Sign in to follow this  
hatfarm

OpenGL How do I handle lots of textures

Recommended Posts

hatfarm    224

So, I've been working on a 2d turn based tiled strategy game, think like the original XCom.  However, I was doing it all from an overhead view and doing it 2d via draw functions (in Qt's QWidget if you're wondering).  However, I really don't like that because I want to be able to have land and air based units.  So, I thought maybe OpenGL would be a good option.  I've been working through tutorials and am able to draw a rectangle at the angle I'm wanting, and I'm able to put a couple of different textures on the rectangle.  That's fine.  However, if I want to draw a significant portion of map (say 15x15 tiles) and each of those have their own texture, how can I do that?  Most of the tutorials I've read said that you can expect to have at least 48 textures (source: http://open.gl/textures).  That tutorial even goes so far as to say that we'll never use that many.  I don't understand how he can say that.  If I want a unique texture for each of the tiles, I'm going to need 225 textures for my 15x15 example.  Obviously I'm missing something here, but I would love to know what that is.  I can't find anywhere that people talk about this.  Any help would be much appreciated.

Share this post


Link to post
Share on other sites
zacaj    667

You can only have 48 textures bound AT ONCE, which only happens when you're doing crazy shader based multitasking.  You should be fine with using 100s of textures, as long as only one is bound at a time.  You can probably safely assume most people will have at least 256MB of video ram, which means you can probably have at least 200MB of textures loaded at a time.

Share this post


Link to post
Share on other sites
hatfarm    224

Oh, okay.  That makes much more sense to me.  What would happen if my assumption of 256MB is wrong and they only have 128MB and I've got 200MB of textures I'm trying to load.  What happens?  Does OpenGL have some sort of virtual memory?

Share this post


Link to post
Share on other sites
zacaj    667

256MB is a very conservative estimate.  Any recent computer is going to have at least 1GB.  Check out the Steam Hardware survey at http://store.steampowered.com/hwsurvey under VRAM.  99% of people have at least 256MB, 91% have 512MB.  I'm not sure how high resolution textures you're using, but I doubt you'd even get close to the 256.  256MB gives you space for a 8192x8192 texture.  If each tile is 16x16, that gives you enough space for 262144 tiles.  If you've got more than 1000 different tiles loaded at once, you might want to reevaluate what you're doing.    

 

When your video card runs out of space, the driver will just start using the regular system ram as "virtual memory", which I know causes major slowdowns, but I don't have any first hand experience there.  Don't forget though, using it as virtual memory is going to be pretty much the same as you just uploading your textures from RAM each time they're used.  I'd assume you can do a much better job of switching out the textures that are least likely to be used than the graphics card can, if by some strange turn of events that actually happens.

Share this post


Link to post
Share on other sites
hatfarm    224

Yeah, I don't expect that I'd be filling up that much space, that was mostly just a question because I wanted to know.  Thanks for the information, I'm sure I'll have 100 or so more before I have a full scene drawn :)

Share this post


Link to post
Share on other sites
hatfarm    224

Okay, so I've got multiple textures loaded (and usable) and I'm able to bind them dynamically.  There are a couple of layers to this question.

 

First, what is the best texture strategy if I want to have, for instance, a 100x100 map that could possibly have all unique tiles (probably wouldn't, but it's possible).  You're saying if I have over 1000 textures loaded, I need to rethink things.  Does that mean it would be better for me to do something like Vlad86 is saying and just have a spritesheet of textures?  They're all 32x32, so that wouldn't be hard per se, I'd just rather not have to do that if I don't have to.  I know that 100x100 won't be on the screen (I did the math and I'd like possible 45x25 ish at a time). So, should I load them dynamically?  I wouldn't think that would be efficient at all, but is that a bad assumption?

 

Second, what would be the best way to draw that many tiles?  Currently, I'm drawing 4 tiles with an ibo and using glDrawElements with a buffer offset into my index buffer.  However, this means I'm calling 4 draw functions just for these 4 tiles.  With possibly over a 1000 tiles displayed at a time, that's over 1000 draw calls.  That just doesn't seem doable.  Am I wrong in this assumption?  Since they're all being drawn the same way, I don't want to change the shader just to have different texture uniforms or anything.

 

I thought I had more, but I'm drawing a blank after those two.  If I think of them, I'll post them here.  Thanks for your help so far!

Share this post


Link to post
Share on other sites
zacaj    667

If your map isn't changing at all it'd probably be better to render all your little tiles to a few bigger tiles and just draw those specialized ones.  My "1000 textures" thing might have been exaggerating a bit, but.  32x32x4bppx1000 is only 4MB of VRAM, and shouldn't really be a problem.  I would recommend doing a spritesheet though.  It should be quite easy to have use glTexSubimage to make a few big 2048x2048 textures and just drop any specific tile loaded into one, keeping track of its coordinates in the larger texture, then just have an array that stores the coordinates for you to look up the tile in when you need it.  

 

Although you have one draw call per tile, are you *really* going to have all 1000 tiles on the screen at one time?  (I mean, I guess you could, but)  I doubt 1000 draw calls is going to be very bad.  For most of these things, the best advice really is to just try it.  Think of the different ways it could be done, and just make sure you have a tile drawing interface that could work with any of those, should you find the need to switch somewhere down the line

Share this post


Link to post
Share on other sites
hatfarm    224

So, I've got a working render algorithm going.  The tiles have 3 possible heights and the map is randomly generated.  There are no ramps yet, and the map starts in the center, but I thought it'd be cool to post a progress picture.  Thanks for the help thus far.bvfqfqg.png

Share this post


Link to post
Share on other sites
japro    887

Does that mean it would be better for me to do something like Vlad86 is saying and just have a spritesheet of textures?  They're all 32x32, so that wouldn't be hard per se, I'd just rather not have to do that if I don't have to....

 

Second, what would be the best way to draw that many tiles?  Currently, I'm drawing 4 tiles with an ibo and using glDrawElements with a buffer offset into my index buffer....

There is a connection between those two. Generally one would want to draw as many tiles as possible with a single draw call. This is actually a good use case for instancing. This also means you pretty much have to do the spritesheet thing (which is preferable for performance anyway).

 

 

 

You can only have 48 textures bound AT ONCE...

Uh, you can query how many texture units there are from GL_MAX_TEXTURE_IMAGE units. The spec says it has to be at least 16 but it will vary between vendor/driver/card. On my laptop with AMD gpu it is for example 32. So you can't just state 48 as hard number.

Edited by japro

Share this post


Link to post
Share on other sites
zacaj    667

zacaj, on 03 Apr 2013 - 20:42, said:
You can only have 48 textures bound AT ONCE...
Uh, you can query how many texture units there are from GL_MAX_TEXTURE_IMAGE units. The spec says it has to be at least 16 but it will vary between vendor/driver/card. On my laptop with AMD gpu it is for example 32. So you can't just state 48 as hard number.

Ah. I was remembering GL_MAX_COMBINED_TEXTURE_IMAGE_UNITS. The 3.3 spec requires the card support 16 textures for each stage.

Share this post


Link to post
Share on other sites
hatfarm    224

I think the 48 number came from my original post, because one of the tutorials I was reading said you could count on at least 48 these days.  I used that number in my original post.

 

Another question, instancing was mentioned.  What do you mean by instancing?  I'm always interested in hearing about an increasing performance.

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  

  • Similar Content

    • By pseudomarvin
      I assumed that if a shader is computationally expensive then the execution is just slower. But running the following GLSL FS instead just crashes
      void main() { float x = 0; float y = 0; int sum = 0; for (float x = 0; x < 10; x += 0.00005) { for (float y = 0; y < 10; y += 0.00005) { sum++; } } fragColor = vec4(1, 1, 1 , 1.0); } with unhandled exception in nvoglv32.dll. Are there any hard limits on the number of steps/time that a shader can take before it is shut down? I was thinking about implementing some time intensive computation in shaders where it would take on the order of seconds to compute a frame, is that possible? Thanks.
    • By Arulbabu Donbosco
      There are studios selling applications which is just copying any 3Dgraphic content and regenerating into another new window. especially for CAVE Virtual reality experience. so that the user opens REvite or CAD or any other 3D applications and opens a model. then when the user selects the rendered window the VR application copies the 3D model information from the OpenGL window. 
      I got the clue that the VR application replaces the windows opengl32.dll file. how this is possible ... how can we copy the 3d content from the current OpenGL window.
      anyone, please help me .. how to go further... to create an application like VR CAVE. 
       
      Thanks
    • By cebugdev
      hi all,

      i am trying to build an OpenGL 2D GUI system, (yeah yeah, i know i should not be re inventing the wheel, but this is for educational and some other purpose only),
      i have built GUI system before using 2D systems such as that of HTML/JS canvas, but in 2D system, i can directly match a mouse coordinates to the actual graphic coordinates with additional computation for screen size/ratio/scale ofcourse.
      now i want to port it to OpenGL, i know that to render a 2D object in OpenGL we specify coordiantes in Clip space or use the orthographic projection, now heres what i need help about.
      1. what is the right way of rendering the GUI? is it thru drawing in clip space or switching to ortho projection?
      2. from screen coordinates (top left is 0,0 nd bottom right is width height), how can i map the mouse coordinates to OpenGL 2D so that mouse events such as button click works? In consideration ofcourse to the current screen/size dimension.
      3. when let say if the screen size/dimension is different, how to handle this? in my previous javascript 2D engine using canvas, i just have my working coordinates and then just perform the bitblk or copying my working canvas to screen canvas and scale the mouse coordinates from there, in OpenGL how to work on a multiple screen sizes (more like an OpenGL ES question).
      lastly, if you guys know any books, resources, links or tutorials that handle or discuss this, i found one with marekknows opengl game engine website but its not free,
      Just let me know. Did not have any luck finding resource in google for writing our own OpenGL GUI framework.
      IF there are no any available online, just let me know, what things do i need to look into for OpenGL and i will study them one by one to make it work.
      thank you, and looking forward to positive replies.
    • By fllwr0491
      I have a few beginner questions about tesselation that I really have no clue.
      The opengl wiki doesn't seem to talk anything about the details.
       
      What is the relationship between TCS layout out and TES layout in?
      How does the tesselator know how control points are organized?
          e.g. If TES input requests triangles, but TCS can output N vertices.
             What happens in this case?
      In this article,
      http://www.informit.com/articles/article.aspx?p=2120983
      the isoline example TCS out=4, but TES in=isoline.
      And gl_TessCoord is only a single one.
      So which ones are the control points?
      How are tesselator building primitives?
    • By Orella
      I've been developing a 2D Engine using SFML + ImGui.
      Here you can see an image
      The editor is rendered using ImGui and the scene window is a sf::RenderTexture where I draw the GameObjects and then is converted to ImGui::Image to render it in the editor.
      Now I need to create a 3D Engine during this year in my Bachelor Degree but using SDL2 + ImGui and I want to recreate what I did with the 2D Engine. 
      I've managed to render the editor like I did in the 2D Engine using this example that comes with ImGui. 
      3D Editor preview
      But I don't know how to create an equivalent of sf::RenderTexture in SDL2, so I can draw the 3D scene there and convert it to ImGui::Image to show it in the editor.
      If you can provide code will be better. And if you want me to provide any specific code tell me.
      Thanks!
  • Popular Now