[SPAM] KONYA'NIN EN BÜYÜK FİRMA REHBERİNE KAYIT OLUN
KONYANIN EN BUYUK FiRMA REHBERiNE KAYIT OLUN www.buyukkonyafirmarehberi.com -- The Planet: dedicated and managed hosting, cloud storage, colocation Stay online with enterprise data centers and the best network in the business Choose flexible plans and management services without long-term contracts Personal 24x7 support from experience hosting pros just a phone call away. http://p.sf.net/sfu/theplanet-com___ Musicpd-dev-team mailing list Musicpd-dev-team@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/musicpd-dev-team
[Musicpd-dev-team] MPD 24-bit audio output as 3-byte samples
Hi to all
Today Max Kellerman confirmed to me that mpd is capable of sending
24-bit audio output in true 3-byte sample format (so not the padded 4
byte samples) without the use of ALSA to do any conversions.
As far as I understand it, this native mpd-support for 3-byte samples
is for the S24_3LE format.
I got his answer as a reply to my request to support true 3-byte sample
output I posted on the Mantis bugtracker.
Max replied he did not find Mantis suitable to ask questions, so I
still have a question remaining I wish to post to you here.
How do I configure mpd to make it send 24-bit audio as true 3-byte
samples without using ALSA?
I could not find any examples on the mpd-wiki configuration page. I
assume that mpd is also still capable of sending the 32-bit samples
with the 24-bit audio padded with zeroes. This would mean there has to
be a switch/parameter indicating that a 24-bit sample size is wanted
instead of 32-bit, but I don't know which one that would be.
The audio_output section of the mpd.config now looks like this:
audio_output {
type "smp86xx"
load "/share/Apps/mpd/plugins/smp86xx_plugin.so"
name "My Sigma Chip"
engine "0" # optional
decoder "1" # optional
ramdac "1" # optional
line_count "2048" # optional
fifo_us "50" # optional
hdmi "sii9134"
}
This belongs to mpd as implemented on the PopcornHour mediaplayer with
a plugin developed to send the mpd-output straight into the Sigma
Design SMP86xx processor. As you can see ALSA is not used here as the
output is sent directly to the processor.
This processor requires 24-bit audio to be sent in true unpacked 3
bytes per sample little endian format, as S24_3LE would be as I
understand it, or in a DVD PCM 24-bit linear PCM as described here ( http://wiki.multimedia.cx/index.php?title=PCM#24-Bit_PCM
) which is somewhat more difficult to implement and also limited to
96kHz sample rate.
But Max confirmed that 24-bit audio output in true 3-byte sample format
is available as a native format from mpd, so that would be the best one.
So can anyone instruct me how to make mpd output 24-bit audio as true
3-byte samples without the use of ALSA?
Eagerly looking forward to your reply
-
aleg
--
The Planet: dedicated and managed hosting, cloud storage, colocation
Stay online with enterprise data centers and the best network in the business
Choose flexible plans and management services without long-term contracts
Personal 24x7 support from experience hosting pros just a phone call away.
http://p.sf.net/sfu/theplanet-com___
Musicpd-dev-team mailing list
Musicpd-dev-team@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/musicpd-dev-team
[Musicpd-dev-team] MPD 24-bit audio output as 3-byte samples
Sorry for sending a HTML-mail, I had not thought about the effect that
would have when presented in the list.
So this one as a plain text one.
Hi to all
Today Max Kellerman confirmed to me that mpd is capable of sending
24-bit audio output in true 3-byte sample format (so not the padded 4
byte samples) without the use of ALSA to do any conversions.
As far as I understand it, this native mpd-support for 3-byte samples is
for the S24_3LE format.
I got his answer as a reply to my request to support true 3-byte sample
output I posted on the Mantis bugtracker.
Max replied he did not find Mantis suitable to ask questions, so I still
have a question remaining I wish to post to you here.
How do I configure mpd to make it send 24-bit audio as true 3-byte
samples without using ALSA?
I could not find any examples on the mpd-wiki configuration page. I
assume that mpd is also still capable of sending the 32-bit samples with
the 24-bit audio padded with zeroes. This would mean there has to be a
switch/parameter indicating that a 24-bit sample size is wanted instead
of 32-bit, but I don't know which one that would be.
