USB for a DAC??? Never. I guess you realize that dogmatic statements like this without any context and nothing to back them up are basically worthless and add nothing to the conversation? Noise in the signal will mask any resolution below the noise level. what noise in what signal ? Are you claiming noise in the digital signal somehow makes it thru to the analog output or something else? If you are, isolating the incoming noise from the USB line from the rest of the system is pretty well understood. |
Most USB audio/video devices use the bulk transport because real-time delivery of the data is not necessary.
This simply is not true if "most" includes the topic at hand... USB DACs? As a programmer and electronic interface design engineer, you're not "locked" into using isochronous mode: You have bandwidth and buffers so if i can transfer 500 MBps+ (using SATA SSD = 4 Gbps+) of data without error (well, if there are errors, they are corrected), it should't be a problem to playback music! It doesn't matter what could be done. It matters what is done, and since asynchronous is what is done, errors are not corrected. End of story. |
@antigrunge2
Since you like the Antelope DAC: may I suggest you try it on S/PDIF or AES/EBU? fair enough, but how do you generate the signals? There are a few ways I know of, Ethernet to spdif with a - Raspberry Pi HifiBerry seems like a step back, maybe not
- DCS network bridge, $3500 , more than I paid for the Antelope
- $20K+ Aurender W20, not an option for me
Of course there are USB to spdif but we are trying to eliminate USB. suggestions ?? |
Since Rednet is a proprietary Focusrite interface that limits you to using Dante type pro audio interfaces that no high end DAC uses, doesn't that kind of make it so limited in scope that it really doesn't pertain to a discussion on an audiophile forum? AND it really is intended for a different purpose especially longer runs that USB can't deal with, handling many more channels in a recording studio, , as well as uses where latency is important, not home audio. Also, since there are very few devices that give you a choice of Dante or USB like a Lynx Hilo does, how can you prefer Dante when you can't compare the 2?
That said, I'm all in on pro audio interfaces. You can see the heart of my system is an Antelope Eclipse 384, but it uses USB and sounds wonderful. |
Thanks for the info Furzy, looks interesting, I may try it to compare to USB as the Antelope has spdif and optical inputs. However, even though I doubt it would be significant, those inputs are limited to 192 and 96K respectively whereas I can feed it up to 384k upsampled in Roon from USB.
On the other hand, a Raspberry Pi with digital card will do the ethernet to spdif for about $100
wonder if Roon would see the Rednet as an audio device like it does a Pi? |
Sorry, back to the difference in sending data over USB like we do with file transfers and audio uses.. to further my point above , despite what the angry poster says, .Audio and file transfers are different as file transfer protocols have error correction from Mojo Audio
The combination of asynchronous clocking and data buffering can remove jitter caused by packets of data arriving at irregular intervals, but it can’t correct corrupted data. Though asynchronous USB results in more liquid, more resolving, and more musical sound, if it isn't bidirectional, it has no error correction, and can not assure uncorrupted bit perfect data.
|
USB is a simple serial digital transport with checksumming and buffering! Transmitting audio data in realtime is not the same as transferring TeraBytes of data files. You should look into it before making such statements. |
@herman "Transmitting audio data in realtime is not the same as transferring TeraBytes of data files. You should look into it before making such statements."
Ok go ahead and explain me how it is different!! YOU should study about async digital communication, encoding and buffering... It’s obvious that you just plain don’t know.
https://darko.audio/2016/05/gordon-rankin-on-why-usb-audio-quality-varies/
I will defer to Gordon Rankin who I trust more than somebody angrily ranting on this forum
The three main USB transmission protocols are Bulk, Interrupt and Isochronous. Bulk (used for data transfer to a hard drive) and Interrupt are error-correcting. Isochronous (used for audio) is not.” “Bulk and Interrupt are immediately NAK (negative acknowledgement). The receiver is designed to detect a bad packet immediately and the packet is resent.”
