20 Nov 2019 - TVDB Scraper v3.2.0 is now available which reinstates scraping. TVDB are still in the process of fixing a number of bugs so there may still be further breakages. See this post. 2901570 (post)

  •   
  • 1
  • 89
  • 90
  • 91(current)
  • 92
  • 93
  • 244
  •   
  Thread Closed
Linux Radeon OSS with vdpau (howto)
(2013-12-16, 14:56)fritsch Wrote: Wsnipex is fully right here. We don't have a pulseaudio sink in xbmc yet. So choosing pulse here will a) disable ActiveAE b) disable passthrough completely. Futerhmore the pulse code was not tested nor futher developed (despite some song changing bugs I fixed 6 months ago).

Edit: hacks with pavucontrol I don't consider as sane workarounds.

OK. If I edit the post, then your answer will look strange. What do you prefer?

NB: I still have drops on a E350.
Just edit it and append: "Edit: " ... so that will be clear.

Yeah, i can't see your drops from here. No crystalball available :-) So post a proper logfile, also make sure you got the "nice" correct that I posted yesterday.
First decide what functions / features you expect from a system. Then decide for the hardware. Don't waste your money on crap.
(2013-12-16, 15:49)fritsch Wrote: Just edit it and append: "Edit: " ... so that will be clear.

Yeah, i can't see your drops from here. No crystalball available :-) So post a proper logfile, also make sure you got the "nice" correct that I posted yesterday.

The HTPC used does not start XBMC automatically via init script so the nice setup has to be done differently. The drops/skips I mentionned were just informative. I haven't started to look at log file seriously yet. Anyhow, I will try upgrade to mesa 10.0.1 before hunting performance problems.

I have a lot of "long decoding" and "hurry" in the log file...
I had enabled pulseaudio with my first builds and i got really low volumes (it got tiring to have to launch pulseaudio and adjust the volume from there) and the xbmc volume amplify feature wasnt functional.
So, disabled pulseaudio and i can use the amplification feature.
hoping someone can point me in the right direction.

i've been trying to compile the latest 3.13-rc4 kernel, but it keeps failing on this driver:
Code:
drivers/net/wireless/rtlwifi/rtl8192se/rtl8192se.o
with:
Code:
make[1]: *** [drivers] Error 2

thanks for any help / advice.
Linux Mint 18 LTS 64-bit - Kodi 17 Beta6
Odroid-C2 - Libreelec v7.90.009
(2013-12-16, 18:30)locomot1f Wrote:
Code:
make[1]: *** [drivers] Error 2

This is just the end result.. the actual error should be some lines before that.
For rc4 you don't need to build from source. Ubuntu mainline has it. Though, I did not verify that speakers fix (segfault) went in.
First decide what functions / features you expect from a system. Then decide for the hardware. Don't waste your money on crap.
@fatalsaint, thanks for the reply.

is there some type of log file...?
everything above this is just the typical syntax.

here are the last few lines:
Code:
CC [M]  drivers/net/wireless/rtlwifi/rtl8192se/trx.o
  LD [M]  drivers/net/wireless/ti/wlcore/wlcore.o
  LD [M]  drivers/net/wireless/ti/wlcore/wlcore_spi.o
  LD [M]  drivers/net/wireless/ti/wlcore/wlcore_sdio.o
  LD [M]  drivers/net/wireless/rtlwifi/rtl8723ae/rtl8723ae.o
  LD [M]  drivers/net/wireless/rtlwifi/rtl8192se/rtl8192se.o
make[1]: *** [drivers] Error 2
make[1]: Leaving directory `/root/linux'
make: *** [debian/stamp/build/kernel] Error 2

so, i'm not sure what i should be looking for...
Linux Mint 18 LTS 64-bit - Kodi 17 Beta6
Odroid-C2 - Libreelec v7.90.009
@EricV: Might also be content depending.

Here some rough figures:
1080i50 is max you get done via E350 and temporal deinterlacing or BOB deinterlacing
1080p36 is max you get decoded concerning progressive material

[email protected] is also working, e.g. this: (mo)

Problem is not really the data rate here, but the number of full frames / seconds.

Samples like the jellyfish run fine: http://jell.yfish.us/media/Jellyfish-45-Mbps.mkv

Run a qvdpautest, please.
First decide what functions / features you expect from a system. Then decide for the hardware. Don't waste your money on crap.
(2013-12-16, 19:31)locomot1f Wrote: so, i'm not sure what i should be looking for...

