Jump to content

ToastyMozart

Member
  • Posts

    6
  • Joined

  • Last visited

Everything posted by ToastyMozart

  1. I was referring to the avisynth program he uses for the motion blur.
  2. Hmm, I wonder if you could rig up some pseudo-multithreading by running three instances of the program with higher priority, and having each work on a third of the footage (leaving one core free for running the OS and such, of course).
  3. Eugh, 180-30 any day. Higher framerates are really important for playing games due to things like higher responsiveness and the lack of real films' blur, but watching video is generally fine with a lower framerate, so long as all the motion in each frame's time span is captured. On-the-fly interpolation (or interpolation to artificially increase framerate in general) doesn't look as good as Ross' method.
  4. I figure since it renders in slowmo for him, then that the card isn't capable of rendering 180Hz at 1 second per second, meaning it's running as hard as it can. Trying to composite the frames on it would just end up slowing down the video recording process. Though if the sub-180 framerate is just a Source engine limitation, then yeah, GPU acceleration might be an option. I just figured it made more sense to farm the blending out to the component that's barely being used.
  5. Oh, I meant that I hope it doesn't use the GPU to help with it, since the GPU's busy rendering the game, while the CPU is mostly just sitting there. Unless you can use the CUDA processing independently of the rendering components (I'm not particularly well versed in the details of CUDA).
  6. Wow, the internet really knows how to put its money where it's mouth is! Sorry, I can't really donate cash right now (thanks, tuition) . I've got a couple thousand spare Doge floating around, if you've got a wallet for it. About how long does your method of motion blur take to process? (Assuming it's not GPU accelerated.) If it's capable of compositing it into 30Hz video at the same speed/faster than the demo can render the raw frames, it might be possible to avoid having to store the absurdly gigantuan raw footage in the first place. You might be able to to add some extra memory and set up a RAMdisk,* have the demo recorder save in 1-2GB chunks to the virtual disk, and have your script process them as they come in, and save the finished files to storage. *No worries about write speed While I'm definitely behind the idea of getting a proper RAID card for the setup, RAIDs are limited by the size of the smallest drive, so using all 4 would only give him around 1200GB total, since his boot drive is just 300GB.
×
×
  • Create New...

This website uses cookies, as do most websites since the 90s. By using this site, you consent to cookies. We have to say this or we get in trouble. Learn more.