My question concerns the most efficient way of performing geometric image transformations on the GPU. The goal is essentially to remove lens distortion from aquired images in real time. I can think of several ways to do it, e.g. as a CUDA kernel (which would be preferable) doing an inverse transform lookup + interpolation, or the same in an OpenGL shader, or rendering a forward transformed mesh with the image texture mapped to it. It seems to me the last option could be the fastest because the mesh can be subsampled, i.e. not every pixel offset needs to be stored but can be interpolated in the vertex shader. Also the graphics pipeline really should be optimized for this. However, the rest of the image processing is probably going to be done with CUDA. If I want to use the OpenGL pipeline, do I need to start an OpenGL context and bring up a window to do the rendering, or can this be achieved anyway through the CUDA/OpenGL interop somehow? The aim is not to display the image, the processing will take place on a server, potentially with no display attached. I've heard this could crash OpenGL if bringing up a window.
I'm quite new to GPU programming, any insights would be much appreciated.
Using the forward transformed mesh method is the more flexible and easier one to implement. However performance wise there's no big difference, as the effective limit you're running into is memory bandwidth, and the amount of memory bandwidth consumed does only depend on the size of your input image. If it's a fragment shader, fed by vertices or a CUDA texture access that's causing the transfer doesn't matter.
If I want to use the OpenGL pipeline, do I need to start an OpenGL context and bring up a window to do the rendering,
On Windows: Yes, but the window can be an invisible one.
On GLX/X11 you need an X server running, but you can use a PBuffer instead of a window to get a OpenGL context.
In either case use a Framebuffer Object as the actual drawing destination. PBuffers may corrupt their primary framebuffer contents at any time. A Framebuffer Object is safe.
or can this be achieved anyway through the CUDA/OpenGL interop somehow?
No, because CUDA/OpenGL interop is for making OpenGL and CUDA interoperate, not make OpenGL work from CUDA. CUDA/OpenGL Interop helps you with the part you mentioned here:
However, the rest of the image processing is probably going to be done with CUDA.
BTW; maybe OpenGL Compute Shaders (available since OpenGL-4.3) would work for you as well.
I've heard this could crash OpenGL if bringing up a window.
OpenGL actually has no say in those things. It's just a API for drawing stuff on a canvas (canvas = window or PBuffer or Framebuffer Object), but it doesn't deal with actually getting a canvas on the scaffolding, so to speak.
Technically OpenGL doesn't care if there's a window or not. It's the graphics system on which the OpenGL context is created. And unfortunately none of the currently existing GPU graphics systems supports true headless operation. NVidia's latest Linux drivers may allow for some crude hacks to setup a truly headless system, but I never tried that, so far.
Related
What does gl.glTexImage2D do? The docs say it "uploads texture data". But does this mean the whole image is in GPU memory? I'd like to use one large image file for texture mapping. Further: can I simply use a VBO for uv and position coordinates to draw the texture?
Right, I am using words the wrong way here. What I meant was carrying a 2D array of UV coordinates and a 2D array of model to subsample a larger PNG image (in texture memory) onto individual tile models. My confusion here lies in not knowing how fast these fetches can take. Lets say I have a 5000x5000 pixel image. I load it as a texture. Then I create my own algorithm for fetching portions of it to draw. Where do I save myself the bandwidth for drawing these tiles? If I implement an LOD algorithm to determine which tiles are close, which are far and which are out of the camera frustum how do manage each these tiles in memory? Loaded question I know but I am struggling to find the best implementation to get started. I am developing for mobile devices with OpenGL ES 2.0.
What exactly happens when you call glTexImage2D() is system dependent, and there's no way for you to know, unless you have developer tools that allow you to track GPU and memory usage.
The only thing guaranteed is that the data you pass to the call has been consumed by the time the call returns (since the API definition allows you to modify/free the data after the call), and that the data is accessible to the GPU when it's used for rendering. Between that, anything is fair game. Keep in mind that OpenGL is a very asynchronous API. When you make API calls, the corresponding work is mostly queued up for later execution by the GPU, and is generally not completed by the time the calls return. This can include calls for uploading data.
Also, not all GPUs have "GPU memory". In fact, if you look at them by quantity, very few of them do. Mobile GPUs have caches, but mostly not VRAM in the sense of traditional discrete GPUs. How VRAM and caches are managed is highly system dependent.
With all the caveats above, and picturing a GPU that has VRAM: While it's possible that they can load the data into VRAM in the glTexImage2D() call, I would be surprised if that was commonly done. It just wouldn't make much sense to me. When a texture is loaded, you have no idea how soon it will be used for rendering. Since you don't know if all textures will fit in VRAM (and they often will not), you might have to evict it from VRAM before it was ever used. Which would obviously be very wasteful. As a general strategy, I think it will be much more efficient to load the texture data into VRAM only when you have a draw call that uses it.
