| Index: Source/core/platform/audio/HRTFPanner.cpp
|
| diff --git a/Source/core/platform/audio/HRTFPanner.cpp b/Source/core/platform/audio/HRTFPanner.cpp
|
| index 9234fedb415384ebaf8a34398e2e639b12c9192f..db7f2d886638063627c597c9d0dafaaee4cbd249 100644
|
| --- a/Source/core/platform/audio/HRTFPanner.cpp
|
| +++ b/Source/core/platform/audio/HRTFPanner.cpp
|
| @@ -247,7 +247,7 @@ void HRTFPanner::pan(double desiredAzimuth, double elevation, const AudioBus* in
|
| m_delayLineR.process(segmentSourceR, segmentDestinationR, framesPerSegment);
|
|
|
| bool needsCrossfading = m_crossfadeIncr;
|
| -
|
| +
|
| // Have the convolvers render directly to the final destination if we're not cross-fading.
|
| float* convolutionDestinationL1 = needsCrossfading ? m_tempL1.data() : segmentDestinationL;
|
| float* convolutionDestinationR1 = needsCrossfading ? m_tempR1.data() : segmentDestinationR;
|
| @@ -256,7 +256,7 @@ void HRTFPanner::pan(double desiredAzimuth, double elevation, const AudioBus* in
|
|
|
| // Now do the convolutions.
|
| // Note that we avoid doing convolutions on both sets of convolvers if we're not currently cross-fading.
|
| -
|
| +
|
| if (m_crossfadeSelection == CrossfadeSelection1 || needsCrossfading) {
|
| m_convolverL1.process(kernelL1->fftFrame(), segmentDestinationL, convolutionDestinationL1, framesPerSegment);
|
| m_convolverR1.process(kernelR1->fftFrame(), segmentDestinationR, convolutionDestinationR1, framesPerSegment);
|
| @@ -266,7 +266,7 @@ void HRTFPanner::pan(double desiredAzimuth, double elevation, const AudioBus* in
|
| m_convolverL2.process(kernelL2->fftFrame(), segmentDestinationL, convolutionDestinationL2, framesPerSegment);
|
| m_convolverR2.process(kernelR2->fftFrame(), segmentDestinationR, convolutionDestinationR2, framesPerSegment);
|
| }
|
| -
|
| +
|
| if (needsCrossfading) {
|
| // Apply linear cross-fade.
|
| float x = m_crossfadeX;
|
|
|