DMP-A8: Feature Requests

Is your wired ethernet that "noisy" that you can hear it? Only asking because I have a few runs greater than the maximum 100 metre channel length (with no repeaters) and I'm not hearing any noise, distortion nor getting any dropouts even over these excessively long cables using unshielded Cat6 cables.

 

Seriously?
Apart from it being pointless to explain the absurdness of the claim that fiber improves the sound,
the very least I can say is that adding an SFP port would raise the temperature around it to at least 60°C

 

Oh noew...
Now we are gathering the electrical-noise-over-TCP/IP peasants... Now, we cannot educate them all, can we? OMFG.

1. TCP-IP is a data protocol.
2. There is no way, absolutely, ABSOLUTELY no way electrical noise can be transmitted over a DATA transport. Right? Better be right, because IT IS SO, it's not an analog connection.
The only thing noise can do on a data connection is corrupt packets, which will need to, eventually, be re-transmitted.
In the exceptional case the noise is sooo high in respect to the data signal, it would cause your packets to never be received correctly (that is: too many unsuccessful retransmissions...), hence you'll experience drop-outs, or no sound at all; AKA, there is no way you could hear corrupted music. No way. NO WAY! Either you get it right, or you don't get it at all.

Come on, it's not that hard to understand, is it? Let's all be aware that an analog connection is one thing, while a DATA connection is a totally different thing.
If noise could affect data transmission so badly, don't you think your internet connection, for one thing, would be experiencing it the most, being subjected to kilometers of wire / interference all over the place?!?!?

A little common sense is enough. A little of basic technical knowledge, achievable on the internet (or by help of contributors on public forums, such as this one...) is even more.

Please, do yourself (all of you, this is not, in any sense, personal) a favor: LEARN. This can save you a) a lot of headaches and frustration, and b) possibly also a lot of money.

 

But surely those "audiophile" grade network switches at 20 times the price per port of a standard one remove all this "noise"!

 

I wish we lived on the same alley, in order to ask you out this evening for a few beers on my bill. Unfortunately it is not so.
In this respect I can save such money for a new, audiophile-grade, network switch.

P.S.
The almighty noise-is-everywhere guy, Hans Beekhuyzen has retreated his claims about - listen listen listen - audiophile network switches... here's the specific:

Cheers from one of the oldest towns around, my friend.

 

lol, did you see the whole video? He a actually didn't :-D

 

I did, but knowing the guy, I was kind of expecting the reaction. What's most important to me is the "The BS about" thing in the title.
We can never achieve perfect knowledge in everybody, but we can definitely contribute, what's your position on this, uh?

 

Interesting comments. I can recall the popular misconception some years ago that coaxial SPDIF was "superior" to optical SPDIF (Toslink) for digital interconnects with numerous audiophiles claiming that Toslink was audibly inferior. But here we have a claim that optical fibre is superior to wired ethernet at what amounts to the same relatively low bit rate transmission over short distances. Sure, I can appreciate the technical differences in bandwidth and transmission losses between optical fibre and structured cabling installations, but I seriously doubt any domestic installation is going to span in excess of 100 metres between devices. Admittedly SFP optical transceiver modules and LC patch cables are getting cheaper, but unless your router or network switch supports these I wouldn't bother swapping out gear just to accommodate this whim.

Nearly all network switches, routers and ethernet interfaces offer at least 1000 volts galvanic isolation on the ports so unless your link is exiting the building and you are in a lightning prone area or your channel is more than 100 metres in length the advantages of fibre are mute.

 

"audiophile network switches" => 100% BS - I was laughing when I saw the video(s)

 

Hahaha,
I mean... that's the point... a bucket full of BS.
Then, if we want to enjoy the ride: why not look after cable-suspenders?!?! I must advise, though, that if you depend on a pacemaker, better not caring.
And yet again: LOL

 

It doesn't affect the data transmission. It can't be affected when it is on fiber that's true. Speaker wires and ethernet wires can add noise (or any copper wire). You can have bit perfect data transmission along with noise, they are independent of each other. Fiber makes a difference, try it. You might want to actually watch that video you posted. Seems like the only ones who believe that switches don't make a difference are the only ones who haven't tried it. Or they have bad systems. If you blind a/b or null test, differences are easy to pick out. 10/10

 

Unfortunately the consumer audio industry is rife with outrageous and often unsubstantiated claims and counterclaims about cables, connectors, routers, et al. It's the nature of some factions of the audiophile community to attempt to gain kudos or one upmanship via various means and one of them is "investing" in hype. There is a mindset with some audiophiles that if it cost you an arm, leg or a kidney the product must be superior and making a night and day difference. When someone else begs to differ that the cable, device or whatever doesn't make any audible difference then the counter argument usually tends to go on the lines of -