The audio_output section of the mpd.config now looks like this:
audio_output {
typesmp86xx
load/share/Apps/mpd/plugins/smp86xx_plugin.so
nameMy Sigma Chip
engine 0# optional
decoder 1# optional
ramdac 1# optional
line_count 2048 # optional
fifo_us 50 # optional
hdmisii9134
}
This belongs to mpd as implemented on the PopcornHour mediaplayer with a
plugin developed to send the mpd-output straight into the Sigma Design
SMP86xx processor. As you can see ALSA is not used here as the output is
sent directly to the processor.
This processor requires 24-bit audio to be sent in true unpacked 3 bytes
per sample little endian format, as S24_3LE would be as I understand it,
or in a DVD PCM 24-bit linear PCM as described here (
http://wiki.multimedia.cx/index.php?title=PCM#24-Bit_PCM ) which is
somewhat more difficult to implement and also limited to 96kHz sample rate.
But Max confirmed that 24-bit audio output in true 3-byte sample format
is available as a native format from mpd, so that would be the best one.
So can anyone instruct me how to make mpd output 24-bit audio as true
3-byte samples without the use of ALSA?
Eagerly looking forward to your reply
-
aleg
--
The Planet: dedicated and managed hosting, cloud storage, colocation
Stay online with enterprise data centers and the best network in the business
Choose flexible plans and management services without long-term contracts
Personal 24x7 support from experience hosting pros just a phone call away.
http://p.sf.net/sfu/theplanet-com
___
Musicpd-dev-team mailing list
Musicpd-dev-team@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/musicpd-dev-team
Re: [Musicpd-dev-team] MPD 24-bit audio output as 3-byte samples
On 2010/01/27 12:07, Moimoi cookingher...@gmail.com wrote: How do I configure mpd to make it send 24-bit audio as true 3-byte samples without using ALSA? You can not. You're using an out-of-tree output plugin which we cannot help you with. Ask the author of that plugin. -- The Planet: dedicated and managed hosting, cloud storage, colocation Stay online with enterprise data centers and the best network in the business Choose flexible plans and management services without long-term contracts Personal 24x7 support from experience hosting pros just a phone call away. http://p.sf.net/sfu/theplanet-com ___ Musicpd-dev-team mailing list Musicpd-dev-team@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/musicpd-dev-team
Re: [Musicpd-dev-team] MPD 24-bit audio output as 3-byte samples
On 2010/01/27 13:35, Moimoi cookingher...@gmail.com wrote: Why did you tell me this morning on the Mantis bugtracker ( here http://musicpd.org/mantis/view.php?id=2731 ) that mpd can output S24_3LE as an output format as that was already implemented in mpd and subsequently close my request to add this feature to mpd. You are truly confusing in your communications. Packed 24 bit samples have been implemented in the MPD core and in the ALSA output plugin, this is exactly what I wrote, and this is exactly the truth, and does not contradict with the need for an implementation in the plugin you're using. You are asking for this feature in a plugin which we neither wrote nor support, nor do we know about its capabilities. Which is why I told you to ask that plugin's author. Now please stop bugging us with this issue. It is resolved on our side. -- The Planet: dedicated and managed hosting, cloud storage, colocation Stay online with enterprise data centers and the best network in the business Choose flexible plans and management services without long-term contracts Personal 24x7 support from experience hosting pros just a phone call away. http://p.sf.net/sfu/theplanet-com ___ Musicpd-dev-team mailing list Musicpd-dev-team@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/musicpd-dev-team
Re: [Musicpd-dev-team] MPD 24-bit audio output as 3-byte samples
For the first time, I have banned somebody from using mantis now. The mantis user avanegmond is now downgraded to read-only access, after you have reopened bug 0005291 for the third time, after I have asked you twice not to. -- The Planet: dedicated and managed hosting, cloud storage, colocation Stay online with enterprise data centers and the best network in the business Choose flexible plans and management services without long-term contracts Personal 24x7 support from experience hosting pros just a phone call away. http://p.sf.net/sfu/theplanet-com ___ Musicpd-dev-team mailing list Musicpd-dev-team@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/musicpd-dev-team
Re: [Musicpd-dev-team] persistent streaming to icecast
