So I have read many of these posts and as a network engineer for 25+ years working for a lead company. Please don't tell me that the answer is to purchase and listen myself. Let's stick to logic, science and analysis. First, in regards to ethernet. Ethernet uses en encoding mechanism called Manchester encoding. You can read about it here: en.wikipedia.org/wiki/Manchester_code . It is important to understand this to understand the complexities of return to zero that are used in these switches, routers at INPUT level to discern the zeros and ones. No matter the switch, from $50 monoprice to a $2M linecard used in ISP land, the packet CRC validation is done on input upon receive completion. These CRC calculations are done super fast by ASICS at the port level and in $50 land switches usually completed on a single SOC. Your binary and wave explanation doesn't match what happens on the wire as manchester encoding doesn't happen as ON/OFF you are referring to. The second part is that there is error correction done at Layer 1 ( L1 is the ethernet SIGNAL level ). The first error correction is on the signal itself. There are DSP/ASICS that upon receiving the signal make adjustments to correct the signal pattern. This is because based on length, the signal starts to get weak and introduce deformations that you can see in an oscilloscope. But here is the issue, before the bit pattern is put INTO the buffer this signal error correction is already taking the signal and make it legible to the binary readers. Remember that the manchester encoding process is complex and between the positives and negatives, the encoding mechanisms needs to determine the ones and zeros. So by the time it hits the part of the buffer, the signal had been reconstructed properly. The second layer is called Layer 2 and that has the ethernet data structure. This contains the MAC Addresses that are the destination on this L2 segment as to WHO receives the packet. To double check everything, the packet has a 32bit CRC that is attached from the transmitting port on output appended to L2 frame. This CRC is then used to run a quick calculation against the data payload upon input on the other side of L2. What is done is the same mathematical computation in both side on the same byte data ( that normally is a max of 1500 bytes of length for home networks, but in some cases can be 9000 bytes if the network operator sets it, but not possible for network transmission from the internet). If those two match, the CRC is considered valid and the packet is passed to the next port. The OUTBOUND port then has to recalculate the CRC for the packet as it might have to rebuild the L1 construct depending on what it had to do ( this can happen when going from two separate Layer 2 networks). If the packet reaches it's destination device endpoint, then the packet unfolding begins. Remember at this point for EVERY Layer 2 hop in the network the CRC has been recalculated by the outbound Switch or Router. In a typical network traversal between my computer and Qobuz servers, I am seeing over 14 Layer 2 hops!! That is 14 times we have verified the data integrity of the payload. It gets better though... The transport for Qobuz from their servers in AWS is TCP. TCP stands for transmission control protocol. This is really important to understand because TCP lives on the fourth layer of the OSI model. The whole jitter/latency is just nonsensical. I explain because TCP functions by transmission windows calculated in bytes. In any TCP/IP implementation ( and there are very very few these days [Linux, Windows] ), until the window is closed on the packet transmission the TCP/IP stack is not going to send it up to the application requesting. These chunks need to from a complete chunk that is understood by the higher layers!!!! So this is important because as I mentioned earlier, the application layer of Qobuz doesn't have a clue what TCP/IP is. It simply requested to send a data block/chunk to another device. Why is this important? Because the data block that Qobuz is sending and receiving has a data structure in itself! So that means that until the transmission layer (TCP) hasn't finished getting all the parts of that "block" of data he doesn't send it back up through the RPC channels to the Qobuz application or ROON. This is the reason that when doing packet capture of a Qobuz stream you see "spurts" of byte transfers. From what I have seen usually in windows of about .5MB coming from the server, followed by a pause and continue. The same happens with Netflix, Apple stream etc. Back to TCP. In TCP world and packet transmission you have a WIDE gamut of packet variance delta between packets as it crosses the internet or even your home network. There is NO relation to jitter as until the TCP window is closed the block of data pertaining that request is NOT sent up the application internally in the RPC path internal to the OS. Want t know another fun tidbit? Inside of TCP is another layer of CRC for the data payload of the TCP packet. This TCP CRC is only decoded/checked by the end of the TCP transmission. Network elements in the middle only perceive that as data payload. So if I am ROON core on a MacMINI or sitting on Nucleus (don't have one, but going to assume that it's running some form of Linux), I make API calls to Qobuz or a internal network NAS and receive the packets in the exact same format. If the Qobuz data structure is 1/2MB of data, then until that is all received Qobuz application has nothing to work with as it needs the complete data to have the data structure it needs to process the data block. Oh by the way... If you have a question on what happens on WIFI. Replace L1 ethernet with L1 WIFI and the rest of this explanation is exactly the same!! WIFI utilizes a complete different encoding mechanism and error correction that is much more complex ( called QAM ) to handle the issue that that signal needs to be massively more reconstructed before sending up the stack for its L2 ethernet packet structure work ( including that darn CRC ) at L1. So packets are retransmitting at various layers across a WIFI signal and hence why as you move away from the base station the speeds start dropping as the packet has to be retransmitted for failure of CRC on either end of the wave carrier. So, I don't know what you are listening that you believe is a perceived difference. Without double blind tests, not going to play the game. And doing this correlation that DACS sound different to the fact that ethernet switches should sound different is also a huge leap of faith that doesn't make sense to me. DACs have many components that are related to the conversion of those bits into an analog wave that impact the signal of what we hear. You can hear the difference between a DAC in 1995 and a DAC from 2021 and you can even measure those differences. But packet transmission, while a complex beast that includes hardware packet processing and software packet processing, is very specific and has a ton of safeguards to insure the packet structure is correct on destination upon arrival. It was built this way since the 80's when digital signal processing of data on wire was much less reliable ( think RS-232, X.25 ) and the layers made sure that data integrity was proper. Now a $15 monoprice switch that has a single SOC ethernet switch does everything to forward/receive packets at line rates above 1.5Million PPS without sweating a bat and barely using any power to do so.
@@jamienewton5985 I guess it does. I can understand why someone who's spending 500 or even 1000s or more euros on cables, or network switches WANTS to believe there's an audible difference. If I just spend 2k on a cat7 network cable, I sure hope it makes a difference. But unless someone can PROOF by measuring and showing me the difference on paper, I don't buy it. The same with expensive HDMI cables that show "blacker blacks". I think it speaks volumes Hans has an Aqvox switch on display, which is basically a 30 euro switch, repackaged and sold for 400 euros. LTT opened this switch, only to discover they glued a fake crystal on top of the oscillator, scraped off the original part information, glued everything together and just added a few "illuminati" stickers. Sad to see/hear, since I was under the impression this channel had some good content in the past, which is still relevant and accurate.
Dit you ever listen to those switches on a good stereo. I think not. I can see you know a lot about networking/digital connection. But the answer lies in audio. You might want to watch the presentation I gave on the Noir et Blanc Brussels Hifi Show ua-cam.com/video/TnmsNZb7OeE/v-deo.html
Two things to consider: Ethernet will split the digital sound into multiple packets (fragmenting), so according to your theory, you would have extremly hard jitter at least every 1500 bytes, as there will be a header of the next package and no Audiodata in that timeframe. Therefore buffering has to be implemented in any Ethernet DAC, there is no way arround it! DAC conversion from Buffer -> Analog will always be done with the clock of the DAC device! Galvanic isolation for ethernet ports isn´t just present in the switch, but all Ethernet devices! Common mode on ethernet cable pairs won´t affect the ground plane of a device as it is "galvanically isolated!"
@@TheHansBeekhuyzenChannel it may have to do with buffer size in the equipment (equipment rated for more bandwidth usually has significantly larger buffer sizes, which run faster) and default MTU (max transmission unit) size. (Less fragmentation and smoother transmission.)
Thanks Hans for this interesting video. Jitter is for sure a problem for DA converters, but I doubt that the jitter of a network switch has an impact on the sound quality. A CD quality stream has a sample rate of 44.1kHz. If the player does an upsampling, the sample rate might be 192kHz. The DA converter works with this frequence. Jitter overlays this 192kHz and causes noise. On the network and the switch, the baudrate is 125MHz. This is 651 times quicker than the sample rate of the DA converter. The player has to deal with these two differen speeds. On the network, the music is transferred in packets. Between the packets are breaks. The player must buffer the packets and continuously play the music much slower than it was received. The information of the left and right channel is transportet in sequence through the network. The player has to synchronize them, means it has to delay one channel. There are further tasks that the player has to do to be able to play network music: IEEE decoding, eliminating headers and addresses of UDP/IP, MQA.... The conclusion of all this is, that the player does a lot of dataprocessing. All data that goes to the DA converter is completely processed=rendered by the player in time and amplitude. Any jitter from network won't pass through this processing. I do not know if a switch can impact the soound. But if this happens, the reason for that is not jitter.
