Cannot Play 4K Content - Help Required

General Discussions
,
1

I have recently got a new TV (LG G2 OLED), and 4K content is no longer working.

It worked on my previous TV (a 4K HDR LED).

My current setup is as follows:

Plex Media Server is on a Windows Machine
Plex Media Player is on an Xbox One X

When playing the content, it is showing as Direct Play. There is no transcoding happening.

The stats in task manager are all reasonable, nothing is maxing out at all.

Everything is Ethernet wired, on 1Gb connections. The PC has a 1Gb ethernet port, wired. The router is 1Gb ethernet port, wired. The Xbox One X is a 1Gb ethernet port, wired. These are connected via Tenda PowerLine Adapters, which are also 1Gb ports.

The cables are all Cat 5E (so capable of 1Gbps).

The TV is connected via ARC, to an AV Receiver, which my Xbox One X is going through. Audio settings on the TV set to bitstream. Xbox has passthrough. The setup I have is 5.1 for audio on the receiver. That 5.1 audio track is always being selected when playing the files to ensure Direct Play.

The content which does play, is of a bitrate below 50 Mbps. Anything above 50 Mbps does not play. By “not play”, it either buffers (but still shows as Direct Play, so is NOT a transcoding issue), or the Plex client app freezes at a certain point. All other content plays fine, and it used to play fine on my previous TV. The only thing that has changed in my setup is the TV.

I understand the bitrate shown (for example on one file that does not play, it’s 78 Mbps) is an average. It is as though one part gets a bitrate pushed of let’s say 150 Mbps, it causes it to freeze and crash out.

I did a test of my LAN speed using iperf, and it came back with 50 Mbps.

Does this mean I am not getting the correct speeds to be able to play the content, even though it’s Direct Play, it still uses that bitrate of my LAN? Does upload speed come into it at all?

Does anyone have any recommendations on what I could try to fix the issue?

4K Content Not Playing
2

We should remember correlation is not causation, TV swapping & playback,
but you are right to minimize variables trying to solve this.

If you can swap to the older TV and prove that it works again, let us know.
In the meantime I’ll try to move forward.

You’ve done a great job with the analysis. You gave your own answers.
Yes some films burst the transfer rate that can hit 150 Mbps.
Yes you have a problem if you’re only getting 50 Mbps (which is common for a Tenda).

I’m going to venture that considering the quality of your post and your AV setup,
it’s in your future to have better network hardware.

That’s pretty much it for my solution, but I’ll add some background on video bitrates if you’re curious. When you add a video, Plex starts by calculating the avg bitrate,
total bytes in the video track * 8 / seconds of video
but the avg bitrate and the burst bitrate can differ by a lot.

So Plex’s Deep Analysis examines a video and maps the avg bitrate for eight sections.
The results of Deep Analysis lead to a small piece of information added to your video’s database entry, requiredBandwidths. Below I pasted the relevant piece of XML for my version of 4K Saving Private Ryan, where the bandwidth is too much for your network. You can see the avg bitrate=28432, but the req. bandwidth surges to over 68K

<Stream id="148278" streamType="1" default="1" codec="hevc" index="0" bitrate="28432" bitDepth="10" chromaLocation="topleft" chromaSubsampling="4:2:0" codedHeight="2160" codedWidth="3840" colorPrimaries="bt2020" colorRange="tv" colorSpace="bt2020nc" colorTrc="smpte2084" frameRate="23.976" height="2160" level="153" profile="main 10" refFrames="1" requiredBandwidths="67939,63052,51128,44083,42062,40410,35620,32841" width="3840" displayTitle="4K HDR10 (HEVC Main 10)" extendedDisplayTitle="4K HDR10 (HEVC Main 10)"></Stream>

.
If you click 3 Vertical Dots on a Movie Poster and choose Get Info, then in the Info window that appears you choose in its lower left corner View XML you will see the entries like the one above for your 4K film. If you have chosen to disable Deep Analysis, you won’t see any entry forrequiredBandwidth

Somewhere on these forums I posted once how I ran a script to chart the bitrate of every single frame of a video trying to find what is actually required, and I found it hit 150 Mbps or so. That was within spec for certain video formats, HEVC Main10@L5.1@High, but not for something like HEVC Main10@L5.1@Main. That’s a critical point that some video creators fail to configure. They let it get set to Main not High. See also Wikipedia HEVC Tiers

What would I do? Probably switch to WiFi.

Or I would run fiber optic because it’s so much cheaper than copper CAT 5/6 and isn’t affected by EM capacitative coupling. Amazon Basic sells great 10GBase SR LC LC OM3, and 10GTek sells excellent SFP+ Transceivers. I use a QNAP 10G switch.

EDIT: maths :slight_smile:

3

I think you meant:

(Total bytes * 8) / seconds of video

1 Like
4

Thank you for the info on the bitrates, that is very useful information. Always want to learn more!

I’m not sure if I picked it up from the right bit, but this is what I gathered from the file in question:

size=“78210697162” container=“mkv” deepAnalysisVersion=“6” requiredBandwidths=“215454,91159,85609,82185,80396,79468,77934,76833”

Does that mean I would require 216 Mbps to comfortably play the entire file? There were 64 entries of “RequiredBandwidths”, so I am not fully sure which one I should of been grabbing to get an idea.

I did a test today where I plugged the Plex Media Server PC into the router direct, seperate laptop also direct into the router and did a speed test via iperf and it showed 900Mbps. This does confirm to me the issue is where it is having to go over the powerline adapters, unfortunately. I say unfortunately because it isn’t really an option for me to get the ethernet cables direct. The router is in the hallway, the Plex Media Server in a study upstairs, and the Xbox in a different room upstairs. I’d need two ethernet cables going from the router, 1 into the study, and 1 into the other room.