• You've never tried it therefore you can't possibly know the difference. (Unless the person making this claim has evidence to prove this, how would they possibly know?)
• Your system isn't resolving enough to reveal the differences.
• Your equipment isn't expensive enough so it's incapable of reaching the required standards to realise the differences.
• Your hearing is defective.
• You don't know what to listen for.
• You haven't given the product enough time to "burn in". This a common ploy of snake oil vendors to discourage or somehow allowing the consumer to claim a refund on the promised performance guarantees (or claims) made by some manufacturers or vendors. I guess they work on the principle that they will eventually wear down the consumer with stalling tactics until the "satisfaction guarantee" runs out.
• You didn't place the special stones or pyramids in the correct places, therefore you aren't going to hear the differences.
• "You're never going to hear the difference because of your choice of music."

Now let's just get back to optical fibre vs wired ethernet in the domestic installation context for a moment.

What causes issues on wired ethernet (structured cabling installations) that contribute to degraded data transmission?

• Incorrect fit off of the cable terminations. This can cause Near End (NEXT) or Far End crosstalk (FEXT) between the data pairs. This impacts on the performance of the channel / link. The most common is poor fit off of connectors, splitting pairs and incorrect termination of the shielding (if the cable has shielding).
• Physically damaging the cable during installation. Stretching and minimum cable bending violations will lead to significant impedance variations on the transmission path. This can and does contribute to transmission errors or reduced effective data speed.
• Faulty connectors.
• Water ingress into the cable. (Normally only an issue where the cable exits the building).
• Pest (usually rodent) damage to the cable.
• Incorrect routing of cables. Running your Cat5, Cat6 or Cat7 ethernet cable in parallel with mains wiring, near electrically noisy sources such as SMPS used for LED lighting, fluorescent lighting fixtures.
• The proximity of nearby high power broadcasting stations causing RF induced interference, in which case the use of shielded cable is recommended.
• Faulty patch cables. Actually got caught by this one. The patch lead wire mapped correctly but failed the transmission test with very high crosstalk at both ends of the cable. On performing an "autopsy" the cable had two split and rectified pairs. The cable measured 6 inches too short as well. Looks like someone realised one connector was wired incorrectly and tried to correct the problem by making the same mistake at the other end. (Yeah, two wrongs don't make a right.)
• Mixing Cat5, Cat6 or Cat7 components. A typical mistake novices make is fitting off Cat6 cable with Cat5 connectors and expecting the link or channel to perform to Cat6 transmission standards.
• Compromised earthing where shielded cable was used.
• The installation was never tested or inspected to ensure none of the above were contributing factors to poor wired ethernet performance.

BTW, just because the transmission medium is optical fibre doesn't automatically make it free from issues. Some of the issues that contribute to degraded performance on optical fibre links include -

• Sharp bends in the fibre (minimum bending violations and kinks) which can result in in pulse smearing. However this only becomes an issue on very long links (well beyond what one might encounter on a domestic installation)
• Dirt contamination on the connector ends or within the SFP modules.
• Damage to the optical fibre connector interfaces (usually chips or scratches)
• Fractures in the medium (normally only a problem if the fibre is glass), however a crushed plastic fibre isn't beyond the possibilities.
• SFP optical transmitters don't last forever. The laser can eventually fail, usually from overheating in the higher power units.
• Most if not all domestic installations are rarely tested to verify actual performance standards are met. This is equally true for wired ethernet installations as most consumers don't even realise such objective testing even exists and if they did to what minimum performance benchmarks they should achieve in order to pass to satisfy industry standards.

There are ITU-T standards which cover both wired ethernet and optical fibre installations as well as performance verification testing of such installations, however the average audiophile probably ignores these and even if they were presented with the standards and / or results of testing they may not know how to fully interpret them correctly.

 

Cool story bro.

 

Show me the test results where fiber didn't make a difference where properly implemented fiber run didn't make a measurable improvement in sound.

 

Well seeing you were the one that initially made the claim optical fibre provided superior audio performance compared to a wired ethernet solution in the domestic context I think the onus is on you to prove your assertions are correct.

