|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
216.178.164.160
Is the HDMI interface / standard asynchronous, or, does it operate more akin to standard Toslink or Spdif protocols?
My reason for asking is to better understand to what degree outbound jitter matters in comparison to "standard" Toslink or Spdif legacy digital outputs, and correspondingly, to understand to what degree higher precision clocks impact HDMI sonics.
Please, no diversions around "legacy format jitter doesn't matter because receiver chips overcome that."
Follow Ups:
HDMI is an abortion of an interface that was crammed down our throats by Sony and Hollywood. Silicon Image was the party that made it all possible.
The idea by Sony was to have the audio and the video both on the same cable, to avoid confusing the schmucks who buy their Sony TV sets at Best Buy and can't figure out how to connect it. Hollywood demanded "content protection", and it was decided that HDCP as developed by Intel would suffice. Silicon Image was determined to develop the silicon chips so that they could cash in on the cash cow.
Of the many problems associated with HDMI, the audio quality is totally handicapped for lack of -- a pin! They designed the connector before they finished designing the system. They didn't have enough pins to also have a master audio clock.
So with HDMI, the audio clock is derived from the video clock. For high-def TV, the video clock runs at either 74.25 MHz or 74.25 * (1000/1001) [thank you NTSC!!!]. The audio clock runs at multiples of 48 kHz. Of course, these are not related. So the receiver has a PLL to regenerate an audio clock based on instructions from the transmitter (source) telling it what to do.
The result is the worst jitter of any system yet invented. It truly sucks.
Much later, they added a thing called Audio Rate Control in HDMI 1.3a. This puts a buffer and the master audio clock in the receiver. Then commands are sent upstream on the CEC line telling the player to speed up and slow down as necessary to keep the buffer full.
The only people to use this are Sony (HATS) and Pioneer (PQLS), but both use proprietary implementations that prevent use with other equipment.
And the fee for using this pile of steaming dog dung? $30,000 per year in licensing fees. It's a beautiful world, no?
I confess to being very late to the thread, but what are the prospects now of an open standard?
If not, would Sony or Pioneer still want $30 k pa to access their standard?
Seems a good investment for an audiophile player company. I'd pay more for an Oppo with Audio Rate Control
... and maybe some other more recent formats, are "packet-based" and therefore jitter is irrelevant.
Charles, is this different than what you're talking about? Or is it just wrong?
___
Fearnor's Stereo Configuration
There is a link in the article about the HDMI jitter problems. Have you seen the link?
"Sony's audio rate control - HATS - which is supposed to cure the jitter problem, is discontinued on all currently available products". Sorry for the people who bought a Sony scd-xa5400es sacd player for mch sacd playback.
a
The manufacturers that use HDMI ports and cables to support the I2S interface is not the same as normal HDMI for Audio/Video HT applications. If you plug 1 into the other, you can damage your equipment.
Charles ...
So, if I want to improve the sonics and HDMI is my source then the key is a stable video clock (video clock upgrades would have value) and the highest quality PLL implementation possible? Or, is it only the PLL implementation?
For two channel Blu-ray, the best solution would be to drop ten grand on a product whose name I cannot mention and use the analog outputs.
For multi-channel Blu-ray, both the sending video clock signal and the receiver PLL circuit will have an impact on jitter. The stupidly expensive player whose name I cannot mention has the lowest jitter video clock of any player I know of. For receivers, the latest rage is asynchronous sample rate converters in a perverse attempt to reduce jitter. What you really want is a good double-PLL. There are probably some SSPs that have better PLLs than others, but I don't know which ones.
And what you *really* want is something that uses ARC. But the only choices so far are proprietary versions in low-cost receivers from Pioneer and Sony (if they haven't already discontinued them). The stupidly expensive player whose name I cannot mention has the generic ARC in it if some SSP manufacturer will ever do their end of the work.
There are at least two good ways to get high quality, 2 channel audio, out of BluRay discs: The Ayre DX-5, or going the computer audio route and ripping the audio from the BR disc. BR disc audio can be ripped to a hard drive as full resolution PCM, but the process seems quite tiresome (I have not bothered myself).
Does the Oppo really output in the clear, non sample rate converted, 24/192 PCM from its SPDIF output from BR disc?
It is not the two channel tracks that I am trying to access. Those I can get up to 24/96 via the spdif outs of the Oppo. Or, via the analog outs of a great player, Brand to remain unnamed :)(Charles)
It is the latest multichannel Dolby etc. "lossless" formats that I can only get via the HDMI outputs that I am inquiring about.
Technically it's junk. Its only reason for existence is evil. Why would anyone want to use it?
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
Edits: 12/09/10
there is now no choice with graphics cards connection to screen. Agree not to use for audio just yet. May be there will be a hdmi 2 with audio clock pins.
They had problems implementing the audio when the system was first designed. I told them to use a separate audio clock line. They said the connector had already been designed!
Then he admitted that there was a spare pin. I told him to use that. He said that they had to keep a spare in case they *really* needed (as if audio didn't matter). The ironic thing is that here it is, nearly ten years later, and even with HDMI version 1.4, that spare pin is still not used for anything.
