eeva/jamulus
1
0
Fork 0
Code Issues Pull requests Releases Wiki Activity

go back to original factor for auto jitter buffer calculation

Browse source
This commit is contained in:
Volker Fischer 2009-09-09 19:41:06 +00:00
parent 8734ee8092
commit 519590d764
1 changed files with 2 additions and 2 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

4
src/client.cpp
View file

@ -585,7 +585,7 @@ void CClient::UpdateSocketBufferSize()
// We use the time response measurement for the automatic setting. // We use the time response measurement for the automatic setting.
// Assumptions: // Assumptions:
// - the audio interface/network jitter is assumed to be Gaussian // - the audio interface/network jitter is assumed to be Gaussian
// - the buffer size is set to 3 times the standard deviation of // - the buffer size is set to 3.3 times the standard deviation of
// the jitter (~98% of the jitter should be fit in the // the jitter (~98% of the jitter should be fit in the
// buffer) // buffer)
// - introduce a hysteresis to avoid switching the buffer sizes all the // - introduce a hysteresis to avoid switching the buffer sizes all the
@ -607,7 +607,7 @@ void CClient::UpdateSocketBufferSize()
// divide by MIN_SERVER_BLOCK_DURATION_MS because this is the size of // divide by MIN_SERVER_BLOCK_DURATION_MS because this is the size of
// one block in the jitter buffer // one block in the jitter buffer
const double dEstCurBufSet = ( dAudioBufferDurationMs + dServerJitterMs + const double dEstCurBufSet = ( dAudioBufferDurationMs + dServerJitterMs +
3 * ( Channel.GetTimingStdDev() + CycleTimeVariance.GetStdDev() ) ) / 3.3 * ( Channel.GetTimingStdDev() + CycleTimeVariance.GetStdDev() ) ) /
SYSTEM_BLOCK_DURATION_MS_FLOAT + 0.5; SYSTEM_BLOCK_DURATION_MS_FLOAT + 0.5;
// upper/lower hysteresis decision // upper/lower hysteresis decision