Some of the paths were wrong in the enviromental paths etc…īasically unless you all are experts you cant get it installed. Now people have helped me in the past by telling me to get a few other apps like the microsoft SDK or something…i really forget the exact name but the end result was some of the DX10 examples would work but the DX9 examples wouldnt. I dloaded that guys raytracing app and it worked fine on my cpu and on the gpu. Im running windows vista premium 32bit (FULLY UPDATED) on an NF4 motherboard with the latest DX10 redist (the ones that are released bi-monthly) and i believe my current video driver is 179.13 from laptop2go. If you have searched any of my posts in the past i am the guy who could not get cuda to install or work correctly by the instructions provided in the SDK and TOOLKIT. Then you would use the GPU accellerated compression cuda app to compress it.Īlthough it would be nicer to highlight all the files you would want to compress and then you would right click and select “add to winGPU archive” :) ISO image of all the files you would want to compress. I would imagine the proper way to do it would be to create an. I believe that process is called “streaming” which someone previously mentioned. You asked about it being all stored in texture memory area…badaboom does not do this…as far as i know it does not load a DL DVD (what is it 9.4GB) into the texture memory of a vcard. There are different compression algorythems in winzip, winrar 7zip…that can increase the cpu usage (for HIGH compression) and increae the disk I/O for (light compression)įor a data compression cuda app you would just have to make sure it is more GPU bound (very high compression) so it would not be limited by the disk I/O I was thinking to do it exactly like badaboom does it for transcoding. Right now, it’s a bit hard to help, as you have asked a very broad question with thousands of possible answers. Read up on existing parallel compression algorithms and start implementing one in CUDA yourself! When you’ve got specific questions about how best to get a particular data structure into the GPUs memory, we can certainly help you out. And I haven’t seen any previous threads discussing zip type compression algorithms in CUDA.
I believe someone pointed out on the first page that you could split a big stream into many sections and compress them all separately in parallel which then digressed into texture caches and then finally got on to Larrabee… Still, no one here is going to do all your homework for you. Assuming you can do the compression on the GPU, the PCIe transfers are likely to be a big bottleneck.Īre you compressing many independent streams at once, or just one big stream? Thousands of streams in parallel could be done in CUDA without too much trouble. But I’ll start with a few questions to get you thinking about where some bottlenecks might be.Ĭorrect me if I’m wrong, but aren’t file compression routines basically disk I/O bound? What gain would you hope to see using CUDA? Or are you thinking of speeding compression on data that is already in memory? In this case, you are going to be limited by the PCIe bandwidth: copying the data out to the GPU and back again at ~4 GiB/s. I’m no expert in file compression myself, so I can’t offer you any great ideas on putting it on the GPU off the top of my head. Unfortunately, tmurray is the closest thing we have to a moderator right now, and he was participating in the discussion on Larrabee External Media