betacentauri

In another forum biki3 told me that he has problems with mkv with ac3 and aac tracks. If you start with ac3 you can hear sound. Then switch to aac (which also works) and then back to ac3. Then you cannot here anything. If you start with aac track (works) and switch to ac3, you cannot hear anything.

Does anybody have an idea, what the problem is and how to solve it?

 

I found a solution for this . It's a little bug in the dvbaudiosink. Once an aac track was selected, self->aac_adts_header_valid was set to true. Changing audio track back to ac3 track didn't change that flag. This had the effect that always an aac adts header was added to the pes packet.

Attached is a patch to solve this problem.

Hi together,

my et-9200 hardware(and I suggest the hw of all boxes) only supports AAC with LC profile. Main and SSR profiles are not supported.
So currently if these profiles are used in a mkv you can see video and hear no audio.
I have created a patch to enable playback of all gstreamer supported AAC profiles. In case of Main or SSR profile, the audio is decoded by gstreamer software decoder (this costs around 10% CPU time).

1. Patch: 0001-Set-only-by-hardware-supported-audio-mpeg-4-profile.patch.txt
This is for gst-plugin-dvbmediasink. It sets mpeg capabilities to "profile = lc". With this only AAC audio with LC profile will be decoded by the hardware. This only affects mpeg4 audio playback and not mpeg 1 or 2 playback.

2. Patch: 0001-Add-faad-gstreamer-plugin.patch.txt
This patches Openpli Oe Core. In gst-plugins-bad faad plugin is build and will be delivered in the image. Also faad rank is lowered to prevent using software decoder, when hardware decoding is possible.

3. Patch: 0001-Set-audio-profile-caps-field-for-aac-audio-in-mkvs.patch.txt
This patches Openpli Oe Core. For mkv's the recognized profile was not included in the caps. Without profile information dvbmediasink is used for all mpeg4 audio even if it cannot handle it. I created a patch, which was included in the latest gstreamer 1.x git version. So the patch seems to be correct. https://bugzilla.gno...g.cgi?id=703312

 

I have compiled a complete new image without problems. All my testfiles work.

 

I have one question to pli team: I have included libfaad.so.2.0.0 and libgstfaad.so in the image. These 2 files take around 290kb. Should I revert this, because there are boxes with little flash memory?

I included the files, because if you want to playback media files with e.g. AAC main profile it won't work (no video and no audio) and the user gets no proper error message and so he don't know that he needs to install these 2 libs.

 

@Dima73: This solves the problem with your mkv file.

Read this thread:
http://openpli.org/f...intained/page-6

Here's my version of the patch, which prevents the compiler warning.

 

It must be committed AFTER the other patch from athoik, because of the included PR.

I found a solution for the mentioned mp3 problem. Babsy98 has tested several media files and everything works fine. He has added the patch to their openATV image (latest version from today). I have also build an OpenPli image for et9x00 boxes which includes the latest gstreamer 0.10 version and my patch.
Here is a link to the download:
http://www.file-uplo...43_usb.zip.html
(download button is right below the download counter)
 
Please test it with all your media files and report here, if it works properly. Thanks!
 
Attached are the 2 patches, to switch to latest gstreamer version.

Openatv is the old openAAF team.

Hi together,

 

I took gutemines "challenge" and have build a new binary, which can flash images while the box is running. Sources and binary are attached. There's currently no python plugin. Perhaps somebody wants to build one...

 

To use it, copy ofg_bin.zip to the box and unzip it. Usage is quite similar to gutemines binary:

root@et9x00:~# ./ofgwrite 

ofgwrite Utility v0.1
Author: Betacentauri
Based upon: mtd-utils-native-1.4.9
Use at your own risk! Make always a backup before use!

Usage: ofgwrite <parameter> <image_directory>
Options:
   -k --kernel  flash kernel (default)
   -r --rootfs  flash root (default)
   -n --nowrite show only found image and mtd partitions (no write)
   -h --help    show help

I have tested it with my et-9200. I expect that it works with all other et-xx00 boxes and the XP1000. Afaik on VU+ boxes it currently don't run, because of other filenames.

 

WARNING:

• Before use, please make a backup on a usb stick, to be able to flash it, if something goes wrong.
• Don't flash rootfs while a recording is running. It will be interrupted!
• Flashing rootfs stops enigma2 and reboots the box after flashing.

 

Technical things:

• It's a static binary which uses the normal flash_erase, nandwrite, ubiformat sources.
• I had to modify ubiformat a little bit to get it running without ubidetach.
• pivot_root is currently not included. It's quite difficult to test it without using single user mode (init 1). And I can't use single user mode, because it stops remote connection and I have no null modem cable (yet).

Hi Bastelbox,

 

this is a patch for the source code. You cannot apply it to the keymap file.

 

Attached is the patched library (it's compiled for a et9x00, I don't know whether it works on other boxes).

To install it do this:

1. Put the file on the box.

2. Do a gunzip libtuxtxt32bpp.so.0.0.0.gz

3. Make a backup of the original lib. E.g. cp /usr/lib/libtuxtxt32bpp.so.0.0.0 /usr/lib/libtuxtxt32bpp.so.0.0.0.orig

4. Copy new libtuxtxt32bpp.so.0.0.0 to /usr/lib/  E.g. cp /tmp/libtuxtxt32bpp.so.0.0.0 /usr/lib/

5. Restart enigma2

 

If it don't work rename .orig file again.

Hi,

 

perhaps somebody is interested in this:

The attached binary (it's gzipped) simulates a keypress, so you can control enigma2 e.g. via telnet.

 

At first unzip it on the box. Then execute: ./send_key <key>

You can find the key list in the cpp file or in linux/input.h (use google, but not all keys are usable in my program).

E.g. KEY_0, KEY_1, ..., KEY_9, KEY_MENU, KEY_TEXT, KEY_EXIT,...

 

Attached is also the source code. It's free for use.

(currently I use /dev/input/event1. I don't know whether it works for all boxes).

@hwads: And does rewind also work? I have tested it here again. I have switch to a channel, waited some seconds so that timeshift has started in background and have pressed <<. Result: Rewinding starts.

Sorry, but where do I find the log?

It's not so simple. Don't do this if a recording is running!! Login via telnet on your box. Execute "init 4". Then execute "/usr/bin/enigma2". Then you see the logs in telnet window. To abort it press ctrl+c. Then type "init 3" plus enter to start enigma again.

Yes, I meant it in the same way.
What receiver do you have? Do you have MultiQuickButton plugin installed?

Hey guys, can you stop this useless(buyukbang don't have to publish his sources if he don't want it) discussion?

 

And buyukbang thanks for the plugin (even when I can't use it, because I have no solo2).

@Pli:

 

Ich habe die Bibliothek nur neu compiliert und sie funktioniert jetzt. Ich habe nichts am Code geändert!

Macht mal für die Plugins ein bitbake -c cleansstate, damit die Bibliothek neu erstellt wird.

Here is a patch for automatic timeshift start. It's disabled by default. You can enable it in the settings (user interface).

 

This might lead into a discussion:

The "stop timeshift" dialog is not shown until the user really used the timeshift (= pressed pause or did a fast rewind). So you can zap without problems. Once you have really used timeshift (regardless of whether you're currently at the live point or not), the dialog is displayed.

Reason for this: I like the dialog, but I don't like to see it everytime I zap and I also want to start timeshift almost immediately after zap to a channel to be able to save the whole program I was watching and not only a part.

 

If you don't like it, you can remove it without problems by removing the variable ts_min_once_activated.