The silent fallback stream solution described http://mpd.wikia.com/wiki/Icecast2 is not sufficient? Q On 01/27/2010 11:40 AM, Max Arnold wrote: Hello, folks I've already asked this question on the forum [1], but both solutions are not optimal. Is it possible to make streams between mpd and icecast persistent? I.e. even when player paused or stopped, can it generate null samples with specified bitrate to keep clients connected? The only time when it can stop feeding to icecast is when corresponding output get disabled. I do some programmatic playlist rotation and when particular playlist is finished (and before next one will be loaded), player goes to stop state and drops connection to icecast, forcing all listeners to disconnect. Capturing output from hardware soundcard is not an option anymore (I moved my setup to hosted server and plus I need to run several mpd instances), and liquidsoap eats too much processing power with unnecessary transcoding. Also I tried pulseaudio, gstreamer and jack by using them as intermediate layer and found no way to solve this problem. They all stop their output as soon as input is gone. [1] http://www.musicpd.org/forum/index.php?topic=2013.0 -- The Planet: dedicated and managed hosting, cloud storage, colocation Stay online with enterprise data centers and the best network in the business Choose flexible plans and management services without long-term contracts Personal 24x7 support from experience hosting pros just a phone call away. http://p.sf.net/sfu/theplanet-com ___ Musicpd-dev-team mailing list Musicpd-dev-team@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/musicpd-dev-team -- The Planet: dedicated and managed hosting, cloud storage, colocation Stay online with enterprise data centers and the best network in the business Choose flexible plans and management services without long-term contracts Personal 24x7 support from experience hosting pros just a phone call away. http://p.sf.net/sfu/theplanet-com ___ Musicpd-dev-team mailing list Musicpd-dev-team@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/musicpd-dev-team
Re: [Musicpd-dev-team] MPD 24-bit audio output as 3-byte samples
Hi Max Why did you tell me this morning on the Mantis bugtracker ( here http://musicpd.org/mantis/view.php?id=2731 ) that mpd can output S24_3LE as an output format as that was already implemented in mpd and subsequently close my request to add this feature to mpd. You are truly confusing in your communications. - aleg On 27-1-2010 13:29, Max Kellermann wrote: On 2010/01/27 12:07, Moimoicookingher...@gmail.com wrote: How do I configure mpd to make it send 24-bit audio as true 3-byte samples without using ALSA? You can not. You're using an out-of-tree output plugin which we cannot help you with. Ask the author of that plugin. -- The Planet: dedicated and managed hosting, cloud storage, colocation Stay online with enterprise data centers and the best network in the business Choose flexible plans and management services without long-term contracts Personal 24x7 support from experience hosting pros just a phone call away. http://p.sf.net/sfu/theplanet-com ___ Musicpd-dev-team mailing list Musicpd-dev-team@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/musicpd-dev-team
Re: [Musicpd-dev-team] persistent streaming to icecast
On Wed, 27 Jan 2010 17:40:58 +0700, Max Arnold lwa...@gmail.com wrote: I do some programmatic playlist rotation and when particular playlist is finished (and before next one will be loaded), player goes to stop state and drops connection to icecast, forcing all listeners to disconnect. I don't think you're going to get what you want anytime soon. Maybe you could try setting up a playlist that contains tracks of silence and switching to that. You need to make sure MPD never stops playing. Cheers, Tim. -- The Planet: dedicated and managed hosting, cloud storage, colocation Stay online with enterprise data centers and the best network in the business Choose flexible plans and management services without long-term contracts Personal 24x7 support from experience hosting pros just a phone call away. http://p.sf.net/sfu/theplanet-com ___ Musicpd-dev-team mailing list Musicpd-dev-team@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/musicpd-dev-team
Re: [Musicpd-dev-team] persistent streaming to icecast
To be clear, all I'm trying to say here is that there's significant motivation to implement a feature like this someday, since it'd help mpd better perform certain roles, some of which (multiple users controlling) are unique to mpd. I don't think you're going to get what you want anytime soon. Maybe you could try setting up a playlist that contains tracks of silence and switching to that. You need to make sure MPD never stops playing. -- The Planet: dedicated and managed hosting, cloud storage, colocation Stay online with enterprise data centers and the best network in the business Choose flexible plans and management services without long-term contracts Personal 24x7 support from experience hosting pros just a phone call away. http://p.sf.net/sfu/theplanet-com ___ Musicpd-dev-team mailing list Musicpd-dev-team@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/musicpd-dev-team
Re: [Musicpd-dev-team] persistent streaming to icecast
It seems to me that while the silent output works, it'd be nice to not have to rely on it - in particular, it'd be great if the operator of MPD couldn't accidentally disconnect everyone. As a concrete example, I've already used httpd output + web client (+FTP) as a way to listen to music with friends online, letting everyone control the player. It's all too easy for a less technical friend to forget they're not supposed to click on that pause button (or even do it by accident!). Jeffrey On Wed, Jan 27, 2010 at 6:06 AM, Qball Cow qb...@qballcow.nl wrote: The silent fallback stream solution described http://mpd.wikia.com/wiki/Icecast2 is not sufficient? Q -- The Planet: dedicated and managed hosting, cloud storage, colocation Stay online with enterprise data centers and the best network in the business Choose flexible plans and management services without long-term contracts Personal 24x7 support from experience hosting pros just a phone call away. http://p.sf.net/sfu/theplanet-com ___ Musicpd-dev-team mailing list Musicpd-dev-team@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/musicpd-dev-team
