我已检测功能 AV codec_de code_audio3
作品使用MP4格式慢,在这里我的code周期解码音频:
I have detected that function avcodec_decode_audio3
works slow with mp4 format, here my code cycle for decoding audio:
while (av_read_frame(av_format_context, &packet) >= 0 && is_play == 1) {
if (av_codec_context->codec_type == AVMEDIA_TYPE_AUDIO
&& is_play == 1) {
int out_size = AVCODEC_MAX_AUDIO_FRAME_SIZE;
int size = packet.size;
int n;
int dataLength = size;
int decoded = 0;
while (size > 0) {
//start measure time
gettimeofday(&tvBegin, NULL);
int len = avcodec_decode_audio3(av_codec_context,
(int16_t *) pAudioBuffer, &out_size, &packet);
//stop measure time
gettimeofday(&tvEnd, NULL);
timeval_subtract(&tvDiff, &tvEnd, &tvBegin);
LOGI("%d", tvDiff.tv_usec / 1000);
LOGI("len='%d'", len);
LOGI("out_size='%d'", out_size);
if (len < 0) {
break;
return 1;
}
if (out_size > 0) {
jbyte *bytes = (*env)->GetByteArrayElements(env, array,
NULL);
memcpy(bytes, (int16_t *) pAudioBuffer, out_size);
(*env)->ReleaseByteArrayElements(env, array, bytes, 0);
(*env)->CallVoidMethod(env, obj, play, array, out_size,
is_play);
}
size -= len;
}
}
if (packet.data)
av_free_packet(&packet);
}
但与其他格式,如FLAC和MP3,它工作正常。 AV codec_de code_audio3
采取解码MP3架约1-2 milisecounds与 out_size
= 4608,但随着MP4的解码相同的帧大小大约需要6-7 millisecounds。我接到的这里。
But with other formats like flac and mp3 it works fine. avcodec_decode_audio3
take about 1-2 milisecounds for decoding mp3 frame with out_size
= 4608 but with the same frame size in mp4 decoding take about 6-7 millisecounds. I got my build script from here.
它是否正常行为?什么办法可以提高解码MP4的表现呢?
Does it normal behavior? Is any way to increase performance of decoding mp4?
ü可以硬件加速搜索...
u could search on hardware acceleration...
H / W加速H.264解码Android上的
"H/W Accelerated H.264 Decoding on Android"
例如:
How在Android上使用硬件加速的视频解码?
http://readlist.com/lists/mplayerhq。胡/ mplayer的用户/ 5 / 26751.html
IMO - ffmpeg的仍主要集中在软件,但在Android的新版本,你可以找到一些stagefright的API提供GPU加速的方式来访问帧缓冲??
IMO - ffmpeg is still mostly in software but on the newer versions of android, you may be able to find some stagefright API's providing GPU accelerated way to access the framebuffer??