[mythtv-users] Troubleshooting video problems on my combined FE/BE
Jerry
mythtv at hambone.e4ward.com
Thu Mar 12 16:09:16 UTC 2015
On Mon, Mar 9, 2015 at 9:17 PM, Jerry <mythtv at hambone.e4ward.com> wrote:
> I'm experiencing some video glitches on my FE/BE. Typically, they occur
> when there is a long shot of video where the camera pans an area for ten
> seconds or so. I also see other glitches at other times. It is connected
> to a 2014 Samsung Smart TV by an HDMI cable. The system is a Core 2 Quad
> Q9400 with 8 GB ram, and the OS runs on an SSD drive. The recordings are
> on a 3 TB hard drive.
>
> I have tried switching video playback modes. I have a GTX 260 card and I
> have been trying to use VDPAU Slim. I have tried Normal as well, and Slim.
>
> I am not seeing these issues on my frontend running on a Core 2 Duo
> machine on the wired network. It has a GTX 9600 card. I only just
> installed it tonight to troubleshoot the problem.
>
> I produced this log with -v playback. I apologize if it is too much
> information, but I was unable to see any problems with the default
> logging. I'd be happy to run it again with different levels of debugging.
>
> I put a GTX 560 Ti in this machine and I still got the same dropping of
> frames. I followed JYA's guide for judder free playback in case it is due
> to issues with the refresh rate. I have all of the standard refresh rates
> incorporated into my xorg.conf as per the instructions on the wiki.
>
> Does anyone have any ideas? A few days ago, I was only a click away from
> getting a new video card, but this does not seem to be the problem if my
> oldest computer with the slowest video card has no problems with playback.
>
> I am not seeing any spikes in CPU usage during playback. I do have a
> number of other services running at the moment such as Plex, BubbleUPnP
> Server and Serviio. I disabled all but Plex just now and I'm still seeing
> the video glitches.
>
> Thanks for any help!
> Jerry
>
> http://filebin.ca/1uHt1b5x3ADK/mythfrontend.20150310002102.29005.log.bz2
>
Replying to myself here.
Is this CPU usage normal? I clipped it here, but basically one CPU hits
100% for each instance (60 instances) save for the first instance of 4 CPUs
at 100%
Search "jitterometer" (60 hits in 1 file)
C:\Users\freetacos\Desktop\mythfrontend.20150310002102.29005.log (60 hits)
Line 442: 2015-03-09 20:21:37.496352 I [29005/29005] CoreContext
jitterometer.cpp:119 (RecordEndTime) - Player(0): FPS: 24.35 Mean: 41076
Std.Dev: 9575 CPUs: 100% 100% 100% 100%
Line 487: 2015-03-09 20:21:41.717137 I [29005/29005] CoreContext
jitterometer.cpp:119 (RecordEndTime) - Player(0): FPS: 28.20 Mean: 35466
Std.Dev: 5568 CPUs: 5% 100% 9% 6%
Line 506: 2015-03-09 20:21:45.804480 I [29005/29005] CoreContext
jitterometer.cpp:119 (RecordEndTime) - Player(0): FPS: 29.12 Mean: 34345
Std.Dev: 7950 CPUs: 3% 100% 8% 7%
Line 685: 2015-03-09 20:21:50.725992 I [29005/29005] CoreContext
jitterometer.cpp:119 (RecordEndTime) - Player(0): FPS: 24.18 Mean: 41354
Std.Dev: 10016 CPUs: 5% 100% 10% 9%
Line 865: 2015-03-09 20:21:55.730891 I [29005/29005] CoreContext
jitterometer.cpp:119 (RecordEndTime) - Player(0): FPS: 23.78 Mean: 42055
Std.Dev: 10504 CPUs: 4% 100% 8% 7%
Line 1044: 2015-03-09 20:22:00.685684 I [29005/29005] CoreContext
