•   
  • 1
  • 302
  • 303
  • 304(current)
  • 305
  • 306
  • 355
  •   
  Thread Closed
v18 LibreELEC Testbuilds for x86_64 (Kodi 18.0)
(2018-10-20, 20:58)theboy181 Wrote: I'm not sure if I encountered a BUG or if there was a development change regarding my WIFI PCIE card.

I installed LibreELEC then updated to LibreELEC-Generic.x86_64-9.0-Milhouse-20181018210245-#1018-g86e23ee via the UPDATE folder, using this installed (Rosewill RNX-AC1900PCE) WIFI card.

https://wikidevi.com/wiki/Rosewill_RNX-AC1900PCE

Please upload the latest zip file from the Logfiles Samba share - you'll need to do this over a wired network connection.

(2018-10-20, 20:58)theboy181 Wrote: After the update the WIFI stopped working. I have tested several builds to see where it broke, and going all the way back to AUGUST 2018 was still a no go.

Try going back all the way to October 2017 (oldest is currently #1012 [2017]). Identifying when this stopped working might be useful if it's not obvious from the zip files.
Texture Cache Maintenance Utility: Preload your texture cache for optimal UI performance. Remotely manage media libraries. Purge unused artwork to free up space. Find missing media. Configurable QA check to highlight metadata issues. Aid in diagnosis of library and cache related problems.
New LibreELEC.tv Leia build #1020: Generic
(Supercedes previous build)

SHA256 Checksum: 96d1b6da1c502aab7c4d9b4d39e9b02622887d3618b176fd5f0b8488b2330dc2 (Generic)

# uname -a
Linux NUC 4.18.15 #1 SMP Sat Oct 20 21:04:58 BST 2018 x86_64 GNU/Linux

# lsb_release
LibreELEC (Milhouse): devel-20181020210325-#1020-g86e23ee [Build #1020]

# Kodi version
(18.0-BETA4 Git:938146c). Platform: Linux x86 64-bit

Based on tip of LibreELEC.tv master (86e23ee, changelog) and tip of XBMC master (938146c, changelog) with the following modifications: Build Highlights:
  1. [jsonrpc] set movie collection plots and get/set season titles
  2. DVB Updates: add rc/ir again
Build Details:
  1. XBMC:
    • [jsonrpc] set movie collection plots and get/set season titles (PR:14523, 2 commits, 4 files changed)
    • [fix][win10] relocate InitOSKeymap after #14511 (PR:14651, 1 commit, 2 files changed)
  2. pvr.mythtv:
    • sync upstream cppmyth (2.10.2) (51a9cd3)
    • bump version 5.8.9 (9b543d1)
  3. Additional commits/pull requests/changes not yet merged upstream:
    • Updated: [env] PR:3001 (perma): DVB driver addons update to latest
Texture Cache Maintenance Utility: Preload your texture cache for optimal UI performance. Remotely manage media libraries. Purge unused artwork to free up space. Find missing media. Configurable QA check to highlight metadata issues. Aid in diagnosis of library and cache related problems.
Hi @fritsch I tried today test 8K channels from satellite (some test) but Kodi when I'm trying play constantly crash Sad 
same story I had with with 4K channels when I didn't have custom xorg.conf for add missing modes like [email protected]/60, after I added 4K channels start playing.

(im tried delisted from whitelist mode all modes, switch off picture sync etc. nothing help)

But in that case Kodi suppose do downscaling ? I noticed that problem start happen when white modelist future appear in Kodi, before even on old Apollo Lake I tested some
samples 8k file and works smooth, now is not possible Sad

here some samples: tvraid.zapto.org:9999 - try also LG4K-3D picture flickering :/


Update:
i didn't found info about limitation of VAAPI in source but here:
i media studio is https://github.com/intel/media-driver#su...g-features is table with max res. and from KabyLake for HEVC and VP9 is 8k.

