Recherche avancée

Sur d’autres sites (13619)

• lavc/aacpsdsp : use LMUL=2 and amortise strides

  19 novembre 2023, par Rémi Denis-Courmont

  lavc/aacpsdsp : use LMUL=2 and amortise strides
  The input is laid out in 16 segments, of which 13 actually need to be
  
  loaded. There are no really efficient ways to deal with this :
  
  1) If we load 8 segments wit unit stride, then narrow to 16 segments with
  
     right shifts, we can only get one half-size vector per segment, or just 2
  
     elements per vector (EMUL=1/2) - at least with 128-bit vectors.
  
     This ends up unsurprisingly about as fas as the C code.
  
  2) The current approach is to load with strides. We keep that approach,
  
     but improve it using three 4-segmented loads instead of 12 single-segment
  
     loads. This divides the number of distinct loaded addresses by 4.
  
  3) A potential third approach would be to avoid segmentation altogether
  
     and splat the scalar coefficient into vectors. Then we can use a
  
     unit-stride and maximum EMUL. But the downside then is that we have to
  
     multiply the 3 (of 16) unused segments with zero as part of the
  
     multiply-accumulate operations.
  In addition, we also reuse vectors mid-loop so as to increase the EMUL
  
  from 1 to 2, which also improves performance a little bit.
  Oeverall the gains are quite small with the device under test, as it does
  
  not deal with segmented loads very well. But at least the code is tidier,
  
  and should enjoy bigger speed-ups on better hardware implementation.
  Before :
  
  ps_hybrid_analysis_c :       1819.2
  
  ps_hybrid_analysis_rvv_f32 : 1037.0 (before)
  
  ps_hybrid_analysis_rvv_f32 :  990.0 (after)
  

  • [DH] libavcodec/riscv/aacpsdsp_rvv.S

• Pipe input in to ffmpeg stdin

  3 juin 2021, par AbstractDissonance

  I am trying to use ffmpeg to decode audio data. While it works to load from a file, I would like to avoid using files because to do so, means I would have to use a temporary. Instead, I'd like to pipe in the data(which I've previously loaded) using stdin.

  



  Is this possible ?

  



  e.g.,

  



  

  1. Manually load mp3 file

  


  2. Pipe it in to the ffmpeg spawned process

  


  3. Get raw output

  


  



  (it should work with ffprobe and ffplay also)

  


• Ffmpeg "no directory found" when passing in a path to the mp3 file

  21 février 2020, par Eugene Levinson

  How my code should work :

  Join the voice channel with the user who sent the command

  Download the video using the link

  Save it as random 16 digit number .mp3

  Then pass the path to the FFmpeg player to play

  My code :

  The random 16 digit number

  def get_digits(amount):
  
      st = ""
  
      for i in range(amount):
  
          n = random.randint(0,9)
  
          st = st + str(n)
  
  
  
      return int(st)

  Downloading the file

  def get_path(url):
  
  
  
      #checking if the directory exists
  
      os.makedirs('Music', exist_ok=True)
  
  
  
      title = YouTube(url).streams.get_highest_resolution().title
  
      current_directory = pathlib.Path(__file__).parent.absolute()
  
  
  
      print(str(current_directory))
  
  
  
      #name for the music
  
      name = str(get_digits(16))
  
  
  
      YouTube(url).streams.filter(only_audio=True).order_by("bitrate").desc().first().download("Music",name )
  
      return str(str(current_directory) + "/Music/" + name + ".mp3")

  This gets called on command play

  #function to connect to a voice chat
  
  async def join_auth(ctx):
  
      try:
  
          channel = ctx.author.voice.channel
  
          vc = await channel.connect()
  
          return vc
  
      except Exception as e:
  
          logg("Exception occured when joining a voice channel: " + str(e),"error",str(ctx.guild.name), str(ctx.guild.id))

  The error I get :
  C:\Users\Eugene\Desktop\Discord bot/Music/4343941300524002.mp3: No such file or directory
  But the directory C:\Users\Eugene\Desktop\Discord bot\Music exists and it does contain the 4343941300524002.mp3 file. Does anyone know why do I get the error ?