On Sep 24, 2005, at 4:43 PM, Isaac Richards wrote:
Not complaining, just fyi, this change completely breaks ff/rw on all of my frontends. Granted they are not anywhere near a 4400+, but I don't think it should be checking to see if the cpu is keeping up while ff/rw. The symptom is as follows:On Saturday 24 September 2005 05:18 pm, Daniel Kristjansson wrote:Its harder to get consistent timings at the moment because front- end cpu load has increased (<~35% at 6910 to > 84% now) to the extent that thecode in NuppelVideoPlayer::DisplayNormalFrameinvoking "playing slower due to falling behind..." is now being invoked.( The effect here is probably not what was intended because cpu loadreduces a little when it plays slower so the "Playing at normal speed from falling behind..." is triggered - which takes cpu load up and we goround again :) - which also means the osd displaying the timing is almost always visible.A secondary consequence is that the only way to exit is to change channelas the exit request is ignored whilst the above is happening)This is something Issac is working on to improve LiveTV, and is not part ofthis patch. It is a work in progress...Yeah. Works great* on my X2 4400+, not so hot for those with, ah, lessermachines. =)* this and the other minor changes in that revision cut channel change time bya little over 60% on my machine.
1) press ff / rw2) video goes forwards or backwards about 1/2 second then returns to play mode.
3) log contains: 2005-09-25 13:13:18.535 playing slower due to falling behind... 2005-09-25 13:13:18.632 Changing time stretch to 0.8 2005-09-25 13:13:19.118 prebuffering pause 2005-09-25 13:13:19.225 prebuffering pause . . . 2005-09-25 13:13:20.075 prebuffering pause 2005-09-25 13:13:21.025 playing at normal speed from falling behind... 2005-09-25 13:13:21.538 Changing time stretch to 14) At this point we are in play mode even though it should be in ff and we never pressed anything to stop ff.
Again, this probably doesn't occur on faster machines. Geoff
smime.p7s
Description: S/MIME cryptographic signature
_______________________________________________ mythtv-dev mailing list [email protected] http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-dev
