Sign in to follow this  
sth

OpenGL High performance texture splatting?

Recommended Posts

Is there a way to improve the performance of texture reads in OpenGL (ES)?
I'd like to do texture splatting but I'm running into severe performance problems when using multiple texture reads in my fragment shader. With a single texture, I get a constant 60fps (vsync limited, with room to spare). Adding a second texture read, my performance drops down to 40fps and with the shader provided below, performance is down to 15fps.

Any ideas?

[source lang="plain"]# Simplified fragment shader for demonstration purposes

varying lowp vec2 v_texcoord;
varying lowp vec4 v_color;

uniform sampler2D texture0;
uniform sampler2D texture1;
uniform sampler2D texture2;
uniform sampler2D texture3;
uniform sampler2D texture4;


void main()
{
lowp vec4 alpha = texture2D(texture4, v_texcoord);

lowp vec4 color = texture2D(texture0, v_texcoord);
color = mix(color, texture2D(texture1, v_texcoord), alpha[0]);
color = mix(color, texture2D(texture2, v_texcoord), alpha[1]);
color = mix(color, texture2D(texture3, v_texcoord), alpha[2]);

gl_FragColor = v_color * color;
}[/source] Edited by sth

Share this post


Link to post
Share on other sites
If you throw out the vsync sample, your frame times are 40Hz/25ms for 2 samples, and 15Hz/66.6..ms for 5 samples. Divide the frame times by the number of samples and you get 12.5ms and 13.3..ms, which are pretty similar.
From that, it looks like you're just using more texture bandwidth than your GPU can cope with... draw less pixels using these shaders -- use distance material-LOD maybe?

You can also reduce data in many ways, such as packing 4 different monochrome textures together along with 4 uniform colours.

Share this post


Link to post
Share on other sites
Perhaps try avoiding the mix and optimize the code manually:

[code]
void main()
{
lowp vec4 alpha = texture2D(texture4, v_texcoord);

lowp vec4 color0 = texture2D(texture0, v_texcoord);
lowp vec4 color1 = texture2D(texture1, v_texcoord);
lowp vec4 color2 = texture2D(texture2, v_texcoord);
lowp vec4 color3 = texture2D(texture3, v_texcoord);

gl_FragColor = v_color * (alpha[0] * color0 + alpha[1] * color1 + alpha[2] * color2 + alpha[3] * color3);
}[/code]

(I reindexed the way how the indices of the alpha vector affect the read color texture for straightforwardness). The idea is that alpha[0] is already precomputed to be 1.0f - alpha[1] - alpha[2] - alpha[3] in the texture, so one doesn't need to compute that in the shader. I feel this would be faster than using mix(), but can't be sure without profiling. Let me know how it compares.

Something that's potentially optimizable is to drop one or two texture channels to splat, and subdivide your mesh down by which splat textures it is using at each triangle. Also, if the splat texture is low-frequency, try storing the splat weights as vertex attributes and pass them through to pixel shader, which will avoid you one texture read.

Finally, if the splat texture is very low frequency, you can try just decaling the contents, i.e. manually generate geometry planes that you alphablend on top of the terrain.

Share this post


Link to post
Share on other sites
[quote name='Hodgman' timestamp='1340028032' post='4950245']
If you throw out the vsync sample, your frame times are 40Hz/25ms for 2 samples, and 15Hz/66.6..ms for 5 samples. Divide the frame times by the number of samples and you get 12.5ms and 13.3..ms, which are pretty similar.
From that, it looks like you're just using more texture bandwidth than your GPU can cope with...[/quote]
I think you're right, it looks pretty much bandwidth-limited right now.
Today I realized that I still had trilinear filtering enabled. Disabling it cut the frame times [i]in half[/i]. I can't recall trilinear filtering ever having such a profound effect in any of my projects, but I guess that's just what happens when you're running at the limit.

Anyway, at least I'm now back to playable framerates.

[quote name='clb' timestamp='1340028652' post='4950252']
Perhaps try avoiding the mix and optimize the code manually:
[/quote]
Thanks for the suggestion. I tried it but it didn't have any effect on the performance.

[quote name='dpadam450' timestamp='1340044660' post='4950331']
Also, Why not try just 3 splat channels instead of 4. 3 should be sufficient especially if it is for a small screen embedded device.
[/quote]
I'm considering it, but I'm still not sure if I really want it.
The thing is that these days it's not just phone screens, it's also tablet screens and (wireless) TV output.