Aye - perfect recap, and thank you for taking the time. I also do not see how it is possible that a properly working network switch can negatively impact playback of an audio-file. The obvious test here would be to a/b a file that's streamed over the network, vs a file that resides on an drive of the PC used for playback. Needless to say, I could never get any of the uptone-proponents to report the results of such a simple test.
I have 45 years of reviewing both domestic and studio equipment behind me and I do have the UpTone. I’m curious why you question my findings... I’m not offended, just curious.
Thank you for this really clear explanation of Jitter and digital to analog audio conversion. I really believe in the effect this can have on what we hear out of our audio system. And that’s what I’m after. Not numbers. So it’s needless to say that our ears are the most valuable measurement tool we have on our quest to find the best sonic experience. As we push the limits of our audio system, we also push the limits of our ear. And thus, it’s important to trust in the things we hear, yet stay critical and sceptical. (That’s why I think we can benefit from sceptical comments) For me being an audiophile is more about training your own hearing then about the best setup. And that’s what’s beautiful about this community. One set of ears is amazing, but a thousand ears giving eachother feedback is more amazing. And even the more stoic feedback from brains without ears can strengthen our hearing So lets be respectful with our opinions here and listen and consider other people’s opinions
OMG Hans and everyone. Please come to your senses! The reason any digital tweak or gadget affects the sound of a proximate DAC is due to radiated or conducted RF noise. The data through a digital chain is always 100% perfect. It's only deep in a DACs d/a section that all the RF noise becomes manifest as subtle SQ differences. Hans I have a 90dB RF isolation chamber I am selling and when the digital chain in placed inside there is NO sound quality difference with any tweak.
@@TheHansBeekhuyzenChannel Hans: If you have a RF isolation chamber (like what I provide ...and like what's used for mobile device certification), anything that's placed inside has no RF profile outside. My unit has a waveguide port for optical I/O and a filtered AC connection. No galvanic signals. So you place your SOURCE and a FIBER MEDIA CONVERTER inside and pass Fibre Ethernet and TOSLINK outside. Run the Fibre ethernet as distant as you can to your (ordinary) switch/hub and TOSLINK to your DAC. Turn off any and all other proximate digital electronics and remove their associated switching power supplies from the AC mains. So you now have your DAC fairly well RF isolated from the upstream digital. Perform a listening test: Then swap your SOURCE: use the units you said you could here a difference. Well since they are inside the RF isolation chamber ..you will hear no difference. You can extend this experiment with USB cables (say from SOURCE to an USB/SPDIF converter) or insert an Audiophile ethernet switch or try an audiophile power supply on the SOURCE, etc. Again, with any component change that takes place inside the RF chamber ...there is no audible difference. But if you open the lid to the RF chamber the audible effects of changes return. Of course it's all incredulous but true. So everything you are hearing when you do digital tweaks in open air and galvanic cables is all about RF impinging on a DAC. RF noise affects the ground level, the reference voltage, the clocking ...and this is all immeasurable in the audio band. ....even to my 150dB test equipment. Yet the corollary is definitely real: a DAC in the presence of less (zero) RF noise energy sounds better and unchanging with upstream tweaks.
@@peterzuiderveld2979 Peter. Sorry, no. The industry wants to sell you goods ...not make you understand what the problem is. You can't string a set of technical words together and make sense. I know what I am talking about ...its not about gross deficiencies in the sound. We don't hear dropouts, pops or clicks? right? Pick a song, any song ...the bits from the distant server that hosts that song travels across the internet, makes it to your home, your switch, your source/streamer and your DAC with 100% fidelity. Are you doubting this? So when you hear a diminished soundstage or treble that's off or a raspy background ..what causes this? And you change a USB cable or server power supply and the sound changes again. Well if the bits are always 100% correct, what is the mechanism that changes the sound in these subtle ways? Hint: its perturbations in the final D/A stage of the DAC caused by extreme sensitivity to RF noise. Get rid of the noise and the sound is glorious. Trust me.
Perhaps I'm misunderstanding this - How can a network cable or network switch have an impact on audio quality? Most digital formats/protocols employ checksums to ensure data integrity, including TCP. The only way something like this could conceivably happen is if all the audio is being sent via UDP, where the source-device sends traffic, but requires no acknowledgement from the destination, so its can easily get dropped/lost etc. TCP is incredibly reliable, if data was being corrupted, the destination-device would see that packet checksum and contents don't match, and the source would be asked to re-transmit it. Likewise, a packet is a packet, a switch does not know (or care) if its part of a word document or audio file, if its alternating/corrupting the audio, it would also be doing the same to all content on your network - the net result being, either all your content ends up corrupted, or your network will saturated by all the retransmits (An it would feel more like a 10Mb network, rather than 1Gb or even 10Gb). I've 20 years experience managing networks, working with everything from $50 desktop switches, to $5m chassis switches for backbones, so my scepticism doesn't come from ignorance, but experience - I often call BS when I read an article claiming a $1000 network cable gives better audio than a $30 CAT6a/CAT7 cable. But I would seriously like to know, why people think it is and the science behind it? also has anybody tried to a packet-capture to see if there is a difference?
Its got _virtually nothing_ to do with packets the protocols and checksums; its got _everything_ to do with noise on the ground plane of the *analog voltages* of digital signals causing timing errors, clock phase noise and increased jitter.
@@petekavanagh9186 Effectively and to a large part, yes. To a large part, it's due high-source-impedance leakage current from SMPS causing increased clock phase noise and jitter. Its also due to sh*tty clocks on generic devices, e.g. consumer-grade switches, routers, NAS', fiber media convertors, etc. Han's animation shows this really well.
Hi Kelvin, I was thinking the same thing too, but was not able to confirm this in my tests. Two findings, 1) playing a wav file from a local hard disk, I get exact amplitudes and no jitter. When playing a wav file from a network mapped drive (TCP) I see samples dropped and arbitrary samples added at times. This can only be the player, a Roon player in this case. 2) When I look at the traffic between the source-device (player) and the destination device on ethernet, I do not see enough UDP traffic to account for the audio stream. See my more detailed description in the most recent posts on this video.
Brilliant, Hans. I was aware of most of the details here, but the way you put it all in the same context made it come up a couple of lights for me. This was helpful.Thanks
My initial tests of comparing digital amplitudes of songs show that when playing a song hosted on a Roon server and received on another LAN device through a Linksys router via ethernet, and no other traffic, that some samples are dropped (amp = 0). It was a very small amount but you can hear it (a couple of skips in a 43 second song). All samples that did go through came across with the exact amplitude values but did experience a small amount if jitter (small time shift). I also did the test with Wifi on a network with many devices on it and in addition to many dropped samples I saw non zero amplitudes overriding actual samples values with about 10 samples out of 2000 being just plain wrong. Since I am using Roon, do you think this can be attributed to UDP which doesn't do error checking? Tomorrow I will run a test with a player using a network storage (CIFS) via TCP with error checking. If there are no dropped samples or incorrect amplitudes then would you consider UDP as an explanation for why networks have a 'sound'?
For the first test I had my Roon Core running on Windows (where I capture the samples) playing a file on the local hard disk, it was perfect, no dropped samples. I then put the same file on a mapped network drive (Synology NAS) and ran many tests. I consistently got 2-3 gaps (dropped samples all zero amplitudes) per play. In one of the tests I got a small section of non-zero amplitudes (~10 samples) causing a slight shift in time. My test song is only 43 seconds long. When I look at the samples when played from the Roon Server over Ethernet, I see a lot more gaps and many 10-sample-long regions with incorrect amplitudes. My tests reading the file from a network drive were not what I was expecting so I don't think this is noise caused by the non-error checking UDP protocol. Actually, putting a network sniffer on the traffic from the Roon server showed there was very small amount of UDP traffic between to/from my Windows machine, not nearly enough to account for the 8MB song stream. My theory is the Roon player is not handling the Ethernet input fast enough, whether from a Roon server or a file served over CIFS. I do not think this noise is coming from the network. Given these findings and your testimonial to the improvement with the high-end network switch leads me to believe the Etheregen is able to get the packets to the Roon player faster (possible through improved time sync). This allows the Roon player to play the song to the sound card with no gaps or modified samples. This is just a theory as I cannot duplicate your setup exactly and capture what you are hearing. One way to fix the problem I see with the Roon player is for it to buffer enough samples so that it is not replacing samples or dropping them all together while it is reading from a slower source. Perhaps there is a setting for this, but I couldn't find it. Audirvana seems to have no problem with even slower WAN streams, so I am sure it can be done.
Those suckers do sound - a little. A friend and I recently had two Zyxel GS108 and the Bonn - which is actually a modded Zyxel - for a test session. We modded one of the Zyxel step by step with capacitors (POSCAPs) mostly. We agreed that our modded Zyx did not fall behind the Bonn for less money. My older modded Zyxel GS108 again sounded a little different again. Zyxel puts into the switches whatever they get from OEMs, it seems. The name and case looks the same, the boards are totally different. I highly doubt that anything has to do with data errors. That does make no sense at all, because the IP protocol does not let anything go wrong and the devices have buffers for a number of seconds. My dCS Network bridge keeps on playing for 5 seconds or more after I unplug the switch. Meaning: The switch cannot have any influence at this moment. There are no data errors nor jitter induced by the switch. It is all about high frequency noise travelling across the cables. That's my theory.
Hans, if you stream data from a NAS to a streamer you use a protocol. SMB as an example supports multiple parallel streams and encryption. So in the digital domain the 0s and 1s and their jitter is compensated by the protocol. The parallel TCP streams needs to be reconstructed in a buffer as the data requires decryption. So it heavily relies on the capability of the receiving node, in this case the streamer. The streamer will reclock the digital information from it’s buffer to the analog output. So, why can an audio switch still result in a better sound quality? The only thing I can imagine is making the difference in what noise can enter into your system but not in the 0s and 1s themselves becasue these are handled by source, the protocol and the network target.
In economics scientists turned inquisitors during the eighties and ninieties when big data analyses showed their trusted theorie unsufficient, but they kept their trust in their tried and tested. Their teachers werd the best, they could't be wrong. Right? Additional input from psychologists and biologists gave economics the broader view that restored relevance. Neuroscience, chemistry and even psychology are now filtering through in audio. So inquisitors, turn inquisitive. If you want to describe what can be heard, be VERY sure you understand how hearing works. Hans does, and he is inquisitive. Hans, spot on. Like always. Chapeau.
You need to be careful relying on purely engineering driven knuckle headed group running the audio science review forum. I’m an engineer but strongly believe in the impact of SQ by audio equipment, including network switches. The ASR forum is polluted with measurements only loyalists that will wage war on anyone that disagrees with them or voices an opinion. What they fail to understand is that measuring instruments including their APxxx they use to measure audio equipment are made by humans who also have ears and should be able to discern differences in SQ.
What he doesn't understand (or talk about) is the parallel effect in the complete live chain where especialy ground is a big issue. It's not only about the 1's and 0´s but everything attached to the switch and back to the power line that "can" impact the sound quality. I say "can" because it all depends on the quality of the rest of the system. (everything is build for a price remember that) Even with fiber you can't solve all of these problems because there is always a connection to the power line as well and converting the signal brings hf noise. It starts with the switch mode power supplies and like any other device (audio or video related) can be improved with better filtering to minimize the parallel influence. In general you can improve any signal because voltages are still analog.
Do they even listen to the equipment they trash or praise over there? I installed a $40 switch, this one, NETGEAR 8-Port Gigabit Ethernet Plus Switch (GS108Ev3) Even my 80 yo mother immediately noticed the betterment of both the SQ and PQ of my system. I said nothing to her about doing anything to my setup, but she was still impressed. Quite frankly, the difference this inexpensive switch made is still blowing me away. This is the same switch used in many $500+ "audiophile" switches.
Dear Mr. Beekhuyzen. this has nearly reached the point of a which hunt. and i would like to excuse for all the wanne be network specialist that are not able to understand the described problem. They are mixing different layers of the OSI Model to argue, but do not realise that you are mostly talking about the first layer wich is physical transport. there are more problems in ethernet data transmission that can cause signal distortion in the physical transport besides jitter introduced by hardware. for example the problem of "cross talking" or "echo effects" there a special measuring tools to test these problems in a network. And as you correctly explained, this is not a problem if the signal stil is interpret at a given time as the correct value (1 or 0) if we talk purely about digital information transport. but we are not. and i think you made it very clear that DAC is looking at the exact voltage level of a signal at an exact time point to interprete its value as an analog representation wich than is absolutely prone to distortion in voltage changes. (did i get that right?) Even if i never heard the sound distortion of audio transfered over network. i would not argue that it can't be heard by others. i worked with high class musicians and i know my hearing ability sucks in comparison to theirs (and probably yours too). But i am glad you try to explain it in detail so that others can understand something that they couldn't before. so thank you.
Apologies if this has already been answered but I couldn’t see it anywhere. If audiophile switches sound better than normal switches because of attention paid to electrical noise etc, and galvanic isolation of data and clock lines help the sound, why not just use WiFi instead of an expensive switch? I’d be interested in your opinion, thanks 👍
Just my two cents worth on an old topic,when streaming with a wiim pro plus to my audiolb 8200 cdq dac with WiFi there was an obvious sound improvement when changing from WiFi to ethernet cable which is why I am considering the purchase of an etheregen switch between modem and wiim pro plus .The wiim pro plus is the budget component in my system as rest of system is pretty hi end and quite neutral/revealing and highlights pros/cons of individual component changes.If a roughly $1000 aud switch can make an easily audible difference in a 30 k system I believe that is money well spent!
Like everything on the internet, you will find an opinion and it’s opposite. As you say the only way is use your own ears and form your own opinion. I own a Chord Dave DAC and I can definitely hear a difference using different digital inputs. E.g. if I stream music over USB or optical from my PC or from my recently acquired Lumin U1 Mini the difference is clearly audible in favour of the latter. Bass is deeper and more defined, piano notes suddenly become richer (small variations in a short time, I don’t know how to say this) and in general there seems to be more “air” between instruments. It’s not a small difference at all! .As such I decided to order an Etherregen to feed my Lumin and in principle I am open to accept I might hear a difference even though I can’t technically explain why yet. Thanks Hans, as usual you are very serious and respectful also with people who conversely claim to hold the only truth...
@@TheHansBeekhuyzenChannel The EtherREGEN arrived and boy oh boy am I enjoying my music! I don't understand how all of this works... "And yet it moves".
Hello Hans: Does the quality ot the ethernet cord ( ca. 2M) matter from the Router to the Regen? Shure the quality of the ethernet cord from the Regen to the streamer ist very imortant. for this connection i have Audioquest ethernet dimond. But from the Router to the Regen is also important to have a hi End ethernet cord? I guess the Regen taks the signal with a clock etc.
The EtherREGEN will clean up the signal so as long as you use a normal quality patch cable and you're not close to a really strong HF source, like a TV station transmitter you're fine.
Ok i have no idea in digital transport but i was skeptical on rca wires for a long time and just used $3 to $5 rca cables for as long as i can remember. With watching ps audio, this channel, and steve, i tried a cable thats well received (qed profile) with a modest price. I was shocked how well it improved in detail, returned to my old cable and the difference is really there. I had the option to return it but i got one more set for my rca cable. Needless to say, im a convert. Now with this topic of digital noise, i will try the ifi purifier 3 and see if that cleans it up more than my idac2 can deliver. I am very skeptical about this purchase and dont have any hope that it will change how my music sounds. If it does, i am inclined to give this video a benefit of a doubt on digital transmission does make a difference. The illustration made in the video does make sense but im no expert. It seems to me that it presents the extremes on both sides that is why im a bit reluctant to 100% agreement on the subject.
How you relate jitter of the Ethernet bus to an i2s bus for example ? They are 2 different signals, from 2 different busses and 2 different clocks. Except I'm missing something here
Hi Hans nice video again. I received my EtherREGEN this Friday and it is really a big improvement on different levels. 3d, depth, width, blacks. It is even a bit louder. Very happy. I asked some questions on their forum about adding a second EtherREGEN at the internet router. The router has my Roon ROCK directly connected and I wonder what would be the benefit of using this B-port on mij Roon server. The EtherREGENs would be interconnected by glass fiber of course. I will update if something interesting comes out of that discussion...
So the outcome was that the 2nd EtherREGEN makes no sense but if I wanted to further isolate by using glass fiber the Sonore opticalModule would be a good option at the internet router side. I went along and ordered and install that. Outcome is good but the difference is smaller than the switch to EtherREGEN I did previously. My conclusion is that the EtherREGEN takes away virtually all the 'nasties' as you tend to call them already.
@@TheHansBeekhuyzenChannel Thanks for your answer. In the mean time quite some things changed in my set-up. I have a Hugo TT2 and opticalRendu both with Farad Super3 Linear Power Supply and a switch with SFP behind that. I sold the EtherREGEN because in this configuration it does not have any effect anymore what-so-ever. (Tried Ethernet to B port to Fiber to opticalRendu -so inversed but no audible difference)
If it's not too much to ask, try a cheap switch with metal case, connect the switch case to the amplifier case, and put some ferrite cores around the network cable. (and power cables, speaker cables e.t.c.) Just the amatour radio/ham way of dealing with disturbances.
I sure can hear an difference of switches to the sound. Tried a Innuos phoenix net switch and the sound was deeper, the voice point in the soundstage becomes much more point focused. And the bass sounded more tight
Hans, I thoroughly enjoyed this video. It explains a complex subject in a digestible manner. Although I consider myself an audio sceptic, i have also heard for myself that the tiniest details CAN matter. As an audiophile, I must investigate for myself. The one thing I cannot get behind is cable elevators. Couldn't hear a difference in my highly resolving system.
Hans, I recently purchased an ER and am trying to see if there is an improvement. I’m also using aN UPTONE LPS power supply. My DAC is a PS audio DSD Jr. My amp is an STA 200. My speakers are Tekton double impacts and I have DIY cables. My music is streamed via Qobuz/Tidal from a Mac mini running Audirvana. USB is a curious evolved cable. I am having difficulty hearing a difference. I heard clear differences when trying different USB cables, interconnects, and speaker cables. I also heard a difference with an ISO Regen that I recently sold. Any thoughts on why I’m not hearing a difference? I’ve completed about 50 hours of burn in as well. Thanks for your help.
You the Mac mini as the digital (USB) source. Over time I have learned that, although being better dan many Windows Computers, it pollutes the digital stream more than for instance a streamer/networkbridge by SOtM or Sonore. This eliminates the positive influence of the switch. You might want to watch ua-cam.com/video/G9EreotBL6c/v-deo.html.
@@TheHansBeekhuyzenChannel Thanks for the reply, I have come a long way. Can I ask one follow up question? If I got a Sonore streamer, my Mac mini no longer needs to be hooked up at all. Could I change my Roon Server to be on a much more powerful (but also electrically noisy) computer now that my audio system will be physically detached?
@@TheHansBeekhuyzenChannel Wow! from your experience, you can have a powerful and noisy computer running up conversion software without a detriment to SQ as long as the unit is no physically connected (using a streamer through USB for example)? Why doesn't everyone just use a streamer!? Thanks Hans. You are a rare gem on UA-cam and to the audiophile community!
@@TheHansBeekhuyzenChannel I have used my two routers as network switches with fibre-opitc connections from my PC music server to M1 Mac mini, now the digital stream seems to sound much cleaner and almost jitter-free. I also had a Holo Audio USB filter connect the Mac mini to my Gustard A22 DAC.
Thank you Hans for sticking your neck out. It is not because for many this remains a controversial subject that we should not talk about it and try to explain. For me the deepest digital secrets still haven’t emerged. I too was very skeptical regarding audiophile switches, but I’m glad I did at least tried it. Many don’t even bother to give it a try, but nevertheless have strong opinions. That is beyond comprehension to me. And yes just as you explained, it has nothing to do with data integrity. Even a 20€ switch will manage data integrity just fine. To me it is about not upsetting the DA converter with a dirty signal.
I use the Sonore opticalRendu instead. Perfect galvanic isolation which makes everything in the network before the opticalRendu irrelevant. No need for further audio optimization in the ethernet network. As a result, my Roon server can on my iMac in the normal home/office ethernet zone, without any specific audio optimization devices..
Hi Hans, My Aurender A10 (Dac, Streamer) has built in Hard drive and no WiFi. When I A/B test with and without network cable and I don’t hear any difference, then I can not expect any improvement with an Audiophile network switch, right?
I suspect that if there is a real (and therefore measurable) improvement in the sound you hear after changing the network switch then the 'audiophile' switch is compensating for deficiencies in the box you describe as a 'DAC'. This doesn't mean the network switch is better, just that the 'DAC' is not well designed - specifically the firmware running on the microcontroller, or limitations imposed by the wrong choice of microcontroller.
I can modestly confirm recommendations made by Hans. Thanks to him I bought SOtM SMS200Ultra/SPS500 and the sound improvement was already beyond my imagination (I had played Tidal using Chord 2qute DAC/ULPS Tomanek + Cardas Clear USB cable). Curious, I took the next step and bought the Silent Angel switch + some not expensive LAN cables for my NUC (Roon Rock), Sotm, router etc. Just one combination of these cables worked really well. Supra CAT8 between router and Silent Angel, and Chord C-stream to SMS200Ultra, and Cat7 cable (that comes with Silent Angel) to NUC. This tonal balance and low frequency detail just made the magic for me. Two Supra cables in the chain raised low frequency energy too much. The rest of my system is Densen DM20/30 paired with Dynaudio Contour 1.3SE. Power cables everywhere...
Wouldn't be the right conclusion to avoid switches and anything that follows and just put the music files u want to play onto a ram drive (either virtual or physical). I also don't understand why advocate USB reclockers instead of direct I2S output
Very nice video. I'm new to the channel and in hi-fi setup. I currently use a Synology NAS DLNA Server which holds my Digital Audio Library and is then connected to a LAN Switch via Cat 6 ethernet. My Marantz AVReceiver NR1711 is also connected to the same LAN Switch via Cat6 ethernet cable. I then use an app HEOS from my Phone (connected via Wifi to the same router where the LAN Switch is connected) to fetch the digital audio from the NAS to the AVR. Should I consider the sound I'm hearing as "Altered" or "Dirtier" as compared when I do DAS (as I also watched and learned earlier from one of your older videos). Your experience and thoughts on this will be mostly appreciated. Keep safe and thank you in advance.
If all is well, there should be no difference in sound quality. Audio needs hardly any bandwidth. Using a DAS is simpler and it's easier to solve problems when they occur. But a NAS can work just as good.
We appreciate your work and devotion, Hans. Your standards are high. Not accepting audible differences by digital specialists is like medieval astronomists continuing to argue the earth is flat. Based on knowledge that will be proven wrong afterwards.
i dont have the inclination to test my modem/router, amplitudes, samples, etc. I have reasonable (i think) expectations on sound quality. The question I have is the same one I always have with any piece of hardware, car, bed, etc: how much better is it and how much does that "better" cost? Chasing incremental, possible, improvements that costs thousands of dollars seems, idk, like a sort of mania. Ive reached the point in my system that Im comfortable with the sound and can sit back and enjoy the music without fomo. I certainly wish that for all of you and hope you dont have to spend a college education to get there.
You know once a degree qualified mechanical engineer would say all a turntable needs to do is turn at 33 1/3 rpm. We now know for a fact there is a lot more to it than that. I really don’t care what a “network” expert tells me. All they are saying is they don’t understand. I now have the EtherREGEN in my system feeding a very high end steamer. I’m not deaf and I don’t have money to burn. The difference is overwhelming. Who the hell are these “experts” to tell me it’s not better based on absolutely zero first hand experience! Well done Hans, you have made the enjoyment of music I love much better. Thank you and UpTone Audio.
Once again, it appears that there is no definitive explanation as to why, in certain cases, a difference is audible. Why not ask the manufacturers of "audiophile network switches" to explain their technical design decisions in order to close the debate in a rational manner?
@@TheHansBeekhuyzenChannelPerhaps it would be interesting for the industry to provide some arguments, rather than perpetuating a reputation as snake oil vendors
So I have read many of these posts and as a network engineer for 25+ years working for a lead company. Please don't tell me that the answer is to purchase and listen myself. Let's stick to logic, science and analysis.
First, in regards to ethernet. Ethernet uses en encoding mechanism called Manchester encoding. You can read about it here: en.wikipedia.org/wiki/Manchester_code . It is important to understand this to understand the complexities of return to zero that are used in these switches, routers at INPUT level to discern the zeros and ones. No matter the switch, from $50 monoprice to a $2M linecard used in ISP land, the packet CRC validation is done on input upon receive completion. These CRC calculations are done super fast by ASICS at the port level and in $50 land switches usually completed on a single SOC. Your binary and wave explanation doesn't match what happens on the wire as manchester encoding doesn't happen as ON/OFF you are referring to. The second part is that there is error correction done at Layer 1 ( L1 is the ethernet SIGNAL level ). The first error correction is on the signal itself. There are DSP/ASICS that upon receiving the signal make adjustments to correct the signal pattern. This is because based on length, the signal starts to get weak and introduce deformations that you can see in an oscilloscope. But here is the issue, before the bit pattern is put INTO the buffer this signal error correction is already taking the signal and make it legible to the binary readers. Remember that the manchester encoding process is complex and between the positives and negatives, the encoding mechanisms needs to determine the ones and zeros. So by the time it hits the part of the buffer, the signal had been reconstructed properly. The second layer is called Layer 2 and that has the ethernet data structure. This contains the MAC Addresses that are the destination on this L2 segment as to WHO receives the packet. To double check everything, the packet has a 32bit CRC that is attached from the transmitting port on output appended to L2 frame. This CRC is then used to run a quick calculation against the data payload upon input on the other side of L2. What is done is the same mathematical computation in both side on the same byte data ( that normally is a max of 1500 bytes of length for home networks, but in some cases can be 9000 bytes if the network operator sets it, but not possible for network transmission from the internet). If those two match, the CRC is considered valid and the packet is passed to the next port. The OUTBOUND port then has to recalculate the CRC for the packet as it might have to rebuild the L1 construct depending on what it had to do ( this can happen when going from two separate Layer 2 networks). If the packet reaches it's destination device endpoint, then the packet unfolding begins. Remember at this point for EVERY Layer 2 hop in the network the CRC has been recalculated by the outbound Switch or Router. In a typical network traversal between my computer and Qobuz servers, I am seeing over 14 Layer 2 hops!! That is 14 times we have verified the data integrity of the payload. It gets better though...
