampleDir be (0,0,1)? Cause that is what I was expecting.

  Read the article

< Previous Page | 545 546 547 548 549 550 551 552 553 554 555 556  | Next Page >