Share this post


Link to post
Share on other sites
[quote]The thing is that these days it's not just phone screens, it's also tablet screens and (wireless) TV output.[/quote]
Well the thing really is, you don't have enough power currently on GLES. If you want 20 different types of splats, you could do that as well, just divide your terrain into sections and only allow 1 section 3 splats (rgb). Each section can have any 3 of the 20 splats.

Share this post


Link to post
Share on other sites
[quote name='sth' timestamp='1340061761' post='4950422']I think you're right, it looks pretty much bandwidth-limited right now.Today I realized that I still had trilinear filtering enabled. Disabling it cut the frame times in half. I can't recall trilinear filtering ever having such a profound effect in any of my projects, but I guess that's just what happens when you're running at the limit.[/quote]Yeah, trilinear filtering will double the amount of data that each pixel has to pull into the texture cache, so if you're already texture-bound, it going to make things a lot worse.[quote name='dpadam450' timestamp='1340064858' post='4950429']
Well the thing really is, you don't have enough power currently on GLES. If you want 20 different types of splats, you could do that as well, just divide your terrain into sections and only allow 1 section 3 splats (rgb). Each section can have any 3 of the 20 splats.[/quote]Yep, I've worked on a few games where each triangle could only have 2 layers on it, but the entire terrain could have many, many layers. We'd use a tool to split the terrain up into sections, where each section only had 2 texture layers. As long as you never needed 3 or more materials present at a single vertex, it worked ok.

Share this post


Link to post
Share on other sites
[quote name='dpadam450' timestamp='1340064858' post='4950429']
[quote]The thing is that these days it's not just phone screens, it's also tablet screens and (wireless) TV output.[/quote]
Well the thing really is, you don't have enough power currently on GLES. If you want 20 different types of splats, you could do that as well, just divide your terrain into sections and only allow 1 section 3 splats (rgb). Each section can have any 3 of the 20 splats.
[/quote]
I'll have to see. My current implementation already allows every chunk of the terrain to have its own set of textures, although it would be nicer not to be bound to the fixed chunk structure.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