The transport for Qobuz from their servers in AWS is TCP. TCP stands for transmission control protocol. This is really important to understand because TCP lives on the fourth layer of the OSI model. The whole jitter/latency is just nonsensical. I explain because TCP functions by transmission windows calculated in bytes. In any TCP/IP implementation ( and there are very very few these days [Linux, Windows] ), until the window is closed on the packet transmission the TCP/IP stack is not going to send it up to the application requesting. These chunks need to from a complete chunk that is understood by the higher layers!!!! So this is important because as I mentioned earlier, the application layer of Qobuz doesn't have a clue what TCP/IP is. It simply requested to send a data block/chunk to another device. Why is this important? Because the data block that Qobuz is sending and receiving has a data structure in itself! So that means that until the transmission layer (TCP) hasn't finished getting all the parts of that "block" of data he doesn't send it back up through the RPC channels to the Qobuz application or ROON. This is the reason that when doing packet capture of a Qobuz stream you see "spurts" of byte transfers. From what I have seen usually in windows of about .5MB coming from the server, followed by a pause and continue. The same happens with Netflix, Apple stream etc.
Back to TCP. In TCP world and packet transmission you have a WIDE gamut of packet variance delta between packets as it crosses the internet or even your home network. There is NO relation to jitter as until the TCP window is closed the block of data pertaining that request is NOT sent up the application internally in the RPC path internal to the OS. Want t know another fun tidbit? Inside of TCP is another layer of CRC for the data payload of the TCP packet. This TCP CRC is only decoded/checked by the end of the TCP transmission. Network elements in the middle only perceive that as data payload. So if I am ROON core on a MacMINI or sitting on Nucleus (don't have one, but going to assume that it's running some form of Linux), I make API calls to Qobuz or a internal network NAS and receive the packets in the exact same format. If the Qobuz data structure is 1/2MB of data, then until that is all received Qobuz application has nothing to work with as it needs the complete data to have the data structure it needs to process the data block.
Oh by the way... If you have a question on what happens on WIFI. Replace L1 ethernet with L1 WIFI and the rest of this explanation is exactly the same!! WIFI utilizes a complete different encoding mechanism and error correction that is much more complex ( called QAM ) to handle the issue that that signal needs to be massively more reconstructed before sending up the stack for its L2 ethernet packet structure work ( including that darn CRC ) at L1. So packets are retransmitting at various layers across a WIFI signal and hence why as you move away from the base station the speeds start dropping as the packet has to be retransmitted for failure of CRC on either end of the wave carrier.
So, I don't know what you are listening that you believe is a perceived difference. Without double blind tests, not going to play the game. And doing this correlation that DACS sound different to the fact that ethernet switches should sound different is also a huge leap of faith that doesn't make sense to me. DACs have many components that are related to the conversion of those bits into an analog wave that impact the signal of what we hear. You can hear the difference between a DAC in 1995 and a DAC from 2021 and you can even measure those differences.
But packet transmission, while a complex beast that includes hardware packet processing and software packet processing, is very specific and has a ton of safeguards to insure the packet structure is correct on destination upon arrival. It was built this way since the 80's when digital signal processing of data on wire was much less reliable ( think RS-232, X.25 ) and the layers made sure that data integrity was proper. Now a $15 monoprice switch that has a single SOC ethernet switch does everything to forward/receive packets at line rates above 1.5Million PPS without sweating a bat and barely using any power to do so.
Absence of a reply speaks volumes
@@jamienewton5985 I guess it does. I can understand why someone who's spending 500 or even 1000s or more euros on cables, or network switches WANTS to believe there's an audible difference. If I just spend 2k on a cat7 network cable, I sure hope it makes a difference. But unless someone can PROOF by measuring and showing me the difference on paper, I don't buy it. The same with expensive HDMI cables that show "blacker blacks". I think it speaks volumes Hans has an Aqvox switch on display, which is basically a 30 euro switch, repackaged and sold for 400 euros. LTT opened this switch, only to discover they glued a fake crystal on top of the oscillator, scraped off the original part information, glued everything together and just added a few "illuminati" stickers. Sad to see/hear, since I was under the impression this channel had some good content in the past, which is still relevant and accurate.
Dit you ever listen to those switches on a good stereo. I think not. I can see you know a lot about networking/digital connection. But the answer lies in audio. You might want to watch the presentation I gave on the Noir et Blanc Brussels Hifi Show ua-cam.com/video/TnmsNZb7OeE/v-deo.html
@@TheHansBeekhuyzenChannel so you did not know what to answer.... hahahah
Two things to consider:
Ethernet will split the digital sound into multiple packets (fragmenting), so according to your theory, you would have extremly hard jitter at least every 1500 bytes, as there will be a header of the next package and no Audiodata in that timeframe.
Therefore buffering has to be implemented in any Ethernet DAC, there is no way arround it! DAC conversion from Buffer -> Analog will always be done with the clock of the DAC device!
Galvanic isolation for ethernet ports isn´t just present in the switch, but all Ethernet devices! Common mode on ethernet cable pairs won´t affect the ground plane of a device as it is "galvanically isolated!"
OK, then explain why there is such a difference in sound? For that's undeniable.
@@TheHansBeekhuyzenChannel it may have to do with buffer size in the equipment (equipment rated for more bandwidth usually has significantly larger buffer sizes, which run faster) and default MTU (max transmission unit) size. (Less fragmentation and smoother transmission.)
Thanks for this Hans, I value your experience and history and the time you take to demonstrate your findings.
My pleasure
Thanks Hans for this interesting video. Jitter is for sure a problem for DA converters, but I doubt that the jitter of a network switch has an impact on the sound quality.
A CD quality stream has a sample rate of 44.1kHz. If the player does an upsampling, the sample rate might be 192kHz. The DA converter works with this frequence. Jitter overlays this 192kHz and causes noise.
On the network and the switch, the baudrate is 125MHz. This is 651 times quicker than the sample rate of the DA converter. The player has to deal with these two differen speeds. On the network, the music is transferred in packets. Between the packets are breaks. The player must buffer the packets and continuously play the music much slower than it was received. The information of the left and right channel is transportet in sequence through the network. The player has to synchronize them, means it has to delay one channel. There are further tasks that the player has to do to be able to play network music: IEEE decoding, eliminating headers and addresses of UDP/IP, MQA....
The conclusion of all this is, that the player does a lot of dataprocessing. All data that goes to the DA converter is completely processed=rendered by the player in time and amplitude. Any jitter from network won't pass through this processing.
I do not know if a switch can impact the soound. But if this happens, the reason for that is not jitter.
`it's obvious you didn't try them. Giving an opinion without testing is rather useless, I'm afraid.
Aye - perfect recap, and thank you for taking the time. I also do not see how it is possible that a properly working network switch can negatively impact playback of an audio-file. The obvious test here would be to a/b a file that's streamed over the network, vs a file that resides on an drive of the PC used for playback. Needless to say, I could never get any of the uptone-proponents to report the results of such a simple test.
Than do that first before commenting on my findings 😟
@@TheHansBeekhuyzenChannel I would, but unfortunately I don't own an uptone, or anything like it.
I have 45 years of reviewing both domestic and studio equipment behind me and I do have the UpTone. I’m curious why you question my findings... I’m not offended, just curious.
Thank you for this really clear explanation of Jitter and digital to analog audio conversion. I really believe in the effect this can have on what we hear out of our audio system.
And that’s what I’m after. Not numbers. So it’s needless to say that our ears are the most valuable measurement tool we have on our quest to find the best sonic experience. As we push the limits of our audio system, we also push the limits of our ear. And thus, it’s important to trust in the things we hear, yet stay critical and sceptical.
(That’s why I think we can benefit from sceptical comments)
For me being an audiophile is more about training your own hearing then about the best setup. And that’s what’s beautiful about this community. One set of ears is amazing, but a thousand ears giving eachother feedback is more amazing. And even the more stoic feedback from brains without ears can strengthen our hearing
So lets be respectful with our opinions here and listen and consider other people’s opinions
👍🏼
OMG Hans and everyone. Please come to your senses! The reason any digital tweak or gadget affects the sound of a proximate DAC is due to radiated or conducted RF noise. The data through a digital chain is always 100% perfect. It's only deep in a DACs d/a section that all the RF noise becomes manifest as subtle SQ differences.
