Noisy digital audio on LG TVs

[-Matt-]

And that's exactly where it gets not just warm, but really hot.
Of course, the DSC-88 doesn't comply with the latest SPDIF standards, and that's exactly what it's not supposed to do. This is only about stereo (2-channel, not 2.1) standard SPDIF.
The standard was only expanded, not changed, because that would be fatal for any compatibility.
Tens of thousands or millions of devices that work with this standard in the professional audio sector with device generations well over 30 years apart with virtually no problem, but the really big TV manufacturers can't manage this data output?

I understand your point about backward compatibility, but if I were to consider using optical SPDIF, I'd at least want it to do 5.1. (That has been a long standing capability of SPDIF IIRC). From my point of view (and maybe the TV manufacturers) it would be better still if it could bitstream Atmos (obviously, in an ideal world, adding this shouldn't break anything else).

if eARC works for you and your dac/avr/whatever+tv aren't having dropouts there's no reason to make any changes.
if you're having dropouts, then try a different connection method.

Yes - very sensible suggestion. I may be hearing increased noise/distortion rather than dropouts. Unfortunately I've changed many things at once in my setup (AVR change from TI to ESS DACs, speakers, calibration, etc). So at the moment it is a bit hard to pin down. I was initially suspecting the upmixing or loudness compensation processing as these seem to also effect the audibility of the noise. Anyway - I'm asking the questions in general for anyone who has an AVR with ESS chips and an LG TV.

 

[Roland68]

I understand your point about backward compatibility, but if I were to consider using optical SPDIF, I'd at least want it to do 5.1. (That has been a long standing capability of SPDIF IIRC). From my point of view (and maybe the TV manufacturers) it would be better still if it could bitstream Atmos (obviously, in an ideal world, adding this shouldn't break anything else).

Now you're confusing a few things. Of course, you have been able to send 5.1 to your AVR via SPDIF since the 90s, but of course not at the same time as the stereo signal. Two things cannot occupy the same space.
A stereo DAC can't do anything with a 5.1 signal, an AVR can do anything with it. The AVR has to break down the signal and send the respective part to the respective DAC chip. The individual Dac chips in an AVR can't do anything with the 5.1 signal.

 

[-Matt-]

Now you're confusing a few things. Of course, you have been able to send 5.1 to your AVR via SPDIF since the 90s, but of course not at the same time as the stereo signal. Two things cannot occupy the same space.
A stereo DAC can't do anything with a 5.1 signal, an AVR can do anything with it. The AVR has to break down the signal and send the respective part to the respective DAC chip. The individual Dac chips in an AVR can't do anything with the 5.1 signal.

But my question was specifically about whether this problem could also apply to AVRs with ESS DACs. Where multichannel is more normal than stereo.

If the TV sends the digital audio signal (either over optical or HDMI ARC/eARC) and it uses a new encoding method that looks to the ESS DAC like terrible jitter then potentially it causes problems for those of us with ESS DACs in our AVRs too.

Denon/Marantz seem to be shifting to ESS for their latest higher-end AVRs (AKM previously, then TI as a stopgap following the AKM fire and for the lower end models). And LG TVs are perhaps the most popular with enthusiasts. So it would be a big problem if it proved to be the case that there are incompatibilities.

 

[elvisizer]

I may be hearing increased noise/distortion rather than dropouts.

increased compared to what?
spdif is a stream, unless you're having dropouts you're chasing ghosts.

 

[-Matt-]

increased compared to what?

Relative to my previous AVR with TI DACs.

...unless you're having dropouts you're chasing ghosts.

Another user @Tanquen has reported very serious dropouts with a brand new ESS DAC'd Denon AVR which didn't occur with his previous (AKM? DAC'd one):
https://www.audiosciencereview.com/...x4900h-x6800h-announcement.47000/post-1865130

 

[Roland68]

But my question was specifically about whether this problem could also apply to AVRs with ESS DACs. Where multichannel is more normal than stereo.

If the TV sends the digital audio signal (either over optical or HDMI ARC/eARC) and it uses a new encoding method that looks to the ESS DAC like terrible jitter then potentially it causes problems for those of us with ESS DACs in our AVRs too.

Denon/Marantz seem to be shifting to ESS for their latest higher-end AVRs (AKM previously, then TI as a stopgap following the AKM fire and for the lower end models). And LG TVs are perhaps the most popular with enthusiasts. So it would be a big problem if it proved to be the case that there are incompatibilities.

Only the audio information in the audio stream is sent via HDMI. So only what is present in the source material (music/film).
The problems with certain TVs (SPDIF interface) that affect stereo DACs with PCM don't have much to do with the transmission of 5.1 or Dolby Digital to an AVR.
Normal jitter can always occur and depends on the source. But I think it's unlikely that the HDMI connection is worse than the SPDIF. Then I would assume it is defective.
You should also keep in mind that a lot of effort goes into AVRs to remove jitter before the data gets to the DAC chips. This applies even more to the stereo DACs tested for ASR.

I am not aware of a jitter problem like you suspect with current devices. It would have to be very bad for it to be audible and even worse for there to be failures in the connection. Do you have any further information about this?

What problem do you have exactly with the HDMI connection from the TV to the AVR?
What kind of AVR do you have?

 

[-Matt-]

@Tanquen has a brand new AVR-X6800H (reportedly with ESS ES9017S). He previously had an AVR-X6500H (AKM AK4458?). No problems with X6500 - swap in X6800 and immediately bad audio and video dropout. I'm speculating if when the ESS DAC has to resync could this also cause the HDCP handshake to go wrong?


@wyup earlier in thread reported audio dropout problems via ARC. (Presumably on an SMSL SU-10 with ESS ES9038Pro, although not clear as an Onkyo A-9010 is also mentioned).


Mine is a Denon AVC-X8500HA (8x ES9010K2M). Previously I had an AVC-X6700H (8x TI PCM5102A).

I'm certainly not saying that there is a problem with the X8500HA yet (probably there isn't). As I mentioned I think I may be hearing more noise/distortion relative to the previous AVR, but I can't yet rule out that this might be caused by some other aspect of the configuration.

The title of this thread "Noisy digital audio on LG TVs" seemed to pretty accurately describe the problem I recently had: Centre channel seemed particularly grainy/noisy when we recently watched "The I-Land" from Netflix via the LG CX's app with audio via eARC and especially using the Dolby Atmos + Dolby Surround upmixer. (Would be good if someone else could test this, basically any scene with waves breaking on the shore - speech is distorted). Problem was notably worse with: Audio > Surround Parameter > Loudness Management: On.

As you can probably guess, with an AVR there are a lot of possible other factors due to all of the configuration options. Which is why I'm trying to determine if it is even plausible for a badly formatted digital audio stream from the TV to be involved.

Even if the noise in my setup ends up being a red herring, the matter of non-standard digital audio output from LG/Samsung TVs and possible incompatibility with ESS DACs is an interesting/worrying and important one. As the use of ESS DACs in Denon/Marantz AVRs is a relatively recent development, it is possible that so far not many cases of these problems have come to light; but it could be the tip of an iceburg.

 

[Roland68]

@Tanquen has a brand new AVR-X6800H (reportedly with ESS ES9017S). He previously had an AVR-X6500H (AKM AK4458?). No problems with X6500 - swap in X6800 and immediately bad audio and video dropout. I'm speculating if when the ESS DAC has to resync could this also cause the HDCP handshake to go wrong?


@wyup earlier in thread reported audio dropout problems via ARC. (Presumably on an SMSL SU-10 with ESS ES9038Pro, although not clear as an Onkyo A-9010 is also mentioned).


Mine is a Denon AVC-X8500HA (8x ES9010K2M). Previously I had an AVC-X6700H (8x TI PCM5102A).

I'm certainly not saying that there is a problem with the X8500HA yet (probably there isn't). As I mentioned I think I may be hearing more noise/distortion relative to the previous AVR, but I can't yet rule out that this might be caused by some other aspect of the configuration.

The title of this thread "Noisy digital audio on LG TVs" seemed to pretty accurately describe the problem I recently had: Centre channel seemed particularly grainy/noisy when we recently watched "The I-Land" from Netflix via the LG CX's app with audio via eARC and especially using the Dolby Atmos + Dolby Surround upmixer. (Would be good if someone else could test this, basically any scene with waves breaking on the shore - speech is distorted). Problem was notably worse with: Audio > Surround Parameter > Loudness Management: On.

As you can probably guess, with an AVR there are a lot of possible other factors due to all of the configuration options. Which is why I'm trying to determine if it is even plausible for a badly formatted digital audio stream from the TV to be involved.

Even if the noise in my setup ends up being a red herring, the matter of non-standard digital audio output from LG/Samsung TVs and possible incompatibility with ESS DACs is an interesting/worrying and important one. As the use of ESS DACs in Denon/Marantz AVRs is a relatively recent development, it is possible that so far not many cases of these problems have come to light; but it could be the tip of an iceburg.

Please forget the construct you have in your head, it doesn't exist in reality. And that's not meant in a bad way. You're hunting a white rabbit.
- These DAC chips cannot affect the video signal.
- The Dac chips work autonomously in the AVR and only convert the digital signal into analog.
- Before the signal gets to the DAC chips, it is completely divided, processed and prepared by the HDMI/DSP ICs (the actual function of an AVR) and also re-clocked. Any problems you see on the display or that lead to failures occur in this area. Never in any of the DAC chips. These ICs are very advanced and do not pass on an inconsistent signal to the DAC chips.

In my experience, any external streaming is unsuitable for testing, especially when it comes to sound aspects. I generally reject such attempts among friends and acquaintances, as it only leads to a waste of time without any usable results. Take reliable digital material, DVD, BD, SACD, etc. and always feed it from the same source device. So you can test the difference by playing it directly via SPDIF/Toslink via the TV with HDMI return channel and then via the AVR.

 

[Greenman]

I’ve been using toslink from my TVs to an AV receiver for 15+ Years and never had any issues at all. currently using an LG C1 to old Denon Av3802 receiver via 6m toslink cable. No hint of any jitter artifacts, no dropouts. I don’t use eARC because AVR doesn’t support it, but 6.1 DD seems to be produced perfectly Via toslink.

 

[-Matt-]

I’ve been using toslink from my TVs to an AV receiver for 15+ Years and never had any issues at all. currently using an LG C1 to old Denon Av3802 receiver via 6m toslink cable. No hint of any jitter artifacts, no dropouts. I don’t use eARC because AVR doesn’t support it, but 65.1 DD seems to be produced perfectly Via toslink.

Do you know what DACs the Denon Av3802 uses? I doubt it would have ESS DACs (probably AKM) given the vintage, so it wouldn't be expected to have this problem.

 

[DLS79]

If LG or Samsung are dumping out signals that don't comply to standards, then they are the ones causing the issue.

 

[-Matt-]

If LG or Samsung are dumping out signals that don't comply to standards, then they are the ones causing the issue.

Potentially, LG and Samsung are putting out signals that DO comply with the latest version of the standards, and offer improved performance and new features when used with their own brand soundbars etc. ESS may not have updated to support recent developments/extensions of the standard.

At the end of the day, I'm not really interested where the finger of blame is pointed. I hope all parties are working on a fix. I'd rather see technical details that explain exactly what the problem is here.

 

[DLS79]

Potentially, LG and Samsung are putting out signals that DO comply with the latest version of the standards, and offer improved performance and new features when used with their own brand soundbars etc. ESS may not have updated to support recent developments/extensions of the standard.

I find this unlikely, because one of the reasons standards exist, is so that any given standard can change over time without breaking backwards compatibility. Specially standards at this low a level.

for example the 2021 update does not look significant.

https://webstore.iec.ch/publication/66291

IEC 61937-2:2021 cancels and replaces the second edition published in 2007, Amendment 1:2011 and Amendment 2:2018. This edition constitutes a technical revision. This edition includes the following significant technical changes with respect to the previous edition:
a) new audio data-types of MPEG-D USAC, ACX, ACX HBR2, ACX HBR4 and ACX HBR8 have been added;
b) extended data-type field in Pe has been activated.

At the end of the day, I'm not really interested where the finger of blame is pointed. I hope all parties are working on a fix. I'd rather see technical details that explain exactly what the problem is here.

Imo you should be, because if someone isn't following the protocol correctly, they are the ones responsible for fixing the issue, and in some cases having to deal with the legal ramifications of it (depending on the protocol).

 

[-Matt-]

- The Dac chips work autonomously in the AVR and only convert the digital signal into analog.
- Before the signal gets to the DAC chips, it is completely divided, processed and prepared by the HDMI/DSP ICs (the actual function of an AVR) and also re-clocked. Any problems you see on the display or that lead to failures occur in this area. Never in any of the DAC chips. These ICs are very advanced and do not pass on an inconsistent signal to the DAC chips.

I found this part interesting so went looking for a bit more detail just to satisfy my curiosity. The first service manual I found for free online was one for the AVR-X2500H. (This has an AKM DAC not an ESS one so wouldn't be expected to have the problem we are discussing. I'm just using it to get an idea of what goes on inside an AVR). The service manual has some nice diagrams that show the digital audio signal flow:

Digital Coaxial and Optical inputs:

HDMI inputs (including ARC) for stereo?

HDMI inputs (including ARC) for multichannel?

It seems that both the digital coaxial/optical and the stereo HDMI (including ARC) go via a TI PCM9211. This isn't a DAC, it is a DIX (digital audio interface transceiver). I think this is the part that you mention can re-clock the signal. Indeed it incorporates phase locked loops and will give flag signals when it detects sample rate changes etc. It claims to fully comply with various SPDIF standards and to have good jitter rejection.

Is the 3rd diagram showing that the PCM9211 (and hence the reclocking and jitter rejection) is bypassed for multichannel digital that comes in via HDMI?

