diff --git a/src/rvoice/fluid_rvoice_dsp.c b/src/rvoice/fluid_rvoice_dsp.c index 7931c812..c3a49072 100644 --- a/src/rvoice/fluid_rvoice_dsp.c +++ b/src/rvoice/fluid_rvoice_dsp.c @@ -119,6 +119,27 @@ void fluid_rvoice_dsp_config (void) fluid_check_fpe("interpolation table calculation"); } +/* + * Combines the most significant 16 bit part of a sample with a potentially present + * least sig. 8 bit part in order to create a 24 bit sample. + */ +static int32_t fluid_rvoice_get_sample(const short int* dsp_msb, const char* dsp_lsb, unsigned int idx) +{ + /* cast sample to unsigned type, so we can safely shift and bitwise or + * without relying on undefined behaviour (should never happen anyway ofc...) */ + uint32_t msb = (uint32_t)dsp_msb[idx]; + uint8_t lsb = 0U; + + /* most soundfonts have 16 bit samples, assume that it's unlikely we + * experience 24 bit samples here */ + if(G_UNLIKELY(dsp_lsb != NULL)) + { + lsb = (uint8_t)dsp_lsb[idx]; + } + + return (int32_t)((msb << 8) | lsb); +} + /* No interpolation. Just take the sample, which is closest to * the playback pointer. Questionable quality, but very * efficient. */ @@ -128,6 +149,7 @@ fluid_rvoice_dsp_interpolate_none (fluid_rvoice_dsp_t *voice) fluid_phase_t dsp_phase = voice->phase; fluid_phase_t dsp_phase_incr; short int *dsp_data = voice->sample->data; + char *dsp_data24 = voice->sample->data24; fluid_real_t *dsp_buf = voice->dsp_buf; fluid_real_t dsp_amp = voice->amp; fluid_real_t dsp_amp_incr = voice->amp_incr; @@ -151,7 +173,7 @@ fluid_rvoice_dsp_interpolate_none (fluid_rvoice_dsp_t *voice) /* interpolate sequence of sample points */ for ( ; dsp_i < FLUID_BUFSIZE && dsp_phase_index <= end_index; dsp_i++) { - dsp_buf[dsp_i] = dsp_amp * dsp_data[dsp_phase_index]; + dsp_buf[dsp_i] = dsp_amp * fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index); /* increment phase and amplitude */ fluid_phase_incr (dsp_phase, dsp_phase_incr); @@ -189,13 +211,14 @@ fluid_rvoice_dsp_interpolate_linear (fluid_rvoice_dsp_t *voice) fluid_phase_t dsp_phase = voice->phase; fluid_phase_t dsp_phase_incr; short int *dsp_data = voice->sample->data; + char *dsp_data24 = voice->sample->data24; fluid_real_t *dsp_buf = voice->dsp_buf; fluid_real_t dsp_amp = voice->amp; fluid_real_t dsp_amp_incr = voice->amp_incr; unsigned int dsp_i = 0; unsigned int dsp_phase_index; unsigned int end_index; - short int point; + int32_t point; fluid_real_t *coeffs; int looping; @@ -209,8 +232,8 @@ fluid_rvoice_dsp_interpolate_linear (fluid_rvoice_dsp_t *voice) end_index = (looping ? voice->loopend - 1 : voice->end) - 1; /* 2nd interpolation point to use at end of loop or sample */ - if (looping) point = dsp_data[voice->loopstart]; /* loop start */ - else point = dsp_data[voice->end]; /* duplicate end for samples no longer looping */ + if (looping) point = fluid_rvoice_get_sample(dsp_data, dsp_data24, voice->loopstart); /* loop start */ + else point = fluid_rvoice_get_sample(dsp_data, dsp_data24, voice->end); /* duplicate end for samples no longer looping */ while (1) { @@ -220,8 +243,8 @@ fluid_rvoice_dsp_interpolate_linear (fluid_rvoice_dsp_t *voice) for ( ; dsp_i < FLUID_BUFSIZE && dsp_phase_index <= end_index; dsp_i++) { coeffs = interp_coeff_linear[fluid_phase_fract_to_tablerow (dsp_phase)]; - dsp_buf[dsp_i] = dsp_amp * (coeffs[0] * dsp_data[dsp_phase_index] - + coeffs[1] * dsp_data[dsp_phase_index+1]); + dsp_buf[dsp_i] = dsp_amp * (coeffs[0] * fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index) + + coeffs[1] * fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index+1)); /* increment phase and amplitude */ fluid_phase_incr (dsp_phase, dsp_phase_incr); @@ -238,8 +261,8 @@ fluid_rvoice_dsp_interpolate_linear (fluid_rvoice_dsp_t *voice) for (; dsp_phase_index <= end_index && dsp_i < FLUID_BUFSIZE; dsp_i++) { coeffs = interp_coeff_linear[fluid_phase_fract_to_tablerow (dsp_phase)]; - dsp_buf[dsp_i] = dsp_amp * (coeffs[0] * dsp_data[dsp_phase_index] - + coeffs[1] * point); + dsp_buf[dsp_i] = dsp_amp * (coeffs[0] * fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index) + + coeffs[1] * point); /* increment phase and amplitude */ fluid_phase_incr (dsp_phase, dsp_phase_incr); @@ -278,13 +301,14 @@ fluid_rvoice_dsp_interpolate_4th_order (fluid_rvoice_dsp_t *voice) fluid_phase_t dsp_phase = voice->phase; fluid_phase_t dsp_phase_incr; short int *dsp_data = voice->sample->data; + char *dsp_data24 = voice->sample->data24; fluid_real_t *dsp_buf = voice->dsp_buf; fluid_real_t dsp_amp = voice->amp; fluid_real_t dsp_amp_incr = voice->amp_incr; unsigned int dsp_i = 0; unsigned int dsp_phase_index; unsigned int start_index, end_index; - short int start_point, end_point1, end_point2; + int32_t start_point, end_point1, end_point2; fluid_real_t *coeffs; int looping; @@ -300,23 +324,23 @@ fluid_rvoice_dsp_interpolate_4th_order (fluid_rvoice_dsp_t *voice) if (voice->has_looped) /* set start_index and start point if looped or not */ { start_index = voice->loopstart; - start_point = dsp_data[voice->loopend - 1]; /* last point in loop (wrap around) */ + start_point = fluid_rvoice_get_sample(dsp_data, dsp_data24, voice->loopend - 1); /* last point in loop (wrap around) */ } else { start_index = voice->start; - start_point = dsp_data[voice->start]; /* just duplicate the point */ + start_point = fluid_rvoice_get_sample(dsp_data, dsp_data24, voice->start); /* just duplicate the point */ } /* get points off the end (loop start if looping, duplicate point if end) */ if (looping) { - end_point1 = dsp_data[voice->loopstart]; - end_point2 = dsp_data[voice->loopstart + 1]; + end_point1 = fluid_rvoice_get_sample(dsp_data, dsp_data24, voice->loopstart); + end_point2 = fluid_rvoice_get_sample(dsp_data, dsp_data24, voice->loopstart + 1); } else { - end_point1 = dsp_data[voice->end]; + end_point1 = fluid_rvoice_get_sample(dsp_data, dsp_data24, voice->end); end_point2 = end_point1; } @@ -328,10 +352,11 @@ fluid_rvoice_dsp_interpolate_4th_order (fluid_rvoice_dsp_t *voice) for ( ; dsp_phase_index == start_index && dsp_i < FLUID_BUFSIZE; dsp_i++) { coeffs = interp_coeff[fluid_phase_fract_to_tablerow (dsp_phase)]; - dsp_buf[dsp_i] = dsp_amp * (coeffs[0] * start_point - + coeffs[1] * dsp_data[dsp_phase_index] - + coeffs[2] * dsp_data[dsp_phase_index+1] - + coeffs[3] * dsp_data[dsp_phase_index+2]); + dsp_buf[dsp_i] = dsp_amp * + ( coeffs[0] * start_point + + coeffs[1] * fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index) + + coeffs[2] * fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index+1) + + coeffs[3] * fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index+2)); /* increment