Hans I have a 90dB RF isolation chamber I am selling and when the digital chain in placed inside there is NO sound quality difference with any tweak.
Yes exactly...
Details please for this sounds like you didn't try or did use a poor setup
@@TheHansBeekhuyzenChannel Hans: If you have a RF isolation chamber (like what I provide ...and like what's used for mobile device certification), anything that's placed inside has no RF profile outside. My unit has a waveguide port for optical I/O and a filtered AC connection. No galvanic signals. So you place your SOURCE and a FIBER MEDIA CONVERTER inside and pass Fibre Ethernet and TOSLINK outside. Run the Fibre ethernet as distant as you can to your (ordinary) switch/hub and TOSLINK to your DAC. Turn off any and all other proximate digital electronics and remove their associated switching power supplies from the AC mains. So you now have your DAC fairly well RF isolated from the upstream digital.
Perform a listening test: Then swap your SOURCE: use the units you said you could here a difference. Well since they are inside the RF isolation chamber ..you will hear no difference. You can extend this experiment with USB cables (say from SOURCE to an USB/SPDIF converter) or insert an Audiophile ethernet switch or try an audiophile power supply on the SOURCE, etc. Again, with any component change that takes place inside the RF chamber ...there is no audible difference. But if you open the lid to the RF chamber the audible effects of changes return. Of course it's all incredulous but true.
So everything you are hearing when you do digital tweaks in open air and galvanic cables is all about RF impinging on a DAC. RF noise affects the ground level, the reference voltage, the clocking ...and this is all immeasurable in the audio band. ....even to my 150dB test equipment. Yet the corollary is definitely real: a DAC in the presence of less (zero) RF noise energy sounds better and unchanging with upstream tweaks.
Again, did you do a listening test and if so, how.
@@peterzuiderveld2979 Peter. Sorry, no. The industry wants to sell you goods ...not make you understand what the problem is. You can't string a set of technical words together and make sense. I know what I am talking about ...its not about gross deficiencies in the sound. We don't hear dropouts, pops or clicks? right? Pick a song, any song ...the bits from the distant server that hosts that song travels across the internet, makes it to your home, your switch, your source/streamer and your DAC with 100% fidelity. Are you doubting this?
So when you hear a diminished soundstage or treble that's off or a raspy background ..what causes this? And you change a USB cable or server power supply and the sound changes again. Well if the bits are always 100% correct, what is the mechanism that changes the sound in these subtle ways? Hint: its perturbations in the final D/A stage of the DAC caused by extreme sensitivity to RF noise. Get rid of the noise and the sound is glorious. Trust me.
Good job. I'm impressed by your motivation to get over all the trouble explaining.
🙏🏻
Thanks for your efforts and sharing your results and insights .
My pleasure!
Perhaps I'm misunderstanding this - How can a network cable or network switch have an impact on audio quality?
Most digital formats/protocols employ checksums to ensure data integrity, including TCP. The only way something like this could conceivably happen is if all the audio is being sent via UDP, where the source-device sends traffic, but requires no acknowledgement from the destination, so its can easily get dropped/lost etc.
TCP is incredibly reliable, if data was being corrupted, the destination-device would see that packet checksum and contents don't match, and the source would be asked to re-transmit it. Likewise, a packet is a packet, a switch does not know (or care) if its part of a word document or audio file, if its alternating/corrupting the audio, it would also be doing the same to all content on your network - the net result being, either all your content ends up corrupted, or your network will saturated by all the retransmits (An it would feel more like a 10Mb network, rather than 1Gb or even 10Gb).
I've 20 years experience managing networks, working with everything from $50 desktop switches, to $5m chassis switches for backbones, so my scepticism doesn't come from ignorance, but experience - I often call BS when I read an article claiming a $1000 network cable gives better audio than a $30 CAT6a/CAT7 cable. But I would seriously like to know, why people think it is and the science behind it? also has anybody tried to a packet-capture to see if there is a difference?
Its got _virtually nothing_ to do with packets the protocols and checksums; its got _everything_ to do with noise on the ground plane of the *analog voltages* of digital signals causing timing errors, clock phase noise and increased jitter.
@@stephenscharf6293 So this is the classic problem of noisy household PSUs, or something different?
@@petekavanagh9186 Effectively and to a large part, yes. To a large part, it's due high-source-impedance leakage current from SMPS causing increased clock phase noise and jitter. Its also due to sh*tty clocks on generic devices, e.g. consumer-grade switches, routers, NAS', fiber media convertors, etc. Han's animation shows this really well.
Hi Kelvin, I was thinking the same thing too, but was not able to confirm this in my tests. Two findings, 1) playing a wav file from a local hard disk, I get exact amplitudes and no jitter. When playing a wav file from a network mapped drive (TCP) I see samples dropped and arbitrary samples added at times. This can only be the player, a Roon player in this case. 2) When I look at the traffic between the source-device (player) and the destination device on ethernet, I do not see enough UDP traffic to account for the audio stream. See my more detailed description in the most recent posts on this video.
@@bitrate3691 Thanks Bit Rate - I am sceptic, but not close minded on it, no smoke without fire, as the saying goes.
Brilliant, Hans. I was aware of most of the details here, but the way you put it all in the same context made it come up a couple of lights for me. This was helpful.Thanks
Glad you liked it!
Very interesting, as always many thanks
Glad you enjoyed it
I was a doubter but now I believe. thank you
👍🏼👍🏼
My initial tests of comparing digital amplitudes of songs show that when playing a song hosted on a Roon server and received on another LAN device through a Linksys router via ethernet, and no other traffic, that some samples are dropped (amp = 0). It was a very small amount but you can hear it (a couple of skips in a 43 second song). All samples that did go through came across with the exact amplitude values but did experience a small amount if jitter (small time shift). I also did the test with Wifi on a network with many devices on it and in addition to many dropped samples I saw non zero amplitudes overriding actual samples values with about 10 samples out of 2000 being just plain wrong. Since I am using Roon, do you think this can be attributed to UDP which doesn't do error checking? Tomorrow I will run a test with a player using a network storage (CIFS) via TCP with error checking. If there are no dropped samples or incorrect amplitudes then would you consider UDP as an explanation for why networks have a 'sound'?
Very interesting. Keep at it please.
For the first test I had my Roon Core running on Windows (where I capture the samples) playing a file on the local hard disk, it was perfect, no dropped samples. I then put the same file on a mapped network drive (Synology NAS) and ran many tests. I consistently got 2-3 gaps (dropped samples all zero amplitudes) per play. In one of the tests I got a small section of non-zero amplitudes (~10 samples) causing a slight shift in time. My test song is only 43 seconds long.
When I look at the samples when played from the Roon Server over Ethernet, I see a lot more gaps and many 10-sample-long regions with incorrect amplitudes. My tests reading the file from a network drive were not what I was expecting so I don't think this is noise caused by the non-error checking UDP protocol. Actually, putting a network sniffer on the traffic from the Roon server showed there was very small amount of UDP traffic between to/from my Windows machine, not nearly enough to account for the 8MB song stream. My theory is the Roon player is not handling the Ethernet input fast enough, whether from a Roon server or a file served over CIFS. I do not think this noise is coming from the network.
Given these findings and your testimonial to the improvement with the high-end network switch leads me to believe the Etheregen is able to get the packets to the Roon player faster (possible through improved time sync). This allows the Roon player to play the song to the sound card with no gaps or modified samples. This is just a theory as I cannot duplicate your setup exactly and capture what you are hearing.
One way to fix the problem I see with the Roon player is for it to buffer enough samples so that it is not replacing samples or dropping them all together while it is reading from a slower source. Perhaps there is a setting for this, but I couldn't find it. Audirvana seems to have no problem with even slower WAN streams, so I am sure it can be done.
Those suckers do sound - a little. A friend and I recently had two Zyxel GS108 and the Bonn - which is actually a modded Zyxel - for a test session. We modded one of the Zyxel step by step with capacitors (POSCAPs) mostly. We agreed that our modded Zyx did not fall behind the Bonn for less money. My older modded Zyxel GS108 again sounded a little different again. Zyxel puts into the switches whatever they get from OEMs, it seems. The name and case looks the same, the boards are totally different. I highly doubt that anything has to do with data errors. That does make no sense at all, because the IP protocol does not let anything go wrong and the devices have buffers for a number of seconds. My dCS Network bridge keeps on playing for 5 seconds or more after I unplug the switch. Meaning: The switch cannot have any influence at this moment. There are no data errors nor jitter induced by the switch. It is all about high frequency noise travelling across the cables. That's my theory.
I can tel you the UpTone Audio EtherREGEN does even more, a lot more.