Remember I did qualify my responses by saying that the quality of the installation had a bearing on the performance of the link / channel. (see post #32 above). These caveats apply to both structured cabling (wired ethernet) as well as optical fibre. As a previous auditor for the Australian Communications and Media Authority (ACMA) I found lots of non-compliant installations for both types. Sadly some done by so called licensed professional cablers who should have known better and numerous non-compliant installations done by unlicensed DIY cablers. Also I have tested (in the capacity as both an auditor and as a certifications technical expert) numerous installations which didn't pass the performance tests. 99% of failures which were directly contributed to the use of non-compliant product, poor installation practices, lack of correct certification testing or all of the aforementioned.

I don't know what regulations apply in the USA, but based on a lot of unqualified DIY house wiring it wouldn't surprise me that either medium could go horribly wrong. Many unqualified DIYers watch something on YouTube or read something on the internet and take that information as gospel without ever verifying its authenticity or validity. Oh, the wonders of the misinformation age.

 

Enough said, the poor cabling techniques may lead to signal degradation, and therefore to excessive packet retransmission. Not. though, to data degradation, on this we have to agree, otherwise you could assert that this text of mine could be seen as garbage by someone with a bad cabling at home...

It is of course true that the link quality can be bad / subject to excessive electrical / magnetic / RF noise, but this will ONLY lead to the need of packet retransmission (lowering the speed / bandwidth). It can be otherwise improved (against such noise) using fiber-optic cabling. Excessive packet retransmission, though, DOES NOT MEAN the quality of the music you hear is bad / with higher noise-floor etc.

You COULD, eventually, experience pops / clicks / crackles due to underruns (packets do not arrive in time, or do not arrive at all due to too many transmission errors caused by such noise) BUT the data inside the packets - the actual music - will not be degraded.
As a silly example, in a web page you EITHER SEE an image, or NOT; what doesn't happen, though, is that you won't definitely SEE IT CORRUPTED. So happens to the music: either you hear it, or you don't; but you won't hear it degraded from an acoutic noise perspective; hence the audio-grade cables / switches / routers blatancy does not make any sense.

 

And this is why I enjoy this Forum! Though it makes sense, I hadn’t put 2 & 2 together and realised that Ethernet connectors came in different Cats as well as cable.

Taken from Reddit:
‘I have more money than sense, and figured I’ll upgrade my speaker cables to something equally stupid. I like the idea of a laser plasma channel, but I’m worried about RF. Liquid Mercury is also more conductive than silver, but I’m not sure how I’d braid cables filled with liquid metal … also, not very green. Suggestions?? Also, if I used plasma channel is ordinary air too pedestrian? I feel like a mix of argon and helium would improve both the high and low end.’

That said, I do recall stories in the ‘90’s (probably apocryphal) of Japanese enthusiasts who were using troughs of liquid mercury as speaker cabling and then suffering from mercury poisoning as the mercury heated-up and evaporated. Probably total BS, but audiophiles be crazy.

As we’re now completely off-topic, a quick question from an electronics/streaming layman: I use my A6 as straightforward USB out streaming transport to a Z8 in a traditional stereo set up and volume is fixed. I’m not seeing anything in the A8 specs that will do anything for SQ for my purposes - am I missing anything?

Now, off to rearrange my stones!

 

Hahahahahaha

No you are not

 

Even in some technical circles that deal with data cabling one finds Cat6 cable terminated onto Cat5 sockets and Cat5 patch panels or patched with Cat5 leads and the less knowledgeable in this area wonder why they aren't getting Gigabit performance in both directions of transmission simultaneously (i.e. full duplex transmission at a least 1 Gbits/s). In effect the link slows down to the lowest common denominator. In this example combination of Cat6 and Cat 5 components pulls the link or channel down to Cat5 performance.

Now there will be some who will claim Gigabit performance over Cat5, but this can only be achieved in one direction at a time. To ensure full duplex Gigabit performance ALL components not only have to be at minimum Cat6 components but the cabling installation and socket fit off has to comply with the minimum standards. One of which is maintaining the natural twist of the cable pair to within 5mm of the termination otherwise NEXT and FEXT will be compromised and this in turn will degrade the overall performance of the link or channel. When I was doing performance audit testing of structured cabling installations this was one of the contributing factors to the circuit not meeting the required performance criteria to be certified to the relevant category standard. The general excuse the installers cited was they didn't know the different requirements between Cat5 and Cat6 or Cat7 installations. They mistakenly thought they were all the same and only the cable made the difference. Sad as in order to make the installation pass the transmission tests the sockets need to be re-terminated and / or the correct category sockets installed. This is when the installer hopes they have left a bit of excess cable in the wall or duct to allow a refit off without having to re-run the cable.

Apologies for going off topic but unfortunately there are a lot of misconceptions and even misinformation out there about structured cabling.

 

No multichannel DSD, no purchase