Comments on: Quick Gaussian Filtering http://www.realtimerendering.com/blog/quick-gaussian-filtering/ Tracking the latest developments in interactive rendering techniques Mon, 17 Jun 2013 03:17:13 +0000 hourly 1 http://wordpress.org/?v=3.4.1 By: venzon http://www.realtimerendering.com/blog/quick-gaussian-filtering/comment-page-1/#comment-1745 venzon Thu, 23 Sep 2010 23:26:01 +0000 http://www.realtimerendering.com/blog/?p=1718#comment-1745 The mipmap chain stuff is very useful for very-large-kernel gaussian blurs (for example, you can get results identical to applying a 40x40 gaussian kernel by recursively applying a 5x5 gaussian blur kernel over 4 mip chains). I'm not sure of any other way to do big kernels efficiently on DX9 hardware. The mipmap chain stuff is very useful for very-large-kernel gaussian blurs (for example, you can get results identical to applying a 40×40 gaussian kernel by recursively applying a 5×5 gaussian blur kernel over 4 mip chains). I’m not sure of any other way to do big kernels efficiently on DX9 hardware.

]]> By: castano http://www.realtimerendering.com/blog/quick-gaussian-filtering/comment-page-1/#comment-1743 castano Wed, 22 Sep 2010 07:05:05 +0000 http://www.realtimerendering.com/blog/?p=1718#comment-1743 No, Philip is just being too modest. I suppose he's referring to the mipmap chain stuff which is not really necessary for a gaussian filter, but it's nevertheless useful for some effects (Halo 2 for example uses a similar approach for their bloom filter). In any case, that's completely orthogonal to the main points (pascal triangle, separability and bilinear interpolation). No, Philip is just being too modest. I suppose he’s referring to the mipmap chain stuff which is not really necessary for a gaussian filter, but it’s nevertheless useful for some effects (Halo 2 for example uses a similar approach for their bloom filter). In any case, that’s completely orthogonal to the main points (pascal triangle, separability and bilinear interpolation).

]]> By: Eric http://www.realtimerendering.com/blog/quick-gaussian-filtering/comment-page-1/#comment-1741 Eric Tue, 21 Sep 2010 20:03:22 +0000 http://www.realtimerendering.com/blog/?p=1718#comment-1741 > A similar article that predates these: http://prideout.net/archive/bloom/ Interestingly enough, the author of that article has the first comment on Daniel's blog post, warning people away from it (which begs the question, why not just fix that article?): Sweet. This is similar to my old bloom tutorial, which I wrote when I was young and stupid: http://prideout.net/archive/bloom/ Some aspects of my tutorial are an embarrassment to me now, so I hope folks will encounter your article before mine. Keep up the good work Daniel! > A similar article that predates these: http://prideout.net/archive/bloom/

Interestingly enough, the author of that article has the first comment on Daniel’s blog post, warning people away from it (which begs the question, why not just fix that article?):

Sweet. This is similar to my old bloom tutorial, which I wrote when I was young and stupid: http://prideout.net/archive/bloom/
Some aspects of my tutorial are an embarrassment to me now, so I hope folks will encounter your article before mine. Keep up the good work Daniel!

]]> By: castano http://www.realtimerendering.com/blog/quick-gaussian-filtering/comment-page-1/#comment-1740 castano Tue, 21 Sep 2010 19:12:31 +0000 http://www.realtimerendering.com/blog/?p=1718#comment-1740 A similar article that predates these: http://prideout.net/archive/bloom/ A similar article that predates these: http://prideout.net/archive/bloom/

]]> By: imbusy http://www.realtimerendering.com/blog/quick-gaussian-filtering/comment-page-1/#comment-1739 imbusy Tue, 21 Sep 2010 08:26:03 +0000 http://www.realtimerendering.com/blog/?p=1718#comment-1739 And now there are tricks for saving bandwidth and storing texture access results in thread local storage with OpenCL/DirectX Compute/CUDA. And now there are tricks for saving bandwidth and storing texture access results in thread local storage with OpenCL/DirectX Compute/CUDA.

]]> By: Ivan-Assen http://www.realtimerendering.com/blog/quick-gaussian-filtering/comment-page-1/#comment-1738 Ivan-Assen Tue, 21 Sep 2010 06:11:11 +0000 http://www.realtimerendering.com/blog/?p=1718#comment-1738 Keep in mind that on a certain platform that can't directly texture from its render targets, a non-separable 9x9 Gaussians is faster than a separable 9+9 version with a resolve (transfer from RT to texture) between the H and V passes. Whether it's faster to calculate offsets in the VS and pass down via interpolators to the PS, or directly calculate there should be benchmarked on your particular hardware - some GPUs really hate additional vertex attributes. Keep in mind that on a certain platform that can’t directly texture from its render targets, a non-separable 9×9 Gaussians is faster than a separable 9+9 version with a resolve (transfer from RT to texture) between the H and V passes.

Whether it’s faster to calculate offsets in the VS and pass down via interpolators to the PS, or directly calculate there should be benchmarked on your particular hardware – some GPUs really hate additional vertex attributes.

]]>