0
0

I have a music file (mp3) split into multiple parts i.e.

1 intro,
2. multiple loops – that can be played for infinite time in any order
3. outro

What is the best practice for playing this kind of file?

  • You must to post comments
0
0

Actually is there a better way than this as I’m on iPhone and this method is very wasteful with memory.

Should this not be a bug?

  • You must to post comments
0
0

I decided to have a look at this again as I see the problem still hasn’t been fixed with the latest release.

Anyway to everyone who’s having the problem (I notived there’s a lot of posts regarding it), it can be "solved" by simply increasing the amount of buffers used (the number depends on the files) however obviously that’s not a perfect solution.

  • You must to post comments
0
0

Does no-one have a suggestion for this?

I tried using the END callback but it’s not smooth, I notice that I can specify loop points on the track.

Is that the way to go?

  • You must to post comments
0
0

Ok I have it sorted now. I followed the realtimestitching example.

  • You must to post comments
0
0

Alternatively you could check out the sample accurate sequencer example on the wiki:

http://52.88.2.202/wiki/index.php5?tit … h_setDelay

  • You must to post comments
0
0

Actually the realtimestitching example method hasn’t worked.

It’s fine for the files included however if I replace them with my own mp3 it seems to "swap" at the wrong time.

Could this be with how the FMOD_TIMEUNIT_BUFFERED flag is used, do I need to use a different flag or comibantion perhaps?

I also don’t think I can use the accurate_sequencing example as it swaps channels and I think the wiki says that’s not possible with hardware decoding as only one channel can exist at once. I am on iPhone.

Plus stitching should really work so I think it must be something simple.

  • You must to post comments
0
0

Hi Fmod guys,

We’re trying to get our game out the door and really need this sorted. I’m sure someone has needed to do this with mp3s.

Is there not a solution?

btw – It’s not a bug in my code as I’ve tested it using the sample by replacing where the songs are defined so it now becomes:

<code>
"../media/floating_in_the_atmosphere_intro.mp3", /* Ma- /
"../media/floating_in_the_atmosphere_middle_1.mp3", /
ry /
"../media/floating_in_the_atmosphere_middle_2.mp3", /
had /
"../media/floating_in_the_atmosphere_middle_1.mp3", /
a /
"../media/floating_in_the_atmosphere_middle_2.mp3", /
lit- */
</code>

And I’ve also verified that my mp3s’ format is PCM 16 with two channels and sampled at 44100Hz.

I tried even taking this info from the mp3s using GetFormat.

I also tried to get it working by using 3 sub sounds sentence instead of two.

At the moment I’ve put in a timer to stop it changing within a couple of seconds and that helps it a little however it’s no solution.

  • You must to post comments
0
0

Hi Adversus,

You are right, on iPhone the sequencer wont work with the hardware decoder. I’m not entirely clear what the issue is
[quote:16wbivjw]it seems to "swap" at the wrong time. [/quote:16wbivjw]

How far off is it? A few samples like a ‘pop’ sound? A few miliseconds like a tiny gap or glitch, or many miliseconds?

I think it could possibly related to inaccuracy of MP3 encoding, as an experiment can you try using a different format like PCM or ADPCM and see if that exhibits the same issues.

  • You must to post comments
0
0

Couple of quick questions:

[list=]Are you opening the file with FMOD_ACCURATETIME?
Are the files CBR or VBR?
Have you tried the same with just a WAV file?[/list]

  • You must to post comments
Showing 9 results
Your Answer

Please first to submit.