HDMI vs. Coaxial Digital Interconnects

RBFC

WBF Founding Member
Apr 20, 2010
5,158
45
1,225
Albuquerque, NM
www.fightingconcepts.com
In hooking up my new Krell processor, I spoke to Krell engineers and my local dealer for advice concerning a few details.

One situation that was discussed is the best method to connect my CD transport (Denon 3800 Blu-ray player) to the Krell S-1200 for CD playback. Krell advised the use of the coaxial cable, stating that it is a lower jitter connection. My dealer stated that they had heard about the comparative jitter issue, but still preferred the HDMI cable for CD playback.

I carefully auditioned both connections, and agree with my dealer that the HDMI provides better CD sound quality. The coaxial sounded a bit more bloated in the upper/mid bass, with less frequency extension at both ends of the spectrum. The HDMI opened up the soundstage and allowed more of the unique timbers of instruments to come through. Since it is the same DACs in the Krell being used, the only difference was the cables used.

I am somewhat surprised at this result, but have no complaints about the resulting sound quality. Can anyone shed any light on why they feel there is such a not-subtle difference in sound quality between the two connections? Is this something to do with clocking, etc.?

The levels were matched as well as I could without measurement equipment, and the results were the same even when I purposely made the coaxial connection a bit louder than the HDMI.

Any thoughts would be appreciated.

I will be comparing the analog outputs of the 3800 for CD playback as well, but the DACs there and following analog stage are probably not up to the standard of the Krell's circuitry.

Lee
 

amirm

Banned
Apr 2, 2010
15,813
37
0
Seattle, WA
I have not seen measurements of either unit to see their relative jitter performance on either connection. Perhaps the Denon has poor performance over its Coax.

As to hearing a difference, you are not alone. I always hear a difference when I change connection types. This is a mystery for an engineer in me as I can't quite explain the magnitude of difference and tonal quality :).

BTW, did you turn off all the front panel displays and such on the Denon assuming it has such a capability? If so, try it with that. One advantage of in that mode is that video circuits are shut off for music whereas that is not possible over HDMI where video clock represents audio.
 

RBFC

WBF Founding Member
Apr 20, 2010
5,158
45
1,225
Albuquerque, NM
www.fightingconcepts.com
I have not seen measurements of either unit to see their relative jitter performance on either connection. Perhaps the Denon has poor performance over its Coax.

At to hearing a difference, you are not alone. I always hear a difference when I change connection types. This is a mystery for an engineer in me as I can't quite explain the magnitude of difference and tonal quality :).

BTW, did you turn off all the front panel displays and such on the Denon assuming it has such a capability? If so, try it with that. One advantage of in that mode is that video circuits are shut off for music whereas that is not possible over HDMI where video clock represents audio.

Yes, I ran it in "normal" and "pure direct" modes to compare the effect of turning off non-essential circuitry. Virtually no difference = very little improvement. I'm not kidding when I say that the sound quality difference I'm describing is quite large.

Lee
 

fas42

Addicted To Best
Jan 8, 2011
3,973
3
0
NSW Australia
Can anyone shed any light on why they feel there is such a not-subtle difference in sound quality between the two connections? Is this something to do with clocking, etc.?
In a word, yes. All digital signals are analogue waveforms representing digital information, there is no "it's only 0's and 1's" about it. Once these very high frequency analogue signals enter the Krell, it is then a case of how well the Krell processes these ANALOGUE signals to then extract the analogue audio signal. If the Krell has done a better job processing the HDMI analogue (representing digital) signal and stopping interference effects then it will sound better. Everything that is called digital is always in real life analogue, hence all these peculiarities with changing cables, sources, etc, etc.

Frank
 

hifidelity1

Well-Known Member
Jan 19, 2011
28
0
906
Hi, a couple of questions.
1. Did you let the cable settle?
2. What kind of coax cable?
3. did you go back a couple of days latter and compare again
4. Is the Krell fully run in and which connection has been used the longest coax or the HDMI ?
Reason for this is because each connection even the jacks themselves need burn in and time to settle and open up, that way you can know for sure which is better and give a fair assesment.
 

naturephoto1

Member
May 24, 2010
820
7
16
Breinigsville, PA
www.nelridge.com
Lee,

I can't really say which is better. At one point I thought that the HDMI sounded a little thinner and more mechanical over HDMI than Coax when I was using my Denon 3800 BDCI. I have found though that different Coax cables and quite possibly HDMI cables at this level have an affect on the sound. Also, from some experimenting and we have alluded to this elsewhere on the forum that different length cables have an affect. I was finding that 6 foot Coax cables (and quite possibly 6 foot HDMI cables) were giving better results than 3 foot. At this point, I am using 6' lengths of both of these cables in my system rather than 3'.

