1. Things you need to know about the new ‘Conversations’ PM system:

    a) DO NOT REPLY TO THE NOTIFICATION EMAIL! I get them, not the intended recipient. I get a lot of them and I do not want them! It is just a notification, log into the site and reply from there.

    b) To delete old conversations use the ‘Leave conversation’ option. This is just delete by another name.
    Dismiss Notice

The whole "gapless" thing on digital files...

Discussion in 'audio' started by doctorjuggles, Mar 11, 2019.

  1. doctorjuggles

    doctorjuggles pfm Member

    Right, so gapless playback was sorted on the iPod sometime back in 2006 and was never really an issue unless the file was very badly ripped.

    Yet somehow, with PCs, devices and all the surrounding technology advancements in the intervening years, it continues to be an issue on certain platforms, music apps and files.

    Can someone explain in layman's terms (aside from general manufacturer/consumer apathy towards the issue) what makes it so difficult as a concept to implement? The classic iPods had a dual 80mhz ARM chip, modern DAPs and phones are many times more powerful, so it must be a coding issue, surely? But I don't really understand why it's so hard to get right.

    I know you can find loads of players/apps that do it - but it's still a pain and so many things still get it wrong (Google Music app for a start)
  2. Romulus

    Romulus pfm Member

    Sorry to be a dinosaur but what does gapless playback mean? Why do audiophiles get so heated when discussing gapless? I have a ripping machine (also an amp,storage and internet radio) on the settings I have a kind of an option in regard to gapless, never used it. As I don't know what it is, maybe I don't know what I am missing, but all my playback have been normal without issue, its all so mysterious....
  3. zippy

    zippy pfm Member

    As far as I'm concerned, gapless playback is simply being able to play one track directly after another with no silence in between so that the tracks merge seamlessly together.
    I think most current music doesn't need it as the gap between tracks is there on purpose. Ditto most classical music.
    Where it hits hardest is, e.g. on concept albums - best examples I can think of are Pink Floyd albums.. (e.g. between tracks 5 and 6 on 'Dark Side...')
  4. doctorjuggles

    doctorjuggles pfm Member

    On some files and/or on some players, when you play back an album, there's are small gaps in between the tracks. Literally inserting a short silence in between tracks.
    Now for most pop and many other forms of music, it's not recognisable or noticeable (and if it is, it's often not very annoying). But for other forms of music (classical / prog-rock concept albums / DJ mixes) it's incredibly annoying and interrupts the flow. I listen to a lot of DJ mixes and much of the skill in DJing is in how seamlessly one can blend two tracks. It's not exactly seamless if a clunky gap is introduced. Same issue on, to use and often-quoted example, Pink Floyd's "Dark Side of the Moon" where many of the tracks flow into each other.
  5. Jim Audiomisc

    Jim Audiomisc pfm Member

    For an engineer or programmer it isn't 'difficult' to arrange for a system to play (or rip) gapless. It just requires the relevant programmers to have a clue. So unwanted 'gaps' are a sign of poor digital design / programming of how a system processes the digital audio. Typical reason in ancient times was down to the way someone didn't deal correctly with the way data transfers tend to operate in 'chunks'. But if the designers / programmers / etc know what they are doing it is trivial to avoid the problems and get smoothly gapless replay.

    TBH I'm amazed if there are *still* devices or players that get this wrong.
  6. John Phillips

    John Phillips pfm Member

    It may be true that most classical music is composed with gaps between movements or other segments. But there are many classical compositions where gapless playback is essential to the composer's intention and to the listener's enjoyment.

    In symphonies just listen to the glorious continuous transition between the third and fourth movements of Beethoven's fifth symphony. In opera there are countless examples. Others are providing examples in different genres and also see the Wikipedia entry including:
    "The absence of gapless playback is a source of annoyance to listeners of music where tracks are meant to segue into each other, such as some classical music (opera in particular), progressive rock, concept albums, electronic music, and live recordings with audience noise between tracks."
  7. allthingsanalog

    allthingsanalog pfm Member

    Pink Floyd certainly needs gapless playback!
  8. doctorjuggles

    doctorjuggles pfm Member

    That's the part I'm amazed at too. After all this time. I mean I understand that it's a small cadre of users who get annoyed by it, but it must be so easy to program for this requirement that it should be a non-issue, especially as the group who care about this are likely to be heavy users/spenders and also likely to become fierce advocates. It's a bit baffling considering, as you say, that it's not difficult to engineer/program for
  9. George J

    George J Herefordshire member

    This would be right if the new tracks in classical music always occur at changes of music where the musicians stop playing ... They do not.

    If one considers a forty minute section of music that is continuous, but has many tracks marked during the flow, then gaps become serious!

    I find iTunes manages gapless perfectly with no effort from me.