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

Create an account

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

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

Sign in to follow this  

  • Announcements

  • Forum Statistics

    • Total Topics
      628331
    • Total Posts
      2982113
  • Similar Content

    • By DejayHextrix
      Hi, New here. 
      I need some help. My fiance and I like to play this mobile game online that goes by real time. Her and I are always working but when we have free time we like to play this game. We don't always got time throughout the day to Queue Buildings, troops, Upgrades....etc.... 
      I was told to look into DLL Injection and OpenGL/DirectX Hooking. Is this true? Is this what I need to learn? 
      How do I read the Android files, or modify the files, or get the in-game tags/variables for the game I want? 
      Any assistance on this would be most appreciated. I been everywhere and seems no one knows or is to lazy to help me out. It would be nice to have assistance for once. I don't know what I need to learn. 
      So links of topics I need to learn within the comment section would be SOOOOO.....Helpful. Anything to just get me started. 
      Thanks, 
      Dejay Hextrix 
    • By mellinoe
      Hi all,
      First time poster here, although I've been reading posts here for quite a while. This place has been invaluable for learning graphics programming -- thanks for a great resource!
      Right now, I'm working on a graphics abstraction layer for .NET which supports D3D11, Vulkan, and OpenGL at the moment. I have implemented most of my planned features already, and things are working well. Some remaining features that I am planning are Compute Shaders, and some flavor of read-write shader resources. At the moment, my shaders can just get simple read-only access to a uniform (or constant) buffer, a texture, or a sampler. Unfortunately, I'm having a tough time grasping the distinctions between all of the different kinds of read-write resources that are available. In D3D alone, there seem to be 5 or 6 different kinds of resources with similar but different characteristics. On top of that, I get the impression that some of them are more or less "obsoleted" by the newer kinds, and don't have much of a place in modern code. There seem to be a few pivots:
      The data source/destination (buffer or texture) Read-write or read-only Structured or unstructured (?) Ordered vs unordered (?) These are just my observations based on a lot of MSDN and OpenGL doc reading. For my library, I'm not interested in exposing every possibility to the user -- just trying to find a good "middle-ground" that can be represented cleanly across API's which is good enough for common scenarios.
      Can anyone give a sort of "overview" of the different options, and perhaps compare/contrast the concepts between Direct3D, OpenGL, and Vulkan? I'd also be very interested in hearing how other folks have abstracted these concepts in their libraries.
    • By aejt
      I recently started getting into graphics programming (2nd try, first try was many years ago) and I'm working on a 3d rendering engine which I hope to be able to make a 3D game with sooner or later. I have plenty of C++ experience, but not a lot when it comes to graphics, and while it's definitely going much better this time, I'm having trouble figuring out how assets are usually handled by engines.
      I'm not having trouble with handling the GPU resources, but more so with how the resources should be defined and used in the system (materials, models, etc).
      This is my plan now, I've implemented most of it except for the XML parts and factories and those are the ones I'm not sure of at all:
      I have these classes:
      For GPU resources:
      Geometry: holds and manages everything needed to render a geometry: VAO, VBO, EBO. Texture: holds and manages a texture which is loaded into the GPU. Shader: holds and manages a shader which is loaded into the GPU. For assets relying on GPU resources:
      Material: holds a shader resource, multiple texture resources, as well as uniform settings. Mesh: holds a geometry and a material. Model: holds multiple meshes, possibly in a tree structure to more easily support skinning later on? For handling GPU resources:
      ResourceCache<T>: T can be any resource loaded into the GPU. It owns these resources and only hands out handles to them on request (currently string identifiers are used when requesting handles, but all resources are stored in a vector and each handle only contains resource's index in that vector) Resource<T>: The handles given out from ResourceCache. The handles are reference counted and to get the underlying resource you simply deference like with pointers (*handle).  
      And my plan is to define everything into these XML documents to abstract away files:
      Resources.xml for ref-counted GPU resources (geometry, shaders, textures) Resources are assigned names/ids and resource files, and possibly some attributes (what vertex attributes does this geometry have? what vertex attributes does this shader expect? what uniforms does this shader use? and so on) Are reference counted using ResourceCache<T> Assets.xml for assets using the GPU resources (materials, meshes, models) Assets are not reference counted, but they hold handles to ref-counted resources. References the resources defined in Resources.xml by names/ids. The XMLs are loaded into some structure in memory which is then used for loading the resources/assets using factory classes:
      Factory classes for resources:
      For example, a texture factory could contain the texture definitions from the XML containing data about textures in the game, as well as a cache containing all loaded textures. This means it has mappings from each name/id to a file and when asked to load a texture with a name/id, it can look up its path and use a "BinaryLoader" to either load the file and create the resource directly, or asynchronously load the file's data into a queue which then can be read from later to create the resources synchronously in the GL context. These factories only return handles.
      Factory classes for assets:
      Much like for resources, these classes contain the definitions for the assets they can load. For example, with the definition the MaterialFactory will know which shader, textures and possibly uniform a certain material has, and with the help of TextureFactory and ShaderFactory, it can retrieve handles to the resources it needs (Shader + Textures), setup itself from XML data (uniform values), and return a created instance of requested material. These factories return actual instances, not handles (but the instances contain handles).
       
       
      Is this a good or commonly used approach? Is this going to bite me in the ass later on? Are there other more preferable approaches? Is this outside of the scope of a 3d renderer and should be on the engine side? I'd love to receive and kind of advice or suggestions!
      Thanks!
    • By nedondev
      I 'm learning how to create game by using opengl with c/c++ coding, so here is my fist game. In video description also have game contain in Dropbox. May be I will make it better in future.
      Thanks.
    • By Abecederia
      So I've recently started learning some GLSL and now I'm toying with a POM shader. I'm trying to optimize it and notice that it starts having issues at high texture sizes, especially with self-shadowing.
      Now I know POM is expensive either way, but would pulling the heightmap out of the normalmap alpha channel and in it's own 8bit texture make doing all those dozens of texture fetches more cheap? Or is everything in the cache aligned to 32bit anyway? I haven't implemented texture compression yet, I think that would help? But regardless, should there be a performance boost from decoupling the heightmap? I could also keep it in a lower resolution than the normalmap if that would improve performance.
      Any help is much appreciated, please keep in mind I'm somewhat of a newbie. Thanks!
  • Popular Now