Rich
 

amirm

Banned
Apr 2, 2010
15,813
37
0
Seattle, WA
I've had the Krell for just over a week. I'm using a 3 ft. coaxial cable (Ultralink) and a 4 ft. HDMI cable (Purist). I'll check it again in a few weeks.

Lee
Oh, definitely try to a longer S/PDIF cable: 1.5 meters minimum. Anything less causes reflections back to the receiver, increasing cable induced jitter. This is especially a problem with RCA connectors on both ends which is not properly impedance matched as BNC connectors are.
 

fas42

Addicted To Best
Jan 8, 2011
3,973
3
0
NSW Australia
why HMDI sounds different can be found here
As far as I'm concerned it all comes down to having a clean, very low jitter clock, and a clean low jitter digital input to the DAC. So long as you can get to that you shouldn't have any problems getting good sound. As has been said by technical people many times, this can be solved by GOOD QUALITY (my emphasis!) technical solutions, but often it seems the manufacturers are lazy, or if they put in a good technical solution they forget that they are always still dealing with analogue voltages, and don't take care of the subtle bits, such as power supply quality, interference effects, etc, well enough.

Frank
 

mep

Member Sponsor & WBF Founding Member
Apr 20, 2010
9,481
17
0
Vincent-I take it that you haven't read many of Frank's posts before. If you did, you would understand.
 

fas42

Addicted To Best
Jan 8, 2011
3,973
3
0
NSW Australia
As all electrons are analog by design, pointing out that they are analog is a truism
Vincent, the point of that is that people frequently argue that "correct" digital signals is all you have to worry about, it's all "1's and 0's". I am saying they are wrong: taken as a series of voltages it could be 1.000, .997, 1.012, 0.007, -0.025, -0.005, etc. In other words, a layer of "distortion", interference can easily be added to those "pure" digital signals, which won't upset the digital processing circuitry, but analogue parts in the vicinity may not be as happy!

Nice try, mep ... :D:D:D

Frank
 

Phelonious Ponk

New Member
Jun 30, 2010
8,677
23
0
As all electrons are analog by design, pointing out that they are analog is a truism, isn’t it.

Yes, but given that the electrons running through a digital coax or HDMI cable represent off or on only, not an analog waveform, it is a truism that appears to be misunderstood by many.

Tim
 

Phelonious Ponk

New Member
Jun 30, 2010
8,677
23
0
Vincent, the point of that is that people frequently argue that "correct" digital signals is all you have to worry about, it's all "1's and 0's". I am saying they are wrong: taken as a series of voltages it could be 1.000, .997, 1.012, 0.007, -0.025, -0.005, etc. In other words, a layer of "distortion", interference can easily be added to those "pure" digital signals, which won't upset the digital processing circuitry, but analogue parts in the vicinity may not be as happy!

Nice try, mep ... :D:D:D

Frank

And those insignificant variations are still going to be read as on or off, zero or one, as they are converted to a true analog waveform. If it was as inaccurate as you presume, we'd all be using optical.

Tim
 

Vincent Kars

WBF Technical Expert: Computer Audio
Jul 1, 2010
860
1
0
Nothing wrong by pointing out that everything digital is done with analog electrons.
Indeed a lot of people think that bits are bits and if the bits are right, everything is right.
Unfortunately this is not the case.
On the other hand I do think it is more clear to state explicitly what the consequences of 'digital' are.
The square wave form generating RFI, the imperfection of clocks resulting in jitter, etc are typical consequences of using 'digital' technology.
 

Phelonious Ponk

New Member
Jun 30, 2010
8,677
23
0
Nothing wrong by pointing out that everything digital is done with analog electrons.
Indeed a lot of people think that bits are bits and if the bits are right, everything is right.
Unfortunately this is not the case.
On the other hand I do think it is more clear to state explicitly what the consequences of 'digital' are.
The square wave form generating RFI, the imperfection of clocks resulting in jitter, etc are typical consequences of using 'digital' technology.

Nothing wrong by pointing out that everything digital is done with analog electrons.

No, and correct me if I'm wrong Vincent, as you know this much better than I -- is it not wrong to represent that stream of electrons as an analog waveform, given that exactly the same data, minus the RFI, could be carried in pulses of light with no electrons required? Square waves notwithstanding?

Bits is bits. And timing is timing and noise is noise, and representing digital as if it is analogue confuses the issues and has caused hundreds of audiophiles to wrongly attribute (and hear) analog characteristics where they don't belong or exist.

Tim
 

Vincent Kars

WBF Technical Expert: Computer Audio
Jul 1, 2010
860
1
0
Hi Tim

It is a bit unclear what you mean.
If you do it electrical, it is a (more or less) a square wave.
If you translate this to optical, you get pulses.
In both cases one uses the change of state an not the absolute value of the signal to generate our famous 1 en 0

Your 2d remark is a nice one.
Indeed some people apply what they 'know' about analog also to digital assuming that both the analog and the digital world are equal.
 

amirm

Banned
Apr 2, 2010
15,813
37
0
Seattle, WA
Here is the best way to think of this system: samples are transmitted as digital but the timing is analog.

As an example, let's look at what happens when we send digital pulses over a cable. To create a digital pulse, which is a signal that goes from zero to "one" (whatever that voltage is), we must have infinite bandwidth. That is the definition of a signal that instantly goes from 0 to something else. Another way of defining such a waveform is to say that it has infinite harmonics. So if the digital pulse has a frequency of 1 KHz, to make it a square wave, you need to also have 2, 3, 4, 5,... Khz all the way up to infinity.

No cable has infinite bandwidth. It has capacitance which combined with the impedance of the load, creates a low pass filter which means it starts to attenuate signals, the higher they get. So even if we assumed perfect receiver and transmitter, we cannot transmit all the signals and hence, the notion of transmitting a square wave cannot happen in reality.

Now, beauty of a digital system is that we don't really care what the waveform really looks like. As long as there is distinct difference between zero and one, we can recover our signal. So even a highly distorted (within reason) signal can allow us to extract the samples. Once there, we can put them in computer memory (called a buffer) and from then on, move it around perfectly with no loss.

Unfortunately for audio, we are not done yet. The sampling rate of the source, say, 44,100 of CD, is treated at informative data. It is what your display shows but is NOT the actual rate of the samples. It is perfectly fine to have 44,102 samples per second or even have it go up and down as you play. The DAC must by definition follow the speed of the source and not rely on the sampling rate metadata. To solve this problem a clock signal is embedded in the stream itself in case of S/PDIF and HDMI. That clock tells the DAC when it is to output the sample and how to stay in sync with the source.

Now our job gets more complicated. We need to extract the timing of the signals from those distorted waveforms. Let's look at what happens if the signal is distorted:



As you see in the above example out of Digital Audio Precision manual, if the signal gets distorted, we lose our sense of timing. This is called cable induced jitter.

Sampling theory tells us that we can reproduce original data only if we follow its timing precisely. If we do not, then whole bet is off. We can pray that good and precise timing was used in creation of music and that which lives in our digital format represents that. Our job then is to match that timing as closely as we can. Since only 250 picoseconds is enough to damage a bit of resolution at 20,000 Hertz, the margin of error is exceptionally low. A picosecond is a trillionth of a second btw.
 

Vincent Kars

WBF Technical Expert: Computer Audio
Jul 1, 2010
860
1
0
Oh, definitely try to a longer S/PDIF cable: 1.5 meters minimum. Anything less causes reflections back to the receiver, increasing cable induced jitter. This is especially a problem with RCA connectors on both ends which is not properly impedance matched as BNC connectors are.

If I remember correctly this 1.5 m had everything to do with the rate of the SPDIF signal.
If this is correct, you need a different SPDIF cable for each sample rate…..

Anyway, an interesting discussion about this subject between a couple of engineers:
http://recforums.prosoundweb.com/index.php/mv/msg/11678/0/32/0/
 

About us

  • What’s Best Forum is THE forum for high end audio, product reviews, advice and sharing experiences on the best of everything else. This is THE place where audiophiles and audio companies discuss vintage, contemporary and new audio products, music servers, music streamers, computer audio, digital-to-analog converters, turntables, phono stages, cartridges, reel-to-reel tape machines, speakers, headphones and tube and solid-state amplification. Founded in 2010 What’s Best Forum invites intelligent and courteous people of all interests and backgrounds to describe and discuss the best of everything. From beginners to life-long hobbyists to industry professionals, we enjoy learning about new things and meeting new people, and participating in spirited debates.

Quick Navigation

User Menu

Steve Williams
Site Founder | Site Owner | Administrator
Ron Resnick
Site Co-Owner | Administrator
Julian (The Fixer)
Website Build | Marketing Managersing