I tried webm VP9 - 8k but is softwared
Nobody knows ...
First decide what functions / features you expect from a system. Then decide for the hardware. Don't waste your money on crap.
New LibreELEC.tv Leia build #1021: Generic
(Supercedes previous build)

SHA256 Checksum: b0720cbc0bd12dd05498efd27f75791343aa4b2f335e892ab255b76f1f737f45 (Generic)

# uname -a
Linux NUC 4.18.16 #1 SMP Sun Oct 21 21:05:01 BST 2018 x86_64 GNU/Linux

# lsb_release
LibreELEC (Milhouse): devel-20181021210318-#1021-g86e23ee [Build #1021]

# Kodi version
(18.0-BETA4 Git:fe875cf). Platform: Linux x86 64-bit

Based on tip of LibreELEC.tv master (86e23ee, changelog) and tip of XBMC master (fe875cf, changelog) with the following modifications: Build Highlights:
  1. Scraper/JSON/GUI artwork updates
  2. CCurlFile Exists/Stat changes
  3. kernel: Fix for hw csum failure
Build Details:
  1. XBMC:
    • [artwork] Scrapers set separate artwork for a movie and the collection (PR:13563, 2 commits, 5 files changed)
    • [videolibrary] Assign extra artwork from file system (PR:13859, 1 commit, 5 files changed)
    • [guilib] attach collection artwork to movie items as "set.[arttype]" (PR:13564, 2 commits, 2 files changed)
    • X11: ability to start headless and trigger on change of EDID (PR:14612, 4 commits, 8 files changed)
    • CCurlFile::Exists() to fallback to GET if 405 is returned (PR:14586, 1 commit, 1 file changed)
    • CCurlFile::Stat missing easy_release on early return (PR:14587, 1 commit, 1 file changed)
    • [addons] force a repo refresh (ignore checksum) if repo version changed (PR:14656, 2 commits, 2 files changed)
    • Fix broken controller input in games (PR:14658, 1 commit, 1 file changed)
  2. inputstream.adaptive:
    • [HEVC] Retrieve fps information from HVCC atom payload (8bfd456)
    • [DASH] release live start segment calculation (sling) (d400211)
    • Version bump 2.3.3 (e42f85f)
  3. pvr.mythtv:
    • fix timeshift info for recording being recorded (9686b9f)
    • bump version 5.8.10 (e471707)
  4. Additional commits/pull requests/changes not yet merged upstream:
    • Updated: [env] PR:3006 (perma): linux (Generic/RPi): update to linux-4.18.16
    • Added: [pkg] patch: net: fix pskb_trim_rcsum_slow() with odd trim offset (linux)
Texture Cache Maintenance Utility: Preload your texture cache for optimal UI performance. Remotely manage media libraries. Purge unused artwork to free up space. Find missing media. Configurable QA check to highlight metadata issues. Aid in diagnosis of library and cache related problems.
New LibreELEC.tv Leia build #1022: Generic
(Supercedes previous build)

SHA256 Checksum: 2c5b31caa1ad9db9b51a7a2c797cc20e04c4409da26d69202800501c8d201f58 (Generic)

