vacuum_tubes
pfm Member
and expensive, I stopped renewing it 10 years ago when the company I worked for stopped paying it.
I did mine 20yrs ago via the Voice route so I won't claim to be anywhere near as good as the current crop of CCIEs out there.
Does connecting a streamer to Ethernet make much difference compared to wifi?
- Thread starter JezmondTutu
- Start date
and expensive, I stopped renewing it 10 years ago when the company I worked for stopped paying it.
I did mine 20yrs ago via the Voice route so I won't claim to be anywhere near as good as the current crop of CCIEs out there.
tuga
Legal Alien
So am I right to conclued that you have no direct experience with high-fidelity audio?
vacuum_tubes
pfm Member
tuga
Legal Alien
This is not about data corruption/loss.
chiily
PFM Special Builder
Don't you think it strange though that with all the wires (analogue and digital) and devices the music goes through in a studio, mixing desk, etc, etc where they don't use fancy cables and network switches, that the music itself doesn't turn out unlistenable because of all the noise from the non treated, specialised cables?
Kinda a marketing blessing that the one/two bits of cable in one's home have such an impact, whereas all those other cables don't...
Philip Read
pfm Member
I think some of the people who are buying into this stuff must have massively better hearing than most or just spend all their time time bowing to ASR and have far too much disposable income!!!
Jubal
pfm Member
There’s no point replying. I’m blinded by your tin foil hat.
tuga
Legal Alien
2. INTERFERENCE GENERATION AND COUPLING
Digital audio systems are generally used in electrically noisy environments. Movie studios, broadcast studios and satellite uplinks are examples rich in interference sources. Stray capacitance and inductance provide coupling paths for the interference to enter the digital audio transmission path. Other entry points include radiated EMI, crosstalk between signals and common-mode noise. Noise currents flow in either direction through any input or output.
2.1 Interference Sources
Digital clocks for DSPs and CPUs, video equipment, switching power supplies, wireless and wired computer networks are examples of conducted and radiated interference sources. The digital audio signal is intentionally limited in bandwidth, but out-of-band high frequency noise can nonetheless affect data transmission and clock recovery circuits. In large studios, the use of hundreds of signals on long cables (100-1000M) further increase the noise burden.
Switch-mode power supplies usually operate in therange between 20 kHz and 10 MHz. These supplies must meet regulatory requirements for radiated and conducted EMI, but they can emit sufficient interference to affect high-resolution systems.
(...)
The Effects and Reduction of Common-Mode Noise and Electromagnetic Interference in High-
Resolution Digital Audio Transmission Systems
Jon D. Paul, Vice President Scientific Conversion, Inc. Novato, California, USA [email protected]
Digital audio systems are generally used in electrically noisy environments. Movie studios, broadcast studios and satellite uplinks are examples rich in interference sources. Stray capacitance and inductance provide coupling paths for the interference to enter the digital audio transmission path. Other entry points include radiated EMI, crosstalk between signals and common-mode noise. Noise currents flow in either direction through any input or output.
2.1 Interference Sources
Digital clocks for DSPs and CPUs, video equipment, switching power supplies, wireless and wired computer networks are examples of conducted and radiated interference sources. The digital audio signal is intentionally limited in bandwidth, but out-of-band high frequency noise can nonetheless affect data transmission and clock recovery circuits. In large studios, the use of hundreds of signals on long cables (100-1000M) further increase the noise burden.
Switch-mode power supplies usually operate in therange between 20 kHz and 10 MHz. These supplies must meet regulatory requirements for radiated and conducted EMI, but they can emit sufficient interference to affect high-resolution systems.
(...)
