I’m going to get access to one of the HQ development systems.
I need go spelunking
# lscpu
Architecture: x86_64
CPU op-mode(s): 32-bit, 64-bit
Byte Order: Little Endian
Address sizes: 39 bits physical, 48 bits virtual
CPU(s): 12
On-line CPU(s) list: 0-11
Thread(s) per core: 2
Core(s) per socket: 6
Socket(s): 1
NUMA node(s): 1
Vendor ID: GenuineIntel
CPU family: 6
Model: 166
Model name: Intel(R) Core(TM) i7-10710U CPU @ 1.10GHz
Stepping: 0
CPU MHz: 584.009
CPU max MHz: 4700.0000
CPU min MHz: 400.0000
BogoMIPS: 3199.92
Virtualization: VT-x
L1d cache: 192 KiB
L1i cache: 192 KiB
L2 cache: 1.5 MiB
L3 cache: 12 MiB
NUMA node0 CPU(s): 0-11
Vulnerability Itlb multihit: KVM: Mitigation: Split huge pages
Vulnerability L1tf: Not affected
Vulnerability Mds: Not affected
Vulnerability Meltdown: Not affected
Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl and seccomp
Vulnerability Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2: Mitigation; Enhanced IBRS, IBPB conditional, RSB filling
Vulnerability Srbds: Not affected
Vulnerability Tsx async abort: Not affected
Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acp
i mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_per
fmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64
monitor ds_cpl vmx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic mov
be popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_
fault epb invpcid_single ssbd ibrs ibpb stibp ibrs_enhanced tpr_shadow vnmi flexpriorit
y ept vpid ept_ad fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid mpx rdseed adx s
map clflushopt intel_pt xsaveopt xsavec xgetbv1 xsaves dtherm ida arat pln pts hwp hwp_
notify hwp_act_window hwp_epp md_clear flush_l1d arch_capabilities
@ChuckPa I don’t understand your comment about my kernel version not being able to “hack it”. @emb531 is correct that 5.4 is the 20.04 LTS kernel and 5.4.0-70-generic has no issues with hardware transcoding with PMS 1.22.0.4163. I am using the Intel Compute Runtime.
intel-gmmlib/now 20.3.2 amd64 [installed,local]
intel-igc-core/now 1.0.6646 amd64 [installed,local]
intel-igc-opencl/now 1.0.6646 amd64 [installed,local]
intel-level-zero-gpu/now 1.0.19310 amd64 [installed,local]
intel-ocloc/now 21.11.19310 amd64 [installed,local]
intel-opencl/now 21.11.19310 amd64 [installed,local]
I’m sorry for my choice of words.
As I now better understand , which I’m waiting for absolute confirmation about,
PMS now seems to be built, rather incorrectly IMHO, to Ubuntu 20.04 minimum for HW transcoding.
^^ THIS is what I’m working on resolving.
That as my previous foundation and where my mind was at, Ubuntu 18 or that kernel wouldn’t “hack it” (versions too old). It was not directed at you in any way. I was incorrectly writing what I was thinking (sometimes multi-tasking is detrimental – and that time it was)
I know what has to be done and what Engineering needs to correct.
I’m going to pack my sleeping bag and camp outside their door until done 
No worries and I promise that I didn’t take your comments personally. 
I was concerned from a PMS support perspective, that I would have to consider upgrading my OS/kernel to a non-LTS version in order to get/maintain hardware transcoding.
Thank you for the clarification and please let me/us know if there is any additional information I/we can provide to assist.
Don’t upgrade yet. LTS kernels are supported. This is an issue for Engineering to resolve, not you.
1 Like
Excellent, because I absolutely hate looking at this: 
Not REALLY a solution …
BUT
You could back down to 1.22.0.4163
and give them time to sort it out.
Probem solved.
I bought a new 4K TV and an Apple TV 4K. Please pass my thanks to the engineering team for helping to justify the purchases.
No really, but I am keeping the old TV and non-4K Apple TV. I’m still invested in seeing this issue resolved and I’m holding at 1.22.0.4163.
That’s one way! 
Yes, I will pass along your thanks
Any update on a fix @ChuckPa? Interested to try out the new home screen management, but don’t want to lose HW transcoding for HDR → SDR.
i dont know if this is related, but im using an nvidia shield with a tv (no surround or whatever)
with an hdr 4k movie, plex tries to transcode it reason being “eac3 not supported.”
If I use the default version of plex that came with the shield (not the latest update) it plays fine with no transcoding
Because of issues with PMS 1.21.1.4200 - which was only Beta,
I would like to ask everyone to upgrade to PMS 1.21.1.4288
There is nothing to be gained by debugging a version which had problems and has since been superseded.
PMS 1.21.1.4200 was released on the main download page, so that is weird if it was a beta. I mentioned above that the issue persists with PMS 1.21.1.4288 and provided updated logs.
For now, I’ve been holding at PMS 1.22.0.4163 aside from updating temporarily to provide logs and help test your preinst/postinst scripts.
That’s correct. Holding where you currently are at is correct.
I don’t know if you’ve seen but I have updated them again (Friday night).
I’m trying to work out that last little bit of reporting (the messaging) so it works on all distros.
Were you able to review? I corrected what you caught before that with the 4 errors.
What I’ve done is, as instructed, migrated all -9xxx and above CPUs to Intel Compute Runtime.
