Mercurial > hg > qm-dsp

comparison dsp/onsets/DetectionFunction.cpp @ 331:6dd7c8bb5e3a

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Restore the accidentally-removed line in which all the work happens! Fixes onset detector
author Chris Cannam <c.cannam@qmul.ac.uk>
date Thu, 05 Sep 2013 12:33:35 +0100
parents 31f22daeba64
children f3c69325cca2
comparison
equal deleted inserted replaced
330:3f7b4b282df1 331:6dd7c8bb5e3a
98 // If we have to use this implementation (as opposed to the 98 // If we have to use this implementation (as opposed to the
99 // version of process() below that operates on frequency domain 99 // version of process() below that operates on frequency domain
100 // data directly), we will have to use the next smallest power of 100 // data directly), we will have to use the next smallest power of
101 // two from the block size. Results may vary accordingly! 101 // two from the block size. Results may vary accordingly!
102 102
103 int actualLength = MathUtilities::previousPowerOfTwo(m_dataLength); 103 int actualLength = MathUtilities::previousPowerOfTwo((int)m_dataLength);
104 104
105 if (actualLength != (int)m_dataLength) { 105 if (actualLength != (int)m_dataLength) {
106 // Pre-fill mag and phase vectors with zero, as the FFT output 106 // Pre-fill mag and phase vectors with zero, as the FFT output
107 // will not fill the arrays 107 // will not fill the arrays
108 for (int i = actualLength/2; i < (int)m_dataLength/2; ++i) { 108 for (int i = actualLength/2; i < (int)m_dataLength/2; ++i) {