Recherche avancée
Médias (1)
-
1 000 000 (wav version)
26 septembre 2011, par
Mis à jour : Avril 2013
Langue : English
Type : Audio
Autres articles (66)
-
Submit enhancements and plugins
13 avril 2011If you have developed a new extension to add one or more useful features to MediaSPIP, let us know and its integration into the core MedisSPIP functionality will be considered.
You can use the development discussion list to request for help with creating a plugin. As MediaSPIP is based on SPIP - or you can use the SPIP discussion list SPIP-Zone. -
Publier sur MédiaSpip
13 juin 2013Puis-je poster des contenus à partir d’une tablette Ipad ?
Oui, si votre Médiaspip installé est à la version 0.2 ou supérieure. Contacter au besoin l’administrateur de votre MédiaSpip pour le savoir -
Encoding and processing into web-friendly formats
13 avril 2011, parMediaSPIP automatically converts uploaded files to internet-compatible formats.
Video files are encoded in MP4, Ogv and WebM (supported by HTML5) and MP4 (supported by Flash).
Audio files are encoded in MP3 and Ogg (supported by HTML5) and MP3 (supported by Flash).
Where possible, text is analyzed in order to retrieve the data needed for search engine detection, and then exported as a series of image files.
All uploaded files are stored online in their original format, so you can (...)
Sur d’autres sites (5920)
-
ffmpeg - operation not permitted error while conversion
20 février 2012, par JomoosI am developing an android app. My requirement is that to implement an rtsp streaming server on android. It has to live stream video and audio captured using
MediaRecorder. Another requirement is that I have to use live555 as the streaming server. What I get fromMediaRecorderis in MP4 or 3GP format. live555 cannot able to stream both. But it can stream audio if I recorded it only in 'RAW_AMR' format. Since live555 support 'mpg' format for streaming, I decided to put someone in middle who can convert 'mp4' or '3gp' to 'mpg', and I chose ffmpeg.I have ported live555 and ffmpeg to android. ffmpeg is able to convert the file recorded by
MediaRecorderonce it is finished. But the problem is that ffmpeg cannot be able to do it concurrently. That is, ffmpeg is not able to convert the file while recording. It shows anOperation not permittederror. I tried the same on my linux machine, using VLC to record instead ofMediaRecorderon android. The result is same. ffmpeg is able to convert once the recording is finished, and not able to do the same while recording.Here is the ffmpeg command I issued on my linux box :
ffmpeg -v 9 -loglevel 99 -i test.mp4 test.mpgWhere
test.mp4is the file to which VLC is recording inmp4format. andtest.mpgis my destination file. The following is the output by ffmpeg on terminal.ffmpeg version 0.8.9, Copyright (c) 2000-2011 the FFmpeg developers
built on Feb 1 2012 18:29:27 with gcc 4.6.2 20111027 (Red Hat 4.6.2-1)
configuration: --prefix=/usr --bindir=/usr/bin --datadir=/usr/share/ffmpeg --incdir=/usr/include/ffmpeg --libdir=/usr/lib64 --mandir=/usr/share/man --arch=x86_64 --extra-cflags='-O2 -g -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector --param=ssp-buffer-size=4 -m64 -mtune=generic' --enable-bzlib --enable-libcelt --enable-libdc1394 --enable-libdirac --enable-libfreetype --enable-libgsm --enable-libmp3lame --enable-libopenjpeg --enable-librtmp --enable-libschroedinger --enable-libspeex --enable-libtheora --enable-libvorbis --enable-libvpx --enable-libx264 --enable-libxvid --enable-x11grab --enable-avfilter --enable-postproc --enable-pthreads --disable-static --enable-shared --enable-gpl --disable-debug --disable-stripping --shlibdir=/usr/lib64 --enable-runtime-cpudetect
libavutil 51. 9. 1 / 51. 9. 1
libavcodec 53. 8. 0 / 53. 8. 0
libavformat 53. 5. 0 / 53. 5. 0
libavdevice 53. 1. 1 / 53. 1. 1
libavfilter 2. 23. 0 / 2. 23. 0
libswscale 2. 0. 0 / 2. 0. 0
libpostproc 51. 2. 0 / 51. 2. 0
[mov,mp4,m4a,3gp,3g2,mj2 @ 0x1672600] Format mov,mp4,m4a,3gp,3g2,mj2 probed with size=2048 and score=100
[mov,mp4,m4a,3gp,3g2,mj2 @ 0x1672600] ISO: File Type Major Brand: isom
[mov,mp4,m4a,3gp,3g2,mj2 @ 0x1672600] moov atom not found
test.mp4: Operation not permittedWould anyone please tell me what is causing the problem ? Or is the scenario above is possible by ffmpeg. That is, is ffmpeg is able to do the conversion at the same time as that of recording ? If it is not possible by ffmpeg, would you please suggest any alternative solutions ?
NOTE : I am putting a
Ctag because if it possible by some tweaking in C on ffmpeg, I am ready to do that(I want the solution that badly). But please provide some pointers to the right direction. -
Trying to sync audio/visual using FFMpeg and openAL
22 août 2013, par user1379811hI have been studying dranger ffmpeg tutorial which explains how to sync audio and visual once you have the frames displayed and audio playing which is where im at.
Unfortunately, the tutorial is out of date (Stephen Dranger explaained that himself to me) and also uses sdl which im not doing - this is for Blackberry 10 application.
I just cannot make the video frames display at the correct speed (they are just playing very fast) and I have been trying for over a week now - seriously !
I have 3 threads happening - one to read from stream into audio and video queues and then 2 threads for audio and video.
If somebody could explain whats happening after scanning my relevent code you would be a lifesaver.
The delay (what I pass to usleep(testDelay) seems to be going up (incrementing) which doesn't seem right to me.
count = 1;
MyApp* inst = worker->app;//(VideoUploadFacebook*)arg;
qDebug() << "\n start loadstream";
w = new QWaitCondition();
w2 = new QWaitCondition();
context = avformat_alloc_context();
inst->threadStarted = true;
cout << "start of decoding thread";
cout.flush();
av_register_all();
avcodec_register_all();
avformat_network_init();
av_log_set_callback(&log_callback);
AVInputFormat *pFormat;
//const char device[] = "/dev/video0";
const char formatName[] = "mp4";
cout << "2start of decoding thread";
cout.flush();
if (!(pFormat = av_find_input_format(formatName))) {
printf("can't find input format %s\n", formatName);
//return void*;
}
//open rtsp
if(avformat_open_input(&context, inst->capturedUrl.data(), pFormat,NULL) != 0){
// return ;
cout << "error opening of decoding thread: " << inst->capturedUrl.data();
cout.flush();
}
cout << "3start of decoding thread";
cout.flush();
// av_dump_format(context, 0, inst->capturedUrl.data(), 0);
/* if(avformat_find_stream_info(context,NULL) < 0){
return EXIT_FAILURE;
}
*/
//search video stream
for(int i =0;inb_streams;i++){
if(context->streams[i]->codec->codec_type == AVMEDIA_TYPE_VIDEO)
inst->video_stream_index = i;
}
cout << "3z start of decoding thread";
cout.flush();
AVFormatContext* oc = avformat_alloc_context();
av_read_play(context);//play RTSP
AVDictionary *optionsDict = NULL;
ccontext = context->streams[inst->video_stream_index]->codec;
inst->audioc = context->streams[1]->codec;
cout << "4start of decoding thread";
cout.flush();
codec = avcodec_find_decoder(ccontext->codec_id);
ccontext->pix_fmt = PIX_FMT_YUV420P;
AVCodec* audio_codec = avcodec_find_decoder(inst->audioc->codec_id);
inst->packet = new AVPacket();
if (!audio_codec) {
cout << "audio codec not found\n"; //fflush( stdout );
exit(1);
}
if (avcodec_open2(inst->audioc, audio_codec, NULL) < 0) {
cout << "could not open codec\n"; //fflush( stdout );
exit(1);
}
if (avcodec_open2(ccontext, codec, &optionsDict) < 0) exit(1);
cout << "5start of decoding thread";
cout.flush();
inst->pic = avcodec_alloc_frame();
av_init_packet(inst->packet);
while(av_read_frame(context,inst->packet) >= 0 && &inst->keepGoing)
{
if(inst->packet->stream_index == 0){//packet is video
int check = 0;
// av_init_packet(inst->packet);
int result = avcodec_decode_video2(ccontext, inst->pic, &check, inst->packet);
if(check)
break;
}
}
inst->originalVideoWidth = inst->pic->width;
inst->originalVideoHeight = inst->pic->height;
float aspect = (float)inst->originalVideoHeight / (float)inst->originalVideoWidth;