# uname -a
Linux NUC 4.18.16 #1 SMP Mon Oct 22 21:06:09 BST 2018 x86_64 GNU/Linux

# lsb_release
LibreELEC (Milhouse): devel-20181022210409-#1022-g86e23ee [Build #1022]

# Kodi version
(18.0-BETA4 Git:d92d5d1). Platform: Linux x86 64-bit

Based on tip of LibreELEC.tv master (86e23ee, changelog) and tip of XBMC master (d92d5d1, changelog) with the following modifications: Build Highlights:
  1. Game API: v1.1.0 - Sync with RetroPlayer test builds
  2. libssh: update to 0.8.4
Build Details:
  1. XBMC:
    • [addons] sync with repo (PR:14655, 1 commit, 15 files changed)
    • Fix NDK r18 Android x86 ffmpeg dependency build (PR:14615, 1 commit, 1 file changed)
    • Game API: v1.1.0 - Sync with RetroPlayer test builds (PR:14660, 3 commits, 12 files changed)
    • Fix cue sheet iteration ignoring the last line (PR:14642, 1 commit, 1 file changed)
    • Cleanup includes - Part 3 (PR:14644, 19 commits, 168 files changed)
  2. inputstream.rtmp:
    • [depends] bump openssl to 1.1.0h (PR:41, 3 commits, 9 files changed)
  3. Additional commits/pull requests/changes not yet merged upstream:
    • Updated: [env] PR:3006 (perma): linux (Generic/RPi): update to linux-4.18.16
    • Added: [env] PR:3036 (perma): libssh: update to 0.8.4
Texture Cache Maintenance Utility: Preload your texture cache for optimal UI performance. Remotely manage media libraries. Purge unused artwork to free up space. Find missing media. Configurable QA check to highlight metadata issues. Aid in diagnosis of library and cache related problems.
(2018-10-19, 21:29)menno Wrote:
(2018-10-19, 10:20)menno Wrote:
(2018-10-07, 22:36)fritsch Wrote: Do you have custom configured OVERSCAN values? E.g. in kodi's gui calibration? Can you reset them please?    
 Hi Fritsch,
Woud this be considered custom settings?
Code:
    <resolutions>
        <resolution>
            <description>[email protected] 60.00 - Full Screen</description>
            <subtitles>1015</subtitles>
            <pixelratio>1.000000</pixelratio>
            <refreshrate>60.000000</refreshrate>
            <output>HDMI1</output>
            <xrandrid>0x48</xrandrid>
            <overscan>
                <left>21</left>
                <top>13</top>
                <right>1826</right>
                <bottom>1054</bottom>
            </overscan>
        </resolution>
        <resolution>
            <description>HDMI1: 1920x1080 @ 60.00Hz</description>
            <subtitles>1042</subtitles>
            <pixelratio>1.000000</pixelratio>
            <refreshrate>60.000000</refreshrate>
            <output>HDMI1</output>
            <xrandrid>0x48</xrandrid>
            <overscan>
                <left>0</left>
                <top>0</top>
                <right>1920</right>
                <bottom>1080</bottom>
            </overscan>
        </resolution>

I do remember back in the days you had some special settings for this chipset and hw scaling and something i believe with 20% etc, I cannot find that in the current version though anymore. This is a snippit from guisettings.xml   
Ok ill drop it tomorrow.
  Latest nightly didnt fix the problem. I am back to "d" for now and drop the guisettings.xml stuff tomorrow for further test  
Running latest nightly 22 oct, I still have the borders on bottom and right side of the screen. Removing Overscan values in guisettings didnt help. The custom version "d"  released earlier works like a charm.
   It seems the overscan values mentioned before are some sort of hard setting cause they return in guisettings.xml after reboot.... so that might be a factor why removing them doesnt do anything. On the other hand with those seemingly standerd values and release "d"  it works like a charm.

// to comment a little more. I dont think this has to do with overscan values. Ill explain why,
     On video calibration mode, the so called artificats remain within the corners of video calibration. So the pixels on the bottom and right side of the screen which remain old are arent refreshed do still fall within the video calibration overscan values. 
   If the pixels would be outside the whole tweaking area I would assume it could be an overscan problem.
(2018-10-03, 15:33)Milhouse Wrote: Those with the artefact problem, can you please test the following builds:

1002b: Generic - same as #0813
1002c: Generic - 1002b + PR14291
1002d: Generic - 1002c + PR14282
1002e: Generic - 1002d + PR14295

1002b should not have the artefact issue, while 1002e should have the artefact issue. I'm not expecting c & d to have it, but who knows for sure so please test them too.

Assuming the issue is confirmed as starting with 1002e then I might try and narrow it down to the specific commit with additional test builds.