The Effects and Reduction of Common-Mode Noise and Electromagnetic Interference in High-
Resolution Digital Audio Transmission Systems
Jon D. Paul, Vice President Scientific Conversion, Inc. Novato, California, USA [email protected]
Last edited:
Fourlegs
Trade: WAVE High Fidelity digital cables
I suspect that there are others who will conclude that you have very little understanding of the design and implementation of digital audio devices and the impact of noise on their sound quality. Your day job is unfortunately not relevant to the finer points of audio design. I am sure that someone such as Rob Watts would struggle with the design of networks for safety critical environments but I know who I would rather listen to in discussions regarding digital audio circuits and their susceptibility to out of audio band noise in dac analogue circuits.
chiily
PFM Special Builder
So that says there are lots of interference and noise coming out of the studio. In a studio lots more sources of interference than at home. The problem space is significantly smaller at home, yet a single cable at home has such an impact...wow...and that cheap stuff in studios doesn't. That doesn't compute.
So how do studios get away with it?
vacuum_tubes
pfm Member
What are you on about, my job is all about audio.
VoiP uses ADC and DAC to convert analogue waveforms into digital signals that are then encspsulated into ethernet frames.
VoiP is 8 Khz bandwidth and music recording/streaming is 22-96 Khz bandwidth but the science or principles of conversion are exactly the same i.e Nyquist etc.
If we take the flow from recording studio to your HiFi it is no different to VoiP
analogue > mic > ADC > digital stream > FLAC/WAV file > IP stream > your home > router > switch > streamer > DAC > analogue
analogue > mic > ADC > digital stream > IP stream > IP LAN/WAN > digital stream > DAC > analogue
All the science of digital music was based on the science developed from telecomms, signal processing was developed for telephony/telecomms decades before it was used in recording studios etc.
Please explain the difference to me then?
You certainly without doubt have very little understanding of the design and implementation of digital audio devices. What single digital audio device have you ever designed? What single qualification in digital audio engineering or any related discipline have you achieved?
vacuum_tubes
pfm Member
Beyond selling foo cables he thinks that Rob Watts invented the DAC
TheFlash
Reiki Audio
And, one might politely ask, back at you: what single digital audio device have you ever designed? What single qualification in digital audio engineering have you achieved? Do all of us here need to have such experience to comment with any validity? I hope not!
The trouble with asking Nick (or anyone) that question is that it seems to suggest that the questioner doesn't understand digital audio devices at all; that he/she presumes that they are primarily about digital. A superficially digital device can be designed primarily to address non-digital issues in digital playback, as the venerable @tuga has politely pointed out.
For example, if we have say 5m of ethernet cable from our router to our streamer, then through a digital lens there is no difference in placing a network switch just after the router or in placing it just before the streamer. But in sound quality terms, there is usually a clearly audible difference and placing it just before the streamer is the better option. This is not because it does anything digitally; it's because it does something analogue-ly... as in massively reducing the RFI reaching the streamer and ultimately the DAC.
It's easily demonstrated.
TheFlash
Reiki Audio
If you see the world through a digital lens, sure. I have decades in IT and can vouch for the inherent robustness of ethernet transmission. But this is completely missing the point. With respect.
Jubal
pfm Member
Stick the kettle on buddy! Settle the matter? It’s going in circles because people keep saying RFI without explaining.
It’s very hard to understand what analogue-ly is occurring for a digital guy. Logic tells me that what leaves the streamer reaches the DAC 1:1 regardless of what you put between. Assuming no digital network errors. You agree with that. So what, exactly, is happening at the DAC to make any RFI on the Ethernet wiring an audible factor? Data only becomes analogue at the DAC. After it’s left the network completely. What am I missing?
Fourlegs
Trade: WAVE High Fidelity digital cables
I think we are fast approaching the point where there is very little advantage of discussing this any more. It is a pity because I am sure if were were all in the same room discussing the subject and listening to music which demonstrated the matters in question it would be easy to come to some agreement.
If one cannot win the argument with logic and reason then playing the ‘don’t you know who I am and what I do‘ card very rarely works.
The best way though to discuss this is over a cuppa and whilst listening to music because then one can demonstrate this sound and how the music improves when steps are taken to address it.
