RGBCube/serenity
1
Fork 0
mirror of https://github.com/RGBCube/serenity synced 2025-05-25 18:55:08 +00:00
Code Issues Activity Actions

LibAudio/Piano: Replace floats with doubles

Browse source 
We should default to double-precision so that clients can make the
choice to use float or double.
This commit is contained in:
William McPherson 2020-02-10 23:45:10 +11:00 • committed by Andreas Kling
parent d55d2b2794
commit aa149b9330
5 changed files with 30 additions and 30 deletions
Download patch file Download diff file Expand all files Collapse all files

10
Applications/Piano/AudioEngine.cpp
View file

@ -158,10 +158,10 @@ String AudioEngine::set_recorded_sample(const StringView& path)
m_recorded_sample.clear();
m_recorded_sample.resize(wav_buffer->sample_count());
float peak = 0;
double peak = 0;
for (int i = 0; i < wav_buffer->sample_count(); ++i) {
float left_abs = fabs(wav_buffer->samples()[i].left);
float right_abs = fabs(wav_buffer->samples()[i].right);
double left_abs = fabs(wav_buffer->samples()[i].left);
double right_abs = fabs(wav_buffer->samples()[i].right);
if (left_abs > peak)
peak = left_abs;
if (right_abs > peak)
@ -228,8 +228,8 @@ Audio::Sample AudioEngine::recorded_sample(size_t note)
int t = m_pos[note];
if (t >= m_recorded_sample.size())
return 0;
float w_left = m_recorded_sample[t].left;
float w_right = m_recorded_sample[t].right;
double w_left = m_recorded_sample[t].left;
double w_right = m_recorded_sample[t].right;
if (t + 1 < m_recorded_sample.size()) {
double t_fraction = m_pos[note] - t;
w_left += (m_recorded_sample[t + 1].left - m_recorded_sample[t].left) * t_fraction;