Page 16 of 20

Re: testing MrMC 3.2.0

Posted: 13 Oct 2017, 19:12
by timstephens24
I would suggest changing the "Allow HEVC/HDR" wording, since I thought I needed to have that checked to get HDR but in actuality it's completely the opposite.

Re: testing MrMC 3.2.0

Posted: 13 Oct 2017, 20:07
by davilla
well, it follows the same order wording for mediacodec/interlaced on android.

I can't think of anything else that fits the logic, if vtb does not decode it, and afv is unchecked, then ffmpeg gets it. It's a trickle down effect.

A
B
C

If A does not handle, the passes to B, if not B, then C. It B is disabled, A -> C.

Re: testing MrMC 3.2.0

Posted: 13 Oct 2017, 21:06
by timstephens24
Just say "Use VideoToolBox for HEVC/HDR" and then add a note in the bottom saying AVFoundation is preferred instead of that. Or something.

Because A -> B -> C doesn't work but A -> C does.

Re: testing MrMC 3.2.0

Posted: 13 Oct 2017, 21:08
by timstephens24
Side note: can I get an iOS build to test on my iPad Pro with HDR support, please?

Re: testing MrMC 3.2.0

Posted: 13 Oct 2017, 21:35
by davilla
timstephens24 wrote: 13 Oct 2017, 21:06 Just say "Use VideoToolBox for HEVC/HDR" and then add a note in the bottom saying AVFoundation is preferred instead of that. Or something.

Because A -> B -> C doesn't work but A -> C does.
A little redundant as the setting IS under videotoolbox. We could change the description from '"Allow hardware decoding of HEVC/HDR content" to "Use VideoToolBox for HEVC/HDR content but AVFoundation is preferred to get correct color mapping"

Re: testing MrMC 3.2.0

Posted: 13 Oct 2017, 22:05
by timstephens24
Yeah that would work, too, just something to help clear up any confusion.

Re: testing MrMC 3.2.0

Posted: 14 Oct 2017, 10:44
by Everbrave
davilla wrote: 13 Oct 2017, 20:07 well, it follows the same order wording for mediacodec/interlaced on android.

I can't think of anything else that fits the logic, if vtb does not decode it, and afv is unchecked, then ffmpeg gets it. It's a trickle down effect.

A
B
C

If A does not handle, the passes to B, if not B, then C. It B is disabled, A -> C.
From the logical point of view, if-then-else describes an Implication relation. If A then B does not mean, however, If NOT-A then NOT-B. The simplest way is to use an OR-conjunction (Union), i.e., the checked methods are all enabled and can all be used. The selection of which method can be decided based on some "designer preference", e.g. energy efficiency on (e.g. iOS devices), frame-rate, Codec, etc.
My personal opinion is to reduce the menu selections to a minimum and handle such choices internally by the program based on the media and device information or global user setting such as screen resolution and/or compatibility (e.g. HDR), This not only enhances the user experience but also reduces the the -combinatorial- sources of malfunction.
I am aware of the challenges involved in rewriting the UI but it is an important issue for current generation of media players in general.

Re: testing MrMC 3.2.0

Posted: 14 Oct 2017, 20:31
by Lollypop
I am seeing some wierd audio on some of my AC3 files after installing the latest testflight.

Here is a sample: https://www.dropbox.com/s/bmkmoedaw1y6n ... o.m4a?dl=0

Has something changed?

Re: testing MrMC 3.2.0

Posted: 14 Oct 2017, 20:36
by davilla
(20171014-c27c925)
skin.opacity: extend top bar label
darwin: tweaks to AVFoundation/VideoToolBox regarding number of retained demux packets

Re: testing MrMC 3.2.0

Posted: 14 Oct 2017, 20:38
by davilla
Lollypop wrote: 14 Oct 2017, 20:31 I am seeing some wierd audio on some of my AC3 files after installing the latest testflight.

Here is a sample: https://www.dropbox.com/s/bmkmoedaw1y6n ... o.m4a?dl=0

Has something changed?
Please define " installing the latest testflight", were you previously running the 3.2 Testflight series or where you previously running some other version. If latter, which version ?