If you need custom filters you can always write them and add to GPUImage project. Just grab some easy filter like GPUImageRGBFilter or GPUImageLineGenerator and experiment. You can also modify OpenGL calls directly to inject your custom effects in front of a movie. Take a look at CubeExample. 1 solved GPU Image Filter [closed]
This might be more of an extended comment, bet hear me out … As pointed out in the comments CUDA Samples are not meant as performance measuring tools. The second benchmark you provided does not actually use tensor cores, but just a normal instruction executed on FP32 or FP64 cores. for(int i=0; i<compute_iterations; i++){ tmps[j] … Read more
Your code is correctly printing the value of c as 9. You need to clarify on the environment you are running this code. 2 solved (CUDA C) Why is it not printing out the value copied from device memory?
Short answer, Yes Running this type of data will out-perform CPU. Long answer, if its is ( stream-able), what I mean is that if you can have your data structured in such a way that its iterable and can be emitted one by one to some consuming end, then yes. But about the programing language … Read more