phase and amplitude */ fluid_phase_incr (dsp_phase, dsp_phase_incr); @@ -343,10 +368,11 @@ fluid_rvoice_dsp_interpolate_4th_order (fluid_rvoice_dsp_t *voice) for ( ; dsp_i < FLUID_BUFSIZE && dsp_phase_index <= end_index; dsp_i++) { coeffs = interp_coeff[fluid_phase_fract_to_tablerow (dsp_phase)]; - dsp_buf[dsp_i] = dsp_amp * (coeffs[0] * dsp_data[dsp_phase_index-1] - + coeffs[1] * dsp_data[dsp_phase_index] - + coeffs[2] * dsp_data[dsp_phase_index+1] - + coeffs[3] * dsp_data[dsp_phase_index+2]); + dsp_buf[dsp_i] = dsp_amp * + ( coeffs[0] * fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index-1) + + coeffs[1] * fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index) + + coeffs[2] * fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index+1) + + coeffs[3] * fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index+2)); /* increment phase and amplitude */ fluid_phase_incr (dsp_phase, dsp_phase_incr); @@ -363,10 +389,11 @@ fluid_rvoice_dsp_interpolate_4th_order (fluid_rvoice_dsp_t *voice) for (; dsp_phase_index <= end_index && dsp_i < FLUID_BUFSIZE; dsp_i++) { coeffs = interp_coeff[fluid_phase_fract_to_tablerow (dsp_phase)]; - dsp_buf[dsp_i] = dsp_amp * (coeffs[0] * dsp_data[dsp_phase_index-1] - + coeffs[1] * dsp_data[dsp_phase_index] - + coeffs[2] * dsp_data[dsp_phase_index+1] - + coeffs[3] * end_point1); + dsp_buf[dsp_i] = dsp_amp * + ( coeffs[0] * fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index-1) + + coeffs[1] * fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index) + + coeffs[2] * fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index+1) + + coeffs[3] * end_point1); /* increment phase and amplitude */ fluid_phase_incr (dsp_phase, dsp_phase_incr); @@ -380,10 +407,11 @@ fluid_rvoice_dsp_interpolate_4th_order (fluid_rvoice_dsp_t *voice) for (; dsp_phase_index <= end_index && dsp_i < FLUID_BUFSIZE; dsp_i++) { coeffs = interp_coeff[fluid_phase_fract_to_tablerow (dsp_phase)]; - dsp_buf[dsp_i] = dsp_amp * (coeffs[0] * dsp_data[dsp_phase_index-1] - + coeffs[1] * dsp_data[dsp_phase_index] - + coeffs[2] * end_point1 - + coeffs[3] * end_point2); + dsp_buf[dsp_i] = dsp_amp * + ( coeffs[0] * fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index-1) + + coeffs[1] * fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index) + + coeffs[2] * end_point1 + + coeffs[3] * end_point2); /* increment phase and amplitude */ fluid_phase_incr (dsp_phase, dsp_phase_incr); @@ -402,7 +430,7 @@ fluid_rvoice_dsp_interpolate_4th_order (fluid_rvoice_dsp_t *voice) { voice->has_looped = 1; start_index = voice->loopstart; - start_point = dsp_data[voice->loopend - 1]; + start_point = fluid_rvoice_get_sample(dsp_data, dsp_data24, voice->loopend-1); } } @@ -428,14 +456,14 @@ fluid_rvoice_dsp_interpolate_7th_order (fluid_rvoice_dsp_t *voice) fluid_phase_t dsp_phase = voice->phase; fluid_phase_t dsp_phase_incr; short int *dsp_data = voice->sample->data; + char *dsp_data24 = voice->sample->data24; fluid_real_t *dsp_buf = voice->dsp_buf; fluid_real_t dsp_amp = voice->amp; fluid_real_t dsp_amp_incr = voice->amp_incr; unsigned int dsp_i = 0; unsigned int dsp_phase_index; unsigned int start_index, end_index; - short int start_points[3]; - short int end_points[3]; + int32_t