Stepan Podoskin
2013-08-20 19:23:35 UTC
I’m currently working on sound recording in my project but I haven't yet commited anything related to it.
There are two sound renderers - software and OpenAL. Software renderer allows to grab all output using filter (iSndSysRendererSoftware::AddOutputFilter). That is what I'm currently using. My implementation works well in real-time mode, but in non-realtime mode it fails because sound output is still synchronized with true time. OpenAL renderer doesn't support grabbing of sound output and OpenAL itself doesn't support this, so I think it is impossible to implement sound recording using OpenAL.
I tested my implementation on Windows using sndtest application but on Linux sndtest segfaults (I tried latest trunk on Kubuntu 13.04). Is it only happening to me? Anyway, are there any other
There are two sound renderers - software and OpenAL. Software renderer allows to grab all output using filter (iSndSysRendererSoftware::AddOutputFilter). That is what I'm currently using. My implementation works well in real-time mode, but in non-realtime mode it fails because sound output is still synchronized with true time. OpenAL renderer doesn't support grabbing of sound output and OpenAL itself doesn't support this, so I think it is impossible to implement sound recording using OpenAL.
I tested my implementation on Windows using sndtest application but on Linux sndtest segfaults (I tried latest trunk on Kubuntu 13.04). Is it only happening to me? Anyway, are there any other