PMS build 4200 completed the transition to ICR for those machines.
I’d like to add that I also have this issue. I’m running unraid 6.9.1 with latest docker PMS and having the exact issues as OP. CPU is a Intel i3-10100.
Folks,
I’ve chatted with the engineer who did the work. Yes, everything should be supported with one proviso. Current Intel Compute Runtime release versions must be installed.
When they are, we can generate logs to submit.
We’re going to enable both PMS and Transcoder verbose logging to capture the failure. This requires a bit of manual work for transcoder logging but is easily done.
You’ll want to be able to set to VERBOSE transcoder logs now.
Then probably return to normal DEBUG after.
Here are two scripts I’ve created. They extract your PlexToken on the fly and use it for setting the preference.
[chuck@lizum ~/bin.452]$ cat set-transcoder-debug
#!/bin/sh
Token="$(cat "/var/lib/plexmediaserver/Library/Application Support/Plex Media Server/Preferences.xml" | tail -1 | sed -e 's/.*PlexOnlineToken="//' | sed -e 's/".*//')"
curl http://127.0.0.1:32400/:/prefs/set\?TranscoderLogLevel=debug\&X-Plex-Token=$Token
echo Token=$Token
[chuck@lizum ~/bin.453]$
[chuck@lizum ~/bin.453]$ cat set-transcoder-verbose
#!/bin/sh
Token="$(cat "/var/lib/plexmediaserver/Library/Application Support/Plex Media Server/Preferences.xml" | tail -1 | sed -e 's/.*PlexOnlineToken="//' | sed -e 's/".*//')"
curl http://127.0.0.1:32400/:/prefs/set\?TranscoderLogLevel=verbose\&X-Plex-Token=$Token
[chuck@lizum ~/bin.454]$
Both these presume they are being invoked from the server itself (127.0.0.1) . Adjust as appropriate.
What we’re going to do:
REQ: With PMS 1.22.1.4288 installed
- Set transcoder Prefs to Verbose
- Set PMS logging to VERBOSE
- Recreate the failure
- Download the logs ZIP
- Now restore PMS and Transcoder logging to DEBUG
- Attach ZIP file
- Attach XML of item being played ( the
<media> ..... </media>part which shows the details of the streams found in the file .
Plex Media Server Logs_2021-03-28_18-42-33.zip (2.9 MB)
<Media id="80557" duration="9066112" bitrate="18687" width="3840" height="2160" aspectRatio="1.78" audioChannels="6" audioCodec="eac3" videoCodec="hevc" videoResolution="4k" container="mkv" videoFrameRate="24p" videoProfile="main 10">
<Part accessible="1" exists="1" id="174363" key="/library/parts/174363/1608920021/file.mkv" duration="9066112" file="/mnt/htpcvol2/Movies/Wonder Woman 1984 (2020).mkv" size="21180090476" container="mkv" deepAnalysisVersion="4" requiredBandwidths="27710,27531,27095,26394,25715,25183,22024,19724" videoProfile="main 10">
<Stream id="378459" streamType="1" default="1" codec="hevc" index="0" bitrate="17919" bitDepth="10" chromaSubsampling="4:2:0" codedHeight="2160" codedWidth="3840" colorPrimaries="bt2020" colorRange="tv" colorSpace="bt2020nc" colorTrc="smpte2084" frameRate="23.976" height="2160" level="150" profile="main 10" refFrames="1" requiredBandwidths="26942,26763,26327,25626,24946,24415,21256,18956" width="3840" displayTitle="4K (HEVC Main 10 HDR)" extendedDisplayTitle="4K (HEVC Main 10 HDR)"> </Stream>
<Stream id="378460" streamType="2" selected="1" default="1" codec="eac3" index="1" channels="6" bitrate="768" language="English" languageCode="eng" audioChannelLayout="5.1(side)" requiredBandwidths="768,768,768,768,768,768,768,768" samplingRate="48000" title="English" displayTitle="English (EAC3 5.1)" extendedDisplayTitle="English (EAC3 5.1)"> </Stream>
<Stream id="378461" streamType="3" codec="srt" index="2" bitrate="0" language="English" languageCode="eng" requiredBandwidths="1,1,1,1,1,1,1,1" title="English [SDH]" displayTitle="English (SRT)" extendedDisplayTitle="English [SDH] (SRT)"> </Stream>
<Stream id="378462" streamType="3" codec="srt" index="3" bitrate="0" language="Español" languageCode="spa" requiredBandwidths="1,1,1,1,1,1,1,1" title="Spanish [SDH]" displayTitle="Español (SRT)" extendedDisplayTitle="Spanish [SDH] (Español SRT)"> </Stream>
</Part>
</Media>
… Forwarded
Hi there,
Well, I’m running an Intel i5-9600K in a and it uses none of them all.
I’m using Unraid with GitHub - linuxserver/docker-plex, and it was working for the past 4 months
but it’s failing now.
I’m now using a workaround to recompile and install Intel Beignet driver, which does the fix, but still is not an ideal scenario.
***GUIDE*** Plex Hardware Acceleration using Intel Quick Sync - Page 9 - Docker Containers - Unraid
I actually had the same issue when they introduced this new tone mapping feature and it seems some 9th gen Intel CPU (mine) are not properly recognised by the old Beignet driver, and nor the new ICR one…