Hans, if you stream data from a NAS to a streamer you use a protocol. SMB as an example supports multiple parallel streams and encryption. So in the digital domain the 0s and 1s and their jitter is compensated by the protocol. The parallel TCP streams needs to be reconstructed in a buffer as the data requires decryption. So it heavily relies on the capability of the receiving node, in this case the streamer. The streamer will reclock the digital information from it’s buffer to the analog output. So, why can an audio switch still result in a better sound quality? The only thing I can imagine is making the difference in what noise can enter into your system but not in the 0s and 1s themselves becasue these are handled by source, the protocol and the network target.
You might want to wats my video "Why digital circuits influence the sound quality' at ua-cam.com/video/B-StTplQZys/v-deo.html. We might agree then.
In economics scientists turned inquisitors during the eighties and ninieties when big data analyses showed their trusted theorie unsufficient, but they kept their trust in their tried and tested. Their teachers werd the best, they could't be wrong. Right? Additional input from psychologists and biologists gave economics the broader view that restored relevance. Neuroscience, chemistry and even psychology are now filtering through in audio. So inquisitors, turn inquisitive. If you want to describe what can be heard, be VERY sure you understand how hearing works. Hans does, and he is inquisitive.
Hans, spot on. Like always. Chapeau.
🙏🏻
It's interesting to contrast this with Audio science's view ( and data) on this and other related topics.
You need to be careful relying on purely engineering driven knuckle headed group running the audio science review forum. I’m an engineer but strongly believe in the impact of SQ by audio equipment, including network switches. The ASR forum is polluted with measurements only loyalists that will wage war on anyone that disagrees with them or voices an opinion. What they fail to understand is that measuring instruments including their APxxx they use to measure audio equipment are made by humans who also have ears and should be able to discern differences in SQ.
What he doesn't understand (or talk about) is the parallel effect in the complete live chain where especialy ground is a big issue.
It's not only about the 1's and 0´s but everything attached to the switch and back to the power line that "can" impact the sound quality. I say "can" because it all depends on the quality of the rest of the system. (everything is build for a price remember that) Even with fiber you can't solve all of these problems because there is always a connection to the power line as well and converting the signal brings hf noise. It starts with the switch mode power supplies and like any other device (audio or video related) can be improved with better filtering to minimize the parallel influence.
In general you can improve any signal because voltages are still analog.
Do they even listen to the equipment they trash or praise over there? I installed a $40 switch, this one, NETGEAR 8-Port Gigabit Ethernet Plus Switch (GS108Ev3) Even my 80 yo mother immediately noticed the betterment of both the SQ and PQ of my system. I said nothing to her about doing anything to my setup, but she was still impressed. Quite frankly, the difference this inexpensive switch made is still blowing me away. This is the same switch used in many $500+ "audiophile" switches.
SHORT AND SWEET:
Thank you for what you do.
Learned plenty from you.
Very welcome
L'histoire se repete, indeed! Reminds me of the never ending battle against pops, clicks, skips, wow and flutter during the Age of Vinyl!
Well, clicks and pops were not disputed though...
At least the problems of vinyl and tape were *real* problems.
Maybe do a blind test and document it, see where that gets you.
See my video on double blind testing
Dear Mr. Beekhuyzen.
this has nearly reached the point of a which hunt. and i would like to excuse for all the wanne be network specialist that are not able to understand the described problem. They are mixing different layers of the OSI Model to argue, but do not realise that you are mostly talking about the first layer wich is physical transport.
there are more problems in ethernet data transmission that can cause signal distortion in the physical transport besides jitter introduced by hardware. for example the problem of "cross talking" or "echo effects"
there a special measuring tools to test these problems in a network.
And as you correctly explained, this is not a problem if the signal stil is interpret at a given time as the correct value (1 or 0) if we talk purely about digital information transport. but we are not. and i think you made it very clear that DAC is looking at the exact voltage level of a signal at an exact time point to interprete its value as an analog representation wich than is absolutely prone to distortion in voltage changes. (did i get that right?)
Even if i never heard the sound distortion of audio transfered over network. i would not argue that it can't be heard by others. i worked with high class musicians and i know my hearing ability sucks in comparison to theirs (and probably yours too).
But i am glad you try to explain it in detail so that others can understand something that they couldn't before.
so thank you.
Thank you sir.
Apologies if this has already been answered but I couldn’t see it anywhere.
If audiophile switches sound better than normal switches because of attention paid to electrical noise etc, and galvanic isolation of data and clock lines help the sound, why not just use WiFi instead of an expensive switch?
I’d be interested in your opinion, thanks 👍
I'm looking into that right now.
The Hans Beekhuyzen Channel Thanks for the reply; I look forward to your findings
@@quicky82 great question. Or just the optical fibre option that costs a lot less too.
Just my two cents worth on an old topic,when streaming with a wiim pro plus to my audiolb 8200 cdq dac with WiFi there was an obvious sound improvement when changing from WiFi to ethernet cable which is why I am considering the purchase of an etheregen switch between modem and wiim pro plus .The wiim pro plus is the budget component in my system as rest of system is pretty hi end and quite neutral/revealing and highlights pros/cons of individual component changes.If a roughly $1000 aud switch can make an easily audible difference in a 30 k system I believe that is money well spent!
Like everything on the internet, you will find an opinion and it’s opposite. As you say the only way is use your own ears and form your own opinion. I own a Chord Dave DAC and I can definitely hear a difference using different digital inputs. E.g. if I stream music over USB or optical from my PC or from my recently acquired Lumin U1 Mini the difference is clearly audible in favour of the latter. Bass is deeper and more defined, piano notes suddenly become richer (small variations in a short time, I don’t know how to say this) and in general there seems to be more “air” between instruments. It’s not a small difference at all! .As such I decided to order an Etherregen to feed my Lumin and in principle I am open to accept I might hear a difference even though I can’t technically explain why yet. Thanks Hans, as usual you are very serious and respectful also with people who conversely claim to hold the only truth...
Thanks. Enjoy the music
@@TheHansBeekhuyzenChannel The EtherREGEN arrived and boy oh boy am I enjoying my music! I don't understand how all of this works... "And yet it moves".
Congrets
Hello Hans: Does the quality ot the ethernet cord ( ca. 2M) matter from the Router to the Regen? Shure the quality of the ethernet cord from the Regen to the streamer ist very imortant. for this connection i have Audioquest ethernet dimond. But from the Router to the Regen is also important to have a hi End ethernet cord? I guess the Regen taks the signal with a clock etc.
The EtherREGEN will clean up the signal so as long as you use a normal quality patch cable and you're not close to a really strong HF source, like a TV station transmitter you're fine.
Fantastic Hans! Well done and smart. Thank you!
Thank you too!
Ok i have no idea in digital transport but i was skeptical on rca wires for a long time and just used $3 to $5 rca cables for as long as i can remember. With watching ps audio, this channel, and steve, i tried a cable thats well received (qed profile) with a modest price. I was shocked how well it improved in detail, returned to my old cable and the difference is really there. I had the option to return it but i got one more set for my rca cable. Needless to say, im a convert.
Now with this topic of digital noise, i will try the ifi purifier 3 and see if that cleans it up more than my idac2 can deliver. I am very skeptical about this purchase and dont have any hope that it will change how my music sounds. If it does, i am inclined to give this video a benefit of a doubt on digital transmission does make a difference. The illustration made in the video does make sense but im no expert. It seems to me that it presents the extremes on both sides that is why im a bit reluctant to 100% agreement on the subject.
You take a fair position. And I applaud you for that.
The Hans Beekhuyzen Channel thank you sir hans. I respect and appreciate what you and other audiophiles youtubers do for our very divided community.
How you relate jitter of the Ethernet bus to an i2s bus for example ? They are 2 different signals, from 2 different busses and 2 different clocks.
Except I'm missing something here
Perhaps. I'll get back to this in a video next month.
Hi Hans nice video again. I received my EtherREGEN this Friday and it is really a big improvement on different levels. 3d, depth, width, blacks. It is even a bit louder. Very happy. I asked some questions on their forum about adding a second EtherREGEN at the internet router. The router has my Roon ROCK directly connected and I wonder what would be the benefit of using this B-port on mij Roon server. The EtherREGENs would be interconnected by glass fiber of course. I will update if something interesting comes out of that discussion...
So the outcome was that the 2nd EtherREGEN makes no sense but if I wanted to further isolate by using glass fiber the Sonore opticalModule would be a good option at the internet router side. I went along and ordered and install that. Outcome is good but the difference is smaller than the switch to EtherREGEN I did previously. My conclusion is that the EtherREGEN takes away virtually all the 'nasties' as you tend to call them already.
I just use a professional switch with SFP, works great.
@@TheHansBeekhuyzenChannel Thanks for your answer. In the mean time quite some things changed in my set-up. I have a Hugo TT2 and opticalRendu both with Farad Super3 Linear Power Supply and a switch with SFP behind that. I sold the EtherREGEN because in this configuration it does not have any effect anymore what-so-ever. (Tried Ethernet to B port to Fiber to opticalRendu -so inversed but no audible difference)
Absolutely outstanding. The best and most accurate explanation of why network switches have an audilble impact I've seen so far.
I'm working on even a better one....
@@TheHansBeekhuyzenChannel Good on 'ya, Hans! Thank you. Your recent video of the quality of DACs was really excellent, as well.
🙏
If it's not too much to ask, try a cheap switch with metal case, connect the switch case to the amplifier case, and put some ferrite cores around the network cable. (and power cables, speaker cables e.t.c.)
Just the amatour radio/ham way of dealing with disturbances.
Other (adaption of) rules apply for serious audio. A video is in the making.
I sure can hear an difference of switches to the sound.
Tried a Innuos phoenix net switch and the sound was deeper, the voice point in the soundstage becomes much more point focused. And the bass sounded more tight
Thanks for sharing!
Man, Hans, you rock.
🙏🏽
Hans, I thoroughly enjoyed this video. It explains a complex subject in a digestible manner. Although I consider myself an audio sceptic, i have also heard for myself that the tiniest details CAN matter. As an audiophile, I must investigate for myself. The one thing I cannot get behind is cable elevators. Couldn't hear a difference in my highly resolving system.
Cable elevators work on some cables - especially those that are microphonic - and not on others.
Hans, I recently purchased an ER and am trying to see if there is an improvement. I’m also using aN UPTONE LPS power supply. My DAC is a PS audio DSD Jr. My amp is an STA 200. My speakers are Tekton double impacts and I have DIY cables. My music is streamed via Qobuz/Tidal from a Mac mini running Audirvana. USB is a curious evolved cable. I am having difficulty hearing a difference. I heard clear differences when trying different USB cables, interconnects, and speaker cables. I also heard a difference with an ISO Regen that I recently sold. Any thoughts on why I’m not hearing a difference? I’ve completed about 50 hours of burn in as well. Thanks for your help.
You the Mac mini as the digital (USB) source. Over time I have learned that, although being better dan many Windows Computers, it pollutes the digital stream more than for instance a streamer/networkbridge by SOtM or Sonore. This eliminates the positive influence of the switch. You might want to watch ua-cam.com/video/G9EreotBL6c/v-deo.html.
@@TheHansBeekhuyzenChannel Thanks for the reply, I have come a long way. Can I ask one follow up question? If I got a Sonore streamer, my Mac mini no longer needs to be hooked up at all. Could I change my Roon Server to be on a much more powerful (but also electrically noisy) computer now that my audio system will be physically detached?
Yes
@@TheHansBeekhuyzenChannel Wow! from your experience, you can have a powerful and noisy computer running up conversion software without a detriment to SQ as long as the unit is no physically connected (using a streamer through USB for example)? Why doesn't everyone just use a streamer!? Thanks Hans. You are a rare gem on UA-cam and to the audiophile community!
@@TheHansBeekhuyzenChannel I have used my two routers as network switches with fibre-opitc connections from my PC music server to M1 Mac mini, now the digital stream seems to sound much cleaner and almost jitter-free. I also had a Holo Audio USB filter connect the Mac mini to my Gustard A22 DAC.
Thank you Hans for sticking your neck out. It is not because for many this remains a controversial subject that we should not talk about it and try to explain. For me the deepest digital secrets still haven’t emerged. I too was very skeptical regarding audiophile switches, but I’m glad I did at least tried it. Many don’t even bother to give it a try, but nevertheless have strong opinions. That is beyond comprehension to me. And yes just as you explained, it has nothing to do with data integrity. Even a 20€ switch will manage data integrity just fine. To me it is about not upsetting the DA converter with a dirty signal.
It doesn't feel like sticking my neck out. I just report what I hear and it is up to the viewer what he does with it. But thanks anyway.
Absolutely brilliant Hans!
🙏🏻
I use the Sonore opticalRendu instead. Perfect galvanic isolation which makes everything in the network before the opticalRendu irrelevant. No need for further audio optimization in the ethernet network. As a result, my Roon server can on my iMac in the normal home/office ethernet zone, without any specific audio optimization devices..
👍
have you tested bnc cables on clocks?
No, I don't do cable testing.
Hi Hans, My Aurender A10 (Dac, Streamer) has built in Hard drive and no WiFi. When I A/B test with and without network cable and I don’t hear any difference, then I can not expect any improvement with an Audiophile network switch, right?
That seems to make sense.
I suspect that if there is a real (and therefore measurable) improvement in the sound you hear after changing the network switch then the 'audiophile' switch is compensating for deficiencies in the box you describe as a 'DAC'. This doesn't mean the network switch is better, just that the 'DAC' is not well designed - specifically the firmware running on the microcontroller, or limitations imposed by the wrong choice of microcontroller.
I can modestly confirm recommendations made by Hans.
Thanks to him I bought SOtM SMS200Ultra/SPS500 and the sound improvement was already beyond my imagination (I had played Tidal using Chord 2qute DAC/ULPS Tomanek + Cardas Clear USB cable).
Curious, I took the next step and bought the Silent Angel switch + some not expensive LAN cables for my NUC (Roon Rock), Sotm, router etc. Just one combination of these cables worked really well. Supra CAT8 between router and Silent Angel, and Chord C-stream to SMS200Ultra, and Cat7 cable (that comes with Silent Angel) to NUC. This tonal balance and low frequency detail just made the magic for me. Two Supra cables in the chain raised low frequency energy too much. The rest of my system is Densen DM20/30 paired with Dynaudio Contour 1.3SE. Power cables everywhere...
👍🏼
Wouldn't be the right conclusion to avoid switches and anything that follows and just put the music files u want to play onto a ram drive (either virtual or physical). I also don't understand why advocate USB reclockers instead of direct I2S output
You might want to watch "An introduction to playing high quality music files": ua-cam.com/video/G9EreotBL6c/v-deo.html
Very nice video. I'm new to the channel and in hi-fi setup. I currently use a Synology NAS DLNA Server which holds my Digital Audio Library and is then connected to a LAN Switch via Cat 6 ethernet. My Marantz AVReceiver NR1711 is also connected to the same LAN Switch via Cat6 ethernet cable. I then use an app HEOS from my Phone (connected via Wifi to the same router where the LAN Switch is connected) to fetch the digital audio from the NAS to the AVR. Should I consider the sound I'm hearing as "Altered" or "Dirtier" as compared when I do DAS (as I also watched and learned earlier from one of your older videos). Your experience and thoughts on this will be mostly appreciated. Keep safe and thank you in advance.
If all is well, there should be no difference in sound quality. Audio needs hardly any bandwidth. Using a DAS is simpler and it's easier to solve problems when they occur. But a NAS can work just as good.
We appreciate your work and devotion, Hans. Your standards are high. Not accepting audible differences by digital specialists is like medieval astronomists continuing to argue the earth is flat. Based on knowledge that will be proven wrong afterwards.
👍🏼
Can we hear the equivalent of multiple hurricanes within the quantum existence of zillions of subatomic particles - YES.
That might be only slightly exaggerated 😂
i dont have the inclination to test my modem/router, amplitudes, samples, etc. I have reasonable (i think) expectations on sound quality. The question I have is the same one I always have with any piece of hardware, car, bed, etc: how much better is it and how much does that "better" cost? Chasing incremental, possible, improvements that costs thousands of dollars seems, idk, like a sort of mania. Ive reached the point in my system that Im comfortable with the sound and can sit back and enjoy the music without fomo. I certainly wish that for all of you and hope you dont have to spend a college education to get there.
Good for you.
@@TheHansBeekhuyzenChannel I know right! I live a blessed life
You know once a degree qualified mechanical engineer would say all a turntable needs to do is turn at 33 1/3 rpm. We now know for a fact there is a lot more to it than that. I really don’t care what a “network” expert tells me. All they are saying is they don’t understand. I now have the EtherREGEN in my system feeding a very high end steamer. I’m not deaf and I don’t have money to burn. The difference is overwhelming. Who the hell are these “experts” to tell me it’s not better based on absolutely zero first hand experience! Well done Hans, you have made the enjoyment of music I love much better. Thank you and UpTone Audio.
Thanks for sharing. But we have to live with the fact that there always be people that don't believe it and thus will never try.
Once again, it appears that there is no definitive explanation as to why, in certain cases, a difference is audible. Why not ask the manufacturers of "audiophile network switches" to explain their technical design decisions in order to close the debate in a rational manner?
As if they are going to tell you their developments…
@@TheHansBeekhuyzenChannelPerhaps it would be interesting for the industry to provide some arguments, rather than perpetuating a reputation as snake oil vendors
Perhaps those who call it snake oil should listen to it in a good hifi shop in stead of injustice calling names
lovely stuff
🙏🏻
⭐⭐⭐⭐⭐👍👍👍👍👍
🙏🏻