Edit: We still need a debug log from an affected build which may reveal some specific GL/mesa behaviour that is hardware dependent.
 As stated before "d" is fine, new info "e"  gives me artifacts. Hope this helps in hunting it down Smile
Nobody is currently hunting. Custom offsets in calibration are NOT suggested at all, this entirely drives the 1:1 pixel mapping nuts. Please fix the root-cause with your TV and its Fixed / Just Scan / Whatever Setting.
First decide what functions / features you expect from a system. Then decide for the hardware. Don't waste your money on crap.
New LibreELEC.tv Leia build #1023: Generic
(Supercedes previous build)

SHA256 Checksum: d4c942aeb60ac100a18d70672a010966656fcaed7aa9729a0ed8224f28e198df (Generic)

# uname -a
Linux NUC 4.19.0 #1 SMP Tue Oct 23 23:00:24 BST 2018 x86_64 GNU/Linux

# lsb_release
LibreELEC (Milhouse): devel-20181023225233-#1023-g46f6d4e [Build #1023]

# Kodi version
(18.0-BETA5 Git:6423651). Platform: Linux x86 64-bit

Based on tip of LibreELEC.tv master (46f6d4e, changelog) and tip of XBMC master (6423651, changelog) with the following modifications: Build Highlights:
  1. New 4.19.0 kernel (with DVB Latest addons, but without CrazyCat drivers)
  2. Kodi 18 Beta 5 starts
  3. pvr.iptvsimple: RapidXML new package
Build Details:
  1. LibreELEC.tv:
    • cleanup package.mk (PR:3035, 7 commits, 630 files changed)
    • config/functions: next wave of cleanups (PR:3040, 5 commits, 1 file changed)
    • mesa: update to mesa-18.2.3 (PR:3043, 1 commit, 1 file changed)
    • libssh: update to 0.8.4 (PR:3036, 1 commit, 2 files changed)
    • linux (Generic/RPi): update to linux-4.18.16 (PR:3006, 9 commits, 6 files changed)
    • projects/rpi: move device settings to device options (PR:3046, 1 commit, 5 files changed)
    • dvb-addons: accidentally dropped PKG_SECTION (PR:3047, 1 commit, 3 files changed)
    • busybox: quit countdown timers on keyboard input (PR:3041, 1 commit, 1 file changed)
    • kodi: v18 Beta 4 (PR:3049, 5 commits, 65 files changed)
    • busybox: update to 1.29.3 (#3042) (8d22f6c)
    • xf86-video-nvidia: update to xf86-video-nvidia-410.66 (#3038) (c809a44)
  2. XBMC:
    • changed: remove old shoutcast workaround in curl file (PR:14669, 1 commit, 1 file changed)
    • [windows] bump nfs to 3.0.0 (PR:14668, 1 commit, 5 files changed)
    • [music] Fix compilation with taglib versions < 1.11 (PR:14678, 1 commit, 1 file changed)
    • fixed: do not add rar to subtitle files without support (PR:14648, 1 commit, 1 file changed)
    • added: flag for vfs add-ons with encoded hostnames in urls (PR:14635, 2 commits, 5 files changed)
    • [playlist] Add support for reading XSPF Playlists (PR:14585, 1 commit, 14 files changed)
    • bump to v18.0 beta5 (PR:14633, 1 commit, 1 file changed)
    • [android] bump targetsdk to 26 and handle runtime security (PR:14500, 1 commit, 2 files changed)
    • [Android] switch renderGUI based on surface existance instead onPause/onSuspend (PR:14686, 1 commit, 2 files changed)
    • Fix infinite loop when listitem action is "Play" (PR:14653, 1 commit, 1 file changed)
    • [Estuary] Update addon browser (PR:14687, 1 commit, 1 file changed)
  3. peripheral.joystick:
    • Fix compiler warnings on Windows (PR:156, 1 commit, 1 file changed)
  4. pvr.dvblink:
    • [depends] bump tinyxml2 to 6.2.0 (PR:107, 1 commit, 4 files changed)
  5. pvr.filmon:
    • [depends] add checksum file (PR:87, 1 commit, 1 file changed)
  6. pvr.hdhomerun:
    • [depends] add checksum file (8454ebf)
  7. pvr.iptvsimple:
    • move rapidxml to depends (PR:199, 1 commit, 15 files changed)
  8. pvr.pctv:
    • [depends] add checksum file (e933506)
  9. pvr.stalker:
    • [depends] bump zlib to 1.2.11 (PR:109, 2 commits, 9 files changed)
  10. pvr.vuplus:
    • New options: stream over HTTPS & stream with login (PR:119, 1 commit, 7 files changed)
  11. Additional commits/pull requests/changes not yet merged upstream:
    • Updated: [env] patch: Improve fs-resize, fix delay and simplify
    • Updated: [env] PR:2954 (perma): linux (Generic/RPi): update to linux-4.19
    • Updated: [env] PR:3001 (perma): DVB driver addons update to latest
    • Added: [env] PR:3048 (perma): pvr.iptvsimple update addon and add rapidxml
Texture Cache Maintenance Utility: Preload your texture cache for optimal UI performance. Remotely manage media libraries. Purge unused artwork to free up space. Find missing media. Configurable QA check to highlight metadata issues. Aid in diagnosis of library and cache related problems.
(2018-10-23, 14:43)menno Wrote:
(2018-10-19, 21:29)menno Wrote:
(2018-10-19, 10:20)menno Wrote:  Hi Fritsch,
Woud this be considered custom settings?
Code:
    <resolutions>
        <resolution>
            <description>[email protected] 60.00 - Full Screen</description>
            <subtitles>1015</subtitles>
            <pixelratio>1.000000</pixelratio>
            <refreshrate>60.000000</refreshrate>
            <output>HDMI1</output>
            <xrandrid>0x48</xrandrid>
            <overscan>
                <left>21</left>
                <top>13</top>
                <right>1826</right>
                <bottom>1054</bottom>
            </overscan>
        </resolution>
        <resolution>
            <description>HDMI1: 1920x1080 @ 60.00Hz</description>
            <subtitles>1042</subtitles>
            <pixelratio>1.000000</pixelratio>
            <refreshrate>60.000000</refreshrate>
            <output>HDMI1</output>
            <xrandrid>0x48</xrandrid>
            <overscan>
                <left>0</left>
                <top>0</top>
                <right>1920</right>
                <bottom>1080</bottom>
            </overscan>
        </resolution>

