[Bob Sorel]

Ok, I have an Anthem AVM-40 ordered from my local dealer, but while I am waiting for it to get here, I have been trying to get HDMI working at least partially. Right now I am trying to get 2 devices working - Sony Playstation 3 and Toshiba HD-XA2 players. Since I don't have a HDMI pre/pro yet, I am feeding the 2 HDMI sources into my Crystalio 2 video processor (via HDMI, of course) and then I taking the digital coaxial audio output of the C2 and feeding it to my current pre/pro, the Lexicon MC-12. This is working, but only partially. I am not expecting multichannel PCM to get passed, nor am I expecting Dolby TrueHD or DTS Master HD to go through, but I did think that I should be able to get standard DD 5.1 and DTS to pass.


What happens with the PS3 is that I am getting what sounds like left and right main channels of a 5.1 mix passing through properly, but I am not getting and center, surrounds, or subwoofer information. On the HD-XA2 I am getting what sounds like a stereo downmix which the Lex sees as a stereo signal and is correctly applying Logic 7 post processing. The sound from the PS3 has no dialog (which would normally be in the center), and is not even usable as listenable mix, while the 2 channel mixdown from the XA2 is at least usable, though nowhere near as good as a proper 5.1 channel discrete mix.


I have posted in the PMS (Crystalio 2) forum, but there is very little traffic there, so I figured I would try here. Since most people will not know the specifics of the C2, I though I would ask more generic questions.


1. Are there any HDMI "rules" which are stopping the full 5.1 encoded bitstream from passing into the C2 via HDMI and out of the same processor via coaxial digital or is this an issue specific to the C2?


2. Once I have the Anthem here, which will input HDMI audio and video, should I then be able to pass DD 5.1, DD +, and DTS digital bitstreams, as well as multi channel decoded PCM on to the AVM-40?


3. A little OT, but...Can DD+ be transmitted over standard digital connects and decoded by the Lex? Or does the pre/pro have to specifically support DD+ (as opposed to standard DD)?


4. Any other HDMI "quirks" and "rules" I should be aware of?


My eventual plan is to feed the C2's 4 HDMI inputs with 4 HDMI sources and send both video and audio over those connects. Since the C2 has 2 HDMI outputs that can be configured independently, yet run simultaneously, I would like to send HDMI output #1 from the C2 to my display device. Then I would like to send HDMI output #2 from the C2 to the AVM-40 for audio processing. Since there will be no display device connected at the end of the AVM-40 chain, if I understood correctly there is a way to configure the AVM-40 to "fool" HDMI into believing that there is a display connected anyway.


Will this work? And if so, what and where is that setting in the AVM-40?


I sure hope all of this works...

 

[Jack_T]

I'm interested in seeing feeback on your question, because your setup seems a pretty obvious configuration for someone with an external video processor. (I am planning the same type of setup myself....)

 

[JOHNnDENVER]

generally a coax digital audio connection cannot carry the new high rez audio content. You have to use either multi-channel analog, or HDMI to an AVR to recieve this pre-decoded. So I don't think this can be done to hear the new audio codecs. it probably passed DD, and DTS just fine though. Maybe even 2 channel PCM?

 

[Bob Sorel]



Quote:
generally a coax digital audio connection cannot carry the new high rez audio content.

Yup, I understand that...

Quote:
it probably passed DD, and DTS just fine though.

Nope, no go (please read my original post).

Quote:
Maybe even 2 channel PCM?

That's very possibly what I am hearing from the XA2, but it DEFINITELY is not what is coming through the PS3 (again, please read my original post for more information).

 

[Bob Pariseau]

What video output resolutions are you using?


Multi-channel PCM will be limited to 2 channel at HDMI 480i or HDMI 480p due to the way audio is multiplexed. This is a limitation of the HDMI spec.


I don't know the CyrstalIO, but it may limit you to only 2 channel PCM input as regards what it will then OUTPUT over the SPDIF connecction.


Your symptoms seem consistent with failing to tell the source devices that you want core bitstreams to be passed over the HDMI connection instead of PCM, or the audio processing being limited to 2 channels.

--Bob

 

[Bob Sorel]



Quote:
What video output resolutions are you using?

Everything in and out is at 1080p.

Quote:
I don't know the CyrstalIO, but it may limit you to only 2 channel PCM input as regards what it will then OUTPUT over the SPDIF connecction.

No...Other owners claim to have 5.1 digital working.

Quote:
Your symptoms seem consistent with failing to tell the source devices that you want core bitstreams to be passed over the HDMI connection instead of PCM, or the audio processing being limited to 2 channels.