Re: [Musicpd-dev-team] persistent streaming to icecast
On Wed, Jan 27, 2010 at 11:44:58AM +, Tim Phipps wrote: I don't think you're going to get what you want anytime soon. Maybe you could try setting up a playlist that contains tracks of silence and switching to that. You need to make sure MPD never stops playing. How I'm supposed to detect when one playlist is finished and it is time to load next one? Tracks in playlist are played in random order (with looping disabled) and when mpd finishes it stops. My script immediately notices this, clears old playlist and queues new. Also I have scheduled playlists. At defined time my script does fadeout, remembers current position, clears playlist, optionally plays short transition jingle and launches scheduled playlist. When finished, it restores old playlist and track position with fadein effect. All this song juggling constantly drops connection to icecast server. To me most logical MPD behavior is to establish icecast connection upon daemon startup and drop it only if user disables this particular output. Another useful behaviour is to continue playback even if output (icecast server) is inaccessible, just discarding audio samples to /dev/null and trying to reconnect with defined interval (maybe even with exponential fallback). I'd be glad to implement this feature myself, but unfortunately I have no C programming skills and this low level audio/buffering/threading mechanics is too complex for me. -- The Planet: dedicated and managed hosting, cloud storage, colocation Stay online with enterprise data centers and the best network in the business Choose flexible plans and management services without long-term contracts Personal 24x7 support from experience hosting pros just a phone call away. http://p.sf.net/sfu/theplanet-com ___ Musicpd-dev-team mailing list Musicpd-dev-team@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/musicpd-dev-team
Re: [Musicpd-dev-team] persistent streaming to icecast
On 2010/01/27 11:40, Max Arnold lwa...@gmail.com wrote: I've already asked this question on the forum [1], but both solutions are not optimal. Is it possible to make streams between mpd and icecast persistent? I.e. even when player paused or stopped, can it generate null samples with specified bitrate to keep clients connected? The only time when it can stop feeding to icecast is when corresponding output get disabled. MPD already generates silence during pause (both shout and httpd). But for this special use case, we could think about implementing an always-on mode, which never disrupts streaming clients. -- The Planet: dedicated and managed hosting, cloud storage, colocation Stay online with enterprise data centers and the best network in the business Choose flexible plans and management services without long-term contracts Personal 24x7 support from experience hosting pros just a phone call away. http://p.sf.net/sfu/theplanet-com ___ Musicpd-dev-team mailing list Musicpd-dev-team@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/musicpd-dev-team
Re: [Musicpd-dev-team] persistent streaming to icecast
On Wed, 27 Jan 2010 22:52:51 +0700, Max Arnold lwa...@gmail.com wrote: How I'm supposed to detect when one playlist is finished and it is time to load next one? Tracks in playlist are played in random order (with looping disabled) and when mpd finishes it stops. My script immediately notices this, clears old playlist and queues new. Don't play them in random order. Randomize the playlist into a temporary file before giving it to MPD. Or you could use the shuffle command to make MPD randomize the order. Then your script should be able to tell when the last track is being played. At that point you can use the crop command to remove all the previous tracks from the playlist and do whatever you want to do next. Also I have scheduled playlists. At defined time my script does fadeout, remembers current position, clears playlist, optionally plays short transition jingle and launches scheduled playlist. When finished, it restores old playlist and track position with fadein effect. Don't clear the playlist, crop it so just one track remains then add in the jingles etc.. Hope this helps. Cheers, Tim. -- The Planet: dedicated and managed hosting, cloud storage, colocation Stay online with enterprise data centers and the best network in the business Choose flexible plans and management services without long-term contracts Personal 24x7 support from experience hosting pros just a phone call away. http://p.sf.net/sfu/theplanet-com ___ Musicpd-dev-team mailing list Musicpd-dev-team@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/musicpd-dev-team
[Musicpd-dev-team] Mantis issue 0002449: ReplayGain and FLAC files
Hi All,
This bug is marked as fixed but the fix isn't in GIT trunk. Has it been
forgotten? Can the patch be applied to GIT please? Below is a more up to
date patch:
Cheers,
Tim.
