comparison src/BTrack.h @ 17:a31841af2bbc develop

Before this commit, the OnsetDetectionFunction class used double precision, and the BTrack class used floats. I have now made BTrack use double precision also. This works fine and the only cost of doing this is that we have to perform one array copy into floating point format so that sample rate conversion (which has to be in floating point format) can take place.

 	
     /** Initialise with frame size and set all frame sizes accordingly */
 	void initialise(int fsize);
     
     /** Add new sample to buffer and apply beat tracking */
-	void process(float df_sample);
+	void process(double df_sample);
    
     /** Set the tempo of the beat tracker */
-	void settempo(float tempo);
+	void settempo(double tempo);
     
     /** fix tempo to roughly around some value */
-	void fixtempo(float tempo);
+	void fixtempo(double tempo);
     
     /** do not fix the tempo anymore */
 	void unfixtempo();
 	
 	int playbeat;
-	float cscoreval;
-	float est_tempo;
+	double cscoreval;
+	double est_tempo;
 			
 private:
     
     /** Convert detection function from N samples to 512 */
 	void dfconvert();
     
     /** update the cumulative score */
-	void updatecumscore(float df_sample);
+	void updatecumscore(double df_sample);
 	
     /** predicts the next beat */
     void predictbeat();
     
     /** Calculates the current tempo expressed as the beat period in detection function samples */
     void calcTempo();
     
     /** calculates an adaptive threshold which is used to remove low level energy from detection 
      * function and emphasise peaks 
      */
-	void adapt_thresh(float *x,int N);
+	void adapt_thresh(double *x,int N);
     
     /** calculates the mean of values in an array from index locations [start,end] */
-	float mean_array(float *array,int start,int end);
+	double mean_array(double *array,int start,int end);
     
     /** normalises a given array */
-	void normalise(float *array,int N);
+	void normalise(double *array,int N);
     
     /** calculates the balanced autocorrelation of the smoothed detection function */
-	void acf_bal(float *df_thresh);
+	void acf_bal(double *df_thresh);
     
     /** returns the output of the comb filter */
 	void getrcfoutput();
 	
 	// buffers
-	float *dfbuffer;			/**< to hold detection function */
-	float df512[512];			/**< to hold resampled detection function */
-	float *cumscore;			/**<  to hold cumulative score */
+	double *dfbuffer;			/**< to hold detection function */
+	double df512[512];			/**< to hold resampled detection function */
+	double *cumscore;			/**<  to hold cumulative score */
 	
-	float acf[512];				/**<  to hold autocorrelation function */
+	double acf[512];				/**<  to hold autocorrelation function */
 	
-	float wv[128];				/**<  to hold weighting vector */
+	double wv[128];				/**<  to hold weighting vector */
 	
-	float rcf[128];				/**<  to hold comb filter output */
-	float t_obs[41];			/**<  to hold tempo version of comb filter output */
+	double rcf[128];				/**<  to hold comb filter output */
+	double t_obs[41];			/**<  to hold tempo version of comb filter output */
 	
-	float delta[41];			/**<  to hold final tempo candidate array */
-	float prev_delta[41];		/**<  previous delta */
-	float prev_delta_fix[41];	/**<  fixed tempo version of previous delta */
+	double delta[41];			/**<  to hold final tempo candidate array */
+	double prev_delta[41];		/**<  previous delta */
+	double prev_delta_fix[41];	/**<  fixed tempo version of previous delta */
 	
-	float t_tmat[41][41];		/**<  transition matrix */
+	double t_tmat[41][41];		/**<  transition matrix */
 	
 	
 	// parameters
-	float tightness;
-	float alpha;
-	float bperiod;
-	float tempo;
+	double tightness;
+	double alpha;
+	double bperiod;
+	double tempo;
 	
 	
-	float p_fact;
+	double p_fact;
 	
 	
 	//
 	int m0;				// indicates when the next point to predict the next beat is
 	int beat;