R3 sound quality in Scotland

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts
  • Ein Heldenleben
    Full Member
    • Apr 2014
    • 6580

    #16
    Originally posted by Dave2002 View Post
    One thing I don't know is whether the on demand replays are better or worse than the live output. I listened to the Strauss Tod und Verklarung twice tonight, once as it happened, and then later. If the data was exactly the same then it should have sounded very similar, but it's possible that the on demand servers do something to the data stream which reduces the quality somehow. I have a very slight suspicion that there is a quality loss - but it's very hard to be sure. After all I don't know the particular form of on demand data management which the BBC is using for R3 or any of its output. There's quite a range of possibilities.
    I would be surprised if there’s any difference audio wise because the date requirements are relatively small . The big thing with Radio 3 as Jayne points out is the high stream rate. I’m guess one way of checking would be to record a live stream and then compare file sizes with a recording on iplayer though your recording may introduce extra data increasing the file size. That quality difference is not true however of video which is heavily processed and compressed (for want of a better word ) on iplayer and other stream services. The best video quality around is live sport and outside broadcasts. Most other stuff is really mucked around with - ruined in my view. When you see uncompressed HD it’s so much better than most streamed or broadcast.

    Comment

    • Bryn
      Banned
      • Mar 2007
      • 24688

      #17
      Originally posted by Ein Heldenleben View Post
      I would be surprised if there’s any difference audio wise because the date requirements are relatively small . The big thing with Radio 3 as Jayne points out is the high stream rate. I’m guess one way of checking would be to record a live stream and then compare file sizes with a recording on iplayer though your recording may introduce extra data increasing the file size. That quality difference is not true however of video which is heavily processed and compressed (for want of a better word ) on iplayer and other stream services. The best video quality around is live sport and outside broadcasts. Most other stuff is really mucked around with - ruined in my view. When you see uncompressed HD it’s so much better than most streamed or broadcast.
      Actually, I think you will find that all BBC Sounds Radio channels are now available at 320 kbps AAC-LC. The difference with Radio 3 is that it is the only one not to use dynamic compression in addition to the data compression.

      Comment

      • Ein Heldenleben
        Full Member
        • Apr 2014
        • 6580

        #18
        Originally posted by Bryn View Post
        Actually, I think you will find that all BBC Sounds Radio channels are now available at 320 kbps AAC-LC. The difference with Radio 3 is that it is the only one not to use dynamic compression in addition to the data compression.
        Didn’t know that - not before time. Though with most Rock music I’d be hard pressed to notice the difference as it’s all so compressed these days.

        Comment

        • Dave2002
          Full Member
          • Dec 2010
          • 17958

          #19
          Originally posted by Ein Heldenleben View Post
          I would be surprised if there’s any difference audio wise because the data requirements are relatively small .
          I'm just not sure that that is correct. It generally ought to be lot easier to distribute wide bandwidth data to an audience all tuned in at the same time, as techniques such as multicasting for live broadcasting, and data carousels can be used. However you may be correct, as nowadays it seems that I can download Gbytes of data in a very short time - and I don't know how that works. I live very close to a fibre hub, and there may be parts of the UK where having a 10 Gbyte download taking less than 90 seconds is still a significant problem. Clearly if very large amounts of data can be downloaded quickly, then one model would be to have all the data downloaded in one fell swoop, and then stored locally, and the player in the client machine then just plays it back and handles all interactions. However that kind of model would have what nowadays would be considered unacceptable latency, and wouldn't work for things like news items or sport. It's OK to watch a game on TV which is a minute or so behind actual time - provided you can't hear the groans and/or cheers from neighbours who just may have lower latency due to shorter network delays.

          Some on demand data models, particularly for multimedia distribution, do have strategies for compromising audio or video quality in order to keep the the streams and controls still working. I just don't know what strategies the BBC is using - and indeed it may be a complex hybrid strategy depending on the number of logged in users. Some of the more effective strategies scale up a lot better if there are many users all wanting essentially the same data at the same time, while strategies for very many users wanting different data simultaneously are different.

          Comment

          • Ein Heldenleben
            Full Member
            • Apr 2014
            • 6580

            #20
            Originally posted by Dave2002 View Post
            I'm just not sure that that is correct. It generally ought to be lot easier to distribute wide bandwidth data to an audience all tuned in at the same time, as techniques such as multicasting for live broadcasting, and data carousels can be used. However you may be correct, as nowadays it seems that I can download Gbytes of data in a very short time - and I don't know how that works. I live very close to a fibre hub, and there may be parts of the UK where having a 10 Gbyte download taking less than 90 seconds is still a significant problem. Clearly if very large amounts of data can be downloaded quickly, then one model would be to have all the data downloaded in one fell swoop, and then stored locally, and the player in the client machine then just plays it back and handles all interactions. However that kind of model would have what nowadays would be considered unacceptable latency, and wouldn't work for things like news items or sport. It's OK to watch a game on TV which is a minute or so behind actual time - provided you can't hear the groans and/or cheers from neighbours who just may have lower latency due to shorter network delays.

            Some on demand data models, particularly for multimedia distribution, do have strategies for compromising audio or video quality in order to keep the the streams and controls still working. I just don't know what strategies the BBC is using - and indeed it may be a complex hybrid strategy depending on the number of logged in users. Some of the more effective strategies scale up a lot better if there are many users all wanting essentially the same data at the same time, while strategies for very many users wanting different data simultaneously are different.
            This might be true of video on iPlayer or ITVX - but the amount of simultaneous requests on BBC sounds for a R3 concert is unlikely to mean any degradation.

            Comment

            • Dave2002
              Full Member
              • Dec 2010
              • 17958

              #21
              Originally posted by Ein Heldenleben View Post
              This might be true of video on iPlayer or ITVX - but the amount of simultaneous requests on BBC sounds for a R3 concert is unlikely to mean any degradation.
              I recently visited a relatively remote part of Scotland where even receiving emails was problematic, so there could indeed be problems. It is easy for those of us who are mostly on relatively fast connections to forget that there are parts of the world where even very small messages and low bandwidth links won't get through, or certainly not in a timely manner.

              Interestingly though on the drive back I noticed that the car radio seemed to switch - mostly relatively seamlessly - between DAB and FM - according to the display. I wasn't sure whether the radio was actually receiving FM, but since on at least one occasion there was a kind of swishing noise - the indication was that the radio was actually operating in FM mode at that time. Overall the quality was acceptable, though I did wonder if it was always in stereo. Perhaps the relative (mostly) lack of hiss indicates that the radio tended to reduce the stereo component in favour of keeping the noise levels low. Once or twice there did seem to be a very obvious jump between DAB and FM. I am not sure whether many radios do this kind of alternate switching between DAB and FM, though many FM sets have in the past done alternate frequency selection.

              Of course what might be acceptable in a car regarding audio quality may not apply to home installations and more detailed listening with good amplifiers and speakers.

              Comment

              Working...
              X