I just got stuck with texturing my cubes - i searched web and realized that the only way to give my cube 6 different textures (With glDrawElements) is to create about a 24 indices. It is still faster than just glDrawArrays, but it seems quite illogically and horribly slow. I understand that the purpose of glDrawElements is to deal with complex models where very few indices share different texture coords.
But, i am still pretty confused, because glDrawElements gave me perfomance boost (Without any effects, just shader coloring) from about 50-67ms with 10,000 cubes (glDrawArrays), to 25-33ms with 100,000 cubes.
My question is: i just have to accept it, or there is still some way to come over this?
I do not know what you want to archieve but you can try to reduce the number ob vertices in the scene. This will make is much more faster. I think if your Cube needs more than one texture, you can store them in a texture atlas ans assign each vertex of the cube the corresponding texture coordinate for the image in the texture atlas. This will reduce the number of texture calls, as there is only one texutre. IN addition you will only need to render the vertices/Triangles which are visible by the user, so you do not need to texture the back of the cube.
I've found a good site which explains a lot of speeding up rendering thousands of cubes. They try to generate one big object on the CPU by culling a lot of vertices before loading them to the GPU, but I think for your problem, you can only try to use as less textures as possible (TextureAtlas), this will increase the performance, but for a whole cube you need the 8 Vertices with each a specific TextureCoordinate.
Related
I'm currently struggling with finding a good approach to render many (thousands) slightly different models. The model itself is a simple cube with some vertex offset, think of a skewed quad face. Each 'block' has a different offset of its vertices, so basically I have a voxel engine on steroids as each block is not a perfect cube but rather a skewed cuboid. To render this shape 48 vertices are needed but can be cut to 24 vertices as only 3 faces are visible. With indexing we are at 12 vertices (4 for each face).
But, now that I have the vertices for each block in the world, how do I render them?
What I've tried:
Instanced Rendering. Sounds good, doesn't work as my models are not the same.
I could simplify distant blocks to a cube and render them with glDrawArraysInstanced/glDrawElementsInstanced.
Put everything in one giant VBO. This has a better performance than rendering each cube individually, but has the downside of having one large mesh. This is not desireable as I need every cube to have different textures, lighting, etc... Selecting a single cube within that huge mesh is not possible.
I am aware of frustum culling and occlusion culling, but I already have problems with some cubes in front of me (tested with a 128x128 world).
My requirements:
Draw some thousand models.
Each model has vertices offsets to make the block less cubic, stored in another VBO.
Each block has to be an individual object, as you should be able to place/remove blocks.
Any good performance advices?
This is not desireable as I need every cube to have different textures, lighting, etc... Selecting a single cube within that huge mesh is not possible.
Programmers should avoid declaring that something is "impossible"; it limits your thinking.
Giving each face of these cubes different textures has many solutions. The Minecraft approach uses texture atlases. Each "texture" is really just a sub-section of one large texture, and you use texture coordinates to select which sub-section a particular face uses. But you can get more complex.
Array textures allow for a more direct way to solve this problem. Here, the texture coordinates would be the same, but you use a per-vertex integer to select the correct texture for a face. All of the vertices for a particular face would have an index. And if you're clever, you don't even really need texture coordinates. You can generate them in your vertex shader, based on per-vertex values like gl_VertexID and the like.
Lighting parameters would work the same way: use some per-vertex data to select parameters from a UBO or SSBO.
As for the "individual object" bit, that's merely a matter of how you're thinking about the problem. Do not confuse what happens in the player's mind with what happens in your code. Games are an elaborate illusion; just because something appears to the user to be an "individual object" doesn't mean it is one to your rendering engine.
What you need is the ability to modify your world's data to remove and add new blocks. And if you need to show a block as "selected" or something, then you simply need another per-block value (like the lighting parameters and index for the texture) which tells you whether to draw it as a "selected" block or as an "unselected" one. Or you can just redraw that specific selected block. There are many ways of handling it.
Any decent graphics card (since about 2010) is able to render a few millions vertices in a blinking.
The approach is different depending on how many changes per frame. In other words, how many data must be transferred to the GPU per frame.
For the case of small number of changes, storing the data in one big VBO or many smaller VBOs (and their VAOs), sending the changes by uniforms, and calling several glDraw***, shows similar performance. Different hardwares behave with little difference. Indexed data may improve the speed.
When most of the data changes in every frame and these changes are hard or impossible to do in the shaders, then your app is memory-transfer bound. Streaming is a good advise.
A cube with different colored faces in intermediate mode is very simple. But doing this same thing with shaders seems to be quite a challenge.
I have read that in order to create a cube with different coloured faces, I should create 24 vertices instead of 8 vertices for the cube - in other words, (I visualies this as 6 squares that don't quite touch).
Is perhaps another (better?) solution to texture the faces of the cube using a real simple texture a flat color - perhaps a 1x1 pixel texture?
My texturing idea seems simpler to me - from a coder's point of view.. but which method would be the most efficient from a GPU/graphic card perspective?
