I'm skeptical. The x264 guys looked at CUDA and found it wasn't worth it. The badaboom app required a $600 video card to encode a movie as fast as a Core i7 940. And then the quality output from badaboom wasn't even nearly as good as x264 output, and this test was back in November, and there have been some x264 speed improvements since then. Show me the benchmarks!
It doesn't make sense that the quality would be worse. It isn't as if the graphics card would multiply bits differently than a CPU. If they were doing the same operation on both the Core i7 CPU with a non-CUDA application, and a "lesser" CPU combined with an NVIDIA card and a CUDA enabled application, the resulting video would be the same, all that should differ would be the time it took to get there.
It sounds like your "badaboom" test may not be all you think it is.
They are using a Windows operating system.
They didn't even need an updated operating system like Snow Leopard.
So what's the big deal about Snow Leopard if Nero Movie It can use CUDA with Windows Vista?
You've been able to code against CUDA on Windows for some time now. You've probably been able to do it on OS/X as well. But its hard, and it only works if you have an NVIDIA card installed in your system. Here's a test, do your parents know who manufactures the video card in their current system? Most people don't know. NERO is aiming this product at the geek who probably built their own PC from scratch. Also, it is quite possible that wouldn't benefit from more than 1 NVIDIA card installed in the system - although I could be wrong here, I haven't investigated the app that thoroughly.
Apple is aiming for OpenCL to fix some of those issues. 1) It should work (or at least show some positive effects) on any video card that supports OpenCL. That will include NVIDIA and ATI at least, and likely most or all of the other graphics cards on the market in a short time. 2) Apps coded against OpenCL suffer little performance penalty on systems without high-end GPUs, but scale up on systems with one (or multiple) high end cards without need for the application developer to do anything special.
In short, developers will need to code their apps using the OpenCL standard (which I have investigated, and isn't ridiculously hard), but once that is done, the app should benefit from all possible compute resources available on the machine. Right now, that is mostly limited to CPUs and graphics cards, but in the future, I would not be surprised to dedicated encoder-decoder chips shipped with even low-end hardware. OpenCL apps would get access to that hardware for free.
