Difference between revisions of "GLC"
m (→Interfaces) |
(→Interfaces) |
||
Line 73: | Line 73: | ||
== Interfaces == | == Interfaces == | ||
Two graphical interfaces are available for glc to try and simplify it's usage, with both of them being avalible from the [[AUR]]. | Two graphical interfaces are available for glc to try and simplify it's usage, with both of them being avalible from the [[AUR]]. | ||
− | *{{aur|soulcapture}} is a graphical front-end made in [[Gambas]] | + | *{{aur|soulcapture}} is a graphical front-end made in [[Gambas]] that utilizes [[FFmpeg]] for encoding. |
− | *{{aur|gamecaster}} is a graphical front-end made in [[Python]] | + | *{{aur|gamecaster}} is a graphical front-end made in [[Python]] that utilizes [[FFmpeg]] for encoding; optimized for Ubuntu. |
==See also== | ==See also== | ||
*https://github.com/nullkey/glc - Homepage, and they have a good wiki there as well | *https://github.com/nullkey/glc - Homepage, and they have a good wiki there as well |
Revision as of 19:17, 5 April 2013
GLC is an ALSA & OpenGL capture tool for Linux. It consists of a generic video capture, playback and processing library and a set of tools built around that library. GLC should be able to capture any application that uses ALSA for sound and OpenGL for drawing. It is similar to Fraps on Windows.
Contents
Installation
To install GLC, simply install the glc package from AUR.
If you want to record 32 bit programs such as Wine on a 64 bit system, you will also need to install lib32-glc.
NOTE: GLC will only work with ALSA. If you use Pulseaudio, install the glc-pulseaudio* packages instead. If you use OSS, you will probably need to record the audio separately.
Usage
The basic usage is simple. By default, GLC will save a (large) .glc file in the current directory. You can then play or encode it. Just run this:
glc-capture [application]
Press Shift + F8 to start and stop recording. Otherwise you can use:
glc-capture -s [application]
To start recording immediately. For complete list of available options see:
glc-capture --help
If you want to record from two different audio devices, usually the application and the microphone, you need to use the -a option. For example:
glc-capture -a 'hw:0,48000,1;hw:1,48000,1' [application]
The -a format is device,rate,channels;device2...; you probably want to mix the two audio streams togheter after, so to make it easier keep both sample rate at the same value.
Playback
To play a captured stream directly, execute
glc-play [stream file]
ESC stops playback, f toggles fullscreen and Right seeks forward.
Encoding
In order to use the videos outside of glc-play, you will need to encode it. Here are a few example that work well for encoding. Of course, you can be creative and use any of the formats supported by ffmpeg to get your desired result (mencoder works too, I'm just not familiar with it).
For either script, run with the following context (assuming it's saved as glc-encode.sh):
glc-encode.sh filename.glc
H.264 Ultrafast + FLAC Audio: (fairly quick encoding, high quality, good filesize) This script requires the following packages: ffmpeg
#!/bin/bash glc-play $1 -a 1 -o $1.wav glc-play $1 -o - -y 1 | ffmpeg -i - -vcodec libx264 -preset ultrafast -i glc.wav -acodec flac output.mkv rm glc.wav
It will output as output.mkv
Lossless (usually quickest, no quality loss, huge files, need plenty of HD space): This script requires the following packages: ffmpeg
#!/bin/bash glc-play $1 -a 1 -o glc.wav glc-play $1 -y 1 -o glc.yuv ffmpeg -i glc.wav -i glc.yuv -acodec copy -vcodec copy output.mkv rm glc.yuv rm glc.wav
It will output as output.mkv
WebM: This script requires the following packages: vorbis-tools, ffmpeg, and mkvtoolnix
#!/bin/bash glc-play $1 -a 1 -o - | oggenc - -b 128k -o glc.ogg glc-play $1 -o - -y 1 | ffmpeg -i - -vcodec libvpx glc.webm mkvmerge -o output.webm glc.webm glc.ogg rm glc.ogg rm glc.webm
It will output as output.webm
Note: Sometimes when recording WINE, the audio stream you want won't be #1, so you'll have to find out which one it is and experiment, and edit the encoding script accordingly. You can get some info on the streams using glc-play -i 1 filename.glc
Mixing audio streams
Using glc-play -i 1 filename.glc you get the list of audio tracks, extract them with the command from the previous section and you get .wav files with the audio tracks.
Depending on how the application started the Alsa driver, it is possible there are silent tracks, so listen them and delete the unneeded ones. Once done you can mix using sox (from the package of the same name) using this command:
sox -m -v 0.3 gamesound.wav -v 0.7 voice.wav finalaudio.wav
The -m option asks the mix, instead -v options change the volume of the audio file, try to keep the sum of both to 1.
Once you get a single .wav file with audio as you want, encode and mux normally.
Interfaces
Two graphical interfaces are available for glc to try and simplify it's usage, with both of them being avalible from the AUR.
- soulcaptureAUR is a graphical front-end made in Gambas that utilizes FFmpeg for encoding.
- gamecasterAUR is a graphical front-end made in Python that utilizes FFmpeg for encoding; optimized for Ubuntu.
See also
- https://github.com/nullkey/glc - Homepage, and they have a good wiki there as well