Things would be somewhat different if the driver could be very confident that all texture data will fit in VRAM. But with OpenGL, there's really no reasonable way to know this ahead of time. And things get even more complicated since at least on desktop computers, you can have multiple applications running at the same time, while VRAM is a shared resource.
You are correct.
glteximage2d is the function that actually moves the texture data across to the gpu.
you will need to create the texture object first using glGenTextures() and then bind it using glBindTexture().
there is a good example of this process in the opengl redbook
example
you can then use this texture with a VBO. There are many ways to accomplish this, but interleaving your vertex coordinates, texture coordinates, and vertex normals and then telling the GPU how to unpack them with several calls to glVertexAttribPointer is the best bet as far as performance.
you are on the right track with VBOs, the old fixed pipeline GL stuff is depricated so you should just learn VBO from the outset.
this book is not 100% up to date, but it is complete and free and should serve as a great place to start learning VBO Open GL Book
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 kind of stuck on the logic behind an SDL2 texture. To me, they are pointless since you cannot draw to them.
In my program, I have several surfaces (or what were surfaces before I switched to SDL2) that I just blitted together to form layers. Now, it seems, I have to create several renderers and textures to create the same effect since SDL_RenderCopy takes a texture pointer.
Not only that, but all renderers have to come from a window, which I understand, but still fouls me up a bit more.
This all seems extremely bulky and slow. Am I missing something? Is there a way to draw directly to a texture? What are the point of textures, and am I safe to have multiple (if not hundreds) of renderers in place of what were surfaces?
SDL_Texture objects are stored as close as possible to video card memory and therefore can easily be accelerated by your GPU. Resizing, alpha blending, anti-aliasing and almost any compute-heavy operation can harshly be affected by this performance boost. If your program needs to run a per-pixel logic on your textures, you are encouraged to convert your textures into surfaces temporarily. Achieving a workaround with streaming textures is also possible.
Edit:
Since this answer recieves quite the attention, I'd like to elaborate my suggestion.
If you prefer to use Texture -> Surface -> Texture workflow to apply your per-pixel operation, make sure you cache your final texture unless you need to recalculate it on every render cycle. Textures in this solution are created with SDL_TEXTUREACCESS_STATIC flag.
Streaming textures (creation flag is SDL_TEXTUREACCESS_STREAMING) are encouraged for use cases where source of the pixel data is network, a device, a frameserver or some other source that is beyond SDL applications' full reach and when it is apparent that caching frames from source is inefficient or would not work.
It is possible to render on top of textures if they are created with SDL_TEXTUREACCESS_TARGET flag. This limits the source of the draw operation to other textures although this might already be what you required in the first place. "Textures as render targets" is one of the newest and least widely supported feature of SDL2.
Nerd info for curious readers:
Due to the nature of SDL implementation, the first two methods depend on application level read and copy operations, though they are optimized for suggested scenarios and fast enough for realtime applications.
Copying data from application level is almost always slow when compared to post-processing on GPU. If your requirements are more strict than what SDL can provide and your logic does not depend on some outer pixel data source, it would be sensible to allocate raw OpenGL textures painted from you SDL surfaces and apply shaders (GPU logic) to them.
Shaders are written in GLSL, a language which compiles into GPU assembly. Hardware/GPU Acceleration actually refers to code parallelized on GPU cores and using shaders is the prefered way to achieve that for rendering purposes.
Attention! Using raw OpenGL textures and shaders in conjunction with SDL rendering functions and structures might cause some unexpected conflicts or loss of flexibility provided by the library.
TLDR;
It is faster to render and operate on textures than surfaces although modifying them can sometimes be cumborsome.
Through creating a SDL2 Texture as a STREAMING type, one can lock and unlock the entire texture or just an area of pixels to perform direct pixel operations. One must create prior a SDL2 Surface, and link with lock-unlock as follows:
SDL_Surface surface = SDL_CreateSurface(..);
SDL_LockTexture(texture, &rect, &surface->pixels, &surface->pitch);
// paint into surface pixels
SDL_UnlockTexture(texture);
The key is, if you draw to texture of larger size, and the drawing is incremental ( e.g. data graph in real time ) be sure to only lock and unlock the actual area to update. Otherwise the operations will be slow, with heavy memory copying.
I have experienced reasonable performance and the usage model is not too difficult to understand.
In SDL2 it is possible to render off-screen / render directly to a texture. The function to use is:
int SDL_SetRenderTarget(SDL_Renderer *renderer, SDL_Texture *texture);
This only works if the renderer enables SDL_RENDERER_TARGETTEXTURE.
Can anyone pls tell me how to use hardware memory to create textures in OpenGL ? Currently I'm running my game in window mode, do I need to switch to fullscreen to get the use of hardware ?
If I can create textures in hardware, is there a limit for no of textures (other than the hardware memory) ? and then how can I cache my textures into hardware ? Thanks.