inst->newVideoWidth = inst->originalVideoWidth;
int newHeight = (int)(inst->newVideoWidth * aspect);
inst->newVideoHeight = newHeight;//(int)inst->originalVideoHeight / inst->originalVideoWidth * inst->newVideoWidth;// = new height
int size = avpicture_get_size(PIX_FMT_YUV420P, inst->originalVideoWidth, inst->originalVideoHeight);
uint8_t* picture_buf = (uint8_t*)(av_malloc(size));
avpicture_fill((AVPicture *) inst->pic, picture_buf, PIX_FMT_YUV420P, inst->originalVideoWidth, inst->originalVideoHeight);
picrgb = avcodec_alloc_frame();
int size2 = avpicture_get_size(PIX_FMT_YUV420P, inst->newVideoWidth, inst->newVideoHeight);
uint8_t* picture_buf2 = (uint8_t*)(av_malloc(size2));
avpicture_fill((AVPicture *) picrgb, picture_buf2, PIX_FMT_YUV420P, inst->newVideoWidth, inst->newVideoHeight);
if(ccontext->pix_fmt != PIX_FMT_YUV420P)
{
std::cout << "fmt != 420!!!: " << ccontext->pix_fmt << std::endl;//
// return (EXIT_SUCCESS);//-1;
}
if (inst->createForeignWindow(inst->myForeignWindow->windowGroup(),
"HelloForeignWindowAppIDqq", 0,
0, inst->newVideoWidth,
inst->newVideoHeight)) {
} else {
qDebug() << "The ForeginWindow was not properly initialized";
}
inst->keepGoing = true;
inst->img_convert_ctx = sws_getContext(inst->originalVideoWidth, inst->originalVideoHeight, PIX_FMT_YUV420P, inst->newVideoWidth, inst->newVideoHeight,
PIX_FMT_YUV420P, SWS_BILINEAR, NULL, NULL, NULL);
is = (VideoState*)av_mallocz(sizeof(VideoState));
if (!is)
return NULL;
is->audioStream = 1;
is->audio_st = context->streams[1];
is->audio_buf_size = 0;
is->audio_buf_index = 0;
is->videoStream = 0;
is->video_st = context->streams[0];
is->frame_timer = (double)av_gettime() / 1000000.0;
is->frame_last_delay = 40e-3;
is->av_sync_type = DEFAULT_AV_SYNC_TYPE;
//av_strlcpy(is->filename, filename, sizeof(is->filename));
is->iformat = pFormat;
is->ytop = 0;
is->xleft = 0;
/* start video display */
is->pictq_mutex = new QMutex();
is->pictq_cond = new QWaitCondition();
is->subpq_mutex = new QMutex();
is->subpq_cond = new QWaitCondition();
is->video_current_pts_time = av_gettime();
packet_queue_init(&audioq);
packet_queue_init(&videoq);
is->audioq = audioq;
is->videoq = videoq;
AVPacket* packet2 = new AVPacket();
ccontext->get_buffer = our_get_buffer;
ccontext->release_buffer = our_release_buffer;
av_init_packet(packet2);
while(inst->keepGoing)
{
if(av_read_frame(context,packet2) < 0 && keepGoing)
{
printf("bufferframe Could not read a frame from stream.\n");
fflush( stdout );
}else {
if(packet2->stream_index == 0) {
packet_queue_put(&videoq, packet2);
} else if(packet2->stream_index == 1) {
packet_queue_put(&audioq, packet2);
} else {
av_free_packet(packet2);
}
if(!videoThreadStarted)
{
videoThreadStarted = true;
QThread* thread = new QThread;
videoThread = new VideoStreamWorker(this);
// Give QThread ownership of Worker Object
videoThread->moveToThread(thread);
connect(videoThread, SIGNAL(error(QString)), this, SLOT(errorHandler(QString)));
QObject::connect(videoThread, SIGNAL(refreshNeeded()), this, SLOT(refreshNeededSlot()));
connect(thread, SIGNAL(started()), videoThread, SLOT(doWork()));
connect(videoThread, SIGNAL(finished()), thread, SLOT(quit()));
connect(videoThread, SIGNAL(finished()), videoThread, SLOT(deleteLater()));
connect(thread, SIGNAL(finished()), thread, SLOT(deleteLater()));
thread->start();
}
if(!audioThreadStarted)
{
audioThreadStarted = true;
QThread* thread = new QThread;
AudioStreamWorker* videoThread = new AudioStreamWorker(this);
// Give QThread ownership of Worker Object
videoThread->moveToThread(thread);
// Connect videoThread error signal to this errorHandler SLOT.
