A large part of my decision to buy a BluRay player was the fact that only the Panasonic BD10 offered 7.1 analog outputs (my NAD T773 is not equipped with HDMI). During this search, more than one person said "HDMI is the future - buy a new A/V Receiver". Technically speaking I thought there is no difference between which component actually does the D/A conversion, but then I saw this editorial comment in the latest version of The Perfect Vision (downloadable here (http://www.avguide.com/the-perfect-vision-download/?EMC-Issue76email&attr=bluraytitle) )

We’ve compared the sound of digital audio signals presented via HDMI inter-
faces vs. coaxial or optical digital audio interfaces, and in every case achieved superior-sounding results via coax or optical inputs. What accounts for these sonic differences? We aren’t entirely sure, but at least one performance-minded
manufacturer has told TPV that HDMI audio outputs tend to exhibit high levels of jitter, which ostensibly hamper sound quality.

Anyone with a HDMI equipped receiver ever test the analog/coax/optical inputs of a DVD/HD-DVD/BluRay player? Thoughts would be appreciated.

I have both the PS3 and the Toshiba A2. Neither player has analog outs but they do have HDMI and optical. My experience has been that the HDMI sounds better than the optical. This is likely due to the fact that you can pass PCM over HDMI. I think that the multi channel audio in the PCM feed gives better clarity and definition to the soundtrack, it also gives better imaging.

Well, that throws off their entire marketing premise. Thanks for the info!

I have both the PS3 and the Toshiba A2. Neither player has analog outs but they do have HDMI and optical. My experience has been that the HDMI sounds better than the optical. This is likely due to the fact that you can pass PCM over HDMI. I think that the multi channel audio in the PCM feed gives better clarity and definition to the soundtrack, it also gives better imaging.
TPV gave no test data to back up their claim, but I'm amazed this is the first I've heard about it since they wrote the statement as if it was something they mention in every review. I'd love to hear from other people with HDMI who have compared it to Coax & optical.

We’ve compared the sound of digital audio signals presented via HDMI inter-
faces vs. coaxial or optical digital audio interfaces, and in every case achieved superior-sounding results via coax or optical inputs.

Mmm! I love the smell of snake oil in the morning!

Here's a block diagram of a typical receiver:

HDMI decoding
> DSP processing > D/A converters
Coax decoding


You will note that the digital input format doesn't matter -- unless there is something wrong in the clocking of either signal, at which point you'll actually receive the wrong data, and should re-design your inputs. (You'd also see pixel noise on your TV if your HDMI had too much jitter)

Here's a block diagram of a typical receiver:

HDMI decoding
> DSP processing > D/A converters
Coax decoding


You will note that the digital input format doesn't matter -- unless there is something wrong in the clocking of either signal, at which point you'll actually receive the wrong data, and should re-design your inputs. (You'd also see pixel noise on your TV if your HDMI had too much jitter)
So if TPV is to be believed, they have tested HDMI vs Coax/Optical with components whose HDMI implementation does not meet the HDMI spec?

If this is the case - why don't they EVER say this in their reviews, instead of always giving a nice write up? I'm so sick of A/V magazines.

Hang on, hang on. I've never been fully convinced by the "HDMI=jitter" stuff, but the last two posts misunderstand what the (alleged) problem is.

No-one is suggesting that the HDMI receiver is failing to declock the raw HDMI signal stream correctly, and that there's enough video clock jitter to corrupt the data.

But the problem with HDMI... Ahem. One of HDMI's many problems is that the audio is sent between video data. It's sent in little bursts in the borders of the video signal, at video clock rate (25+MHz). So the audio data isn't coming in as a nice, regular stream, it's being delivered in irregular meaty chunks.

The receiver has to be able to retime the samples in those chunks, and make sure they all join back together nicely - each chunk has to start playing at the right time and end at the right time to segue into the next chunk. If, as alleged, HDMI suffers from "jitter", that's where it's going to arise.

This is all somewhat different to properly designed audio connections like i.Link, where the clocking is locked between transmitter and receiver, and based firmly around a nice continuous audio stream. For HDMI, audio is very much an afterthought.

Now, I have no real knowledge about how bad HDMI implementations are in this respect, but from the spec it's clear that it would be very easy to design an implementation that does a bad job. Whereas with i.Link, it would be hard to do a bad job.

The receiver has to be able to retime the samples in those chunks, and make sure they all join back together nicely - each chunk has to start playing at the right time and end at the right time to segue into the next chunk. If, as alleged, HDMI suffers from "jitter", that's where it's going to arise.
Thanks for the post. If you are correct, the problem is not with the manufactures but with HDMI in general - calling into question the very concept of HDMI until they produce a better spec. Sounds like we'll have to wait for HDMI 1.4.

One of HDMI's many problems is that the audio is sent between video data. It's sent in little bursts in the borders of the video signal, at video clock rate (25+MHz). So the audio data isn't coming in as a nice, regular stream, it's being delivered in irregular meaty chunks.

Here's a piece of news for you: The data from your CD discs also arrives in little bursts of 98 frames (each frame is 6 stereo 16 bit samples), from each sector on the CD disc. The CD player then has to string these back together as a byte stream. Given that CD players have done this since 1982, it seems unlikely that a manufacturer in 2007 would get this part wrong.

Having written device drivers for professional audio hardware, I can tell you that failure to re-assemble these blocks by even a single sample is also quite noticeable as "sparkles" in the sound. Any good set of speakers will let you hear that; it's not unlike the glitches you'd get from dust on a vinyl record.

Assuming the HDMI input is clocked and decoded correctly, then the digital data can very easily be double-buffered for output, and that output mechanism is no different depending on the source (coax, hdmi, or even the samples from an analog input). Anything wrong in that stage would be quite audible, and relegate the device to zero-stars review status (assuming the reviewers knew what they were doing, of course).

In my 2nd system I tried both optical out and HDMI out from my HD PVR. Sounds the same to me. The practical benefit of HDMI is you have one cable for both audio and video.