start_points[3], end_points[3]; fluid_real_t *coeffs; int looping; @@ -455,14 +483,14 @@ fluid_rvoice_dsp_interpolate_7th_order (fluid_rvoice_dsp_t *voice) if (voice->has_looped) /* set start_index and start point if looped or not */ { start_index = voice->loopstart; - start_points[0] = dsp_data[voice->loopend - 1]; - start_points[1] = dsp_data[voice->loopend - 2]; - start_points[2] = dsp_data[voice->loopend - 3]; + start_points[0] = fluid_rvoice_get_sample(dsp_data, dsp_data24, voice->loopend - 1); + start_points[1] = fluid_rvoice_get_sample(dsp_data, dsp_data24, voice->loopend - 2); + start_points[2] = fluid_rvoice_get_sample(dsp_data, dsp_data24, voice->loopend - 3); } else { start_index = voice->start; - start_points[0] = dsp_data[voice->start]; /* just duplicate the start point */ + start_points[0] = fluid_rvoice_get_sample(dsp_data, dsp_data24, voice->start); /* just duplicate the start point */ start_points[1] = start_points[0]; start_points[2] = start_points[0]; } @@ -470,13 +498,13 @@ fluid_rvoice_dsp_interpolate_7th_order (fluid_rvoice_dsp_t *voice) /* get the 3 points off the end (loop start if looping, duplicate point if end) */ if (looping) { - end_points[0] = dsp_data[voice->loopstart]; - end_points[1] = dsp_data[voice->loopstart + 1]; - end_points[2] = dsp_data[voice->loopstart + 2]; + end_points[0] = fluid_rvoice_get_sample(dsp_data, dsp_data24, voice->loopstart); + end_points[1] = fluid_rvoice_get_sample(dsp_data, dsp_data24, voice->loopstart + 1); + end_points[2] = fluid_rvoice_get_sample(dsp_data, dsp_data24, voice->loopstart + 2); } else { - end_points[0] = dsp_data[voice->end]; + end_points[0] = fluid_rvoice_get_sample(dsp_data, dsp_data24, voice->end); end_points[1] = end_points[0]; end_points[2] = end_points[0]; } @@ -494,10 +522,10 @@ fluid_rvoice_dsp_interpolate_7th_order (fluid_rvoice_dsp_t *voice) * (coeffs[0] * (fluid_real_t)start_points[2] + coeffs[1] * (fluid_real_t)start_points[1] + coeffs[2] * (fluid_real_t)start_points[0] - + coeffs[3] * (fluid_real_t)dsp_data[dsp_phase_index] - + coeffs[4] * (fluid_real_t)dsp_data[dsp_phase_index+1] - + coeffs[5] * (fluid_real_t)dsp_data[dsp_phase_index+2] - + coeffs[6] * (fluid_real_t)dsp_data[dsp_phase_index+3]); + + coeffs[3] * (fluid_real_t)fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index) + + coeffs[4] * (fluid_real_t)fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index+1) + + coeffs[5] * (fluid_real_t)fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index+2) + + coeffs[6] * (fluid_real_t)fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index+3)); /* increment phase and amplitude */ fluid_phase_incr (dsp_phase, dsp_phase_incr); @@ -515,11 +543,11 @@ fluid_rvoice_dsp_interpolate_7th_order (fluid_rvoice_dsp_t *voice) dsp_buf[dsp_i] = dsp_amp * (coeffs[0] * (fluid_real_t)start_points[1] + coeffs[1] * (fluid_real_t)start_points[0] - + coeffs[2] * (fluid_real_t)dsp_data[dsp_phase_index-1] - + coeffs[3] * (fluid_real_t)dsp_data[dsp_phase_index] - + coeffs[4] * (fluid_real_t)dsp_data[dsp_phase_index+1] - + coeffs[5] * (fluid_real_t)dsp_data[dsp_phase_index+2] - + coeffs[6] * (fluid_real_t)dsp_data[dsp_phase_index+3]); + + coeffs[2] * (fluid_real_t)fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index-1) + + coeffs[3] * (fluid_real_t)fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index) + + coeffs[4] * (fluid_real_t)fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index+1) + + coeffs[5] * (fluid_real_t)fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index+2) + + coeffs[6] * (fluid_real_t)fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index+3)); /* increment