connect(videoThread, SIGNAL(error(QString)), this, SLOT(errorHandler(QString)));
// Connects the thread’s started() signal to the process() slot in the videoThread, causing it to start.
connect(thread, SIGNAL(started()), videoThread, SLOT(doWork()));
connect(videoThread, SIGNAL(finished()), thread, SLOT(quit()));
connect(videoThread, SIGNAL(finished()), videoThread, SLOT(deleteLater()));
// Make sure the thread object is deleted after execution has finished.
connect(thread, SIGNAL(finished()), thread, SLOT(deleteLater()));
thread->start();
}
}
} //finished main loop
int MyApp::video_thread() {
//VideoState *is = (VideoState *)arg;
AVPacket pkt1, *packet = &pkt1;
int len1, frameFinished;
double pts;
pic = avcodec_alloc_frame();
for(;;) {
if(packet_queue_get(&videoq, packet, 1) < 0) {
// means we quit getting packets
break;
}
pts = 0;
global_video_pkt_pts2 = packet->pts;
// Decode video frame
len1 = avcodec_decode_video2(ccontext, pic, &frameFinished, packet);
if(packet->dts == AV_NOPTS_VALUE
&& pic->opaque && *(uint64_t*)pic->opaque != AV_NOPTS_VALUE) {
pts = *(uint64_t *)pic->opaque;
} else if(packet->dts != AV_NOPTS_VALUE) {
pts = packet->dts;
} else {
pts = 0;
}
pts *= av_q2d(is->video_st->time_base);
// Did we get a video frame?
if(frameFinished) {
pts = synchronize_video(is, pic, pts);
actualPts = pts;
refreshSlot();
}
av_free_packet(packet);
}
av_free(pic);
return 0;
}
int MyApp::audio_thread() {
//VideoState *is = (VideoState *)arg;
AVPacket pkt1, *packet = &pkt1;
int len1, frameFinished;
ALuint source;
ALenum format = 0;
// ALuint frequency;
ALenum alError;
ALint val2;
ALuint buffers[NUM_BUFFERS];
int dataSize;
ALCcontext *aContext;
ALCdevice *device;
if (!alutInit(NULL, NULL)) {
// printf(stderr, "init alut error\n");
}
device = alcOpenDevice(NULL);
if (device == NULL) {
// printf(stderr, "device error\n");
}
//Create a context
aContext = alcCreateContext(device, NULL);
alcMakeContextCurrent(aContext);
if(!(aContext)) {
printf("Could not create the OpenAL context!\n");
return 0;
}
alListener3f(AL_POSITION, 0.0f, 0.0f, 0.0f);
//ALenum alError;
if(alGetError() != AL_NO_ERROR) {
cout << "could not create buffers";
cout.flush();
fflush( stdout );
return 0;
}
alGenBuffers(NUM_BUFFERS, buffers);
alGenSources(1, &source);
if(alGetError() != AL_NO_ERROR) {
cout << "after Could not create buffers or the source.\n";
cout.flush( );
return 0;
}
int i;
int indexOfPacket;
double pts;
//double pts;
int n;
for(i = 0; i < NUM_BUFFERS; i++)
{
if(packet_queue_get(&audioq, packet, 1) < 0) {
// means we quit getting packets
break;
}
cout << "streamindex=audio \n";
cout.flush( );
//printf("before decode audio\n");
//fflush( stdout );
// AVPacket *packet = new AVPacket();//malloc(sizeof(AVPacket*));
AVFrame *decodedFrame = NULL;
int gotFrame = 0;
// AVFrame* decodedFrame;
if(!decodedFrame) {
if(!(decodedFrame = avcodec_alloc_frame())) {
cout << "Run out of memory, stop the streaming...\n";
fflush( stdout );
cout.flush();
return -2;
}
} else {
