AudiogeneratorMOD: use full 16bit with slightly better accuracy (#479)

* Fix a tricky bug that causes delays in playback

... this bug was already present in stellaplayer. Obviously "effectNumber" must be checked, because "effectParameterX" just holds the "upper nibble" of effectParameter

* fix for issue #474

- do channel mixing in 32bit, to avoid noise from over/underflow
- write audio output as "signed int" (not unsigned)
- add int16 saturation check ("clipping")

* avoid crash when MOD file has to many CHANNELS

- avoid guru meditation, by logging an error and refusing playback in case MOD.numberOfChannels > CHANNELS
- correct small typo

* internal DAC: avoid possible overflow in conversion int16 -> uint16

* update constants

.. we can do 16bit and 8 channels, so let's do it ;-)

This adds some debugging code to better understand what is going on inside the mixer code (AudioGeneratorMOD::GetSample). And it also prints out some usefull information from the MOD file.

Only active when do_MIXER_DEBUG is defined.

* MOD generator: 12 bits of "real" resolution (instead of 10)

With this change, we gain 2 additional bits of "real resolution" from the sample interpolation step.
Still need some testing to be sure that nothing get "lost" at the same time.

* limit to 4 channels on ESP8266

As discussed in https://github.com/earlephilhower/ESP8266Audio/pull/479#issuecomment-1007708703
This commit is contained in:
Frank authored and GitHub committed 2022-01-09 08:39:34 -08:00
1 parent c55d240ade
commit 7a82ba2ab3
2 files changed
+25 -8

No files matched your search

+17 -6
View File
@@ -69,6 +69,11 @@ bool AudioGeneratorMOD::stop()
free(FatBuffer.channels[i]);
FatBuffer.channels[i] = NULL;
}
if(running && (file != NULL) && (file->isOpen() == true)) {
output->flush(); //flush I2S output buffer, if the player was actually running before.
}
if (file) file->close();
running = false;
output->stop();
@@ -803,14 +808,20 @@ void AudioGeneratorMOD::GetSample(int16_t sample[2])
current = FatBuffer.channels[channel][(samplePointer - FatBuffer.samplePointer[channel]) /*& (FATBUFFERSIZE - 1)*/];
next = FatBuffer.channels[channel][(samplePointer + 1 - FatBuffer.samplePointer[channel]) /*& (FATBUFFERSIZE - 1)*/];
out = current;
// preserve a few more bits from sample interpolation, by upscaling input values.
// This does (slightly) reduce quantization noise in higher frequencies, typically above 8kHz.
// Actually we could could even gain more bits, I was just not sure if more bits would cause overflows in other conputations.
int16_t current16 = (int16_t) current << 2;
int16_t next16 = (int16_t) next << 2;
out = current16;
// Integer linear interpolation - only works correctly in 16bit
out += (next - current) * (Mixer.channelSampleOffset[channel] & ((1 << FIXED_DIVIDER) - 1)) >> FIXED_DIVIDER;
out += (next16 - current16) * (Mixer.channelSampleOffset[channel] & ((1 << FIXED_DIVIDER) - 1)) >> FIXED_DIVIDER;
// Upscale to BITDEPTH
out32 = (int32_t)out << (BITDEPTH - 8);
// Upscale to BITDEPTH, considering the we already gained two bits in the previous step
out32 = (int32_t)out << (BITDEPTH - 10);
// Channel volume
out32 = out32 * Mixer.channelVolume[channel] >> 6;
@@ -819,7 +830,7 @@ void AudioGeneratorMOD::GetSample(int16_t sample[2])
sumL += out32 * min(128 - Mixer.channelPanning[channel], 64) >> 6;
sumR += out32 * min(Mixer.channelPanning[channel], 64) >> 6;
}
// Downscale to BITDEPTH - a bit faster because the compiler can replaced division by constants with proper "right shift" + correct handling of sign bit
if (Mod.numberOfChannels <= 4) {
// up to 4 channels