Re: [web-audio-api] OfflineAudioContext needs a way to handle audio of arbitrary duration (#21)

> [Original comment](https://www.w3.org/Bugs/Public/show_bug.cgi?id=21311#0) by Jussi Kalliokoski on W3C Bugzilla. Sat, 16 Mar 2013 18:49:14 GMT

> Chris Rogers stated in teleconference 14 Mar 2013 that it is in fact feasible to mix down typical track lengths of several minutes with the single oncomplete call.

One thing is that situations where you might expect to get long renderings (several hours) are studio software and scientific applications. Neither are typically run on devices with little memory, and if you think that uncompressed stereo CD quality gives you 80min @ 700MB, the problem may or may not be very serious. Either way, I think that the offline context should have an error callback as well, and if you run out of memory during the rendering the callback should be invoked.

However, to speculate a bit, another approach for handling long renderings is to direct the output to a MediaStreamRecorder which will hopefully in the future allow extracting compressed formats as well. This would allow the UA take care of the render/compress cycle and block sizes, and the API would just hand the developer a readily rendered blob to use however needed. I'm not quite up to date on MediaStreamRecorder though, perhaps Roc knows more about whether this is feasible or not.

---
Reply to this email directly or view it on GitHub:
https://github.com/WebAudio/web-audio-api/issues/21#issuecomment-24244089