avcodec_get_frame_defaults(decodedFrame);
}
int len = avcodec_decode_audio4(audioc, decodedFrame, &gotFrame, packet);
if(len < 0) {
cout << "Error while decoding.\n";
cout.flush( );
return -3;
}
if(len < 0) {
/* if error, skip frame */
is->audio_pkt_size = 0;
//break;
}
is->audio_pkt_data += len;
is->audio_pkt_size -= len;
pts = is->audio_clock;
// *pts_ptr = pts;
n = 2 * is->audio_st->codec->channels;
is->audio_clock += (double)packet->size/
(double)(n * is->audio_st->codec->sample_rate);
if(gotFrame) {
cout << "got audio frame.\n";
cout.flush( );
// We have a buffer ready, send it
dataSize = av_samples_get_buffer_size(NULL, audioc->channels,
decodedFrame->nb_samples, audioc->sample_fmt, 1);
if(!format) {
if(audioc->sample_fmt == AV_SAMPLE_FMT_U8 ||
audioc->sample_fmt == AV_SAMPLE_FMT_U8P) {
if(audioc->channels == 1) {
format = AL_FORMAT_MONO8;
} else if(audioc->channels == 2) {
format = AL_FORMAT_STEREO8;
}
} else if(audioc->sample_fmt == AV_SAMPLE_FMT_S16 ||
audioc->sample_fmt == AV_SAMPLE_FMT_S16P) {
if(audioc->channels == 1) {
format = AL_FORMAT_MONO16;
} else if(audioc->channels == 2) {
format = AL_FORMAT_STEREO16;
}
}
if(!format) {
cout << "OpenAL can't open this format of sound.\n";
cout.flush( );
return -4;
}
}
printf("albufferdata audio b4.\n");
fflush( stdout );
alBufferData(buffers[i], format, *decodedFrame->data, dataSize, decodedFrame->sample_rate);
cout << "after albufferdata all buffers \n";
cout.flush( );
av_free_packet(packet);
//=av_free(packet);
av_free(decodedFrame);
if((alError = alGetError()) != AL_NO_ERROR) {
printf("Error while buffering.\n");
printAlError(alError);
return -6;
}
}
}
cout << "before quoe buffers \n";
cout.flush();
alSourceQueueBuffers(source, NUM_BUFFERS, buffers);
cout << "before play.\n";
cout.flush();
alSourcePlay(source);
cout << "after play.\n";
cout.flush();
if((alError = alGetError()) != AL_NO_ERROR) {
cout << "error strating stream.\n";
cout.flush();
printAlError(alError);
return 0;
}
// AVPacket *pkt = &is->audio_pkt;
while(keepGoing)
{
while(packet_queue_get(&audioq, packet, 1) >= 0) {
// means we quit getting packets
do {
alGetSourcei(source, AL_BUFFERS_PROCESSED, &val2);
usleep(SLEEP_BUFFERING);
} while(val2 <= 0);
if(alGetError() != AL_NO_ERROR)
{
fprintf(stderr, "Error gettingsource :(\n");
return 1;
}
while(val2--)
{
ALuint buffer;
alSourceUnqueueBuffers(source, 1, &buffer);
if(alGetError() != AL_NO_ERROR)
{
fprintf(stderr, "Error unqueue buffers :(\n");
// return 1;
}
AVFrame *decodedFrame = NULL;
int gotFrame = 0;
// AVFrame* decodedFrame;
if(!decodedFrame) {
if(!(decodedFrame = avcodec_alloc_frame())) {
cout << "Run out of memory, stop the streaming...\n";
//fflush( stdout );
cout.flush();
return -2;
}
} else {
avcodec_get_frame_defaults(decodedFrame);
}
int len = avcodec_decode_audio4(audioc, decodedFrame, &gotFrame, packet);
if(len < 0) {
cout << "Error while decoding.\n";
cout.flush( );
is->audio_pkt_size = 0;
return -3;
}
is->audio_pkt_data += len;
is->audio_pkt_size -= len;