This should be covered by almost all texture tutorials for OpenGL. For example here, here and here.
For every texture you first need a texture name. A texture name is like a unique index for a single texture. Every name points to a texture object that can have its own parameters, data, etc. glGenTextures is used to get new names. I don't know if there is any limit besides the uint range (2^32). If there is then you will probably get 0 for all new texture names (and a gl error).
The next step is to bind your texture (see glBindTexture). After that all operations that use or affect textures will use the texture specified by the texture name you used as parameter for glBindTexture. You can now set parameters for the texture (glTexParameter) and upload the texture data with glTexImage2D (for 2D textures). After calling glTexImage you can also free the system memory with your texture data.
For static textures all this has to be done only once. If you want to use the texture you just need to bind it again and enable texturing (glEnable(GL_TEXTURE_2D)).
The size (width/height) for a single texture is limited by GL_MAX_TEXTURE_SIZE. This is normally 4096, 8192 or 16384. It is also limited by the available graphics memory because it has to fit into it together with some other resources like the framebuffer or vertex buffers. All textures together can be bigger then the available memory but then they will be swapped.
In most cases the graphics driver should decide which textures are stored in system memory and which in graphics memory. You can however give certain textures a higher priority with either glPrioritizeTextures or with glTexParameter.
Edit:
I wouldn't worry too much about where textures are stored because the driver normally does a very good job with that. Textures that are used often are also more likely to be stored in graphics memory. If you set a priority that's just a "hint" for the driver on how important it is for the texture to stay on the graphics card. It's also possible the the priority is completely ignored. You can also check where textures currently are with glAreTexturesResident.
Usually when you talk about generating a texture on the GPU, you're not actually creating texture images and applying them like normal textures. The simpler and more common approach is to use Fragment shaders to procedurally calculate the colors of for each pixel in real time from scratch for every single frame.
The canonical example for this is to generate a Mandelbrot pattern on the surface of an object, say a teapot. The teapot is rendered with its polygons and texture coordinates by the application. At some stage of the rendering pipeline every pixel of the teapot passes through the fragment shader which is a small program sent to the GPU by the application. The fragment shader reads the 2D texture coordinates and calculates the Mandelbrot set color of the 2D coordinates and applies it to the pixel.
Fullscreen mode has nothing to do with it. You can use shaders and generate textures even if you're in window mode. As I mentioned, the textures you create never actually occupy space in the texture memory, they are created on the fly. One could probably think of a way to capture and cache the generated texture but this can be somewhat complex and require multiple rendering passes.
You can learn more about it if you look up "GLSL" in google - the OpenGL shading language.
This somewhat dated tutorial shows how to create a simple fragment shader which draws the Mandelbrot set (page 4).
If you can get your hands on the book "OpenGL Shading Language, 2nd Edition", you'll find it contains a number of simple examples on generating sky, fire and wood textures with the help of an external 3D Perlin noise texture from the application.
To create a texture on GPU look into "render to texture" tutorials. There are two common methods: Binding a PBuffer context as texture, or using Frame Buffer Objects. PBuffer render to textures are the older method, and have the wider support. Frame Buffer Objects are easier to use.
Also you don't have to switch to "fullscreen" mode for OpenGL to be HW accelerated. In fact OpenGL doesn't know about windows at all. A fullscreen OpenGL window is just that: A toplvel window on top of all other windows with no decorations and the input focus grabed. Some drivers bypass window masking and clipping code, and employ a simpler, faster buffer swap method if the window with the active OpenGL context covers the whole screen, thus gaining a little performance, but with current hard- and software the effect is very small compared to other influences.
I've read that FBOs can be used for fast image manipulation using the OpenGL drawing actions. Does anyone know the basics of how to do this? or has some very simple example code illustrating it?
Before you can use FBOs for image manipulation you need to know how to handle OpenGL, as a FBO can simply be used as a render target (output buffer for rendering operations). Once you're fluent with OpenGL and probably know how to do shader programming, you can do virtually everything with images in an FBO, and do it extremely fast.
A simpler approach might be to employ CUDA (NVidia) or Stream Computing (ATI) to harness a GPU's power for image manipulation, because these APIs are much closer to regular array-based C++ programming. Image manipulation may be somewhat slower that way than with OpenGL, but still way faster than with traditional CPU driven code.
Framebuffer Objects (FBO) are just a basic tool that cannot be used to manipulate images directly. If you know how to render your image manipulations in OpenGL to the screen, you can then use FBOs to render them off-screen. So they are in fact useful for this task, since you are not limited by the resolution of your screen and don't have to distract the user with thousands of flashing images. However, the manipulation itself happens in OpenGL, probably in the fragment shader.
Visit to the OpenGL forum to get some advice how to start with OpenGL basics. They also have quite a few links to sample code.