I do remember back in the days you had some special settings for this chipset and hw scaling and something i believe with 20% etc, I cannot find that in the current version though anymore. This is a snippit from guisettings.xml    
Ok ill drop it tomorrow.
  Latest nightly didnt fix the problem. I am back to "d" for now and drop the guisettings.xml stuff tomorrow for further test   
Running latest nightly 22 oct, I still have the borders on bottom and right side of the screen. Removing Overscan values in guisettings didnt help. The custom version "d"  released earlier works like a charm.
   It seems the overscan values mentioned before are some sort of hard setting cause they return in guisettings.xml after reboot.... so that might be a factor why removing them doesnt do anything. On the other hand with those seemingly standerd values and release "d"  it works like a charm.

// to comment a little more. I dont think this has to do with overscan values. Ill explain why,
     On video calibration mode, the so called artificats remain within the corners of video calibration. So the pixels on the bottom and right side of the screen which remain old are arent refreshed do still fall within the video calibration overscan values. 
   If the pixels would be outside the whole tweaking area I would assume it could be an overscan problem. 
The problem I'm seeing is simular I think. It looks to me like the black bars are not scaling properly so you see the previous window at the bottom and to the right hand side. The video it's self scales perfectly!
I think the problem started when someone introduced video rotation but I can't be sure. Smile
CPU: I3-6100 - MB: B150M-HD3 - MEM: 8GB DDR4 - HDD: 120GB SSD - MEDIA: 2 x 6T WD RED - TUNER: Hauppauge HVR-2210 - OS: LibreELEC x86_64
New LibreELEC.tv Leia build #1024: Generic
(Supercedes previous build)

SHA256 Checksum: 3f6aabc0d2a1341a113be3c4e772959f7e20d940fe958dfc06100ee7d3c21586 (Generic)