if(packet->size <= 0) {
/* No data yet, get more frames */
//continue;
}
if(gotFrame) {
pts = is->audio_clock;
len = synchronize_audio(is, (int16_t *)is->audio_buf,
packet->size, pts);
is->audio_buf_size = packet->size;
pts = is->audio_clock;
// *pts_ptr = pts;
n = 2 * is->audio_st->codec->channels;
is->audio_clock += (double)packet->size /
(double)(n * is->audio_st->codec->sample_rate);
if(packet->pts != AV_NOPTS_VALUE) {
is->audio_clock = av_q2d(is->audio_st->time_base)*packet->pts;
}
len = av_samples_get_buffer_size(NULL, audioc->channels,
decodedFrame->nb_samples, audioc->sample_fmt, 1);
alBufferData(buffer, format, *decodedFrame->data, len, decodedFrame->sample_rate);
if(alGetError() != AL_NO_ERROR)
{
fprintf(stderr, "Error buffering :(\n");
return 1;
}
alSourceQueueBuffers(source, 1, &buffer);
if(alGetError() != AL_NO_ERROR)
{
fprintf(stderr, "Error queueing buffers :(\n");
return 1;
}
}
}
alGetSourcei(source, AL_SOURCE_STATE, &val2);
if(val2 != AL_PLAYING)
alSourcePlay(source);
}
//pic = avcodec_alloc_frame();
}
qDebug() << "end audiothread";
return 1;
}
void MyApp::refreshSlot()
{
if(true)
{
printf("got frame %d, %d\n", pic->width, ccontext->width);
fflush( stdout );
sws_scale(img_convert_ctx, (const uint8_t **)pic->data, pic->linesize,
0, originalVideoHeight, &picrgb->data[0], &picrgb->linesize[0]);
printf("rescaled frame %d, %d\n", newVideoWidth, newVideoHeight);
fflush( stdout );
//av_free_packet(packet);
//av_init_packet(packet);
qDebug() << "waking audio as video finished";
////mutex.unlock();
//mutex2.lock();
doingVideoFrame = false;
//doingAudioFrame = false;
////mutex2.unlock();
//mutex2.unlock();
//w2->wakeAll();
//w->wakeAll();
qDebug() << "now woke audio";
//pic = picrgb;
uint8_t *srcy = picrgb->data[0];
uint8_t *srcu = picrgb->data[1];
uint8_t *srcv = picrgb->data[2];
printf("got src yuv frame %d\n", &srcy);
fflush( stdout );
unsigned char *ptr = NULL;
screen_get_buffer_property_pv(mScreenPixelBuffer, SCREEN_PROPERTY_POINTER, (void**) &ptr);
unsigned char *y = ptr;
unsigned char *u = y + (newVideoHeight * mStride) ;
unsigned char *v = u + (newVideoHeight * mStride) / 4;
int i = 0;
printf("got buffer picrgbwidth= %d \n", newVideoWidth);
fflush( stdout );
for ( i = 0; i < newVideoHeight; i++)
{
int doff = i * mStride;
int soff = i * picrgb->linesize[0];
memcpy(&y[doff], &srcy[soff], newVideoWidth);
}
for ( i = 0; i < newVideoHeight / 2; i++)
{
int doff = i * mStride / 2;
int soff = i * picrgb->linesize[1];
memcpy(&u[doff], &srcu[soff], newVideoWidth / 2);
}
for ( i = 0; i < newVideoHeight / 2; i++)
{
int doff = i * mStride / 2;
int soff = i * picrgb->linesize[2];
memcpy(&v[doff], &srcv[soff], newVideoWidth / 2);
}
printf("before posttoscreen \n");
fflush( stdout );
video_refresh_timer();
qDebug() << "end refreshslot";
}
else
{
}
}
void MyApp::refreshNeededSlot2()
{
printf("blitting to buffer");
fflush(stdout);
screen_buffer_t screen_buffer;
screen_get_window_property_pv(mScreenWindow, SCREEN_PROPERTY_RENDER_BUFFERS, (void**) &screen_buffer);
