Not really a bug, but I often notice when exporting offline from a sound file, e.g. with a FilePlayer set to start playing, if the file is for example, 10:00.499 and I set the time of the exported file to be 10:00.500 (a fraction longer than the input file), the FilePlayer's "play" triangle doesn't get all the way to the end, even though the output file will be exactly the right length, 10:00.500. Not really a probem, but seems a little strange.
Do you have audio enabled when you perform the export? I suspect that it's just the FilePlayer's user interface not updating.
SoundIn should be synchronised exactly with export, but there is no guarantee about exact time alignment between other FilePlayers and Export to SoundFile. FilePlayer start will be subject to the same time resolution as automation (64 samples or approx. 1ms as of AudioMulch 2.2.0). This could also account for what you're seeing.
Pretty sure it happens with both audio enabled, and audio not enabled. Ah, OK, yes, probably just the graphic interface not updating. Sound files are always the exact length specified! I always automate the file player, and had forgotten about the 64 samples of delay, so in future will account for that. Thanks Ross!
Well, it's not that there is 64 samples of delay, it's that in most cases automation is quantized to 64 sample boundaries.
Clock-based contraptions are still perfectly synchronised to the clock of course. But automation cuves and asynchronous triggers, such as starting a file player, are not aligned to the beat.
I think you will find that if you don't use pre-roll, and you trigger the file player at time 0, then it will start straight away. It's easy enough to check if you have a file with a single sample tick at the start.
Sorry for being vague. I'm reluctant to make strong statements about behavior that is not guaranteed to work in a particular way.
No worries Ross, thanks for the help and explanations! I always use a three second pre roll, but it's just to start a File Player playing, which always has a little silence at the beginning anyway. So it's probably safe to use no Pre-Roll in this situation? Will check some things today!
If the idea is to start from silence then no pre-roll is needed. I'd only use pre-roll if you need to "prime" some delays, reverb etc. It's useful when creating loops for example. But I think for your situation it's not needed.
Ah cheers, that's great to know, will turn it off, and that will also cut down on all the rendering time! :)