SHOUTcast Transcoder 2 Beta (Build 54) 10/07/2011

[DrO]

UPDATE 11/04/2014 - sc_trans is not currently available to download due to licensing issues and it is uncertain if it will be made available again or not.

Presenting the latest BETA drop of the SHOUTcast Transcoder for broadcasters, developers and testers. This build is our latest GM release candidate and introduces new features as listed in the “Changes” section below, as well as addresses bugs reported in the previous BETA builds.

This BETA release is now available for the following platforms:

• Windows 32-bit (Windows 2000 and up)
• Windows 64-bit (Windows XP64 and up)
• Linux
• Linux 64-bit
• Mac OS X (Intel)
• BSD 8.x

Downloads

You can download the updated version of the transcoder from the direct downloads for the operating system version you require:

• Windows (32-bit)
• Windows (64-bit)
• Linux (32-bit)
• Linux (64-bit)
• Mac OS X (Intel)
• BSD 8.x

Changes

• Changed some of the metadata filtering to improve utf-8 character compatibility (build 48-51 specific)
• Changed 'uvoxstreamid' to 'streamid' to match the v2 DNAS name (if found it will be removed if using configrewrite=1)
• Changed default for 'genre' to now be 'Misc' (now matches with Config Builder and the Source DSP)
• Removed 'uvoxauth' and associated usage to use 'password' for any endpoint (if found it will be removed if using configrewrite=1)
• Fixed some ID3v2.3 tag handling quirks to improve metadata and connection reliability
• Fixed ID3v2.x tag parsing failing due to not processing syncsafe frame data correctly
• Fixed relaying of SHOUTcast 2 stream not working correctly due to not fully following the SC2.0 protocol
• Fixed DJ connections on 'djport' not being url escaped in all cases (could cause wrong titles on v1 servers)
• Fixed more config files being shown than allowed on non-Windows versions when no configuration file specified
• Fixed crash on Windows versions after choosing a config file to use when run without write permissions to create 'logfile' when no config file is specified
• Fixed some relative path issues when using 'include' in some scenarios
• Upgraded libraries to use libogg-1.3.0
• Other miscellaneous code changes, improvments and related documentation updates

Getting Started

If you already have a running instance of the transcoder then with this build you should make some changes to the configuration file due to the changing of some of the config option names (though existing configurations should load whilst compatibility mappings are present). This change has been done to make it easier to match up the configuration options between the DNAS and the Transcoder.

If you have the following on the left then they will need to be changed to what is shown on the right:

uvoxstreamid -> streamid
uvoxauth -> password


Otherwise there should not be any issues with replacing your current version with this new version.

If this is a new install then make sure to read through the information in the 'docs' folder along with the example configuration files provided in this new version which should make it easier to get started.



Reporting Issues

If you do come across an issue with the transcoder, then please do post in this thread with as much information as possible about what you're doing at the time, the system you are using and anything else which will make it easier to understand what is or isn't going on with your install.



MP3 Broadcasting License

In order to unlock the MP3 broadcasting capabilities of this transcoder, you will need to purchase a MP3 license key. Click here to purchase.



Known Issues

The following are known issues with the v2 DNAS raised in previous releases but not currently fixed / fully confirmed as needing to be fixed (i.e. intended behaviour):

• With some metadata (primarily if running in v2 mode), when they are read / received then they may have characters incorrectly removed / changed. This is due to some additional processing happening on the metadata which is causing some of the utf-8 encoding to be incorrectly changed / lost which leads them to be filtered / replaced with a different value at a later stage in handling. This will be resolved in a later build and is related to some of the build 51 fixes to prevent invalid metadata being sent to the DNAS which will cause the connection to be dropped.
• Line-in support for Mac builds is likely to be broken (may crash the DNAS) and is currently pending full investigation into the cause of the issue
• In some cases the 'inheritconfig' option will not correctly handle the automatic setting of the 'serverip' value which can lead to the Transcoder not being able to connect to the DNAS server. The solution if using the Config Builder to create the config file is to uncheck the 'Transcoder Inherit mode' option at the top of the page or manually enter the values which are read from the DNAS's config file as specified at http://wiki.winamp.com/wiki/SHOUTcas...#Miscellaneous
• Setting mp3mode to anything other than 0 i.e. mp3mode=0 will cause errors containing mp3enc::mp3Setup Error opening mp3 encoder instance. code=2 which will prevent mp3 encoding from working. This option is deemed deprecated and will not be present in the next release.
• When using scripts in playlists, if the playlist does not return the number of 'next' items as requested then the resulting 'next' output may not be correct or may even reference itself. Additionally the 'next' items obtained are not correctly inserted into the playback queue so if multiple items are returned then it is possible that they will be skipped.

