Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[blgentry]

If you're really picky about digital connections, you won't use HDMI for audio.  It's a very convoluted transport type compared to USB, Toslink, or Coaxial Digital. 

But that's only if you're really picky and care about possible sources of jitter.

Brian.

[JimH]

If you're really picky about digital connections, you won't use HDMI for audio.  It's a very convoluted transport type compared to USB, Toslink, or Coaxial Digital. 

I don't know why you say that.  HDMI is just as much a digital connection as the others.  It's my preferred connection.

[blgentry]

USB, Toslink, Coaxial Digital, and presumably AES all transmit a unified stream of digital audio.  It's just audio and it's sequential and regular.

HDMI, on the other hand, does something completely whacky: Namely, it embeds the digtial audio stream(s) inside the video blanking intervals of the video frames!  It's a very unusual and (in my opinion) convoluted way of transmitting audio.  It means that the HDMI receiver has to disassemble each video frame, grab the relevant audio portion, then reassemble these audio pieces into some sort of coherent stream that can be further processed, decoded, played back, etc. 

This disassembly and reassembly introduces the possibility of digital jitter.  At this point, jitter is a well known, real effect, as opposed to being audiophile crockery. 

Now, do you care about digital jitter?  The answer depends on how serious you are about audio quality, how much of a concentrated listener you are, and how high resolution your audio equipment is.  For casual listening, or even movie, TV, etc, you might not ever hear the consequences of digital jitter.

This is why I said you might care about it if you were "picky".  Others could care less.

Brian.

[JimH]

HDMI, on the other hand, does something completely whacky: Namely, it embeds the digtal audio stream(s) inside the video blanking intervals of the video frames! 

Here is the wikipedia article:
https://en.wikipedia.org/wiki/HDMI

Please find the reference to the use of VBI.

Even if what you say were true, "assembly" and "disassembly" are things computers and CE devices can do in their sleep.  Data is data.  It doesn't matter how data gets to the end point as long as it isn't modified.

Jitter is a favorite audiophile topic, but not normally a problem.

[blgentry]

Here is the wikipedia article:
https://en.wikipedia.org/wiki/HDMI

Please find the reference to the use of VBI.

I'm confused.  VBI seems to be about Closed Captioning.  I'm unfamiliar with it, but that's the only reference to VBI I found on that page.

Even if what you say were true, "assembly" and "disassembly" are things computers and CE devices can do in their sleep.  Data is data.  It doesn't matter how data gets to the end point as long as it isn't modified.

Jitter is a favorite audiophile topic, but not normally a problem.

What I say is true for sure.  But hey, I don't want to fight about it.  The guy in the other thread said "best audiophile quality", so I told him about how crazy the HDMI audio transmission method is.  That's all.

You've made it clear that you don't believe in almost any audiophile type topic and that's cool too.  I'm confident in my statement about the HDMI transmission method and will provide references if you really want.  But I'm guessing you don't actually care because you've heard HDMI sounding "just fine" and probably think I'm making something of nothing. 

I'll consider this topic closed unless someone else wants to discuss it.  I hope that seems fair and reasonable.  I'm still smiling on this side of the internet. 

Brian.

[mbagge]

Having opened that can of worms, I will select one of them, the Coaxial Digital with or without the optical Toslink connector.
Being invented together with the CD-player in the early 1980s, it offers no certainty of the digital audio stream reaching the other end. It is all up to the receiving end to interpret what it can and make up the rest. There is no error correction at all as we are used to expect every where else in the digital domain.
And furthermore it does only support two channels and then it is a problem if you want to transmit multi channel audio.

For me, this type of connection will always be the last resort and only used to connect two devices that for some reason does not have the same ground potential using the Toslink connector. 

Having dragged a "worm" out in the open, you are now allowed to disagree as much as you like. 

[BillT]

AIUI VBI stands for vertical blanking interval (as opposed to horizontal blanking interval). In analogue days all sorts of stuff was put up there, test signals, broadcast control signals, teletext data (as we call it in the UK). Vertical blanking isn't really meaningful in digital systems, as there is no need for blanking in the data stream.

I don't know how data is embedded in the HDMI data stream, but it isn't important. It can be extracted and processed to produce as jitter free stream as you wish. Jitter in the recording/transmission stream is unimportant as long as the signal is good enough for the jitter to be reduced to insignificant levels before the DAC, and I can see no reason to believe that HDMI is uniquely bad. After all, it can transmit 8 audio channels at 192kHz/24 bits and audiophile grade equipment uses HDMi for multi-channel, high sampling rate audio. You can't use SPDIF for those data rates, the most you can do is 96/24 with 4 SPDIF channels

HDMI does have drawbacks, the data handshaking is a nightmare and there are loads of HDMI standards to choose from!

[blgentry]

AIUI VBI stands for vertical blanking interval (as opposed to horizontal blanking interval).

Oh Ok, so Jim was "challenging me" to find the reference to Video Blanking Interval and audio.  Well, here you go:

https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=5&ved=0ahUKEwitrM_qr9rOAhUNxWMKHfXFAAsQFgg-MAQ&url=http%3A%2F%2Fwww.aes-media.org%2Fsections%2Fuk%2FConf2011%2FPresentation_PDFs%2F14%2520-%2520john%2520dawson%2520-%2520Audio%2520Transport%2520over%2520HDMI%2520-%2520AES%25202011.pdf&usg=AFQjCNGoufdoPidk6kuyyGlh8XOiA0KDCQ&sig2=3ZahG8Yi3GlmUBWbZGKcUQ&cad=rja

See pages 12 and 13.

Not that either of you care, but at least my statement about the audio data being mixed in with the video, in the blanking (sync) area is factual.  I still think it's a rather weird way to encapsulate and transmit audio.  Whether or not it leads to jitter would seem to be implementation specific.

Brian.

[rudyrednose]

The name of the game in digital communications is the quality of the recovered clock, directly affecting jitter (which can have audible effects).  I don't care if the actual audio bits are a short burst drowned in the whole signal.

Comm engineers refer to the received signal quality as the "eye diagram".


So, you need a good, stable clock at the source (in our case the computer's circuitry source of the digital stream), a low noise datapath and a very good clock recovery (at the DAC end). 
If you have those, channel being HDMI, TOS, SPDIF or USB is irrelevant because your analog circuitry will have a quality base to build good sound from...

picture source, this article: http://www.edn.com/design/test-and-measurement/4389368/Eye-Diagram-Basics-Reading-and-applying-eye-diagrams

[eddyshere]

So, you need a good, stable clock at the source (in our case the computer's circuitry source of the digital stream), a low noise datapath and a very good clock recovery (at the DAC end). 

If you have those, channel being HDMI, TOS, SPDIF or USB is irrelevant because your analog circuitry will have a quality base to build good sound from...

exactly .........or even with clock data over RJ45...if you remember that one http://www.denon.com/pages/GlossaryDetail.aspx?GId=14 which furthermore also sounded great