On 13/02/07 00:20:19, Nico Sabbi wrote:
lynx.abraxas@freenet.de wrote:
Ah, well. That's a pitty. Probably the TU BS is researching on what I want but if it is so high development it won't be free:(
what is it?
A rough summary translation of http://www.ibr.cs.tu-bs.de/theses/brandt/mpeg4transcoder.html?backurl=%2Fuse... : Because re-encoding (first decoding then encoding) is very cpu intensive and therefor mostly not realtime they try to develop a technique that will give the same result without full decoding. The work is finished and there should be a c/c++ implementation as a programme and as a library. And then http://www.ibr.cs.tu-bs.de/theses/brandt/mdtranscoding.html?backurl=%2Fusers... builds on previous work: It was discovered this adjusting can be done in various domains as time, space and ditail of sharpness. The task is to code a transcoding-module to mpeg-4 that can ajust the video stream in different domains at the same time. This transcoding shouldn't be done sequential but combined to reduce cpu time. It has to be handed in again as a programme and as a module. There are more details and a chart at http://www.ibr.cs.tu-bs.de/theses/brandt/mdtranscoding/?backurl=%2Fusers%2Fb... Sadly there seems to be no English version available.
Did it with mplayer -fs -v -vo cvidix dvb://ard &> mplayer.out to have cvidix and to see when the second channel is visible. Again pressing the h changes it but then the q for quitting appears in the tty and I have to close mplayer with Ctrl-C. mplayer.out attached.
Regards, Lynx
I'll give a look at the output whan I have some time
Can I do any other debugging meanwhile? Regards, Lynx