Additional Information

Windows versions of the transcoder are built with a dependency against the Microsoft Visual C++ 2008 SP1 Redistributable Package. If sc_trans is unable to start due to a dependency issue then you will need to install the correct version of the package so it can run which depends on the version of sc_trans you are attempting to run:

32-bit - http://www.microsoft.com/downloads/e...d-3802b2af5fc2
64-bit - http://www.microsoft.com/downloads/e...4-157cfdffee4e


This new version has been tested on a number of current Linux versions (Ubuntu, Fedora, CentOS) so should hopefully run on other versions as well and without any dependency or locale related issues (as affected prior versions).


Discussion about the previous version of the server including changelogs can be found in the following threads

• SHOUTcast Transcoder 2 (Build 51)
• SHOUTcast Transcoder 2 (Build 48)
• SHOUTcast Transcoder 2 (Build 40)
• SHOUTcast Transcoder 2 (Build 39)

-daz

[Matrix2000]

hy, can this version relay the AIM,Genre,Stations or ICQ from SAM over the DJ Port to the DNASv2?
or only from the sc_trans.conf file?
greets

[DrO]

there's been no changes in that handling (if it was then it would have been mentioned in the changelog) and as mentioned in the build 51 thread, it needs to be considered, though i currently i think it's swinging more towards not implementing changes to forward that information on - though as i've not actually looked into it, that view is still likely to change.

as the v2 system doesn't use that information (and will use what has been set against the authhash on streams when listed in the directory, relaying that information would only be appropriate for non-Directory listed stations.

-daz

[MrSinatra]

Originally Posted by DrO

This BETA release is now available for the following platforms:

• Windows 32-bit and 64-bit versions) (Windows 2000 and up)
• Windows 64-bit (Windows XP64 and up)

hey DrO, glad to see this come out, been baking a while. i still don't understand even a little bit what its for or the usage scenario for it, but i guess i don't need it yet...

thought i'd point out the typo above, or at least, the redundance.

also, i was thinking, if thinktink didn't mind of course, why not host his config builder on winamp.com? that way you guys could tweak it / update it in one place, and make clear whats essential to fill out and what isn't, for differing usage cases. users could then DL their config files, get auth-hash, etc... one online process.

anyway, hope you can do other things now!

[thinktink]

Originally Posted by MrSinatra

...also, i was thinking, if thinktink didn't mind of course, why not host his config builder on winamp.com? that way you guys could tweak it / update it in one place, and make clear whats essential to fill out and what isn't, for differing usage cases. users could then DL their config files, get auth-hash, etc... one online process....

I do not mind.

[DrO]

Originally Posted by MrSinatra

thought i'd point out the typo above, or at least, the redundance.

have fixed it and also in the DNAS thread (was a left over from when there was only a 32-bit windows version).

Originally Posted by MrSinatra

also, i was thinking, if thinktink didn't mind of course, why not host his config builder on winamp.com? that way you guys could tweak it / update it in one place, and make clear whats essential to fill out and what isn't, for differing usage cases. users could then DL their config files, get auth-hash, etc... one online process.

isn't quite as simple as just uploading it (as much as it should be - will probably temporarily host it on my hosting once i'm back up to speed). however, there's other things which need to be changed around as it's still deemed too complicated and i'm not happy with the flash side of things, hence trying to move some aspects over to running the tools in a 'setup' mode like http://forums.winamp.com/attachment....5&d=1317203532 for the DNAS which just does what's needed to get going instead of covering DNAS + Transcoder.

main reason for that is too many people complaining about setting up the DNAS (not many mention the Transcoder) and so in that case, only something specific to the DNAS is generally needed. yes having the config builder able to setup both tools at the same time is something i don't want to drop (as it's very useful), but with the Transcoder able to determine things from the DNAS's config file instead of explicit setting of it in the Transcoder config file lessens the need for such a setup.

also i'm thinking the design of the config builder needs to be changed to make it more of a page like setup (as per the screenshot for the pending DNAS setup mode) as lot of the complaints are that it's just too over-whelming (even with all of the help information, etc).

-daz

[MrSinatra]

thats all right on DrO, i just thought i'd float the idea out there, in case it hadn't been considered. thinktink is kind to make it available!

just to be clear, i was suggesting at least two links, each a useage case. one for dsp and dnas only, and one for dsp, dnas, and transcoder. (i can't tell if u plan to ultimately drop the second one or not?)

if i could suggest something, it would be that each entry where the user inputs something be verbosely explained as to why/what it is, and if it is mandatory or optional, and if the dsp has to have matching info.

thx!

[thinktink]

"Page setup"

Actually, I had thought about that, kind of like a "wizard" style version of the Config Builder. I think it might be better to combine the two versions into one where you'd have like a JavaScript written floating div that acts as the "wizard" where it fills in the original Config Builder options as it advances per page. This way you have the ability to do-it-yourself and a "wizard" way all in one package instead of trying to maintain separate versions, and, of course, code reuse.

[DrO]

Originally Posted by MrSinatra

just to be clear, i was suggesting at least two links, each a useage case. one for dsp and dnas only, and one for dsp, dnas, and transcoder. (i can't tell if u plan to ultimately drop the second one or not?)

well currently there's not much for the DSP since that should be simple enough to use (which i know is naive based on the support posts about setting it up, hence some of the ui / handling changes i'm looking to make for the next version like making 'user id' be a 'dj / user id' option which is usable in v1 mode as well to remove the dj:name setup (which seemed to confuse people).

Originally Posted by MrSinatra

if i could suggest something, it would be that each entry where the user inputs something be verbosely explained as to why/what it is, and if it is mandatory or optional, and if the dsp has to have matching info.

well that is what the upper right part of the config builder is meant to do though a lot of it was pulled from the docs which probably isn't a good idea for some of the options. hence what i've shown in the DNAS setup version screenshot which has a more specific block of information for the 'server port' option in this case.

Originally Posted by thinktink

I think it might be better to combine the two versions into one where you'd have like a JavaScript written floating div that acts as the "wizard" where it fills in the original Config Builder options as it advances per page. This way you have the ability to do-it-yourself and a "wizard" way all in one package instead of trying to maintain separate versions, and, of course, code reuse.

aye, trying to keep it simple + maintainable is the main thing - hadn't thought too much about it so good suggestion on a layered div over the top. splitting out the existing config builder to make just a reduced DNAS specific version probably wasn't a good idea on the maintainability aspect, but then that is intended as a quick setup and go so it's better that is just targeted for what is needed (which is what's in the screenshot, common stream options is enabled, summary of what will be set and then confirmation of it being saved or not, with info on what to do to get the authhash afterwards).

-daz

[adisorn]

Current Song: 46 4D!H6-+%'D
H

not support thai languages

[DrO]

adisorn: with what DNAS and version? just via standard Transcoder to DNAS, or via a DJ connection to the Transcoder and then to the DNAS? also it could just be a poorly encoded tag on the file you're using - would need an example to work with really and is this a new issue or has it been present with the older Transcoder builds?

either way, you're really not giving much to work on. and i've had test files with japenese in which have worked so thai should also be working, but without knowing anything of your setup it's impossible to know if you have a real bug or expected behaviour, etc.

-daz

[adisorn]

SC_TRANS v2.0.0 Build 54 (Beta) connect to SC_SERV v2 Build 29 Current Song: 46 4D!H6-+%'D
H
DJ connect to SC_TRANS v2.0.0 Build 54 (Beta) Current Song: 46 4D!H6-+%'D
H
standard Transcoder to DNAS


DJ (dsp 2.2.3) connect to SC_SERV v2 Build 29 Current Song: OK

[DrO]

and that's a DJ connection using the DSP? also what version of Winamp is being used?

though really sounds like i'm going to need a test file which is exhibiting the issue - can be emailed to me from my address on my site via my signature.

-daz

[adisorn]

I sent an email to you.

[DrO]

i've received it but haven't had a chance yet to look at it. when i've done so then i'll reply here or by email (though as i've a site deploy to do today it'll probably be tomorrow i can look at the file).

-daz

[Elus1ve]

would like to know if there is a option to control the following scenario:

dj stops live set, autodj returns but starts to stream the next file on the playlist from the begining

[DrO]

Elus1ve: there is no option for doing that - the behaviour currently is to resume playlist playback where it was stopped. i've already got a note about making it advance to the next playlist item, but the Transcoder is low on my priority of jobs (this new release was only done to alleviate my spending time on support issues from the previous few builds).