int attribs[] = { SCREEN_BLIT_SOURCE_WIDTH, newVideoWidth, SCREEN_BLIT_SOURCE_HEIGHT, newVideoHeight, SCREEN_BLIT_END };
int res2 = screen_blit(mScreenCtx, screen_buffer, mScreenPixelBuffer, attribs);
printf("dirty rectangles");
fflush(stdout);
int dirty_rects[] = { 0, 0, newVideoWidth, newVideoHeight };
screen_post_window(mScreenWindow, screen_buffer, 1, dirty_rects, 0);
printf("done screneposdtwindow");
fflush(stdout);
}
void MyApp::video_refresh_timer() {
testDelay = 0;
// VideoState *is = ( VideoState* )userdata;
VideoPicture *vp;
//double pts = 0 ;
double actual_delay, delay, sync_threshold, ref_clock, diff;
if(is->video_st) {
if(false)////is->pictq_size == 0)
{
testDelay = 1;
schedule_refresh(is, 1);
} else {
// vp = &is->pictq[is->pictq_rindex];
delay = actualPts - is->frame_last_pts; /* the pts from last time */
if(delay <= 0 || delay >= 1.0) {
/* if incorrect delay, use previous one */
delay = is->frame_last_delay;
}
/* save for next time */
is->frame_last_delay = delay;
is->frame_last_pts = actualPts;
is->video_current_pts = actualPts;
is->video_current_pts_time = av_gettime();
/* update delay to sync to audio */
ref_clock = get_audio_clock(is);
diff = actualPts - ref_clock;
/* Skip or repeat the frame. Take delay into account
FFPlay still doesn't "know if this is the best guess." */
sync_threshold = (delay > AV_SYNC_THRESHOLD) ? delay : AV_SYNC_THRESHOLD;
if(fabs(diff) < AV_NOSYNC_THRESHOLD) {
if(diff <= -sync_threshold) {
delay = 0;
} else if(diff >= sync_threshold) {
delay = 2 * delay;
}
}
is->frame_timer += delay;
/* computer the REAL delay */
actual_delay = is->frame_timer - (av_gettime() / 1000000.0);
if(actual_delay < 0.010) {
/* Really it should skip the picture instead */
actual_delay = 0.010;
}
testDelay = (int)(actual_delay * 1000 + 0.5);
schedule_refresh(is, (int)(actual_delay * 1000 + 0.5));
/* show the picture! */
//video_display(is);
// SDL_CondSignal(is->pictq_cond);
// SDL_UnlockMutex(is->pictq_mutex);
}
} else {
testDelay = 100;
schedule_refresh(is, 100);
}
}
void MyApp::schedule_refresh(VideoState *is, int delay) {
qDebug() << "start schedule refresh timer" << delay;
typeOfEvent = FF_REFRESH_EVENT2;
w->wakeAll();
// SDL_AddTimer(delay,
}I am currently waiting on data in a loop in the following way
QMutex mutex;
mutex.lock();
while(keepGoing)
{
qDebug() << "MAINTHREAD" << testDelay;
w->wait(&mutex);
mutex.unlock();
qDebug() << "MAINTHREAD past wait";
if(!keepGoing)
{
break;
}
if(testDelay > 0 && typeOfEvent == FF_REFRESH_EVENT2)
{
usleep(testDelay);
refreshNeededSlot2();
}
else if(testDelay > 0 && typeOfEvent == FF_QUIT_EVENT2)
{
keepGoing = false;
exit(0);
break;
// usleep(testDelay);
// refreshNeededSlot2();
}
qDebug() << "MAINTHREADend";
mutex.lock();
}
mutex.unlock();Please let me know if I need to provide any more relevent code. I'm sorry my code is untidy - I still learning c++ and have been modifying this code for over a week now as previously mentioned.