# uname -a
Linux NUC 4.19.0 #1 SMP Wed Oct 24 21:04:19 BST 2018 x86_64 GNU/Linux

# lsb_release
LibreELEC (Milhouse): devel-20181024210252-#1024-g1dcd149 [Build #1024]

# Kodi version
(18.0-BETA5 Git:5b7080a). Platform: Linux x86 64-bit

Based on tip of LibreELEC.tv master (1dcd149, changelog) and tip of XBMC master (5b7080a, changelog) with the following modifications: Build Highlights:
  1. Fix invalid "not available" add-on dependency message
  2. Update libdrm, xf86-video-amdgpu and xf86-video-ati
Build Details:
  1. LibreELEC.tv:
    • visualization.projectM: update to 8f442c5 (PR:2951, 2 commits, 14 files changed)
    • pvr.iptvsimple add rapidxml (PR:3048, 2 commits, 2 files changed)
  2. XBMC:
    • [cmake/Xcode] - fix missing text and icon resources when using target Xcode (PR:14693, 1 commit, 1 file changed)
    • fixed: add-on creation is not thread safe (PR:14679, 1 commit, 2 files changed)
    • add mime types for heif/heic (PR:14680, 1 commit, 1 file changed)
    • Fix crash on exit when checking of libgpg-error lock object size. (PR:14683, 1 commit, 1 file changed)
    • OSX: fix fullscreen on Mojave (PR:14613, 1 commit, 2 files changed)
    • EGLUtils: breakup EGL init process (PR:14652, 1 commit, 7 files changed)
    • windowing/gbm: use CFileHandle to handle the fd (PR:14673, 1 commit, 4 files changed)
    • [GBM] support 10bit output modes and other fixes (PR:14515, 4 commits, 8 files changed)
  3. inputstream.adaptive:
    • [TS] allow short 001 start sequence before SPS (missing byte in extra-data) (a216c05)
  4. pvr.hts:
    • 4.4.0: RDS support and minor changes (PR:381, 3 commits, 6 files changed)
  5. pvr.iptvsimple:
    • [depends] rapidxml: convert patch to crlf line endings (PR:201, 1 commit, 1 file changed)
  6. Additional commits/pull requests/changes not yet merged upstream:
    • Updated: [env] PR:2954 (perma): linux (Generic/RPi): update to linux-4.19
    • Added: [env] PR:3050 (perma): Update libdrm, xf86-video-amdgpu and xf86-video-ati
    • Added: [env] PR:3057 (perma): kodi: Hack around 'not available' notification with built-in dependenies
Texture Cache Maintenance Utility: Preload your texture cache for optimal UI performance. Remotely manage media libraries. Purge unused artwork to free up space. Find missing media. Configurable QA check to highlight metadata issues. Aid in diagnosis of library and cache related problems.
hello i have a problem with builds > #1019. #1019 was the last one without any problems.
when i try to update to #1020 i get a checksum error and it won't update so i skipped that build to #1021 and that one is updating fine but
the screen is black as soon as kodi starts. same behaviour for builds > #1021 to present.

here is a log from #1021:
kodi.log https://del.dog/KODI_log
system.log https://del.dog/system_log
hardware.log https://del.dog/hardware_log
(2018-10-25, 03:58)winux Wrote: hello i have a problem with builds > #1019. #1019 was the last one without any problems.
when i try to update to #1020 i get a checksum error and it won't update so i skipped that build to #1021 and that one is updating fine but
the screen is black as soon as kodi starts. same behaviour for builds > #1021 to present.

here is a log from #1021:
kodi.log https://del.dog/KODI_log
system.log https://del.dog/system_log
hardware.log https://del.dog/hardware_log

I've just downloaded #1020 and upgraded to it, and there is no checksum problem on an i5 Skylake NUC, so any invalid checksum behaviour is down to either a corrupted download, or a hardware error.

The sha256 hash for the tar file is 96d1b6da1c502aab7c4d9b4d39e9b02622887d3618b176fd5f0b8488b2330dc2 which will help rule out a corrupted download.

Once you are sure your download is not corrupted, if you continue to experience checksum errors then I would suggest you run MemTest86 on your system to verify you don't have any RAM issues. There is also the built-in memtester but I'm not sure how good that is for x86 - I would suggest MemTest86 is likely to be more thorough. If you continue to experience checksum errors then I would strongly suggest there is a hardware error of some kind.