jitterometer.cpp:119 (RecordEndTime) - Player(0): FPS: 24.02 Mean: 41634
Std.Dev: 10688 CPUs: 3% 100% 9% 8%
Line 1224: 2015-03-09 20:22:05.657227 I [29005/29005] CoreContext
jitterometer.cpp:119 (RecordEndTime) - Player(0): FPS: 23.94 Mean: 41775
Std.Dev: 9301 CPUs: 2% 100% 9% 10%
Line 1378: 2015-03-09 20:22:10.478599 I [29005/29005] CoreContext
jitterometer.cpp:119 (RecordEndTime) - Player(0): FPS: 24.68 Mean: 40513
Std.Dev: 9399 CPUs: 3% 100% 12% 10%
Line 1455: 2015-03-09 20:22:14.866176 I [29005/29005] CoreContext
jitterometer.cpp:119 (RecordEndTime) - Player(0): FPS: 27.12 Mean: 36868
Std.Dev: 6817 CPUs: 4% 100% 9% 9%
Line 1602: 2015-03-09 20:22:19.654269 I [29005/29005] CoreContext
jitterometer.cpp:119 (RecordEndTime) - Player(0): FPS: 24.85 Mean: 40234
Std.Dev: 8815 CPUs: 4% 100% 10% 10%
Line 1782: 2015-03-09 20:22:24.609198 I [29005/29005] CoreContext
jitterometer.cpp:119 (RecordEndTime) - Player(0): FPS: 24.02 Mean: 41635
Std.Dev: 8314 CPUs: 5% 100% 8% 10%
Line 1962: 2015-03-09 20:22:29.580878 I [29005/29005] CoreContext
jitterometer.cpp:119 (RecordEndTime) - Player(0): FPS: 23.94 Mean: 41776
Std.Dev: 8390 CPUs: 3% 100% 10% 9%
Line 2141: 2015-03-09 20:22:34.535486 I [29005/29005] CoreContext
jitterometer.cpp:119 (RecordEndTime) - Player(0): FPS: 24.02 Mean: 41633
Std.Dev: 8375 CPUs: 3% 100% 10% 11%
Line 2328: 2015-03-09 20:22:39.507092 I [29005/29005] CoreContext
jitterometer.cpp:119 (RecordEndTime) - Player(0): FPS: 23.94 Mean: 41775
Std.Dev: 8333 CPUs: 3% 100% 13% 10%
How about this bit? Is the video skipping from progressive to interlaced?
I don't know much of what is going on. This fills up most of the above log.
2015-03-09 20:21:44.186046 I [29005/29005] CoreContext mythplayer.cpp:723
(AutoDeint) - Player(0): progressive frame seen after 166 interlaced frames
2015-03-09 20:21:44.219490 I [29005/29005] CoreContext mythplayer.cpp:706
(AutoDeint) - Player(0): interlaced frame seen after 1 progressive frames
2015-03-09 20:21:44.219499 I [29005/29005] CoreContext mythplayer.cpp:710
(AutoDeint) - Player(0): Locking scan to Interlaced.
2015-03-09 20:21:44.269431 I [29005/29005] CoreContext mythplayer.cpp:723
(AutoDeint) - Player(0): progressive frame seen after 2 interlaced frames
2015-03-09 20:21:44.302852 I [29005/29005] CoreContext mythplayer.cpp:706
(AutoDeint) - Player(0): interlaced frame seen after 1 progressive frames
2015-03-09 20:21:44.302863 I [29005/29005] CoreContext mythplayer.cpp:710
(AutoDeint) - Player(0): Locking scan to Interlaced.
2015-03-09 20:21:45.520706 I [29005/29005] CoreContext mythplayer.cpp:723
(AutoDeint) - Player(0): progressive frame seen after 37 interlaced frames
2015-03-09 20:21:45.554065 I [29005/29005] CoreContext mythplayer.cpp:706
(AutoDeint) - Player(0): interlaced frame seen after 1 progressive frames
2015-03-09 20:21:45.554073 I [29005/29005] CoreContext mythplayer.cpp:710
(AutoDeint) - Player(0): Locking scan to Interlaced.
2015-03-09 20:21:45.604103 I [29005/29005] CoreContext mythplayer.cpp:723
(AutoDeint) - Player(0): progressive frame seen after 2 interlaced frames
2015-03-09 20:21:45.637487 I [29005/29005] CoreContext mythplayer.cpp:706
(AutoDeint) - Player(0): interlaced frame seen after 1 progressive frames
2015-03-09 20:21:45.637496 I [29005/29005] CoreContext mythplayer.cpp:710
(AutoDeint) - Player(0): Locking scan to Interlaced.