-daz

[neXor]

hi again, just swiched to build 54(linux x64) on tuesdays, and since than my sc_serv got several crashes.

i dont know why, the logs throw no errors. the sc_serv simply cuts after

code:
2011-10-27 09:49:00	I	msg:[YP2] Connecting to yp.shoutcast.com

Note: im not running any debug output!

i think might be a sc_trans related problem as the problem came up with the latest sc_trans.


neX

edit:
and nearly forgot this...
sc_trans is continiously using the whole power of one CPU is this normal?
(CPU% 99,0% one one of the cores)

[neXor]

okay... tha crashing is because of incorect filenames (ascii to ansi or so...)
got some german letters in my filenames which should look like "ä" and so on, but they are as "ä"
going to rename them asap

[DrO]

that would be due to an issue in the linux builds of the v2 DNAS where it can terminate when receiving such characters - it is fixed for the next release of the DNAS (or should be as it's been a while since i checked the fix was still ok).

as for the high cpu usage, the Transcoder is probably not as efficient as it could be on non-Windows builds (no specific cpu optimisations are enabled). though it also depends on the machine (real or VPS and any other number of factors as to how much resources it will use).

-daz

[Rothgar]

Hi DrO,

Nice work on the new release.

Do you know if streamtitle's are now multi in this release?

Thanks.

[DrO]

if it's not mentioned in the changelogs / docs then it's a no for the time being - meant to but then it slipped down the list of things. have re-noted it so it's a bit higher up the todo list.

-daz

[Matrix2000]

sorry I have to use the google translator!

I have the problem that if we make the task is to listen for a short time sc_trans2. How can I change that he is not equal between mixed. that is already disturbing the in between is always to listen sc_trans2.
is there a sleep function or something in between which he is not on?
is always between the handover!
greetz, Kurt

[audioprobe]

Is it just me or is the sample rate incorrectly detected for a DJ's incoming AAC stream? I'm sending a 64kb/s aacPlusv2 in stereo with a 44.1khz sample rate from Sam Broadcaster. My reason for thinking something is broken is because of this message in the log and the terrible audio quality that results from this particular build (as well as an older build that doesn't have this problem).

code:
resampling 22050/2 == 44100/2

Here's the relevant config info:

code:
; setup encoders

encoder_1=mp3

bitrate_1=192000

samplerate_1=44100

channels_1=2

serverip_1=192.168.0.101

serverport_1=80

password_1=xxxxxxxxxxxx

outprotocol_1=1

Yes, I know this is a waste of bandwidth... Unfortunately, it's what I'm limited to right now while I'm testing out new servers.

For your reference, here are two test streams: http://listen.audioprobe.net:80 and http://63.247.93.168:80
The first is running the old build that works well (aside from a few annoying bugs...my reason for testing out this build). The second is running the new build. The config is basically the same for both and they both have the same source configured exactly the same.

[DrO]

what builds of the Transcoder are you comparing between?

-daz

[audioprobe]

Originally Posted by DrO

what builds of the Transcoder are you comparing between?

-daz

Builds 40 and 54.

[DrO]

k, so it's likely i broke something when we changed the AAC library used *joy* have noted it to investigate but it's not high up my hectic todo list (and if i do get out a new Transcoder update soon it'd just be for fixing some small config handling issues).

-daz

[audioprobe]

Originally Posted by DrO

k, so it's likely i broke something when we changed the AAC library used *joy* have noted it to investigate but it's not high up my hectic todo list (and if i do get out a new Transcoder update soon it'd just be for fixing some small config handling issues).

-daz

I look forward to it. Thanks for putting so much effort into this!

Edit: Note that I'm reverting the test server to build 40 so I can begin evaluating that provider's network.

[Paderman]

Hello,

I would like to translate builder config into German, but it may be that the builder is not yet finally finished and still be changes made?

By the way: the HTML code contains unnecessary and senseless spaces   in the code

[DrO]

there's nothing stopping you translating it but it's likely to change again (either with the next DNAS release or the one after that) and that a translated version would not be officially shipped in the distribution or be supported.

-daz

[Paderman]

OK, but can you add the label tags in the code for fields and so?

Example:
<label for="GenerateMinimalCheckBox">Text</label>

Is it possible that you can add a language interface?
The default language is English and the user can switch in his language

Thank you

[DrO]

doing that then means essentially supporting localisation of the config builder and as things are with regards to supporting the tools, i really don't want to do that, sorry.

that view may change (no guarantees) as i'm all for localistion (having done the Winamp system) but it's only best done once things with the config builder are stabilised which is not the case at the moment.

-daz

[Paderman]

It's ok

[DrO]

Originally Posted by Rothgar

Do you know if streamtitle's are now multi in this release?

As an update, i've just changed streamtitle to be a <multi> option for the next drop (no eta) so it'll act like before but if you do streamtitle_x (where x is the endpoint configuration) then it'll use that instead. note: streamtitle or streamtitle_1 will be used as the default title if there's not a streamtitle_x specified for the endpoint configuration.

hopefully that makes sense as i've tried to do it so that existing configs will work without having to add another new / specific configuration entry to do things.

