Recherche avancée

Sur d’autres sites (12802)

• Windows 2003 using php popen "start /b" because of executing ffmpeg.exe

  1er juillet 2013, par Oh Seung Kwon

  I have some problem.

  There is a php code that convert audio(wav) to mp3 file with using ffmpeg.exe.

  Here is some code.

  $cmd = "./inc/ffmpeg.exe -i ".$file_name." -acodec mp3 -y -ac 1 -ab 96k ".$mp3_file_name;
  
  
  
  echo $cmd;
  
  echo "Windows";
  
  $handle = popen("start /B ".$cmd, "r");
  
  while(!feof($handle)) {
  
      $read = fread($handle, 2096);
  
      echo $read;
  
  }
  
  pclose($handle);

  Problem is when I execute this code, ffmpeg.exe process isn't terminated. And not gonna die when I stop process with using Windows task manager.

  Do you have a solution for this situation ?

• FFMPEG - Palettegen for animations is not working

  18 juillet 2018, par deanhodges

  I have a script taken from the correct answer at https://superuser.com/questions/1002562/convert-multiple-images-to-a-gif-with-cross-dissolve, but I am getting no results.

  Even copying the code word for word does not work for me.

  However, if I run $ffmpeg = shell_exec("ffmpeg -i images/image001.jpg -vf palettegen palette_test.png"); on a single image, it works fine. But because I need it for an animation, I need to create a pallete for all images.

  Unless of course, on each image in the GIF sequence I can load a new palette ?

  $q = "ffmpeg -f image2 -framerate 0.5 -i ../view/client/files/".$dir."/%*.jpg -i palettePro.png -lavfi paletteuse -y .." . $save_path . '/' . $filename . " -report";

• how the packuswb instruction works ? (in low level bit operation)

  17 avril 2019, par MaikonNascimento

  Studying ffmpeg convertion from yuv to rgb , I came across the equation being implemented in assembly, yuv2rgb_template.c located in ffmpeg/libswscale/x86.

  I want to know how the instruction packuswb works ? Google says it :

  Converts 4 signed word integers from mm and 4 signed word integers from mm/m64 into 8 unsigned byte integers in mm using unsigned saturation.

      DEST[7:0] ← SaturateSignedWordToUnsignedByte DEST[15:0];
  
      DEST[15:8] ← SaturateSignedWordToUnsignedByte DEST[31:16];
  
      DEST[23:16] ← SaturateSignedWordToUnsignedByte DEST[47:32];
  
      DEST[31:24] ← SaturateSignedWordToUnsignedByte DEST[63:48];
  
      DEST[39:32] ← SaturateSignedWordToUnsignedByte SRC[15:0];
  
      DEST[47:40] ← SaturateSignedWordToUnsignedByte SRC[31:16];
  
      DEST[55:48] ← SaturateSignedWordToUnsignedByte SRC[47:32];
  
      DEST[63:56] ← SaturateSignedWordToUnsignedByte SRC[63:48];

  Since I dont know Assembly x86, it is even hard to simulate it or debug the current code.

  I want to know how they fit 16 bits in 8 bits ?