My only other option, that I looked into is the Devolo Magic 2 LAN PowerLine Adapters, which apparently operate a different way to the standard PowerLine adapters (The Tenda ones I have), and at a higher speed?

Has anyone had experience with these Devolo’s and know whether having these instead, would resolve the issue?

5

Yes that’s an extremely high bit rate. If it were a HEVC file, it would have to be Main10@L6.1@High to be compliant and let players know what to expect. That Profile, Tier, Level combination is standard for an 8K@60, but is far beyond what Plex Players expect.

Have you considered WiFi? You didn’t mention it. Most access points / routers that have WiFi use those rod shaped antennas that point vertically. Those send a flat, circular shaped signal (think frisbee). Those are designed for a 1-story house.

Occasionally you’ll find an Access Point / Router that has antennas you can move.
Pointing those horizontally is how you direct the signal vertically and reach the 2nd floor.
Worth a thought. Lots of stores offer free returns (of the WiFi and Devolo).

1 Like
6

Another potential option, if your home is wired with coax (and has connections near your router, server, and clients) would be MoCA adapters. These can be used to carry Ethernet over coax; my experience is that this is much more stable and performant than powerline adaptors.

I actually use a pair of MoCA 2.5 adaptors now to connect two switches in my environment and am able to achieve 1 Gb speeds over them (the limit of my switches). The only real downside is that they add ~3ms additional latency, but you’re likely experiencing more than that with powerline and would experience similar with Wi-Fi as well.

If coax is an option for you, check out GoCoax’s MoCA adaptors.

7

The PC that the Plex Media Server is on does not have WiFi unfortunately. So in any case, there would be a bottleneck there.

I did plug the PC direct into the router, then tried WiFi on 5GHz to the Xbox. It buggered. Was still direct play. But buffered still, annoyingly.

8

The room where the Plex Media Server is does not have Coax. The room where the Xbox is does, but moot as I would still need a connection out of the Plex Media Server room.

If it was up to me, I’d get everything wired with Ethernet etc. but unfortunately that isn’t an option.

Has anyone had experience with the Devolo Magic LAN 2’s? They are meant to provide better speeds than other PowerLine Adapters. Anyone have experience of this resolving a similar issue for them?

9

LATEST UPDATE:

I plugged Ethernet cable direct into Plex Media Server PC, plugged Ethernet cable direct into Xbox. Files now play. They buffered initially, and at some points during play it buffered very slightly, but for the main, the file played.

So it does seem the powerline adapters are not capable of playing it at these current specs/speed of the new setup. I’m wondering if perhaps the new TV means it requires more bitrates compared to my old tv. Only thing I can think of? Didn’t think it worked like that however.

Has anyone used the Devolo Magic LAN 2’s? Would this provide better speeds and resolve the issue?

10

Powerlan is susceptible to electrical interference, which can be caused by a lot of things. If something causes interferences to increase, speed over powerlan will decrease. And that “something” can be almost anything.
And the sad truth is that you almost have no way to determine what’s causing the interference and how to reduce it.
That’s why real Ethernet wires are still king. Powerlan, or MoCA are crutches.

11

okay so interference bad, but moca good because coax is a shielded signal cable
now back to wifi :slight_smile: where the OP said

What about a dongle?

That’s an affiliate site, but they’re decent.
Maybe Tom’s Hardware has a thing.

12

I will see if I have a WiFi dongle here to test.

Does anyone know how I can test my local LAN WiFi speeds? I know how to test my local LAN for Ethernet, but is there any way I can test LAN WiFi speed between two devices?

13

Normally it shouldn’t make any difference. The usual testing apps target network adapters/network connections. Which means they’re oblivious to the type of network adapters in use.

14

What I mean is, I would like to see what actual speeds my WiFi is producing on the local LAN, is there a way I can test to see those results?

15

If you have a speed test app which can run on both the XBOX and the Plex server device, sure.
But finding these for the XBOX could be a challenge. Most are only designed to test against a server on the internet. But if you perform such a kind of test on both your server and the xbox, you should at least be able to make an educated guess about the connection speed between the two local devices.

If you can at least connect one of the devices per Ethernet wire, you are usually better off. If both have a WiFi antenna, it doesn’t mean that they are communicating directly with each other. They’ll still use your router (i.e. its WiFi antenna) as a relay station.

In general when testing WiFi speeds, you need to keep in mind that the number you see as a test result is not static and is in no way guaranteed. It simply reflects the conditions at the time when the test was run.

16

I used iperf between two laptops, on WiFi, next to eachother, and it reported back 70 Mbps.

Seems like to only solution here is to have everything ethernet hard wired, but that is really tricky, given the location of the router and the other devices.

17

Keep in mind that you can always add a cheap Ethernet switch if you need to connect several devices to your network. Then there is only one Ethernet line required between your router and the Ethernet switch.
A strategically positioned Switch (or maybe even two of them) can save you from running several Ethernet wires side by side over longer distances.

18

besides the iPerf3 that you’re already using? curl is great.

You could have significant interference on certain bands WiFi channels or bands.
For channel testing try changing them but usually Auto works best.
For band testing, it’s worth considering 2.5 GHz is more susceptible but travels farther.
Things to test are enable only 2.5 GHz or 5 GHz bands, not both at the same time.
What you really want is real time S:N graphing. That way you can see bad spots.

Certain WiFI adapters and operating systems and device combinations will let you see signal to noise ratios. On my Mac desktop I hold down the Option key and click the WiFI icon. On the iPad and iPhone, it turns out that Apple disabled S:N Ratio readout years ago for every app on the Appstore, because it caused AntennaGate. Linux ofc is your friend here.

nib.

19

This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.