Yeah.. Not sure. Typically make will output the error to the screen. It's possible it was an error previously outputted a while earlier but wasn't realized until that step? I can try rc4 when I get home.. I just did rc3 last night.

But, if you're using Ubuntu, I would do as fritsch suggested.
(2013-12-16, 19:42)fritsch Wrote: @EricV: Might also be content depending.

Here some rough figures:
1080i50 is max you get done via E350 and temporal deinterlacing or BOB deinterlacing
1080p36 is max you get decoded concerning progressive material

Thanks for the info.

Content was 1080p24 (23.9xx in fact). TV supports 24p but for some reason it does not work very well when is force vsync to always on. Did not check yet if it did correctly switch to 24 from 60. Removing vsync lead to nice playback. I mostly have 108024p or 1080i50. More tommorow.
(2013-12-16, 19:56)fatalsaint Wrote: Yeah.. Not sure. Typically make will output the error to the screen. It's possible it was an error previously outputted a while earlier but wasn't realized until that step? I can try rc4 when I get home.. I just did rc3 last night.

But, if you're using Ubuntu, I would do as fritsch suggested.

i'll hold tight... there seems to be some things still getting worked out on the 3.13-rc4 kernel.
there's only one file in the mainline -- http://kernel.ubuntu.com/~kernel-ppa/mai...c4-trusty/
Linux Mint 18 LTS 64-bit - Kodi 17 Beta6
Odroid-C2 - Libreelec v7.90.009
(2013-12-16, 19:42)fritsch Wrote: @EricV: Might also be content depending.

Here some rough figures:
1080i50 is max you get done via E350 and temporal deinterlacing or BOB deinterlacing
1080p36 is max you get decoded concerning progressive material

[email protected] is also working, e.g. this: (mo)

Problem is not really the data rate here, but the number of full frames / seconds.

Samples like the jellyfish run fine: http://jell.yfish.us/media/Jellyfish-45-Mbps.mkv

Run a qvdpautest, please.

qvdpautest 0.5.2
AMD E-350 Processor
Unknown GPU

VDPAU API version : 1
VDPAU implementation : G3DVL VDPAU Driver Shared Library version 1.0


MPEG DECODING (1920x1080): 59 frames/s
MPEG DECODING (1280x720): 92 frames/s
H264 DECODING (1920x1080): 39 frames/s
H264 DECODING (1280x720): 70 frames/s
MPEG4 DECODING (1920x1080): 54 frames/s

MIXER WEAVE (1920x1080): 275 frames/s
MIXER BOB (1920x1080): 460 fields/s
MIXER TEMPORAL (1920x1080): 579 fields/s
MIXER TEMPORAL + IVTC (1920x1080): 580 fields/s
MIXER TEMPORAL + SKIP_CHROMA (1920x1080): 578 fields/s
MIXER TEMPORAL_SPATIAL (1920x1080): 580 fields/s
MIXER TEMPORAL_SPATIAL + IVTC (1920x1080): 578 fields/s
MIXER TEMPORAL_SPATIAL + SKIP_CHROMA (1920x1080): 579 fields/s
MIXER TEMPORAL_SPATIAL (720x576 video to 1920x1080 display): 665 fields/s

MULTITHREADED MPEG DECODING (1920x1080): 54 frames/s
MULTITHREADED MIXER TEMPORAL (1920x1080): 574 fields/s
Nice figures, more than an average E350 by 10% especially 1080p h264.
First decide what functions / features you expect from a system. Then decide for the hardware. Don't waste your money on crap.
(2013-12-17, 10:29)fritsch Wrote: Nice figures, more than an average E350 by 10% especially 1080p h264.

hardware is Zotac AD02. Linux 3.13-rc4, 64 bits debian unstable + experimental parts (as given in initial post) + own cooked kernel with all drivers build in except radeon FTM because of firmware loading (will probably put them in when finished).

BTW I checked the 24p stuff. Xrandr shows 24 and 23.976080 possible frame rate for 1920x1080 for the TV. The log shows that when I sync to display, 23.976080 is selected when I play the film but then the skips appears on a regular basis. Mediainfo says the video stream is at 23.976 =>No obvious bug and no clue on my side for the moment.
  •   
  • 1
  • 89
  • 90
  • 91(current)
  • 92
  • 93
  • 244
  •   
  Thread Closed
 
Thread Rating:
  • 26 Vote(s) - 4.58 Average



Logout Mark Read Team Forum Stats Members Help
Radeon OSS with vdpau (howto)4.5826