Assuming you are able, can you install #1020 and confirm if that has the same black screen problem, so that we can focus on #1021?

In the #1021 log I see these errors:

01:33:48.131 T:140039425247360  NOTICE: Checking resolution 16
01:33:48.205 T:140039425247360 NOTICE: Using visual 0x20
01:33:48.226 T:140039425247360 ERROR: failed to create EGL context
01:33:48.288 T:140039425247360 ERROR: CYUVShaderGLSL::CYUVShaderGLSL - failed to open file gl_shader_frag_texture_lim.glsl
01:33:48.289 T:140039425247360 NOTICE: GL: No pixel shader, fixed pipeline in use
01:33:48.301 T:140039425247360 NOTICE: GL_VENDOR = Intel Open Source Technology Center
01:33:48.301 T:140039425247360 NOTICE: GL_RENDERER = Mesa DRI Intel® 945G
01:33:48.301 T:140039425247360 NOTICE: GL_VERSION = 1.4 Mesa 18.2.3
01:33:48.301 T:140039425247360 NOTICE: GL_SHADING_LANGUAGE_VERSION = 1.20
01:33:48.301 T:140039425247360 NOTICE: GL_EXTENSIONS = GL_ARB_multisample GL_EXT_abgr GL_EXT_bgra GL_EXT_blend_color GL_EXT_blend_minmax GL_EXT_blend_subtract GL_EXT_copy_texture GL_EXT_subtexture GL_EXT_texture_object GL_EXT_vertex_array GL_EXT_compiled_vertex_array GL_EXT_texture GL_EXT_texture3D GL_IBM_rasterpos_clip GL_ARB_point_parameters GL_EXT_draw_range_elements GL_EXT_packed_pixels GL_EXT_point_parameters GL_EXT_rescale_normal GL_EXT_separate_specular_color GL_EXT_texture_edge_clamp GL_SGIS_generate_mipmap GL_SGIS_texture_border_clamp GL_SGIS_texture_edge_clamp GL_SGIS_texture_lod GL_ARB_multitexture GL_IBM_multimode_draw_arrays GL_IBM_texture_mirrored_repeat GL_3DFX_texture_compression_FXT1 GL_ARB_texture_cube_map GL_ARB_texture_env_add GL_ARB_transpose_matrix GL_EXT_blend_func_separate GL_EXT_fog_coord GL_EXT_multi_draw_arrays GL_EXT_secondary_color GL_EXT_texture_env_add GL_EXT_texture_filter_anisotropic GL_EXT_texture_lod_bias GL_INGR_blend_func_separate GL_NV_blend_square GL_NV_light_max_exponent GL_NV_texgen_reflection GL_NV_texture_env_combine4 GL_S3_s3tc GL_SUN_multi_draw_arrays GL_ARB_texture_border_clamp GL_ARB_texture_compression GL_EXT_framebuffer_object GL_EXT_texture_compression_s3tc GL_EXT_texture_env_combine GL_EXT_texture_env_dot3 GL_MESA_window_pos GL_NV_packed_depth_stencil GL_NV_texture_rectangle GL_ARB_depth_texture GL_ARB_shadow GL_ARB_texture_env_combine GL_ARB_texture_env_crossbar GL_ARB_texture_env_dot3 GL_ARB_texture_mirrored_repeat GL_ARB_window_pos GL_EXT_stencil_two_side GL_EXT_texture_cube_map GL_APPLE_packed_pixels GL_ARB_draw_buffers GL_ARB_fragment_program GL_ARB_fragment_shader GL_ARB_shader_objects GL_ARB_vertex_program GL_ARB_vertex_shader GL_ATI_draw_buffers GL_ATI_texture_env_combine3 GL_EXT_shadow_funcs GL_EXT_stencil_wrap GL_MESA_pack_invert GL_MESA_ycbcr_texture GL_ARB_half_float_pixel GL_ARB_point_sprite GL_ARB_shading_language_100 GL_ARB_sync GL_ARB_texture_non_power_of_two GL_ARB_vertex_buffer_object GL_ATI_blend_equation_separate GL_EXT_blend_equation_separate GL_OES_read_format GL_ARB_pixel_buffer_object GL_ARB_texture_rectangle GL_EXT_pixel_buffer_object GL_EXT_texture_compression_dxt1 GL_EXT_texture_rectangle GL_EXT_texture_sRGB GL_ARB_framebuffer_object GL_EXT_framebuffer_blit GL_EXT_packed_depth_stencil GL_APPLE_object_purgeable GL_ARB_vertex_array_object GL_ATI_separate_stencil GL_EXT_gpu_program_parameters GL_EXT_texture_sRGB_decode GL_OES_EGL_image GL_ARB_copy_buffer GL_ARB_map_buffer_range GL_ARB_ES2_compatibility GL_ARB_debug_output GL_ARB_draw_elements_base_vertex GL_ARB_explicit_attrib_location GL_ARB_provoking_vertex GL_ARB_sampler_objects GL_EXT_provoking_vertex GL_ARB_get_program_binary GL_ARB_robustness GL_ARB_separate_shader_objects GL_ANGLE_texture_compression_dxt3 GL_ANGLE_texture_compression_dxt5 GL_ARB_compressed_texture_pixel_storage GL_ARB_internalformat_query GL_ARB_map_buffer_alignment GL_ARB_texture_storage GL_AMD_shader_trinary_minmax GL_ARB_clear_buffer_object GL_ARB_explicit_uniform_location GL_ARB_invalidate_subdata GL_ARB_program_interface_query GL_ARB_vertex_attrib_binding GL_KHR_debug GL_ARB_multi_bind GL_ARB_get_texture_sub_image GL_KHR_context_flush_control GL_KHR_no_error
01:33:48.305 T:140039425247360 ERROR: CYUVShaderGLSL::CYUVShaderGLSL - failed to open file gl_shader_frag_texture_lim.glsl
01:33:48.306 T:140039425247360 NOTICE: GL: No pixel shader, fixed pipeline in use
01:33:48.322 T:140039425247360 ERROR: CYUVShaderGLSL::CYUVShaderGLSL - failed to open file gl_shader_frag_texture_lim.glsl
01:33:48.323 T:140039425247360 NOTICE: GL: No pixel shader, fixed pipeline in use
01:33:48.594 T:140039383140096 NOTICE: Running database version Addons27