One problem is that if they used it for an audio clock that it wouldn't be backwards compatible with all of the older stuff.
They have added the Audio Rate Control (ARC) in version 1.3a, which solves the jitter problem better than any dedicated audio clock line ever could. But the only people that have used it so far are:
a) Sony -- as "HATS" -- proprietary, only works with a Sony player and receiver.
b) Pioneer -- as "PQLS" -- proprietary, only works with a Pioneer player and receiver.
c) Ayre -- as generic ARC -- will work with any compliant SSP or receiver, but nobody makes one!
Hey Charles - I'm not questioning your knowledge, but just came across a discrepancy. The HDMI Wiki says "Audio Return Channel" was added in version 1.4. On both the Denon and Pioneer websites they refer to Audio Return Channel being supported on their latest receivers as part of the 1.4a spec. There's nothing about it being proprietary nor mention of "PQLS" on the latest receiver. (I saw it mentioned on an older model).
So I'm just wondering if this Audio Rate Control you refer to is the same as Audio Return Channel and you just mixed up your words? And also, is it in fact a part of version 1.4a and any device (i.e. from Denon or Pioneer) supporting 1.4a automatically provides the Audio Return Channel ? Or is just "supporting" it not enough? IOW, are you saying there's a proprietary implementation that goes beyond just "supporting" 1.4a?
Thanks. Not busting your balls, just trying to clarify this for myself.
Audio Return Channel is pretty useless, especially for anything high end. It's a way to send an audio signal from you cable TV set back to your hi-fi system with a relatively low resolution.
Audio Rate Control is a feedback pipeline on the HDMI cable, implemented on the CEC (Consumer Electronics Control) line. Instead of the player being the master and the audio processor with the DAC chips slaving to the player, the system is turned around. The master clock is in the receiver and signals are sent to the player telling it to speed up or slow down (the rate of the spinning disc essentially) to match the audio clock in the receiver.
ARC was included in the HDMI 1.3a specification. Pioneer and Sony use(d?) proprietary implementations. If you analyze the traffic on the CEC line, it is all prefaced with a code for "PIONEER" or "SONY" so that only their equipment will respond to the commands from the receiver.
And in this case, receiver is the correct word. Pioneer makes no SSPs, only surround receivers. I don't think they would be a good match for our video disc player source....
Ahh. OK, I get it now. The documents I was reading was confusing since Audio Return Channel was also referred to as ARC in some places. Thanks for clarifying that Audio Rate Control is part of the CEC line. This makes sense to me now.
Thanks Charles!
Yes, they took an interface that was fine for its intended purpose -- sending video signals a few feet from the computer to the nearby monitor. They stretched its effective length to the breaking point, and then they piled endless heaps of dung on top of it:
- HDCP "content protection"
- Audio (but with no audio clock!)
- Consumer Electronics Control (a poorly thought out pathetic attempt at linking the control of various components)
- Power (intended only for turning on the memory chip of a downstream receiver that was off)
- Extended Display Identification Data (again, poorly thought out)
And that was just for starters. After that they added:
- Higher bandwidth for video signals
- More color space encoding schemes
- Additional content protection for DVD-Audio content
- Support for DSD audio
- Audio Rate Control
- Support for higher audio data rates, including new lossless formats
- Ethernet (!)
- Audio Return Channel
- Four additional connector types (!)
- Support for even more video bandwidth, including formats that don't exist (4K x 2K)
- Support for 3D video
It is such a huge pile that it is just a big pile of dung.
For video HD-SDI is a much better solution. For audio *anything* is better, even multiple S/PDIF lines. But we're stuck with it. I would hazard a guess that probably 20% of the total cost and development time of a video disc player and/or a surround processor is just devoted to working out all of the problems and incompatibilities of HDMI...
I can't access the latest highest bit rate lossless audio tracks on BluRay disks any other way than via the HDMI link. My Oppo won't send those out via Spdif. (I think, but don't know, because of copy protection schemes.)
You need to buy a blu ray player with analog outs.
In my experience the best way to get uncompromised audio from an Oppo player is to use the analogue outputs. I don’t see any advantage of carrying the sound in digital form to another device for D/A conversion. HDMI and SPDIF introduce limitations and flows. The DACs and analogue circuits of the Oppo players are state of the art. It is really simple to feed the analogue outputs to analogue multichannel amplifier.
It was an exciting experience for me to hear the full potential of my system with the receiver DSP eliminated from the sound processing chain.
I have got better audio quality in stereo (only) from the "equivalent" DVD (to the BD, same title) in many cases, from the Oppo. Using the S/PDIF output to an external DAC. Much more satisfying, whether 96/24 or 192/24. The point is the Oppo will put those out in an unprotected digital format from DVD. I have been very disappointed with ALL stereo audio via HDMI. However, the *surround* audio via HDMI from BD/SACD/DVD-A is often pretty impressive. Nothing else here for me to compare it with though.
Agreed,
It is the latest multichannel, surround, hi rez lossless format audio that is accompanying movie soundtracks, that I am after. HDMI is the only choice for those.
Good reason not to buy stuff that has DRM. It is hostile to equipment. That's what it's designed to be.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: