[bug] HDMI 2.1 chip (Panasonic Solutions)

[Vasr]

I have no idea who GROBI.TV is, but apparently they got it to work and think the issue can be corrected via firmware:

Firmware of the graphics card? Not a practical solution to this problem even if this were legit. All graphics cards and consoles are not going to get a firmware update to be able to play nicely with these AVRs. Since they all work just fine going directly to TV, the problem is unlikely to be with the graphics cards or any HDMI source that can output 8k@120 HDR.

Even if this can be achieved via an inverse firmware fix at the HDMI receiver in the AVR, not all the firmware on a multi-component device like an AVR is necessarily field upgradable for that particular faulty firmware depending on where it is embedded. May require a HDMI transceiver card swap with new firmware or with a fixed chip.

Don't know if Anthem tested this for their decision to hold back on the 8k for their new line up or they just got lucky.

 

[Nathan Raymond]

Clearly, Denon and all the other manufacturers affected did not test this use case at all which is a strike on them.

There wasn't much for them to test with I think. I bought my Denon AVR-X3700H before there were any HDMI 2.1 source devices on the market, which means Denon R&D and the bulk of their testing had been completed months before that. The NVIDIA RTX 3000 series (which supports HDMI 2.1) didn't come out until late September of this year. As of right now I think it is still not possible to buy a certified Ultra High Speed HDMI Cable - you can only buy non-certified cables that claim to be HDMI 2.1 compliant (but may not be). There's also the question of 40Gbps vs 48Gbps HDMI 2.1 (the new receivers and some TVs only support 40Gbps). I'm not sure who to blame since I haven't read any whitepapers/detailed engineering specs and lack any industry insider info.

 

[Vasr]

There wasn't much for them to test with I think. I bought my Denon AVR-X3700H before there were any HDMI 2.1 source devices on the market, which means Denon R&D and the bulk of their testing had been completed months before that.

You realize that Denons and other manufacturers don't need to wait for consumer equipment to come to market to test their HDMI 2.1, right?

This s not a compatibility issue with source devices. The 8k@120 HDR mode from any source doesn't work (for whatever reason). They will/should have their own test bench to test these different use cases before bringing their devices to market. The rush to be first with 8k will cost them.

 

[Nathan Raymond]

You realize that Denons and other manufacturers don't need to wait for consumer equipment to come to market to test their HDMI 2.1, right?

This s not a compatibility issue with source devices. The 8k@120 HDR mode from any source doesn't work (for whatever reason). They will/should have their own test bench to test these different use cases before bringing their devices to market. The rush to be first with 8k will cost them.

I've read claims that it is a failure to pass through uncompressed 4k/120Hz HDR and 8k/60 HDR, and that the bug doesn't affect those resolutions/rates when passing through compressed. HDMI 2.1 introduces VESA Display Stream Compression (DSC) 1.2a, and DSC makes resolutions of over 8K with HDR 10-bit color possible, at refresh rates of both 60Hz and 120Hz. I agree that the AVR makers should have tested all possibilities, and I'm not trying to say they're not responsible - I'm just trying to point out that their test matrix for HDMI 2.1 is by far the largest video test matrix in history, so it's not entirely surprising that something was overlooked:

 

[circumstances]

I've read claims that it is a failure to pass through uncompressed 4k/120Hz HDR and 8k/60 HDR, and that the bug doesn't affect those resolutions/rates when passing through compressed. HDMI 2.1 introduces VESA Display Stream Compression (DSC) 1.2a, and DSC makes resolutions of over 8K with HDR 10-bit color possible, at refresh rates of both 60Hz and 120Hz. I agree that the AVR makers should have tested all possibilities, and I'm not trying to say they're not responsible - I'm just trying to point out that their test matrix for HDMI 2.1 is by far the largest video test matrix in history, so it's not entirely surprising that something was overlooked:

I don't want to have to send my unit back!

Please let it get fixed via firmware of some sort! lol

I'm glad the GROBI guys got it to work at all.

 

[Logan Nolag]

Depends on who developed the chip, who signed the contract with the manufacturers, what the terms of the sale contract with the manufacturer was, the root cause of the problem and what liabilities were passed down as part of the sale to Nuvoton if Nuvoton had nothing to do with the root cause of the problem. If Nuvoton had competent lawyers and they had some clout in the negotiations of the sale, they will not have indemnified Panasonic for any liabilities for any assets or IP purchased.

If it cannot be fixed by firmware, it is going to be a very expensive recall.

Clearly, Denon and all the other manufacturers affected did not test this use case at all which is a strike on them.

They couldn't test it since up until the RTX 30 series cards there were no HDMI 2.1 sources.

 

[Vasr]

I'm just trying to point out that their test matrix for HDMI 2.1 is by far the largest video test matrix in history, so it's not entirely surprising that something was overlooked:

It is sloppy QC/Product management but may be par for the course in these primarily hardware shops. I will give you that.

Interesting that according to your information, the newly introduced DSC feature works but not the default uncompressed feature common to all formats prior to that. That is odd.

I don't know the technology to see if DSC just compresses the content of video frames inside or it compresses the entire HDMI stream.

I wonder if anyone has tried the AVR stand-by pass through of HDMI or to Zone 2 without audio extraction for these problematic modes to see if it works. If it does, that would eliminate a few things.

 

[Vasr]

They couldn't test it since up until the RTX 30 series cards there were no HDMI 2.1 sources.

They couldn't test compatibility with commercial devices but that is different from testing their implementation with test equipment wihich would have exposed this bug. Both the chip maker (if the problem is within the chip) and transceiver card manufacturer should have done this before certifying it for 8k.

 

[Chromatischism]

 

[Vasr]

Nice informative video. Looks like the problem is with transceiver chips that need to receive, process and forward rather than just receiver chips like in the TV that work fine. Odd that they would have problems with uncompressed rather than compressed. I would think adding overlays from the AVR would require them to uncompress the stream, add the overlay and then compress again or would they be sending uncompressed out? So if they already have the ability to uncompress and process, not sure why it is such a huge problem to handle uncompressed sources. That needs a deeper insight into this issue than what is on the video.

So, if the solution/patch were to be at the graphics card or console side, they would need to compress the signal before sending and everything would work.

BUT, how much computing power is needed to compress it? This would eat into their performance even if it can be done via firmware. And they have no obligation to do so.

If the compression wasn't as complex as a typical video codec encoding (I am guessing it isn't) and could be done in real-time, then perhaps the interim solution would be a HDMI dongle that takes an uncompressed signal and outputs compressed signal.

I am sure the HDFury folks will be happy to supply one to the AVR manufacturers so they could make it available free to any of their customers that requested it.

It is probably not that simple.

Are the AVR manufacturers still selling this as a 8K capable AVR? With an asterisk mark? With specs quitely changed to indicate no 120hz HDR? Interesting ethical dilemma.

Might explain why Anthem rushed their product announcement prematurely, hoping to cash in during this PR issue and confusion surrounding the competition.

 

[circumstances]

I can't see me having 4K 120 or 8K 60 sources any time in the near future (or distant future), but I'm looking forward to getting this resolved regardless.

 

[mcdull]

Nice informative video. Looks like the problem is with transceiver chips that need to receive, process and forward rather than just receiver chips like in the TV that work fine. Odd that they would have problems with uncompressed rather than compressed. I would think adding overlays from the AVR would require them to uncompress the stream, add the overlay and then compress again or would they be sending uncompressed out? So if they already have the ability to uncompress and process, not sure why it is such a huge problem to handle uncompressed sources. That needs a deeper insight into this issue than what is on the video.

I'm wondering if they can just supply a free transceiver box that can compress the uncompress data and feed that to the AVR, seems a much cheaper fix than fixing the unit.

 

[Nathan Raymond]

I'm wondering if they can just supply a free transceiver box that can compress the uncompress data and feed that to the AVR, seems a much cheaper fix than fixing the unit.

I do not want any sort of additional hardware. There is a Denon dealer network (I bought mine from a nearby dealer). If it needs a hardware fix, I would want the dealer to replace the board with the flawed chip on it with a revised one, which I'm sure is something Denon can handle. (My Suzuki SX4 Crossover had a faulty airbag seat sensor on the passenger side, and even though Suzuki exited the American car market, there was a recall and I got the seat replaced at a former dealer at no charge, which is how these type of things should be handled).

 

[Vasr]

I'm wondering if they can just supply a free transceiver box that can compress the uncompress data and feed that to the AVR, seems a much cheaper fix than fixing the unit.

In theory yes.

That is the dongle I was referring to in my post. It is only feasible if it does not require much processing power and can be done on a live stream in real time. So, it depends on on what kind of compressing algorithm is in use. The amount of data needed to be handled in UHD video streams at high frame rates is enormous.

I do not want any sort of additional hardware. There is a Denon dealer network (I bought mine from a nearby dealer). If it needs a hardware fix, I would want the dealer to replace the board with the flawed chip on it with a revised one, which I'm sure is something Denon can handle.

According to the video linked above such a chip may not be available until late 2021.

 

[Logan Nolag]

I do not want any sort of additional hardware. There is a Denon dealer network (I bought mine from a nearby dealer). If it needs a hardware fix, I would want the dealer to replace the board with the flawed chip on it with a revised one, which I'm sure is something Denon can handle. (My Suzuki SX4 Crossover had a faulty airbag seat sensor on the passenger side, and even though Suzuki exited the American car market, there was a recall and I got the seat replaced at a former dealer at no charge, which is how these type of things should be handled).

Seriously part of the reason for getting a receiver is to simplify my setup not complicate it with dongles or other jerry rigged solutions. The only acceptable fix is to replace the defective board/chip or the entire receiver anything less would be really lame.

 

[brettjv]

Do we need to register our products somehow to make sure we get taken care of if there's a hardware repair involved? I couldn't find any way to do so on the Denon site (and as aside, man their site is a JOKE for such a major company). Has anyone else done so?

 

[Logan Nolag]

Do we need to register our products somehow to make sure we get taken care of if there's a hardware repair involved? I couldn't find any way to do so on the Denon site (and as aside, man their site is a JOKE for such a major company). Has anyone else done so?

I tried to register my NR1711 but only the Gold/Silver version was listed and I have a black one so I wasn't able to register mine either.

 

[circumstances]

I tried to register my NR1711 but only the Gold/Silver version was listed and I have a black one so I wasn't able to register mine either.

i have my unit connected to the internet (and have done some firmware upgrades that way since I've had it).

would that have required registering my unit, or is there some further step i should be taking?

i believe i needed the serial number of the unit to connect to the internet.

 

[Urgo]

Do we need to register our products somehow to make sure we get taken care of if there's a hardware repair involved? I couldn't find any way to do so on the Denon site (and as aside, man their site is a JOKE for such a major company). Has anyone else done so?

I had a hard time finding how to register the serial number.

I did the registration in my case on the Marantz Europe pages, but it should take you to the same final address for Denon and Marantz, because when you enter they ask you if it is a Denon or Marantz product.
What you have to do, if you do not already have an account, is to open a user account with your email address and a password, and once you pass this procedure you can register the product with the serial number.
Since I did it 10 days ago, I have not received any notification.

This is where I made it, you can locate its equivalent according to your Technical Service (USA, Europe, ...), or if it is a Denon or a Marantz, searching for "support / customer support":

https://marantz-uk.custhelp.com/app/utils/login_form/redirect/account%2Fpurchprod%2Falist


Edit:
I've searched for it, but haven't been able to find the equivalent Denon page.
But in that direction it is possible to register both Marantz and Denon products, of course it is the Technical Service of Europe.

 

[Urgo]

This is the only thing I have found for Denon.

You can try here:

https://denon.custhelp.com/app/utils/login_form/redirect/ask

At least it looks like you can ask