(2015-12-14, 02:57)J_E_F_F Wrote: I am running #1128 Jarvis nightly on a rPi2 with no addons. There seems to be quite a few problems with addons in these 17/k builds. Is this a continuation of the nightlies from the 16/j series, or a completely new 17/k build? IE, if I was mostly happy with the 16 nightlies for the past few months with the occasional nightly hiccup, can I expect the same from these early stage 17/k builds on a stock rPi2?
(2015-12-14, 13:58)Kougami Wrote: Changing the source works ,but switching off-/on my Samsung TV and then there is no connection to the PI2. Solution: Yatse + input, and then there is a connection. Same Problem with hdmi_edid_file=1 and hdmi_force_hotplug=1.
(2015-12-14, 14:01)popcornmix Wrote:(2015-12-14, 13:58)Kougami Wrote: Changing the source works ,but switching off-/on my Samsung TV and then there is no connection to the PI2. Solution: Yatse + input, and then there is a connection. Same Problem with hdmi_edid_file=1 and hdmi_force_hotplug=1.
First thing we need to know is if you are reporting a regression.
Did an earlier Milhouse build behave better? Does the official OE 6.0 build behave better?
(2015-12-14, 14:18)Kougami Wrote: Only with OE 5 works. I have the problem since OE 6+.
(2015-08-28, 16:25)Milhouse Wrote: Unfortunately scrolling text is a known performance issue - the simplest solution is to highlight a shorter label that won't cause scrolling.
Perhaps it's possible to stop text from scrolling when a player is active on single-core platforms such as the Pi1 that are more sensitive to GUI updates.
(2015-08-28, 16:16)MrNice Wrote: Isengard 5.945.4
CPU usage very high when horizontal text scroll happen.
To reproduce, create an empty folder, name it with a long character string or/and rename a music/video file with a long string.
Compare CPU usage between no scrolling and a scrolling name with debug or ssh/top
With my RPi1-512 overclocked at 1000 I get:
Low: 15-20%
High:65-75%
When I play HD multichannel music I get drops.
(2015-12-14, 14:21)popcornmix Wrote:(2015-12-14, 14:18)Kougami Wrote: Only with OE 5 works. I have the problem since OE 6+.
I'd suggest reporting it, with a debug log with CEC component specific debugging enabled, in both the working OE 5 and non-working OE 6 case here:
https://github.com/Pulse-Eight/libcec/issues
Use OE stable builds, not experimental builds for the logs as it is easier to track the versions of libcec is use.
.
(2015-12-13, 18:23)popcornmix Wrote:I have just installed and confirm that ITV Player catchup content is working normally in build #1213(2015-12-13, 15:04)MikeKL Wrote: Been using ITV Player (plugin.video.itv-2.2.6) successfully for ITV catchup until switching from last testbuild of (Kodi 16.0) to latest testbuild of (Kodi 17.0)Thanks. I can reproduce the issue and have identified the commit (from fernetmenta's branch) that causes the problem.
Hope information is useful/helpful.
I've reverted for now, and reported it, so ITV player should be working in the next build.
# uname -a
Linux rpi512 4.3.2 #1 Mon Dec 14 21:06:33 GMT 2015 armv6l GNU/Linux
# vcgencmd version
Dec 8 2015 14:44:44
Copyright (c) 2012 Broadcom
version e591b5eb05e2cdb1b5ae25512b27d33127d7bee9 (clean) (release)
# lsb_release
OpenELEC (Milhouse) - Version: devel-20151214210542-#1214-g564e22f [Build #1214]
# vcdbg log msg 2>&1 | grep DTOK
001710.818: Kernel trailer DTOK property says yes
# Kernel device tree status: Enabled
(2015-12-14, 19:59)popcornmix Wrote:Just checked and ITV Player catchup content appears to be working normally in build #1214 with quick test(2015-12-14, 19:50)MikeKL Wrote: I have just installed and confirm that ITV Player catchup content is working normally in build #1213Cool. You might want to test again tonight. That reinstates the reverted commit and adds a fix from FernetMenta.
ITV player is still working fine for me.