The next IC in the chain does the format decoding. Then comes the DAC.

Curiosity mostly satisfied for now!

Edit: The X8500HA, X6700H and several other models apparently also use a Cirrus Logic CS2100 for additional jitter reduction. So that gives another possible rabbit hole!

 

[DLS79]

@-Matt- ESS also has a dedicated SOC for soundbars etc.

https://www.esstech.com/wp-content/uploads/2022/09/Crescendo-VII-Product-Brief-v0.5_2.1.pdf

you can get all the datasheets for their standard DAC chips here.

SABRE Audiophile DACs – ESS Technology, Inc.

www.esstech.com

 

[wyup]

@wyup seems to report that there are also problems via HDMI ARC:

I'd like to know for sure if this was ARC or if eARC was also tested.

I used this Amazon ARC extractor:

https://www.amazon.es/gp/aw/d/B07QLTDDHT

I don't remember how I tested It, whether I enabled earc or not on the tv, but surely I tried every option. Dropouts improved but still there. My solution was to connect my dac by USB to the tv. It also has dropouts from time to time but less (every few minutes)

Surprisingly I discovered that if I plug out the TV antena cable all dropouts disappear. Plug in again and return after a few seconds. (Playing from USB media or internet apps).
It must be the tv tuner processing on the operative system that gets the jitter because it acts as a trigger.

I thought of incoming voltage on tv antena but dont think so. Signal is good. There are more people on Samsung forums with Sabre dacs complaining on the same thing. It is a firmware issue from their operative system (hrt timer, clock samplerate sync or whatever).

 

[Tanquen]

If LG or Samsung are dumping out signals that don't comply to standards, then they are the ones causing the issue.

Anything's possible and digital streaming can be a pain. The one issue I have with saying it's LG or Samsung's fault is the fact that the X6500H was fine with my 4-year-old LG C9 TV and it was fine with my 6 month old LG G3 TV. Then I get the X6800H and it can't handle eARC and it has random once twice three times a day dead video output until power is cycled. I'm thinking it has something to do with the new HDMI chip set that also failed to correctly support QMS. Maybe, or maybe someone forgot a setting somewhere in the firmware, I don't know. I just know that I can swap the old one back in and everything's cool and you put the new one in and you change cables and whatever, it still fails repeatedly.

Support wants me to send it in for repair. I don't think that's going to do anything, especially with some folks saying that Denon is now aware and has a fix for eARC and may add it to a firmware at some point but that still leaves me with the randomly dead video output.

It was somewhat infuriating with support constantly telling me it's the cables or the TV and I would explain everything again and then they would tell me, oh that means it cannot be the AVR. Then for the second, third and fourth time, if I put the old one back everything is good. How can you say it's not the new AVR? Was also not happy with them trying to circumvent the issues and telling me to change my setup so I didn't use eARC. I don't care why I'm using a feature on my $3500 AVR, the feature should work. If they don't want to support it, then they need to remove it from their flipping cut sheet.

 

[Greenman]

Then for the second, third and fourth time, if I put the old one back everything is good. How can you say it's not the new AVR?

This type of thing has been covered elsewhere though hasn’t it. How tolerant an AVR (or dac generally) is to jitter or other variations on the input stream is to some degree a design decision. More tolerance can result in reduced SNR on the output. Sites like this pressure the manufacturers to increase the SINAD of the output, this doesn’t encourage the manufacturers to make them as tolerant as possible to input issues. The likelihood is that your new AVR has tighter input tolerances, doesn’t mean it’s necessarily a fault but determining whether the issue is source or AVR isnt viable for the average consumer which leaves you in no man’s land!!

 

[-Matt-]

This type of thing has been covered elsewhere though hasn’t it.

It would really help everyone if we all got in the habit of providing links or references to substantiate statements like this. I don't recall seeing any previous discussion about jitter sensitivity causing AVRs to be temporarily non-funtional.

Regarding the points you go on to make (and based on the emerging anecdotal evidence such as those linked above), I think we are generally in agreement that:

It currently looks as though Denon's move to using ESS DAC chips in recent AVRs may have reduced jitter tolerance, which could lead to an increased ocurrence of audio dropout type problems.

Of course, in the absence of any actual tests or measurements, it may be some time before sufficient anecdotal evidence has accumulated, to make a convincing case.

 

[ZolaIII]

@-Matt- if you set S/PDIF to PCM stereo only it's stereo only end of discussion. If you set beat streaming for newer Dolby's with metadata receiver needs to be able to decode them properly (to a same standard). But that doesn't influence 2.0 PCM stereo in any way. Simply Corean's made bad S/PDIF IC implementation and it's now funding the market.