phase and amplitude */ fluid_phase_incr (dsp_phase, dsp_phase_incr); @@ -536,12 +564,12 @@ fluid_rvoice_dsp_interpolate_7th_order (fluid_rvoice_dsp_t *voice) dsp_buf[dsp_i] = dsp_amp * (coeffs[0] * (fluid_real_t)start_points[0] - + coeffs[1] * (fluid_real_t)dsp_data[dsp_phase_index-2] - + coeffs[2] * (fluid_real_t)dsp_data[dsp_phase_index-1] - + coeffs[3] * (fluid_real_t)dsp_data[dsp_phase_index] - + coeffs[4] * (fluid_real_t)dsp_data[dsp_phase_index+1] - + coeffs[5] * (fluid_real_t)dsp_data[dsp_phase_index+2] - + coeffs[6] * (fluid_real_t)dsp_data[dsp_phase_index+3]); + + coeffs[1] * (fluid_real_t)fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index-2) + + coeffs[2] * (fluid_real_t)fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index-1) + + coeffs[3] * (fluid_real_t)fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index) + + coeffs[4] * (fluid_real_t)fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index+1) + + coeffs[5] * (fluid_real_t)fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index+2) + + coeffs[6] * (fluid_real_t)fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index+3)); /* increment phase and amplitude */ fluid_phase_incr (dsp_phase, dsp_phase_incr); @@ -558,13 +586,13 @@ fluid_rvoice_dsp_interpolate_7th_order (fluid_rvoice_dsp_t *voice) coeffs = sinc_table7[fluid_phase_fract_to_tablerow (dsp_phase)]; dsp_buf[dsp_i] = dsp_amp - * (coeffs[0] * (fluid_real_t)dsp_data[dsp_phase_index-3] - + coeffs[1] * (fluid_real_t)dsp_data[dsp_phase_index-2] - + coeffs[2] * (fluid_real_t)dsp_data[dsp_phase_index-1] - + coeffs[3] * (fluid_real_t)dsp_data[dsp_phase_index] - + coeffs[4] * (fluid_real_t)dsp_data[dsp_phase_index+1] - + coeffs[5] * (fluid_real_t)dsp_data[dsp_phase_index+2] - + coeffs[6] * (fluid_real_t)dsp_data[dsp_phase_index+3]); + * (coeffs[0] * (fluid_real_t)fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index-3) + + coeffs[1] * (fluid_real_t)fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index-2) + + coeffs[2] * (fluid_real_t)fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index-1) + + coeffs[3] * (fluid_real_t)fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index) + + coeffs[4] * (fluid_real_t)fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index+1) + + coeffs[5] * (fluid_real_t)fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index+2) + + coeffs[6] * (fluid_real_t)fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index+3)); /* increment phase and amplitude */ fluid_phase_incr (dsp_phase, dsp_phase_incr); @@ -583,12 +611,12 @@ fluid_rvoice_dsp_interpolate_7th_order (fluid_rvoice_dsp_t *voice) coeffs = sinc_table7[fluid_phase_fract_to_tablerow (dsp_phase)]; dsp_buf[dsp_i] = dsp_amp - * (coeffs[0] * (fluid_real_t)dsp_data[dsp_phase_index-3] - + coeffs[1] * (fluid_real_t)dsp_data[dsp_phase_index-2] - + coeffs[2] * (fluid_real_t)dsp_data[dsp_phase_index-1] - + coeffs[3] * (fluid_real_t)dsp_data[dsp_phase_index] - + coeffs[4] * (fluid_real_t)dsp_data[dsp_phase_index+1] - + coeffs[5] * (fluid_real_t)dsp_data[dsp_phase_index+2] + * (coeffs[0] * (fluid_real_t)fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index-3) + + coeffs[1] * (fluid_real_t)fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index-2) + + coeffs[2] * (fluid_real_t)fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index-1) + + coeffs[3] * (fluid_real_t)fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index) + + coeffs[4] * (fluid_real_t)fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index+1) + + coeffs[5] * (fluid_real_t)fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index+2) + coeffs[6] * (fluid_real_t)end_points[0]); /* increment phase and amplitude */ @@ -605,11 +633,11 @@ fluid_rvoice_dsp_interpolate_7th_order (fluid_rvoice_dsp_t *voice) coeffs = sinc_table7[fluid_phase_fract_to_tablerow (dsp_phase)]; dsp_buf[dsp_i] = dsp_amp - * (coeffs[0] * (fluid_real_t)dsp_data[dsp_phase_index-3] - + coeffs[1] * (fluid_real_t)dsp_data[dsp_phase_index-2] - + coeffs[2] * (fluid_real_t)dsp_data[dsp_phase_index-1] - + coeffs[3] * (fluid_real_t)dsp_data[dsp_phase_index] - + coeffs[4] * (fluid_real_t)dsp_data[dsp_phase_index+1] + * (coeffs[0] * (fluid_real_t)fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index-3) + + coeffs[1] * (fluid_real_t)fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index-2) + + coeffs[2] * (fluid_real_t)fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index-1) + + coeffs[3] * (fluid_real_t)fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index) + + coeffs[4] * (fluid_real_t)fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index+1) + coeffs[5] * (fluid_real_t)end_points[0] + coeffs[6] * (fluid_real_t)end_points[1]); @@ -627,10 +655,10 @@ fluid_rvoice_dsp_interpolate_7th_order (fluid_rvoice_dsp_t *voice) coeffs = sinc_table7[fluid_phase_fract_to_tablerow (dsp_phase)]; dsp_buf[dsp_i] = dsp_amp - * (coeffs[0] * (fluid_real_t)dsp_data[dsp_phase_index-3] - + coeffs[1] * (fluid_real_t)dsp_data[dsp_phase_index-2] - + coeffs[2] * (fluid_real_t)dsp_data[dsp_phase_index-1] - + coeffs[3] * (fluid_real_t)dsp_data[dsp_phase_index] + * (coeffs[0] * (fluid_real_t)fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index-3) + + coeffs[1] * (fluid_real_t)fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index-2) + + coeffs[2] * (fluid_real_t)fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index-1) + + coeffs[3] * (fluid_real_t)fluid_rvoice_get_sample(dsp_data, dsp_data24, dsp_phase_index) + coeffs[4] * (fluid_real_t)end_points[0] + coeffs[5] * (fluid_real_t)end_points[1] + coeffs[6] * (fluid_real_t)end_points[2]); @@ -652,9 +680,9 @@ fluid_rvoice_dsp_interpolate_7th_order (fluid_rvoice_dsp_t *voice) { voice->has_looped = 1; start_index = voice->loopstart; - start_points[0] = dsp_data[voice->loopend - 1]; - start_points[1] = dsp_data[voice->loopend - 2]; - start_points[2] = dsp_data[voice->loopend - 3]; + start_points[0] = fluid_rvoice_get_sample(dsp_data, dsp_data24, voice->loopend - 1); + start_points[1] = fluid_rvoice_get_sample(dsp_data, dsp_data24, voice->loopend - 2); + start_points[2] = fluid_rvoice_get_sample(dsp_data, dsp_data24, voice->loopend - 3); } }