Mirrored_Repos/LookingGlass
1
0
Fork 0
mirror of https://github.com/gnif/LookingGlass.git synced 2024-11-10 00:28:20 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

Revert "[client] audio: tune the target latency based on the latency jitter"

Browse Source 
This reverts commit febd081202.

This causes severe underruns when the quantum size increases.
This commit is contained in:
Chris Spencer 2022-01-29 09:54:52 +00:00 committed by Geoffrey McRae
parent 35bf30910b
commit ca29fe80a6
1 changed files with 5 additions and 35 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

40
client/src/audio.c
View File

@ -102,7 +102,6 @@ typedef struct
RingBuffer timings;
GraphHandle graph;
float jitter;
// These two structs contain data specifically for use in the device and
// Spice data threads respectively. Keep them on separate cache lines to
@ -173,8 +172,8 @@ static const char * audioGraphFormatFn(const char * name,
{
static char title[64];
snprintf(title, sizeof(title),
"%s: min:%4.2f max:%4.2f avg:%4.2f now:%4.2f jitter:%4.2f",
name, min, max, avg, last, max - min);
"%s: min:%4.2f max:%4.2f avg:%4.2f now:%4.2f",
name, min, max, avg, last);
return title;
}
@ -306,7 +305,6 @@ void audio_playbackStart(int channels, int sampleRate, PSAudioFormat format,
audio.playback.sampleRate = sampleRate;
audio.playback.stride = channels * sizeof(float);
audio.playback.state = STREAM_STATE_SETUP;
audio.playback.jitter = 60.0f; //assume 60ms of jitter initially
audio.playback.deviceData.periodFrames = 0;
audio.playback.deviceData.nextPosition = 0;
@ -331,10 +329,8 @@ void audio_playbackStart(int channels, int sampleRate, PSAudioFormat format,
if (audio.audioDev->playback.mute)
audio.audioDev->playback.mute(audio.playback.mute);
// timings for jitter calculations and display graph
// spice operates on a period size of (sampleRate / 100), so allocate enough
// room for 4 seconds of timing samples.
audio.playback.timings = ringbuffer_new(sampleRate / 100, sizeof(float));
// if the audio dev can report it's latency setup a timing graph
audio.playback.timings = ringbuffer_new(1200, sizeof(float));
audio.playback.graph = app_registerGraph("PLAYBACK",
audio.playback.timings, 0.0f, 100.0f, audioGraphFormatFn);
@ -379,22 +375,6 @@ void audio_playbackMute(bool mute)
audio.audioDev->playback.mute(mute);
}
static bool getMinMax(int index, void * value, void * udata)
{
float ms = *(float *)value;
float * minMax = (float *)udata;
if (index == 0)
{
minMax[0] = minMax[1] = ms;
return true;
}
minMax[0] = min(minMax[0], ms);
minMax[1] = max(minMax[1], ms);
return true;
}
void audio_playbackData(uint8_t * data, size_t size)
{
if (!audio.audioDev || size == 0)
@ -518,7 +498,7 @@ void audio_playbackData(uint8_t * data, size_t size)
// device period size, but that would result in underruns if the period size
// suddenly increases. It may be better instead to just reduce the maximum
// latency on the audio devices, which currently is set quite high
int targetLatencyMs = ceil(audio.playback.jitter);
int targetLatencyMs = 70;
int targetLatencyFrames =
targetLatencyMs * audio.playback.sampleRate / 1000;
@ -585,16 +565,6 @@ void audio_playbackData(uint8_t * data, size_t size)
const float latency = latencyFrames /
(float)(audio.playback.sampleRate / 1000);
ringbuffer_push(audio.playback.timings, &latency);
// if the ringbuffer is full calculate the jitter
if (ringbuffer_getCount(audio.playback.timings) ==
ringbuffer_getLength(audio.playback.timings))
{
float minMax[2];
ringbuffer_forEach(audio.playback.timings, getMinMax, minMax, false);
audio.playback.jitter = minMax[1] - minMax[0];
}
app_invalidateGraph(audio.playback.graph);
}