I'm quite sure there was no failure before 2:22 MYT, after that maybe but it's only speculation based on satcom restart (which doesn't automatically imply failure).
the thing is there are many paths that could fit BFO&BTO values within margin of error, especially since we don't have a clue if any of speed/altitude/heading were constant at all
P.S.
http://www.duncansteel.com/ group has made the new statement
this figure is interesting to me
http://www.duncansteel.com/wp-conten...-Figure-1a.jpg
anyone tell me why they assume constant heading from 19:41 onwards for a flight that drastically changed its heading 3 times in first ~2 hours?