diff --git a/src/decoder/flac_metadata.c b/src/decoder/flac_metadata.c
index 2bfb1a6..e258fa6 100644
--- a/src/decoder/flac_metadata.c
+++ b/src/decoder/flac_metadata.c
@@ -19,7 +19,7 @@
#include config.h
#include flac_metadata.h
-#include replay_gain_info.h
+#include replay_gain_config.h
#include tag.h
#include glib.h
@@ -64,14 +64,18 @@ flac_parse_replay_gain(const FLAC__StreamMetadata
*block)
rgi = replay_gain_info_new();
- found = flac_find_float_comment(block, replaygain_album_gain,
-
rgi-tuples[REPLAY_GAIN_ALBUM].gain) ||
- flac_find_float_comment(block, replaygain_album_peak,
-
rgi-tuples[REPLAY_GAIN_ALBUM].peak) ||
- flac_find_float_comment(block, replaygain_track_gain,
-
rgi-tuples[REPLAY_GAIN_TRACK].gain) ||
- flac_find_float_comment(block, replaygain_track_peak,
-
rgi-tuples[REPLAY_GAIN_TRACK].peak);
+ if (REPLAY_GAIN_ALBUM == replay_gain_mode) {
+ found = flac_find_float_comment(block, replaygain_album_gain,
+
rgi-tuples[REPLAY_GAIN_ALBUM].gain)
+ flac_find_float_comment(block, replaygain_album_peak,
+ rgi-tuples[REPLAY_GAIN_ALBUM].peak);
+ } else if (REPLAY_GAIN_TRACK == replay_gain_mode) {
+ found = flac_find_float_comment(block, replaygain_track_gain,
+
rgi-tuples[REPLAY_GAIN_TRACK].gain)
+ flac_find_float_comment(block, replaygain_track_peak,
+ rgi-tuples[REPLAY_GAIN_TRACK].peak);
+ }
+
if (!found) {
replay_gain_info_free(rgi);
rgi = NULL;
--
The Planet: dedicated and managed hosting, cloud storage, colocation
Stay online with enterprise data centers and the best network in the business
Choose flexible plans and management services without long-term contracts
Personal 24x7 support from experience hosting pros just a phone call away.
http://p.sf.net/sfu/theplanet-com
___
Musicpd-dev-team mailing list
Musicpd-dev-team@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/musicpd-dev-team
Re: [Musicpd-dev-team] Mantis issue 0002449: ReplayGain and FLAC files
On 2010/01/27 17:26, Tim Phipps m...@phipps-hutton.freeserve.co.uk wrote:
Hi All,
This bug is marked as fixed but the fix isn't in GIT trunk. Has it been
forgotten? Can the patch be applied to GIT please? Below is a more up to
date patch:
D'oh, looks like it got lost during the v0.15.x - master branch
merge.
-#include replay_gain_info.h
+#include replay_gain_config.h
+ if (REPLAY_GAIN_ALBUM == replay_gain_mode) {
A decoder shouldn't import this header, and shouldn't use this global
variable. The replay gain mode may change at any time, and the
decision to drop all of the replay gain info just because there's no
data for the current mode may become obsolete then. The decoder
should always supply all data it has, and the MPD core shall decide
the rest.
--
The Planet: dedicated and managed hosting, cloud storage, colocation
Stay online with enterprise data centers and the best network in the business
Choose flexible plans and management services without long-term contracts
Personal 24x7 support from experience hosting pros just a phone call away.
http://p.sf.net/sfu/theplanet-com
___
Musicpd-dev-team mailing list
Musicpd-dev-team@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/musicpd-dev-team
