EDIT: I'm still looking for some help about the use of OpenCL or compute shaders. I would prefer to keep using OGL 3.3 and not have to deal with the bad driver support for OGL 4.3 and OpenCL 1.2, but I can't think of anyway to do this type of shading without using one of the two (to match lights and tiles). Is it possible to implement tile-based culling without using GPGPU?
I wrote a deferred render in OpenGL 3.3. Right now I don't do any culling for the light pass (I just render a full screen quad for every light). This (obviously) has a ton of overdraw. (Sometimes it is ~100%). Because of this I've been looking into ways to improve performance during the light pass. It seems like the best way in (almost) everyone's opinion is to cull the scene using screen space tiles. This was the method used in Frostbite 2. I read the the presentation from Andrew Lauritzen during SIGGRAPH 2010 (http://download-software.intel.com/sites/default/files/m/d/4/1/d/8/lauritzen_deferred_shading_siggraph_2010.pdf) , and I'm not sure I fully understand the concept. (and for that matter why it's better than anything else, and if it is better for me)
In the presentation Laurtizen goes over deferred shading with light volumes, quads, and tiles for culling the scene. According to his data, the tile based deferred renderer was the fastest (by far). I don't understand why it is though. I'm guessing it has something to do with the fact that for each tile, all the lights are batched together. In the presentation it says to read the G-Buffer once and then compute the lighting, but this doesn't make sense to me. In my mind, I would implement this like this:
for each tile {
for each light effecting the tile {
render quad (the tile) and compute lighting
blend with previous tiles (GL_ONE, GL_ONE)
}
}
This would still involve sampling the G-Buffer a lot. I would think that doing that would have the same (if not worse) performance than rendering a screen aligned quad for every light. From how it's worded though, it seems like this is what's happening:
for each tile {
render quad (the tile) and compute all lights
}
But I don't see how one would do this without exceeding the instruction limit for the fragment shader on some GPUs . Can anyone help me with this? It also seems like almost every tile based deferred renderer uses compute shaders or OpenCL (to batch the lights), why is this, and if I didn't use these what would happen?
But I don't see how one would do this without exceeding the instruction limit for the fragment shader on some GPUs .
It rather depends on how many lights you have. The "instruction limits" are pretty high; it's generally not something you need to worry about outside of degenerate cases. Even if 100+ lights affects a tile, odds are fairly good that your lighting computations aren't going to exceed instruction limits.
Modern GL 3.3 hardware can run at least 65536 dynamic instructions in a fragment shader, and likely more. For 100 lights, that's still 655 instructions per light. Even if you take 2000 instructions to compute the camera-space position, that still leaves 635 instructions per light. Even if you were doing Cook-Torrance directly in the GPU, that's probably still sufficient.
Related
In Minecraft for example, you can place torches anywhere and each one effects the light level in the world and there is no limit to the amount of torches / light sources you can put down in the world. I am 99% sure that the lighting for the torches is taken care of on the CPU and stored for each block and so when rendering the light value at that certain block just needs to be passed into the shader, but light sources cannot move for this reason. If you had a game where you could place light sources that could move around (arrow on fire, minecart with a light on it, glowing ball of energy) and the lighting wasn't as simple (color was included) what are the most efficient ways to calculate the lighting effects.
From my research I have found differed rendering, differed lighting, dynamically creating shaders with different amounts of lights available and using a for loop (can't use uniforms due to unrolling), and static light maps (these would probably only be used for the still lights). Are there any other ways to do lighting calculations such as doing what minecraft does except allowing moving lights, or is it possible to take an infinite amount of lights and mathematically combine them into an approximation that only involves a few lights (this is an idea I came up with but I can't figure out how it could be done)?
If it helps, I am a programmer with decent experience in OpenGL (legacy and modern) so you can give me code snippets although I have not done too much with lighting so brief explanations would be appreciated. I am also willing to do research if you can point me in the right direction!
Your title is a bit misleading infinite light implies directional light in infinite distance like Sun. I would use unlimited number of lights instead. Here some approaches for this I know of:
(back) ray-tracers
they can handle any number of light sources natively. Light is just another object in engine. If ray hits the light source it just take the light intensity and stop the recursion. Unfortunately current gfx hardware is not suited for this kind of rendering. There are GPU enhanced engines for this but the specialized gfx HW is still in development and did not hit the market yet. Memory requirements are not much different then standard BR rendering and You can still use BR meshes but mathematical (analytical) meshes are natively supported and are better for this.
Standard BR rendering
BR means boundary representation such engines (Like OpenGL fixed function) can handle only limited number of lights. This is because each primitive/fragment needs the complete list of lights and the computations are done for all light on per primitive or per fragment basis. If you got many light this would be slow.
GLSL example of fixed number of light sources see the fragment shader
Also the current GPU's have limited memory for uniforms (registers) in which the lights and other rendering parameters are stored so there are possible workarounds like have light parameters stored in a texture and iterate over all of them per primitive/fragment inside GLSL shader but the number of lights affect performance of coarse so you are limited by target frame-rate and computational power. Additional memory requirements for this is just the texture with light parameters which is not so much (few vectors per light).
light maps
they can be computed even for moving objects. Complex light maps can be computed slowly (not per frame). This leads to small lighting artifacts but you need to know what to look for to spot it. Light maps and shadow maps are very similar and often computed at once. There are simple light maps and complex radiation maps models out there
look Shading mask algorithm for radiation calculations
These are either:
projected 2D maps (hard to implement/use and often less precise)
3D Voxel maps (Memory demanding but easier to compute/use)
Some approaches uses pre-rendered Z-Buffer as geometry source and then fill the lights via Radiosity or any other technique. These can handle any number of lights as these maps can be computation demanding they are often computed in the background and updated once in a while.
fast moving light sources are usually updated more often or excluded from maps and rendered as transparent geometry to make impression of light. The computational power needed for this depends on the computation method the basic are done like:
set a camera to the larges visible surfaces
render scene and handle the result as light/shadow map
store it as 2D or 3D texture or voxel map
and then continue with normal rendering from camera view
So you need to render scene more then once per frame/map update and also need additional buffers to store the rendered result which for high resolution or Voxel maps can be a big chunk of memory.
multi pass light layer
there are cases when light is added after rendering of the scene for example I used it for
Atmospheric scattering in GLSL
Here comes all multi pass rendering techniques you need additional buffers to store the sub results and usually the multi pass rendering is done on the same view/scene so pre-rendered geometry is used which significantly speeds this up either as locked VAO or as already rendered Z-buffer Color and Index buffers from first pass. After this handle next passes as single or few Quads (like in the Atmospheric scattering link) so the computational power needed for this is not much bigger in comparison to basic BR rendering
forward rendering vs. deferred-rendering
in a google this forward rendering vs. deferred-rendering is first relevant hit I found. It is not very good one (a bit to vague for my taste) but for starters it is enough
forward rendering techniques are usually standard single pass BR renders
deffered rendering is standard multi pass renders. In first pass is rendered all the geometries of the scene into Z buffer, Color buffer and some auxiliary buffers just to know which fragment of the result belongs to which object,material,... And then in the next passes are added effects,light,shadows,... but the geometry is not rendered again instead just single or few overlay QUADs/per pass are rendered so the next passes are usually pretty fast ...
The link suggest that for high lights number is the deffered rendering more suited but that strongly depends on which of the previous technique is used. Usually the multi pass light layer is used (with is one of the standard deffered rendering techniques) so in that case it is true, and the memory and computational power demands are the same see the previous section.
Is there any difference in performance between drawing a scene with full triangles (GL_TRIANGLES) instead of just drawing their vertices (GL_POINTS), on modern hardware?
Where GL_POINTS is initialized like this:
glPointSize(1.0);
glDisable(GL_POINT_SMOOTH);
I have a somewhat low-end graphics card (9600gt) and drawing vertices-only can bring a 2x fps increase on certain sceneries. Not sure if it applies too on more recent gpus.
2x fps increase on
You lose 98% of picture and get only 2x fps increase. That's not impressive. If you take into account that you should be able to easily render 300..500 fps on any decent hardware (with vsync disabled and minor optimizations), that's probably not worth it.
Is there any difference in performance between drawing a scene with full triangles (GL_TRIANGLES) instead of just drawing their vertices (GL_POINTS), on modern hardware?
Well, if your scene has a LOT of alpha-blending and very "heavy" pixel shaders, then, obviously, displaying scene as point cloud will speed things up, because there's less pixels to fill.
On other hand, this kind of "optimization" will be completely useless for any practical task. I mean, if you're using blending and shaders, you probably wouldn't want to display your scene as pointlist in the first place, unless you're doing some kind of debug render (using glPolygonMode), and in case of debug render, you'll probably turn shaders off (because shaded/lit point will be hard to see) and disable lighting.
Even if you're using point sprites as particles or something, I'd stick with triangles - they give more control and do not have maximum size limit (compared to point sprites).
I can display more objects?
If you want more objects, you should probably try to optimzie things elsewhere first. If you stop trying to draw invisible objects (outside of field of view, etc), that'll be a start that can improve performance.
you have a mesh which is very far away from the camera. 1 million triangles and you know it is always in view. At this density ratio, triangles can't be bigger than a pixel,
When triangles are smaller than a pixel, and there are many of them, your mesh start looking like garbage and turns into pixelated mess of points. It will be ugly. Roughly same effect as when you disable mippimapping and texture filters and then render checkboard pattern. Using points instead of triangles might even aggravate effect.
: If you have 1mil triangle mesh that is always visible, you already need different kind of optimization. Reduce number of triangles (level of detail, dynamic tesselation or some solution that can simplify geometry on the fly), use bump mapping(maybe parallax mapping) to simulate extra geometry details that aren't even here, or even turn it into static background or a sprite. That'll work much better. Trying to render it using points will simply make it look ugly.
No, if the number of triangles is similar to the number of their shared vertices (considering the glDrawElements rendering command being used) in both modes the geometry-wise part of the rendering pipeline will be evaluated at roughly the same speed. The only benefit you can get from drawing GL_POINTS relies solely on the percentage of empty screen space you get from not drawing faces, thus only at fragment shader level.
With the GLE Tubing and Extrusion Library (http://www.linas.org/gle/) I am able to extrude 2D countours into 3D objects using OpenGL. The Library does all the work on the CPU and uses OpenGL immediate mode.
I guess doing the extrusion on the GPU using Geometry Shaders might be faster especially when rendering a lot of geometry. Since I do not yet have any experience with Geometry Shaders in OpenGL i would like to know if that is possible and what I have to pay attention to. Do you think it is a good Idea to move those computations to the GPU and that it will increase performance? It should also be possible to get the rendered geometry back to the CPU from the GPU, possibly using "Render to VBO".
If the geometry indeed changes every frame, you should do it on the GPU.
Keep in mind that every other solution that doesn't rely on the immediate mode will be faster than what you have right now. You might not even have to do it on the GPU.
But maybe you want to use shadow mapping instead, which is more efficient in some cases. It will also make it possible to render shadows for alpha tested objects like grass.
But it seems like you really need the resulting shadow geometry, so I'm not sure if that's an option for you.
Now back to the shadow volumes.
Extracting the shadow silhouette from a mesh using geometry shaders is a pretty complex process. But there's enough information about it on the internet.
Here's an article by Nvidia, which explains the process in detail:
Efficient and Robust Shadow Volumes Using Hierarchical Occlusion Culling and Geometry Shaders.
Here's another approach (from 2003) which doesn't even require geometry shaders, which could be interesting on low-end hardware:
http://de.slideshare.net/stefan_b/shadow-volumes-on-programmable-graphics-hardware
If you don't need the most efficient solution (using the shadow silhouette), you can also simply extract every triangle of the mesh on it's own. This is very easy using a geometry shader. I'd try that first before trying to implement silhouette extraction on the GPU.
About the "render to VBO" part of your question:
As far as I know there's no way to read the output of the geometry shader back to the CPU. Don't quote me on this, but I've never heard of a way to do this.
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 ?
I would like to draw voxels by using opengl but it doesn't seem like it is supported. I made a cube drawing function that had 24 vertices (4 vertices per face) but it drops the frame rate when you draw 2500 cubes. I was hoping there was a better way. Ideally I would just like to send a position, edge size, and color to the graphics card. I'm not sure if I can do this by using GLSL to compile instructions as part of the fragment shader or vertex shader.
I searched google and found out about point sprites and billboard sprites (same thing?). Could those be used as an alternative to drawing a cube quicker? If I use 6, one for each face, it seems like that would be sending much less information to the graphics card and hopefully gain me a better frame rate.
Another thought is maybe I can draw multiple cubes using one drawelements call?
Maybe there is a better method altogether that I don't know about? Any help is appreciated.
Drawing voxels with cubes is almost always the wrong way to go (the exceptional case is ray-tracing). What you usually want to do is put the data into a 3D texture and render slices depending on camera position. See this page: https://developer.nvidia.com/gpugems/GPUGems/gpugems_ch39.html and you can find other techniques by searching for "volume rendering gpu".
EDIT: When writing the above answer I didn't realize that the OP was, most likely, interested in how Minecraft does that. For techniques to speed-up Minecraft-style rasterization check out Culling techniques for rendering lots of cubes. Though with recent advances in graphics hardware, rendering Minecraft through raytracing may become the reality.
What you're looking for is called instancing. You could take a look at glDrawElementsInstanced and glDrawArraysInstanced for a couple of possibilities. Note that these were only added as core operations relatively recently (OGL 3.1), but have been available as extensions quite a while longer.
nVidia's OpenGL SDK has an example of instanced drawing in OpenGL.
First you really should be looking at OpenGL 3+ using GLSL. This has been the standard for quite some time. Second, most Minecraft-esque implementations use mesh creation on the CPU side. This technique involves looking at all of the block positions and creating a vertex buffer object that renders the triangles of all of the exposed faces. The VBO is only generated when the voxels change and is persisted between frames. An ideal implementation would combine coplanar faces of the same texture into larger faces.