    With best wishes from George
    camverton likes this.
  10. AndyU

    AndyU pfm Member

    Gapless playback is essential for choral music and opera. An act of an opera may have 30 or so “tracks” on a cd, but those are there to aid navigation and identification, the act itself is one continuous piece. Same with choral works. I don’t care if I am a part of a small cadre of users, gapless playback is essential, non-negotiable! The issue came about with some kinds of streamers where the renderer waited for one “track” to finish before requesting the next “track” from the server, hence a small gap. It is the case though that you can get round the problem by ripping all the relevant tracks as one continuous track, and using a cue sheet to navigate. This solution was more popular in the past than now as most decent players can manage gapless nowadays.
    Strictly Stereo and camverton like this.
  11. Octavian

    Octavian pfm Member

    Try Wagner without gapless.
    Vinniemac likes this.
  12. Mark_R

    Mark_R Well-Known Member

    Gapless playback is essential in classical music even when there is a pause between movements. That pause is not silent. It is the sound of the natural acoustic of the recording venue. The sudden plunge to digital silence when you don’t have gapless playback is most annoying.
  13. richie60

    richie60 pfm Member

    My Pioneer N50 streamer doesn't do gapless via upnp dlna but my raspberry pi does using the same technology.
  14. Romulus

    Romulus pfm Member

    Wow I don't think I have ever heard gapless playback on any of my systems, nor any other systems at friends place or any other third party in all my life. I am trying to imagine listening to my music collection gapless (and I think I have this option) I dont think I like it because most probably I am programmed to expect that short interlude, also I often change music genres, to have it gapless would be a form of torture. However some above posts have raised some interesting points in regard to classical music. In regard to Symphonies I always thought that the short break between movements was how the composer intended and I am trying to remember some classical concerts I attended in London, I am quite sure there was that short break between movements. As to Dark Side of the Moon, I may tolerate gapless playback with the help of influencing substances to get into the spirit of things....
  15. Ian M

    Ian M pfm Member

    I’m afraid you haven’t grasped what gapless playback means.;)
    JensenHealey, Stunsworth and suzywong like this.
  16. Wolfmancatsup

    Wolfmancatsup Empire State Human

    I've used Linn streamers since they first arrived. The early ones didn't support gapless but the company sorted that out very quickly. I've heard of a few other companies over the years who still hadn't sorted it though.
  17. Dave-L

    Dave-L Well-Known Member

    OK, just to explain what gapless playback actually is. It does not mean that all gaps between tracks are removed, causing tracks to run into one another. However, where tracks are meant to fade into one another, like the aforementioned Pink Floyd, gaps are not introduced where there should be none.
    With a player that does not support gapless playback, you will hear a gap of a few milliseconds (or more), between such tracks, which is very annoying.
  18. gints

    gints pfm Member

    Jean Michel Jarre also have this thing when next track music already starting in previuos track and this suddenly silence in middle of music is annoying.
  19. Jim Audiomisc

    Jim Audiomisc pfm Member

    The point you need to bear in mind is that music where it is *expected* or *intended* that there should be a 'slient pause' should be recorded and provided as tracks or files that *include a period of silence* at the start and/or end of each track/file. So when played 'gapless' you still get those audible 'breaks' between the music in each track or file.

    'Gapless' playback means the player does not *add* another gap or silence in between successice tracks or files. But it will play the periods of 'silence' when these are deliberately included in the files or CD tracks because they are required by the music.

    I have bucketloads of CDs. Played on a CD player or as ripped files, I get 'gapless' playback - but also get the silent pauses provided by the starts and ends *where they are intended by the producers of the music*. *Not* added 'gaps' introduced by the playback system.
  20. Barbapapa

    Barbapapa pfm Member

    My understanding (being a computer engineer but with no experience in audio processing) of digital playback is that you send the data in consecutive blocks or chuncks to the DAC, which will convert it for analogue processing (i.e. listenable through earphones/speakers). A single audiofile is always played without any gap, obviously, so it should be relatively straightforward to simple fetch the first block of the next audiofile and send it straight through.

    A delay might occur if the processor is too lightweight and/or the program for some reason takes a lot of time to decide what is the next file and where it is located (or has to do decoding stuff). Possibly simple DAPs only perform the decoding once the next file is started. That could account for a audible delay. To my mind this could be solved in several ways, of which two spring to mind:
    - preloading and decoding all files of a single album/folder (which does require sufficient memory, but that should be available as a single file may also be an hour long),
    - start decoding the next file before the current file has ended.
    The second option is more generic and in that respect preferable, but it may require a more complicated hardware design (such as a buffer between processor and DAC) as it should allow the processor (once it has sent its last block (or the one before that, if the final block is very short) to start converting the next file while also sending the current data out to the DAC. Possibly many DAPs use a simpler design in which the processor sends a continuous data stream to the DAC and thereby isn't free to do other stuff until the current file has finished. In that case the second option would only work if the processor would simultaneously (concurrently) perform the two tasks of sending data and converting the next file. But to do so without audible delays in sending the data to the DAC may be hard as the processor is probably fairly lightweight (to be cheap and not power-hungry). I've seen some remarks on gapless playback and multi-threading that suggest that this is indeed the problem for many DAPs.

    So the first option might be simpler to implement. That requires a somewhat more complicated program structure, but should in itself not be difficult to realise with new firmware.

    Of course, I may overlook complications that lie in some specific aspect of DAP design and firmware.

Share This Page

  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.
    Dismiss Notice