AceInfinity
Emeritus, Contributor
I have a few questions...
First, when calling libvlc_video_set_marquee_int() with libvlc_video_marquee_option_t.libvlc_marquee_{X|Y} for setting the coordinates of the text within a video stream, is there any way to bypass the limitation where the text may not extend outside of the video boundaries? Or would this be an unbreakable limitation of the wrapper and require me to modify and recompile libvlc on my own to suit my own needs? I've set the refresh rates and enabled the filters for the marquee itself, and it all works fine, but to make my goal easier, I need to be able to set the text outside of the video. When I set an X coordinate of -10 though for instance, the backend on that variable keeps it's -10 value seemingly, yet where the actual text is displayed location-wise in the video is still at position 0. Because of this limitation, auto-wrap occurs for longer lines of text if the marquee size specified in pixels is not small enough.
Second, for the RAW data of an MPEG2 stream for video, has anybody ever achieved demuxing before? Before the video stream is sent out through one or more protocols (HTTP/RTSP/etc...) I would like to be able to control the components of the stream through any muxing and/or demuxing process.
Lastly, I've got various perams for the instance itself, ie:
And playing an MP4 at higher resolutions (~4K) seems to be very laggy. I've played around with GPU acceleration and such, but it still doesn't seem to change anything. I've even fooled around with the "--ffmpeg-hw" option, but although better, the output still seems to lag. Any other solutions?
Btw, here is what I have so far (my own workaround): http://aceinfinity.net/tmpfiles/libvlc_marquee_demo.mp4
#vlc, #videolan IRC channels on Freenode didn't help me at all either.
:thumbsup2:
First, when calling libvlc_video_set_marquee_int() with libvlc_video_marquee_option_t.libvlc_marquee_{X|Y} for setting the coordinates of the text within a video stream, is there any way to bypass the limitation where the text may not extend outside of the video boundaries? Or would this be an unbreakable limitation of the wrapper and require me to modify and recompile libvlc on my own to suit my own needs? I've set the refresh rates and enabled the filters for the marquee itself, and it all works fine, but to make my goal easier, I need to be able to set the text outside of the video. When I set an X coordinate of -10 though for instance, the backend on that variable keeps it's -10 value seemingly, yet where the actual text is displayed location-wise in the video is still at position 0. Because of this limitation, auto-wrap occurs for longer lines of text if the marquee size specified in pixels is not small enough.
Second, for the RAW data of an MPEG2 stream for video, has anybody ever achieved demuxing before? Before the video stream is sent out through one or more protocols (HTTP/RTSP/etc...) I would like to be able to control the components of the stream through any muxing and/or demuxing process.
Lastly, I've got various perams for the instance itself, ie:
Code:
-I myinstance --ignore-config --plugin-path={PLUGIN PATH HERE} --vout-filter=deinterlace --deinterlace-mode=blendAnd playing an MP4 at higher resolutions (~4K) seems to be very laggy. I've played around with GPU acceleration and such, but it still doesn't seem to change anything. I've even fooled around with the "--ffmpeg-hw" option, but although better, the output still seems to lag. Any other solutions?
Btw, here is what I have so far (my own workaround): http://aceinfinity.net/tmpfiles/libvlc_marquee_demo.mp4
#vlc, #videolan IRC channels on Freenode didn't help me at all either.
:thumbsup2:
