Page 1 of 1
[Not a bug] Gap on looping MP3s
Posted: Thu Mar 09, 2006 2:54 am
by Gambit37
If you use a MP3 as a looping ambient sound, there is a noticeable silent pause between each loop. This doesn't happen with PCM WAV.
I imagine this is due to continually decompressing the sound or something? Can it be fixed?
EDIT: I opened my MP3 to view it and some empty space has been added to the start and end! If I remove this and resave as MP3, it comes back!
Anyone know how to avoid this?
Posted: Thu Mar 09, 2006 2:57 am
by Gambit37
Seems as if OGG doesn't have this problem! Guess I'll stick with that -- it's files are smaller too!
Posted: Thu Mar 09, 2006 7:03 am
by Sophia
You can't avoid it. It's an artifact of the MP3 encoding process.
Posted: Thu Mar 09, 2006 8:26 am
by beowuuf
yup, always annoying on listening to albums where the songs are supposed to be a continuous flow
Posted: Thu Mar 09, 2006 9:48 am
by Gambit37
Guess I've never noticed it using Windows Media Player as it has intelligent read-ahead buffering.
I noticed this on iTunes though and simply thought the software sucked. Guess it was the files themselves afterall.
Posted: Sat Mar 11, 2006 8:56 pm
by Des
I've just been playing with this and even using a WAV I get a "blip" between each repetition. I was hoping to create a seamless continuous sound using just a short sound clip.
This is the one I used (made with Nero Wave Editor)
Takk Snippet - about a second of Sigur Ros.
I was wondering if it's not possible to get a totally seamless loop or if my file actually has a very short period of silence. I guess maybe the answer is to use a longer snippet with a fade-in and fade-out?
Posted: Sat Mar 11, 2006 9:04 pm
by Gambit37
The looping works for me -- you probably have a small gap of silence at the end beginning. Should be easily spottable in any decent sound editor.
Posted: Sat Mar 11, 2006 9:09 pm
by George Gilbert
For looped sounds, RTC detects that it's stopped once per frame (so about every 1/60th of a second) and restarts it if necessary. If something complex is happening in the dungeon, then I guess it could take up to 1/10th of a second to detect the loop.
I would have thought that that was a short enough gap to be inaudible. If you have a very short sample (much less than a second) however, I could just about believe that the rapid on / off of the sound could be audible.
As Gambit suggests, it's probably the sound itself that has a gap at the start / end. As you suggest, you could always just have a longer sample and fade loop sounds at the start and end so that if there's a tiny silence then it won't be noticed. Realistically though, that shouldn't be needed.