Posts: 78
Joined: Dec 2014
Reputation:
2
Ok, tested with Stop. It works flawlessly tested on 2 movies from "Recent Movies" and on 1 TV Show. Stop resumes properly when I jump forward and hit stop. Re-tested pausing and it still doesn't resume at the proper location (resume point set at where I stopped previously).
Posts: 6
Joined: Feb 2014
Reputation:
0
Hi Guys,
Black screen" on video playback playback on some Android devices with Amlogic SOCs.
No change after installing Helix 14.1
(System)
DroiBox MX
Model: MBX reference board (g18ref)
Android ver 4.2.2
Kernel version 3.0.50
ubuntu@ubuntu-desktop #15
Build number:
JDQ39.20130708
Hardware
CPU: ARMv7 Processor rev 0 (v71)
BogoMips: 2393.70
Hardware: Amlogic Meson6 g02 Customer platform
Posts: 31,445
Joined: Jan 2011
(2015-01-09, 20:04)Player1 Wrote: Hi Guys,
Black screen" on video playback playback on some Android devices with Amlogic SOCs.
No change after installing Helix 14.1
Debug log (wiki) of when you attempt to play a video, please.
Posts: 45
Joined: Jan 2014
Reputation:
1
2015-01-09, 22:46
(This post was last modified: 2015-01-09, 22:57 by quixers.)
Thanks for creating the test builds!
For mac os 64bit PR fix binary, for videos opened from home screen recently added shelf:
- exiting video using stop, resume point retained, video restarts at correct point - this is fixed
- exiting video by pausing and then using back, resume point not kept, video restarts at beginning - still broken, would expect the background paused video to reopen at the same point (this is probably a different bug from PR5842)
Actually I mean PR5958
Reckon you are referencing the wrong PRs Ned Scott?
PR5842 (as referenced in the binary filenames and above) is in the master branch already and actually caused this resume point bug, it'll have been PR5958 which is the fix which hasn't been merged up which must have been patched in... Its really PR5958 we are testing here
Posts: 268
Joined: Sep 2014
Reputation:
6
menakite
Retired Team-Kodi Member
Posts: 268
2015-01-09, 22:53
(This post was last modified: 2015-01-09, 23:02 by menakite.)
The pause behavior is unrelated and would need fixing in a different part of the code (did it work in alphas, betas, RCs?).
Edit: "fix-5842" is the Git branch for PR 5958, because 5842 is the PR that introduced this regression.
Posts: 78
Joined: Dec 2014
Reputation:
2
So is there a PR already created for the pause / resume bug? It would be great to get that also fixed for 14.1 and into master.
Posts: 45
Joined: Jan 2014
Reputation:
1
Resume bug anyway yes - PR5958
The test builds created here will hopefully help to make that happen!
Posts: 78
Joined: Dec 2014
Reputation:
2
Let me ask another way as I probably wasn't clear: Now that we have a test build confirmed to be have a working Stop+resume, do we have a PR that is for Pause+resume which is not working yet?
Posts: 11
Joined: Oct 2011
Reputation:
0
fly89
Junior Member
Posts: 11
I have MXQ S805, there is no black screen, after update the box with firmware 109k4.
Posts: 31,445
Joined: Jan 2011
(2015-01-09, 22:46)quixers Wrote: Actually I mean PR5958
Reckon you are referencing the wrong PRs Ned Scott?
PR5842 (as referenced in the binary filenames and above) is in the master branch already and actually caused this resume point bug, it'll have been PR5958 which is the fix which hasn't been merged up which must have been patched in... Its really PR5958 we are testing here
No. I have made a special build that contains the fix on top of the current Helix branch:
https://github.com/nedscott/xbmc/tree/He...42-testing
These builds do contain the fix.
Posts: 31,445
Joined: Jan 2011
2015-01-10, 15:21
(This post was last modified: 2015-01-10, 15:22 by Ned Scott.)
(2015-01-10, 00:50)choekstr Wrote: Let me ask another way as I probably wasn't clear: Now that we have a test build confirmed to be have a working Stop+resume, do we have a PR that is for Pause+resume which is not working yet?
As far as I know, resume points are only made when stop or shutdown commands are used. I would not expect resume points to be made for just using "pause". If that is happening, it might be due to something else and is just a nice extra bonus that only happens in some situations.
Posts: 31,445
Joined: Jan 2011
New builds posted.
See the first post for links.
A few more things have gone in as you can see at
https://github.com/xbmc/xbmc/commits/Helix
Be sure to install and see if any of the fixes have any negative effect. They shouldn't, in theory, but that's what this thread is all about finding out :)
So far it seems the only issue we have is with ODROID-C1. We'll see what can be found out about that. I have a different amlogic S805 device that might behave the same way and I will test later tonight.
Remember, this thread is only for regressions between v14.0 and v14.1. Please post issues that effect both versions in a new thread.
Thanks again to everyone who is helping test.
Posts: 31,445
Joined: Jan 2011
Oh, and I'll try to get an x86 Android build updated as well. That one needs to be built from a different code branch, IIRC. Not sure how many of those users we have, but might as well make it available.