2015-03-09 20:21:45.687534 I [29005/29005] CoreContext mythplayer.cpp:723
(AutoDeint) - Player(0): progressive frame seen after 2 interlaced frames
2015-03-09 20:21:45.720878 I [29005/29005] CoreContext mythplayer.cpp:706
(AutoDeint) - Player(0): interlaced frame seen after 1 progressive frames
2015-03-09 20:21:45.720885 I [29005/29005] CoreContext mythplayer.cpp:710
(AutoDeint) - Player(0): Locking scan to Interlaced.
This is a recording of Better Caul Saul on AMC from a HDHR Prime (mpeg2
file). Mythfrontend says it is in 1080p. I tried turning off the
interlacing options in the VDPAU Slim profile but I'm still getting the
same results.
I've only recently been able to record in HD. There was some mixup with my
cablecard, and they had all the HD channels switched off. I called up the
cable company to ask them about it, and they told me that I had to pay for
a rental for a set top box. Then, about a week later, I did a scan and the
channels were there. Jackpot!
I found that my system was fine for SD, but I'm wondering if I'm a little
underpowered? To recap, I have a Core 2 Quad Q9400 with 8 GB ram and a GTX
260. I thought that this would be fast enough. My other slower Core 2 Duo
with the 9600 seems to play this video fine. It is connected with the VGA
output. My BE/FE is connected with an HDMI output to the TV. I don't know
if this makes a difference for interlacing or not.
I could solve this problem by throwing a pile of money at a new computer,
but I'd rather not. This machine serves media just fine as a Plex server
and does a few other tasks. The CPU utilization from top after a fresh
reboot and no mythfrontend usage reads:
top - 11:47:46 up 51 min, 2 users, load average: 0.15, 0.12, 0.14
Tasks: 214 total, 2 running, 210 sleeping, 0 stopped, 2 zombie
%Cpu(s): 4.7 us, 1.3 sy, 0.0 ni, 92.9 id, 1.1 wa, 0.0 hi, 0.0 si,
0.0 st
KiB Mem : 8175884 total, 4945172 free, 1528544 used, 1702168 buff/cache
KiB Swap: 8388604 total, 8388604 free, 0 used. 6417796 avail Mem
With that video playing:
top - 11:52:10 up 55 min, 2 users, load average: 0.22, 0.11, 0.13
Tasks: 212 total, 2 running, 208 sleeping, 0 stopped, 2 zombie
%Cpu(s): 1.5 us, 0.7 sy, 0.0 ni, 97.4 id, 0.3 wa, 0.0 hi, 0.0 si,
0.0 st
KiB Mem : 8175884 total, 4808588 free, 1631484 used, 1735812 buff/cache
KiB Swap: 8388604 total, 8388604 free, 0 used. 6313316 avail Mem
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
1649 jerry 20 0 6604608 444136 152696 S 5.0 5.4 0:23.31
mythfrontend
32 root 20 0 0 0 0 S 0.3 0.0 0:00.41 rcuos/3
807 root 20 0 4218064 350772 18644 S 0.3 4.3 1:34.53 java
1090 root 20 0 187604 61196 39824 S 0.3 0.7 0:06.78 Xorg.bin
1394 mythtv 20 0 4667032 84984 56128 S 0.3 1.0 0:02.98
mythbackend
1651 jerry 20 0 113640 3276 2964 S 0.3 0.0 0:10.10
scroll_hdd_led_
9836 plex 35 15 1847428 55956 10840 S 0.3 0.7 0:06.04 python
10994 jerry 20 0 146496 3920 3236 R 0.3 0.0 0:00.59 top
I have tried setting the computer to run in 720p as well as 1080p. Things
might possibly be better in 720p, but I really can't tell the difference.
To my eyes, watching my TV sitting about 12 feet away, I can't tell the
difference between 720p and 1080p anyway. But I do notice the hiccups in
motion.
I guess I'll have to live with the little hiccups until I upgrade this
system. I'm starting to see those little hiccups everywhere now. I
thought that Kodi was playing the media cleanly, but that doesn't seem to
be the case.
I'd appreciate any guidance or things to try as I'm a little lost. I feel
as if I'm hitting my BE with a hammer and screwdriver trying to tweak
things. I've been searching the web like a fiend, but I'm coming up
short. :)
Thanks!
Jerry
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mythtv.org/pipermail/mythtv-users/attachments/20150312/0d71b7d6/attachment.html>
More information about the mythtv-users
mailing list