Obviously the naynay crowd will poo poo the suggestion of what is happening and what can be heard but that is just life.
If one cannot win the argument with logic and reason then playing the ‘don’t you know who I am and what I do‘ card very rarely works.
I am absolutely with you on the matter of what leaves the streamer arrives at the DAC 1:1. If there were any digital errors they would probably be heard anyway as a click or other noise and that rarely or never happens. Unfortunately part of what leaves the streamer and arrives at the DAC can include out of audio spectrum (RF) noise and this can cause modulation in the analogue stages of the DAC and this distortion of the music is audible even when only present in small quantities. Most of us would describe it as a hardening of the sound and an emphasis on the leading edge of percussion. It can also make the sound rather fatiguing.
The best way though to discuss this is over a cuppa and whilst listening to music because then one can demonstrate this sound and how the music improves when steps are taken to address it.
Obviously the naynay crowd will poo poo the suggestion of what is happening and what can be heard but that is just life.
TheFlash
Reiki Audio
Now that, sir, is an excellent question... and I'm probably the wrong guy to answer it! I know what RFI does to degrade the sound of a system if it does reach the analogue circuitry of the DAC so my focus is on killing it. But I don't know the precise mechanisms by which RFI interferes with the waveform. I think it's to do with intermodulation i.e. that despite being in its own right well beyond the audible frequency range, it disrupts those frequencies which are in the audible range.
EDIT: and I see that as I was typing my own reply, Nick has answered your question. Bravo.
The impact is splashy treble, muddy bass, softer leading edges to things like drums, guitar strums, double bass plucks. It's not usually obvious - no hums or pops or fizzes and the degree to which it manifests is sytem-dependent - but when you hear any system for the first time without it it's pretty much of an eye-opener (and an ear-opener). I did this with my music-loving but non-audiophile CAD ninja who was concerned his ear might not be sufficiently experienced to hear a difference and expected that if he could hear any difference at all it would be subtle. It didn't take him long to say "this is not subtle".
The key takeaway for anyone seeking to experiment for thermselves is proximity. If you're going to reduce RFI, whether with cables or filters or network switches or optical or magic beans... do it just before the streamer - like 0.5m cable before if you can make that work in your physical space. That way you will be ridding your system of the maximum amount of RFI accumulated to that point.
I'm a chartered IT guy with decades in the profession and I can tell you there is nothing digitally special about audio over ethernet! The 1s and 0s arrive in the right sequence; the 7-layer ethernet protocol includes error checking and retransmission and other stuff to ensure what is sent arrives, just like it does with any other digital data transmission.
It would be nice if interested Fishies could actually share an actual table and an actual cup of tea. Internet forum jousting can be fun but it also can be tiresome. Speaking of which, I'll be at Audio Show Deluxe at Whittlebury Park on Sunday and will most likely wear my distinctive TheFlash T-shirt; happy to fund the coffees! (Sunday will be the quieter of the two days but the real reason I'm going that day is to hear Antonio Forcione live at noon.
All the best,
Nigel
tuga
Legal Alien
Have you read the snippet I posted above.
This is not about the data, as has been said many times. It's also not about band-limited low-fi VoiP.
It's about (out-of-band UHF) noise affecting D/A chips and clocks (with audible effects).
HQPlayer's developer Jussi Laako (AKA 'Miska') posts a lot about noise in Audiophilestyle and has posted measurements of many DACs at up to 2GHz, might be worth a look if you're interested.
He uses regular UTP cable and enterprise switches, ferrites in USB cables, USB isolators, low-noise medical grade SMPSs and EMI/RFI mains filters. Noise was one of the reasons driving the development of a dedicated buffer network bridge software to use with HQPlayer Desktop or Embedded (called NAA).
Of course no one has/needs to use a switch or a network bridge or a USB isolator. The system still reproduces the recorded signal without them, just not as accurately.