I'm not sure what your overall goal is (e.g. what you're learning to do in the long term), but generally for high performance applications (e.g. games) your goal is to reduce GPU load. Every time you switch certain states (e.g. change textures, render targets, shader uniform values, etc..) the GPU stalls reconfiguring itself to meet your demands.
So, you can pass in a 1x1 pixel texture for each face, but then you'd need six draw calls (usually not so bad, but there is some prep work and potential cache misses) and six texture sets (can be very bad, often as bad as changing shader uniform values).
Suppose you wanted to pass in one texture and use that as a texture map for the cube. This is a little less trivial than it sounds -- you need to express each texture face on the texture in a way that maps to the vertices. Often you need to pass in a texture coordinate for each vertex, and due to the spacial configuration of the texture this normally doesn't end up meaning one texture coordinate for one spatial vertex.
However, if you use an environmental/reflection map, the complexities of mapping are handled for you. In this way, you could draw a single texture on all sides of your cube. (Or on your sphere, or whatever sphere-mapped shape you wanted.) I'm not sure I'd call this easier since you have to form the environmental texture carefully, and you still have to set a different texture for each new colors you want to represent -- or change the texture either via the GPU or in step with the GPU, and that's tricky and usually not performant.
Which brings us back to the canonical way of doing as you mentioned: use vertex values -- they're fast, you can draw many, many cubes very quickly by only specifying different vertex data, and it's easy to understand. It really is the best way, and how GPUs are designed to run quickly.
Additionally..
And yes, you can do this with just shaders... But it'd be ugly and slow, and the GPU would end up computing it per each pixel.. Pass the object space coordinates to the fragment shader, and in the fragment shader test which side you're on and output the corresponding color. Highly not recommended, it's not particularly easier, and it's definitely not faster for the GPU -- to change colors you'd again end up changing uniform values for the shaders.
I'm making a small 2D game demo and from what I've read, it's better to use drawElements() to draw an indexed triangle list than using drawArrays() to draw an unindexed triangle list.
But it doesn't seem possible as far as I know to draw multiple elements that are not connected in a single draw call with drawElements().
So for my 2D game demo where I'm only ever going to draw squares made of two triangles, what would be the best approach so I don't end having one draw call per object?
Yes, it's better to use indices in many cases since you don't have to store or transfer duplicate vertices and you don't have to process duplicate vertices (vertex shader only needs to be run once per vertex). In the case of quads, you reduce 6 vertices to 4, plus a small amount of index data. Two thirds is quite a good improvement really, especially if your vertex data is more than just position.
In summary, glDrawElements results in
Less data (mostly), which means more GPU memory for other things
Faster updating if the data changes
Faster transfer to the GPU
Faster vertex processing (no duplicates)
Indexing can affect cache performance, if the reference vertices that aren't near each other in memory. Modellers commonly produce meshes which are optimized with this in mind.
For multiple elements, if you're referring to GL_TRIANGLE_STRIP you could use glPrimitiveRestartIndex to draw multiple strips of triangles with the one glDrawElements call. In your case it's easy enough to use GL_TRIANGLES and reference 4 vertices with 6 indices for each quad. Your vertex array then needs to store all the vertices for all your quads. If they're moving you still need to send that data to the GPU every frame. You could position all the moving quads at the front of the array and only update the active ones. You could also store static vertex data in a separate array.
The typical approach to drawing a 3D model is to provide a list of fixed vertices for the geometry and move the whole thing with the model matrix (as part of the model-view). The confusing part here is that the mesh data is so small that, as you say, the overhead of the draw calls may become quite prominent. I think you'll have to draw a LOT of quads before you get to the stage where it'll be a problem. However, if you do, instancing or some similar idea such as particle systems is where you should look.
Perhaps only go down the following track if the draw calls or data transfer becomes a problem as there's a lot involved. A good way of implementing particle systems entirely on the GPU is to store instance attributes such as position/colour in a texture. Each frame you use an FBO/render-to-texture to "ping-pong" this data between another texture and update the attributes in a fragment shader. To draw the particles, you can set up a static VBO which stores quads with the attribute-data texture coordinates for use in the vertex shader where the particle position can be read and applied. I'm sure there's a bunch of good tutorials/implementations to follow out there (please comment if you know of a good one).
I have to render large quantities of particles. These particles are simple non-textured quads (squares actually). Oh, and they're moving all the time since they're particles.
I have considered 2 options but as I'm not an OpenGL expert I don't know what's best.
Use VBOs to render them all.
Pros: faster than immediate mode.
Cons: (I don't know much about VBOs but) from what I gather the quads' coordinates need to be stored in some buffer in RAM... and all of these coordinates need to be computed by the CPU. So for particle P1(x,y) I would have to compute 4 other coordinates (P2(x-1,y-1), P3(x-1,y+1), P4(x+1,y+1), P5(x+1,y-1)) - that's a lot of work for the CPU!
Use a display list: First create a tiny display list for a single square quad. Then, to render each particle do some pushMatrix, glTranslate, callList, popMatrix.
Pros: I don't have to compute 4 coordinates manually - glTranlate does that.
Supposedly display lists are faster than VBOs.
Cons: Are they faster than VBOs when they contain just one quad?
Mind you: I'm calling OpenGL stuff from Java so there's no smooth way of transforming Java arrays to GPU arrays (everything has to be stored in intermediary FloatBuffers before transfers).
Display Lists are for static geometry. Rendering just one single quad, then changing the transformation, rinse and repeat is horribly inefficient.
Updating VBOs is better but still not optimal.
You should look into instanced rendering. Here's a tutorial:
http://ogldev.atspace.co.uk/www/tutorial33/tutorial33.html
In the case of simple quads using a geometry shader turning simple GL_POINTS into two GL_TRIANGLES would do the trick as well.
To do particle simulations i think what you are looking for is transform feedback, here is a nice demonstration with some code on how to do it http://prideout.net/blog/?tag=opengl-transform-feedback
I'm making a voxel engine in C++ and OpenGL (à la Minecraft) and can't get decent fps on my 3GHz with ATI X1600... I'm all out of ideas.
When I have about 12000 cubes on the screen it falls to under 20fps - pathetic.
So far the optimizations I have are: frustum culling, back face culling (via OpenGL's glEnable(GL_CULL_FACE)), the engine draws only the visible faces (except the culled ones of course) and they're in an octree.
I've tried VBO's, I don't like them and they do not significantly increase the fps.
How can Minecraft's engine be so fast... I struggle with a 10000 cubes, whereas Minecraft can easily draw much more at higher fps.
Any ideas?
#genpfault: I analyze the connectivity and just generate faces for the outer, visible surface. The VBO had a single cube that I glTranslate()d
I'm not an expert at OpenGL, but as far as I understand this is going to save very little time because you still have to send every cube to the card.
Instead what you should do is generate faces for all of the outer visible surface, put that in a VBO, and send it to the card and continue to render that VBO until the geometry changes. This saves you a lot of the time your card is actually waiting on your processor to send it the geometry information.
You should profile your code to find out if the bottleneck in your application is on the CPU or GPU. For instance it might be that your culling/octtree algorithms are slow and in that case it is not an OpenGL-problem at all.
I would also keep count of the number of cubes you draw on each frame and display that on screen. Just so you know your culling routines work as expected.
Finally you don't mention if your cubes are textured. Try using smaller textures or disable textures and see how much the framerate increases.
gDEBugger is a great tool that will help you find bottlenecks with OpenGL.
I don't know if it's ok here to "bump" an old question but a few things came up my mind:
If your voxels are static you can speed up the whole rendering process by using an octree for frustum culling, etc. Furthermore you can also compile a static scene into a potential-visibility-set in the octree. The main principle of PVS is to precompute for evere node in the tree which other nodes are potential visible from it and store pointers to them in a vector. When it comes to rendering you first check in which node the camera is placed and then run frustum culling against all nodes in the PVS-vector of the node.(Carmack used something like that in the Quake engines, but with Binary Space Partitioning trees)
If the shading of your voxels is kindalike complex it is also fast to do a pre-Depth-Only-Pass, without writing into the colorbuffer,just to fill the Depthbuffer. After that you render a 2nd pass: disable writing to the Depthbuffer and render only to the Colorbuffer while checking the Depthbuffer. So you avoid expensive shader-computations which are later overwritten by a new fragment which is closer to the viewer.(Carmack used that in Quake3)
Another thing which will definitely speed up things is the use of Instancing. You store only the position of each voxel and, if nescessary, its scale and other parameters into a texturebufferobject. In the vertexshader you can then read the positions of the voxels to be spawned and create an instance of the voxel(i.e. a cube which is given to the shader in a vertexbufferobject). So you send the 8 Vertices + 8 Normals (3 *sizeof(float) *8 +3 *sizeof(float) *8 + floats for color/texture etc...) only once to the card in the VBO and then only the positions of the instances of the Cube(3*sizeof(float)*number of voxels) in the TBO.
Maybe it is possibile to parallelize things between GPU and CPU by combining all 3 steps in 2 threads, in the CPU-thread you check the octrees pvs and update a TBO for instancing in the next frame, the GPU-thread does meanwhile render the 2 passes while using an TBO for instancing which was created by the CPU thread in the previous step. After that you switch TBOs. If the Camera has not moved you don't even have to do the CPU-calculations again.
Another kind of tree you me be interested in is the so called k-d-tree, which is more general than octrees.
PS: sorry for my english, it's not the clearest....
There are 3rd-party libraries you could use to make the rendering more efficient. For example the C++ PolyVox library can take a volume and generate the mesh for you in an efficient way. It has built-in methods for reducing triangle count and helping to generate things like ambient occlusion. It's got a good community around it so getting support on the forum should be easy.
Have you used a common display list for all your cubes ?
Do you skip calling drawing code of cubes which are not visible to the user ?