It is very possible that this is the cause of the problem. I will look for settings in both the source devices and the C2 that might affect what kind of information is being communicated over HDMI.


Thanks, Bob! This gives me a good place to start looking!

 

[KMO]



Quote:
Originally Posted by Bob Pariseau /forum/post/0


What video output resolutions are you using?


Multi-channel PCM will be limited to 2 channel at HDMI 480i or HDMI 480p due to the way audio is multiplexed. This is a limitation of the HDMI spec.

Incorrect. There's no inherent limitation to 2-channel at 480i/p. The only reference to 2-channel is that it's the only thing a device is required to support (as is 480, if I recall correctly).


There is a limitation on many devices that don't allow multichannel hi-resolution audio (96/24) with SD pictures, which is partly HDMI's fault, but a properly designed device would get this right. It's just that it takes a little more effort (you have to be able to increase pixel repetition to fit the audio data in).


I'm still trying to find out which devices don't have this restriction, as it's rather critical to DVD-Audio...

 

[M Code]



Quote:
Originally Posted by KMO /forum/post/0


Incorrect. There's no inherent limitation to 2-channel at 480i/p. The only reference to 2-channel is that it's the only thing a device is required to support (as is 480, if I recall correctly).


There is a limitation on many devices that don't allow multichannel hi-resolution audio (96/24) with SD pictures, which is partly HDMI's fault, but a properly designed device would get this right. It's just that it takes a little more effort (you have to be able to increase pixel repetition to fit the audio data in).


I'm still trying to find out which devices don't have this restriction, as it's rather critical to DVD-Audio...

A couple of clarifications..

2-channel PCM was part of the 1.0 spec..

As 2.5 years back...

The 1.0 spec was mainly for the video display sync devices rather than audio as S/PDIF was adequate for 5.1 DD and DTS..


Going forward...

For HDMI 1.1 & 1.2a compliance, the # of channels increased as available bandwidth increased..


However even under HDMI 1.1 and higher burden, multi-channel PCM is possible within the available bandwidth.. The biggest challenge is for HDMI 1.3 and here the bandwidth is 3x that of 1.1..

Up to 400MHz..

But most of the additional bandwidth is capacity for deep colors, as well as the added burden of handling the native HD audio streams of DD Plus, DTS HD for the lossy plus the even greater lossless HD audio streams..


Just my $.02 worth..

 

[Bob Pariseau]



Quote:
Originally Posted by KMO /forum/post/0


Incorrect. There's no inherent limitation to 2-channel at 480i/p. The only reference to 2-channel is that it's the only thing a device is required to support (as is 480, if I recall correctly).


There is a limitation on many devices that don't allow multichannel hi-resolution audio (96/24) with SD pictures, which is partly HDMI's fault, but a properly designed device would get this right. It's just that it takes a little more effort (you have to be able to increase pixel repetition to fit the audio data in).


I'm still trying to find out which devices don't have this restriction, as it's rather critical to DVD-Audio...

Have you found ANY source devices yet that don't have this problem? It seems simpler, for now, just to warn people off HDMI 480i and 480p for multi-channel PCM.


Actually what I think is a more likely fix to expect would be for devices like the Oppo SD players to simply do an automatic switch to 1080i when asked to play DVD-Audio (or SACD via PCM) over HDMI. It is unlikely they will be hooked up to something that can't accept HDMI 1080i. Then go back to 480i for normal DVD playback.


By the way, if the source does this pixel rep stuff, does that mean the receiving end also has to do something unusual? Is it known whether any receivers or TVs will actually accept an HDMI 480i or 480p signal with pixel repeats? I haven't seen this discussed anywhere.

--Bob

 

[JOHNnDENVER]

I was only addressing question number three in my post. I am surprised it does not pass DD and/or DTS. I just got over an odd thing where my HD dish net box wouldn't send DD over HDMI, but last night it was doing it out of the blue. I think for somebody to really have input for you, they would have to own that scaler as well. Be patient, I am sure quite a few on here use that scaler. But you would probably have to post in that forum.

 

[KMO]



Quote:
Originally Posted by Bob Pariseau /forum/post/0


Have you found ANY source devices yet that don't have this problem? It seems simpler, for now, just to warn people off HDMI 480i and 480p for multi-channel PCM.

Woah. Hang on there. I'm not aware of any current device that can't do multichannel audio with 480i/p. The restriction is purely on high-res 96/24. Or is HDMI support worse than I thought?

Quote:
Actually what I think is a more likely fix to expect would be for devices like the Oppo SD players to simply do an automatic switch to 1080i when asked to play DVD-Audio (or SACD via PCM) over HDMI. It is unlikely they will be hooked up to something that can't accept HDMI 1080i. Then go back to 480i for normal DVD playback.

On the devices I've seen that have the restriction, they actually drop down to 48kHz multichannel or 96kHz stereo when outputting SD video. I've not seen anything that changes to a higher resolution automatically.


Personally, I don't have an HD display, and don't think I'm likely to for a while, so this issue of no HD audio without HD video is pretty important for me.

Quote:
By the way, if the source does this pixel rep stuff, does that mean the receiving end also has to do something unusual? Is it known whether any receivers or TVs will actually accept an HDMI 480i or 480p signal with pixel repeats? I haven't seen this discussed anywhere.

It means the receiving device has to understand the concept of pixel repetition, and report that they do. I imagine some don't. 480i inherently requires pixel repetition (everything has to be sent twice to meet the minimum 25MHz clock rate requirement), so any receiving device that supports 480i is likely to get it right, I would hope. High-res multichannel requires 480i to be pixel-quadrupled, and 480p to be pixel-doubled.

 

[KMO]



Quote:
Originally Posted by M Code /forum/post/0


However even under HDMI 1.1 and higher burden, multi-channel PCM is possible within the available bandwidth.. The biggest challenge is for HDMI 1.3 and here the bandwidth is 3x that of 1.1..

The interesting thing about HDMI is that the higher the video resolution/bit-depth, the more space there is for audio. Audio fits in the borders of the video, and if the video's higher resolution/depth, so are the borders, so there's more room.

 

[Bob Pariseau]

KMO,

Yes the restriction is for high bandwidth, multi-channel PCM.


But what seems to be happening more often with these devices is that they cut you down to 2 channels of full bandwidth PCM instead of lowering the bandwidth to maintain multiple channels.


Since, as far as I know, there are no source devices that increase the pixel repeats enough to make this work, I suspect that receivers and TVs that accept either HDMI 480i or 480p will likely not have been tested for that.

--Bob

 

[M Code]



Quote:
Originally Posted by KMO /forum/post/0


The interesting thing about HDMI is that the higher the video resolution/bit-depth, the more space there is for audio. Audio fits in the borders of the video, and if the video's higher resolution/depth, so are the borders, so there's more room.

Yes..

But HD video is driving the market and HDMI not HD audio..

Thats one of the reasons even the HD standard of Dolby Digital + is limited to 48KHz..

Additionally as the video processors move up to 12 bit processing they can handle even higher resolution..


Great HD audio is a strong supporting asset from an HD source but clearly HD video is the driving force behind the market..


This was very evident @ the ICES just held in Las Vegas..

 

[Bob Sorel]

Thanks for the replies, guys!

Quote:
I think for somebody to really have input for you, they would have to own that scaler as well. Be patient, I am sure quite a few on here use that scaler. But you would probably have to post in that forum.

Thanks, John. Here is the one and only reply I have gotten from the PMS board so far:

Quote:
I have essentially the same setup, and I'm getting DD5.1 through just fine. My PS3 audio settings are just DD5.1, DTS5.1, and LPCM 2ch 44.1-192, and the BD/DVD Audio Output Format for HDMI is set to Bitstream.

This reply seems to agree with Bob P.'s suggestion almost exactly. It at least lets me know that using the C2 and the same (or similar) source devices, that at least one person has gotten multichannel digital bitstreams passed through the same path as I am attempting, so I must have missed a setting somewhere that tells the source devices to pass the digital bitstream through HDMI. I will be checking that out carefully tonight.


Here is another more generic question that I really should know, but don't:


Is PCM or LPCM a digital or analog signal? I know that PCM stands for Pulse Code Modulation, but other than that I really don't know much about the nature of the signal. For some reason I always think of PCM as being an analog signal, but I really don't know how it is treated by source devices.

 

[Bob Pariseau]



Quote:
Originally Posted by Bob Sorel /forum/post/0


Is PCM or LPCM a digital or analog signal? I know that PCM stands for Pulse Code Modulation, but other than that I really don't know much about the nature of the signal. For some reason I always think of PCM as being an analog signal, but I really don't know how it is treated by source devices.

Both are digital. It is the digital representation of one channel of audio. For a 5.1 audio setup you would need 6 PCM streams coming over the link.


People have started to use Lossless PCM, or LPCM, to distinguish from compressed forms of PCM which could be lossy -- i.e., you can't uncompress them and get back the original signal with complete fidelity.


PCM represents the digitization of audio, which is inherently analog of course. As with all digitization, the sample rate, and the size of each sample are fundamental determinants of how well the digital stream represents the real world, analog audio. And PCM comes in different bandwidths -- i.e., different sample rates.


But there are practical limits. The higher the sample rate and sample size the more space it takes to store the PCM and the faster transmission media you need to move it around. And a gazillion samples per second, of a bezillion bits each -- even if you could create it -- would be indistinguishable, when played, from much more modest bandwidth PCM. I.e., at some point, going to higher sample rates and sizes doesn't ADD anything to what you can actually HEAR.


The "new audio" formats for HD-DVD and Blue Ray are simply different ways to "zip up" a collection of PCM streams into a more compact package. This allows you to put higher bandwidth PCM into the same amount of space on the disc. And the presumption is that the higher bandwidth PCM will sound better when unzipped into its original form and played back. And maybe it will.


The new formats are also lossless in the sense that the PCM that comes out when they are unzipped is identical to that which went in.


This makes it tough for them to add any QUALITY difference to the audio. All they can really compete on is how tiny they can make the zipped up form (and still be lossless when it is unzipped), and how much it costs the studios to do the encoding and the player manufacturers to add the support in players to do the decoding.

--Bob

 

[Jack_T]

Okay, since we are answering basic questions, here's some:


Typically you will have two different kinds of audio signal passing through your SPDIF input: 2 channel PCM from a CD or 5.1 DD. How does the receiver know what kind of signal it's getting?


Also, is two channels the limit for PCM signals over the digital audio connection (SPDIF)?


DD is a compressed format, right? How is the fidelity compared to, say, mp3?

 

[Bob Sorel]

Thanks again for that excellent explanation, Bob!


I made a LOT of progress tonight and I have things working about as well as possible until the AVM-40 arrives.


The PS3 had a setting in the system menu that allowed me to change the HDMI audio from PCM to bitstream, and once I did the audio behaved as expected. That is, standard DD 5.1 and DTS streams were passed on to the Lex properly. If I chose "uncompressed 5.1 PCM" from the Blu-ray disc menu, I got some form of 2 channel signal. I assume that this is because the Crystalio 2 is connected to the Lex via coaxial digital, not HDMI, and is thus bandwidth limited. Are my assumptions correct?


But strange behavior came from the HD-XA2 HD-DVD player. The HDMI audio output had 3 possible settings:


auto

PCM

downmixed PCM


The default was "auto" and that came out as 2 channel downmixed (as I explained in my original post). I switched to "PCM" and then I got much better results - standard DD 5.1 and DTS were passed to my Lex as bitstreams, not PCM as I would have expected. Then I tried choosing "Dolby TrueHD" from the HD-DVD disc menu and the Lex reported that it was getting a 5.1 analog signal (which is what it reports whenever it gets PCM), so I really don't know what it was getting in reality. Does this behavior sound right to you, or might this be some sort of bug or mislabeling problem in the XA2's menu? By choosing "PCM" in the XA2's menu I would have assumed that any and all signals being transmitted over HDMI would be decoded and sent as separate PCM channels and that the Lex would see ALL of those signals as analog.


At any rate I am now passing full encoded DD 5.1 and DTS over HDMI from the 2 source devices to the C2, and the C2 is successfully sending the same signals back out via SP/DIF to the Lex - so far so good...

The next phase will have to wait for the AVM-40 to arrive so that I can attempt to pass the signal from the C2 to the AVM-40 over HDMI without the need for connecting a display to the output of the AVM-40. This should be interesting...

Quote:
Also, is two channels the limit for PCM signals over the digital audio connection (SPDIF)?

That's what I thought, but if you can believe the display on my Lex, it was getting 5.1 channels of PCM over SP/DIF from the C2...But the question is 5.1 channels of what?

 

[Bob Pariseau]



Quote:
Originally Posted by Bob Sorel /forum/post/0


The PS3 had a setting in the system menu that allowed me to change the HDMI audio from PCM to bitstream, and once I did the audio behaved as expected. That is, standard DD 5.1 and DTS streams were passed on to the Lex properly. If I chose "uncompressed 5.1 PCM" from the Blu-ray disc menu, I got some form of 2 channel signal. I assume that this is because the Crystalio 2 is connected to the Lex via coaxial digital, not HDMI, and is thus bandwidth limited. Are my assumptions correct?

Yes. A traditional optical or coax digital audio connection (an SP/DIF connection) is limited to no more than 2 channels of high bandwidth PCM.

--Bob

 

[Bob Sorel]

Thanks, Bob...That means that the PS3 is behaving exactly as expected.


It is the XA2 that is confusing the heck out of me...