-daz

[Rothgar]

Originally Posted by DrO

As an update, i've just changed streamtitle to be a <multi> option for the next drop (no eta) so it'll act like before but if you do streamtitle_x (where x is the endpoint configuration) then it'll use that instead. note: streamtitle or streamtitle_1 will be used as the default title if there's not a streamtitle_x specified for the endpoint configuration.

hopefully that makes sense as i've tried to do it so that existing configs will work without having to add another new / specific configuration entry to do things.

-daz

Excellent, thanks for the update DrO/daz

By what it sounds it will be backward compatible using "streamtitle" however you can then over-ride this per encode using the multi streamtitle_0 streamtitle_1 etc matching with the encoder multi.

This should mean I can in the future specify different titles per stream/encode/transcode which will allow me to specify the bitrate.

What would be nice in the future daz might be to have variable substitutes such as {bitrate} or [bitrate] etc automatically input from the transcode settings? This would work well for when a DJ logs in etc.

Thanks for adding this though.

[DrO]

Originally Posted by Rothgar

Excellent, thanks for the update DrO/daz

By what it sounds it will be backward compatible using "streamtitle" however you can then over-ride this per encode using the multi streamtitle_0 streamtitle_1 etc matching with the encoder multi.

This should mean I can in the future specify different titles per stream/encode/transcode which will allow me to specify the bitrate.

correct though you can only do from 'streamtitle_2' and up since the existing 'streamtitle' is the same as 'streamtitle_1' which shouldn't be a problem i hope (is the best way to keep existing configs working without having to mess around with new options, etc).

-daz

[Matrix2000]

hello,
there is a function of the sc_trans2 not always play in between?
when handing over to the other dj is always running about 5 sec. the sc_trans2.
Unfortunately, the very disturbing.
kurt, greets

[MADxHAWK]

Hi Matrix2000,

not yet.

If you always have the same times your DJs will broadcast,
example:
DJ A 18:00 - 20:00
DJ B 20:00 - 22:00
DJ C 22:00 - 00:00
you can try the folowing:

1. Create a playliste "silence" with a single mp3. the mp3 should have very low volume, best would be silence.

2. create a playlistevent
1th event using playlist silence, starts at 19:55 with a duration of 10 minutes
2nd event using playlist silence, starts at 21:55 with a duration of 10 minutes
both playlist with shuffle and loop disabled and a priotity higher than other playlists.

3. create a cronjob to force sc_trans to skip current track
1st cron:

code:


55 19 * * * killall -SIGWINCH sc_trans >/dev/null 2>&1

2nd cron:

code:


55 21 * * * killall -SIGWINCH sc_trans >/dev/null 2>&1

Sending a signal SIGWINCH to sc_trans will force sc_trans to skip to the next track

sc_trans will load the silence playlist at 19:55 and 21:55 but wil not start to play it as long as a DJ is connected. The Cronjob simply makes sure that that the current track that would be played by sc_trans will be skipped and the silence playlist will be next when the DJ disconnects. So if DJ A hands over to DJ B between 19:55 and 20:05 the silence playlist will be played. Same for handover from DJ B to DJ C.

Maybe the dev-team will add a skiptrack function triggered by a dj connection to one of the next builds of sc_trans, and maybe they add a option to trigger a special playlist to be played when the dj disconnect. Would be a great feature

-MAD

[Matrix2000]

oh ok mad! thx!
i think i wait to another release
i have a empty mp3 file for this problem in the playlist.

it only works if I restart it daily.
because the next day he ignores the simple and plays from the default playlist. I once the priority set to 3. let's see

<event type="playlist">
<playlist loopatend="1" shuffle="1" priority="3"> leerfile </ playlist>
<calendar starttime="15:50:00" endtime="23:00:00" repeat="128" />
</ event>
<event type="playlist">
<playlist loopatend="0" shuffle="1" priority="2"> standard </ playlist>
<calendar starttime="23:00:00" endtime="15:50:00" repeat="128" />
</ event>
<EventList />

The next problem we have is to catch Sat & Sun by 10 clock to.
greets, Kurt

[MADxHAWK]

Hi Kurt,

try this:

code:


<event type="playlist">

<playlist loopatend="1" shuffle="1" priority="3"> leerfile </ playlist>

<calendar starttime="15:50:00" duration="07:10:00" repeat="128" />

</ event>

<event type="playlist">

<playlist loopatend="0" shuffle="1" priority="2"> standard </ playlist>

<calendar starttime="23:00:00" duration="16:50:00" repeat="128" />

</ event>

Using duration instead of endtime may help.

Greetz
-MAD