I don't really know what is causing this. I don't see any obvious change in #1020, but in #1021 there is a new kernel, and maybe the following could also be responsible:
(2018-10-21, 22:43)Milhouse Wrote: Build Details:
  1. XBMC:
    • X11: ability to start headless and trigger on change of EDID (PR:14612, 4 commits, 8 files changed)

Or it's a hardware failure.
Texture Cache Maintenance Utility: Preload your texture cache for optimal UI performance. Remotely manage media libraries. Purge unused artwork to free up space. Find missing media. Configurable QA check to highlight metadata issues. Aid in diagnosis of library and cache related problems.
(2018-10-23, 20:26)fritsch Wrote: Nobody is currently hunting. Custom offsets in calibration are NOT suggested at all, this entirely drives the 1:1 pixel mapping nuts. Please fix the root-cause with your TV and its Fixed / Just Scan / Whatever Setting.
Finally found the option! On LG it's called 'Just Scan' and found within the aspect ratio menu (had dutch language before so couldn't find it). It's on 'Automatic' by default, but you need to set it to 'On' to change the screen size to somewhat smaller. I can confirm with latest build (#1024) i don't have any artifacts on the screen borders. Looks like fritsch is right; it's a TV issue not a kodi issue.
  •   
  • 1
  • 302
  • 303
  • 304(current)
  • 305
  • 306
  • 355
  •   
  Thread Closed
 
Thread Rating:
  • 24 Vote(s) - 4.5 Average



Logout Mark Read Team Forum Stats Members Help
LibreELEC Testbuilds for x86_64 (Kodi 18.0)4.524