Recherche avancée

Sur d’autres sites (17834)

• swscale/x86/yuv2rgb : add ssse3 yuv42{0,2}p -> gbrp unscaled colorspace converters

  6 août 2024, par Ramiro Polla

  swscale/x86/yuv2rgb : add ssse3 yuv420,2p -> gbrp unscaled colorspace converters
  Note : this implementation is limited to x86_64 due to general purpose
  
        register pressure.
  checkasm —bench on an Intel(R) Core(TM) i5-5300U CPU @ 2.30GHz :
  
  yuv420p_gbrp_8_c : 118.5
  
  yuv420p_gbrp_8_ssse3 : 93.3
  
  yuv420p_gbrp_128_c : 1068.3
  
  yuv420p_gbrp_128_ssse3 : 319.3
  
  yuv420p_gbrp_1080_c : 8841.8
  
  yuv420p_gbrp_1080_ssse3 : 2211.8
  
  yuv420p_gbrp_1920_c : 15903.8
  
  yuv420p_gbrp_1920_ssse3 : 3814.3
  
  yuv422p_gbrp_8_c : 144.8
  
  yuv422p_gbrp_8_ssse3 : 93.8
  
  yuv422p_gbrp_128_c : 1395.8
  
  yuv422p_gbrp_128_ssse3 : 313.0
  
  yuv422p_gbrp_1080_c : 11551.5
  
  yuv422p_gbrp_1080_ssse3 : 2240.8
  
  yuv422p_gbrp_1920_c : 20585.3
  
  yuv422p_gbrp_1920_ssse3 : 5249.5
  
  yuva420p_gbrp_8_c : 117.5
  
  yuva420p_gbrp_8_ssse3 : 92.0
  
  yuva420p_gbrp_128_c : 1593.0
  
  yuva420p_gbrp_128_ssse3 : 319.3
  
  yuva420p_gbrp_1080_c : 8694.5
  
  yuva420p_gbrp_1080_ssse3 : 2186.0
  
  yuva420p_gbrp_1920_c : 15946.5
  
  yuva420p_gbrp_1920_ssse3 : 3805.3
  

  • [DH] libswscale/x86/yuv2rgb.c
  • [DH] libswscale/x86/yuv_2_rgb.asm

• lavc/lpc : R-V V compute_autocorr

  8 décembre 2023, par Rémi Denis-Courmont

  lavc/lpc : R-V V compute_autocorr
  The loop iterates over the length of the vector, not the order. This is
  
  to avoid reloading the same data for each lag value. However this means
  
  the loop only works if the maximum order is no larger than VLENB.
  The loop is roughly equivalent to :
      for (size_t j = 0 ; j < lag ; j++)
  
          autoc[j] = 1. ;
      while (len > lag) 
  
          for (ptrdiff_t j = 0 ; j < lag ; j++)
  
              autoc[j] += data[j] * *data ;
  
          data++ ;
  
          len— ;
  
      
      while (len > 0) 
  
          for (ptrdiff_t j = 0 ; j < len ; j++)
  
              autoc[j] += data[j] * *data ;
  
          data++ ;
  
          len— ;
  
      
  Since register pressure is only at 50%, it should be possible to implement
  
  the same loop for order up to 2xVLENB. But this is left for future work.
  Performance numbers are all over the place from 1.25x to 4x speedups,
  
  but at least they are always noticeably better than nothing.
  

  • [DH] libavcodec/riscv/lpc_init.c
  • [DH] libavcodec/riscv/lpc_rvv.S

• FFmpeg - Can't redirect standard output stream

  4 janvier 2017, par Ryan Vancity Gosling

  I’m running FFmpeg throught my C# application to encode an video. To try an get the progress of the video, I’ve redirected the output but there seems to be no output. I tried the same with FFProbe and it worked perfectly. Just doesn’t work for FFmpeg.

   var FFmpeg = new Process
  
          {
  
              StartInfo = new ProcessStartInfo
  
              {
  
                  FileName = Directory.GetCurrentDirectory() + "\\FFmpeg\\ffmpeg.exe",
  
                  Arguments = CurrentArgs,
  
                  UseShellExecute = false,
  
                  RedirectStandardOutput = true,
  
                  CreateNoWindow = false
  
              }
  
          };
  
          FFmpeg.Start();
  
  
  
          string line;
  
          int i = 0;
  
          while (!FFmpeg.StandardOutput.EndOfStream)
  
          {
  
              i = i + 1;
  
          }

  And the weird thing is that I have the console window enabled on my WPF project and the output is shown although there isnt any code to write it to the console. And the loop is not iterating. The FFProbe output didn’t show up on console like this.

  Can someone help me. I need to get a percentage of the FFmpeg encoding process.