Just added a sample of output I'm seeing from print outs I do to console - I can't get my head around it (it's almost too complicated for my level of expertise) but when you see the frames being played and audio playing it's very difficult to give up especially when it took me a couple of weeks to get to this stage.
Please someone give me a hand if they spot the problem.
MAINTHREAD past wait
pts after syncvideo= 1073394046
got frame 640, 640
start video_refresh_timer
actualpts = 1.66833
frame lastpts = 1.63497
start schedule refresh timer need to delay for 123pts after syncvideo= 1073429033
got frame 640, 640
MAINTHREAD loop delay before refresh = 123
start video_refresh_timer
actualpts = 1.7017
frame lastpts = 1.66833
start schedule refresh timer need to delay for 115MAINTHREAD past wait
pts after syncvideo= 1073464021
got frame 640, 640
start video_refresh_timer
actualpts = 1.73507
frame lastpts = 1.7017
start schedule refresh timer need to delay for 140MAINTHREAD loop delay before refresh = 140
pts after syncvideo= 1073499008
got frame 640, 640
start video_refresh_timer
actualpts = 1.76843
frame lastpts = 1.73507
start schedule refresh timer need to delay for 163MAINTHREAD past wait
pts after syncvideo= 1073533996
got frame 640, 640
start video_refresh_timer
actualpts = 1.8018
frame lastpts = 1.76843
start schedule refresh timer need to delay for 188MAINTHREAD loop delay before refresh = 188
pts after syncvideo= 1073568983
got frame 640, 640
start video_refresh_timer
actualpts = 1.83517
frame lastpts = 1.8018
start schedule refresh timer need to delay for 246MAINTHREAD past wait
pts after syncvideo= 1073603971
got frame 640, 640
start video_refresh_timer
actualpts = 1.86853
frame lastpts = 1.83517
start schedule refresh timer need to delay for 299MAINTHREAD loop delay before refresh = 299
pts after syncvideo= 1073638958
got frame 640, 640
start video_refresh_timer
actualpts = 1.9019
frame lastpts = 1.86853
start schedule refresh timer need to delay for 358MAINTHREAD past wait
pts after syncvideo= 1073673946
got frame 640, 640
start video_refresh_timer
actualpts = 1.93527
frame lastpts = 1.9019
start schedule refresh timer need to delay for 416MAINTHREAD loop delay before refresh = 416
pts after syncvideo= 1073708933
got frame 640, 640
start video_refresh_timer
actualpts = 1.96863
frame lastpts = 1.93527
start schedule refresh timer need to delay for 474MAINTHREAD past wait
pts after syncvideo= 1073742872
got frame 640, 640
MAINTHREAD loop delay before refresh = 474
start video_refresh_timer
actualpts = 2.002
frame lastpts = 1.96863
start schedule refresh timer need to delay for 518MAINTHREAD past wait
pts after syncvideo= 1073760366
got frame 640, 640
start video_refresh_timer
actualpts = 2.03537
frame lastpts = 2.002
start schedule refresh timer need to delay for 575 -
Revision 9f5fd31d7f : Expand UMV border to 96 pixels Ensures that the full 64 pixel border is availab
29 avril 2013, par John KoleszarChanged Paths :
Modify /vpx_scale/yv12config.h
Expand UMV border to 96 pixelsEnsures that the full 64 pixel border is available for prediction (need a
minimum of
64+INTERP_EXTEND on all sides, and 32+INTERP_EXTEND on UV). Value also must be a
multiple of 32 to keep UV stride alignment. The smaller border was causing the
prediction
to read outside the frame, which can cause a mismatch.TODO : Get rid of this explicit border and use edge emulation instead.
Change-Id : I3f68453a088ec0ab4349d0f5cc02b573be06d7c4
Syndiquer tout le site
