guys, ive been getting an extreamly loud digital glitch noise in-game after changing a source wav file from 32bit float to 16 bit, (sounds fine in designer tho)
normaly i have to tweek the automatic sample rate optimisation in the banks tab for fmod to realise the bit depth has changed and play it back properly ingame.
but when i do that now, fmod fails to build, it just hangs and i have to quit designer.
- simon kamakazi asked 7 years ago
The source format has no impact on what is output by FMOD Designer. We decode, adjust bitdepth, and encode all the data during the build process, so you’re going to have the exact same run-time performance regardless of what format you use for source data. Internally we use 32-bit float PCM so you’re probably best off sticking with the 32-bit float wav files.
That was a little vague, basically PCM and ADPCM are samplerate based. There compression is very simple and the compression ratio is quite low.
For long sounds like music and dialog it’s generally better to have a high compression format like MP3/MP2/Celt. These will save you quite a lot of memory and make sure you’re using the "Stream from disk" bank type. MP3 is a good codec for PC, but things can get hairy with regards to licencing if it’s a commercial product. I would reccomend MP2/Celt since they’re totally free.
An update on the bug report, it is fixed in the new FMOD Designer 4.31.04 with the new build system.
not sure ill be able to send u the source file (NDA)
but it happens when i replace a 32 bit source file to 16bit. it will play back in fmod fine, but in tyhe game build it is messed up. (not everytime either maby 20% of the time)
its as tho fmod hasnt compleatly realised that i changed the bit depth, because when i adjust the sound’s automatic sample rate it works in game.
hope that helps.
Ok, I will need a bit more information from you before I can try to reproduce the error:
What version of designer did you experience this issue with?
Just to make sure I understand the steps involved:
1/ Create an event which uses float format WAV source file
2/ Build project (including soundbank)
3/ Overwrite the original wav with a 16-bit version
4/ Build project (including soundbank)
Is that correct?
This seems like an unusual use case, out of interest – why are you changing the format of your source files?
By the way, we have recently completely rewritten our build system including encoders, so I reccomend you try out the latest dev branch designer if you can.
im using version
yep those 4 steps are correct.
well my sequencer spits out floating point files as default, so i just use those, untill i have to optimise for better game performance, then i batch convert the sounds to 16 bit.
Please login first to submit.