TakeV/dxx-rebirth
1
0
Fork 0
Code Issues Pull requests Packages Projects Releases Wiki Activity

Use auto & for coeffs, since it is never changed

Browse source
This commit is contained in:
Kp 2022-11-12 19:00:21 +00:00
parent 713b3baf99
commit e42cd300af
1 changed files with 5 additions and 6 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

11
similar/arch/sdl/digi_mixer.cpp
View file

@ -188,14 +188,14 @@ static int32_t coeffs_halfband[FILTER_LEN] =
// Fixed-point FIR filtering
// Not optimal: consider optimization with 1/4, 1/2 band filters, and symmetric kernels
static void filter_fir(int16_t *signal, int16_t *output, int signalLen, int32_t *coeffs, int coeffsLen)
static void filter_fir(int16_t *signal, int16_t *output, int signalLen, const int32_t (&coeffs)[FILTER_LEN])
{
// A FIR filter is just a 1D convolution
// Keep only signalLen samples
for(int nn = 0; nn < signalLen; nn++)
{
// Determin start/stop indices for convolved chunk
int min_idx = std::max(0, nn-coeffsLen+1);
int min_idx = std::max(0, nn - FILTER_LEN + 1);
int max_idx = std::min(nn, signalLen-1);
int32_t cur_output = 0; // Increase bit size for fixed point expansion
@ -254,13 +254,12 @@ static void convert_audio(uint8_t *input, int16_t *output, int inputLen, int upF
upsample(input, stage1.get(), inputLen, upFactor);
// We expect a 4x upscaling 11025 -> 44100
int32_t *coeffs = &coeffs_quarterband[0];
if(upFactor ) // But maybe 2x for d2x in some cases
coeffs = &coeffs_halfband[0];
// But maybe 2x for d2x in some cases
auto &coeffs = upFactor ? coeffs_halfband : coeffs_quarterband;
// Apply LPF filter to smooth out upscaled points
// There will be some uniform amplitude loss here, but less than -3dB
filter_fir(stage1.get(), stage2.get(), upsampledLen, coeffs, FILTER_LEN);
filter_fir(stage1.get(), stage2.get(), upsampledLen, coeffs);
// Replicate channel
replicateChannel(stage2.get(), output, upsampledLen, chFactor);