“For USB audio, the receiving device is basically translating a serial stream of data with a clock interwoven throughout. At the end of the packet sits some sort of block check. If the block check does not match the data then that packet is flagged as an error.” “With Isoschronous USB transmission, packets are sent without any error correction / resending. But guess what? This is the USB protocol used for audio frames. The bad news is they are not error-free. The good news is these Isochronous frames are afforded the highest priority in the system.” |
@jaytor sed:I think what @rixthetrick was getting at is that any signal on a cable is inherently analog How’s that? With all due respect, do you actually understand what makes analog signals analog and how digital signals are different. ABout the only similarity is that they can both use wire as a transport media. If a car operates on roads and a elephant walks down the same road, does that make the elephant a car? It depends on how you look at it. "The only similarity is that they can both use wire" isn’t quite right. Digital is by definition binary.... high/low, 1/0, true/false, whatever. Electrically, those 2 states can be represented by 2 different frequencies, 2 phases, etc. . With USB those 2 digital states are 2 voltage levels. Analog information is represented by constantly changing voltage levels with an infinite number of possibilities. So yes, they are the same in that both analog and digital information are represented by voltage levels. Since the digital voltage level can’t change instantaneously (a perfect square wave) it bears some resemblance to an analog signal’s infinite voltage levels. However, does any of that really matter? With analog, errors are a given. It is impossible to maintain 100% accuracy throughout the analog chain. On the other hand, digital errors are very, very low and in a well designed system, for all practical purposes they are zero. In any case, it is a red herring that distracts from the topic of whether or not USB sucks BTW.... it does not |
Since the digital voltage level can’t change instantaneously (a perfect square wave) it bears some resemblance to an analog signal’s infinite voltage levels.
Not really. That the square wave isn't perfect doesn't make it an analog signal.
yes, we agree, I am only trying to make the point that the digital signal does not strictly meet the definition of 2 and only 2 distinct states since it takes some time to transition. Like I also said, it really makes no difference to the discussion of whether or not USB sucks = a red herring Analog information is represented by constantly changing voltage levels with an infinite number of possibilities. That’s a common misnomer. It could only be true if the bandwidth of the analog system were infinite, which is impossible, of course. So just like digital, analog is bandwidth limited. Its resolution is finite.
It is not a misnomer, my statement is absolutely correct. There are an infinite number of voltage levels between any 2 levels. It has nothing to do with bandwidth. |
You're arguing that analog systems have infinite resolution, which is simply false. It's a commonly held belief, though. I'm not arguing, I'm stating a fact, which has nothing to do with bandwidth or resolution. You are confusing bandwidth which you brought up earlier and resolution now and whatever else with the simple fact that an analog waveform will, at any point in time, be at any one of an infinite number of possible levels. Anything continuous like an analog waveform has an infinite number of possible points. Q: how many point are there on a line? A: an infinite number Let's say for example your DAC has a minimum output level of zero volts and a maximum of 2 V and the output is a sine wave varying between those 2 levels. If you are correct, tell me what level between and 0 and 2 is it incapable of producing? What voltage levels does it skip on its way from 0 to 2v? .0000001 ? .000001 ? .00001 ? .0001 ? .001 ? .01 ? .1 ? 1 ? 1.1 ? and so forth none of these and none of the other possible infinite number of possible levels between these |
My experience...
Cables make a huge difference for sure.
External clockers make a huge difference.
Jitter is the plague of digital audio. It can come from transports and servers. There are devices that go in the audio path to clean it up.
How do I know all this? I use my ears.
Not to discount your experience or your ears, but I have tried all manner of these devices and find they make no difference .. in my system.... to me.. I have tried among others - Uptone USB regen
- Uptone EtherRegen
- Cisco 2960 switch modified with a TCXO oscillator
- Mutec MC-3+ USB reclocker with Mutec Ref 10
- Roon Nucleus server
- Innous Zenith server
- Small Green Computer i5
- various Mac Minis (which I currently use with the the Uptone linear DC/fan control mod)
- Innous Phoenix USB reclocker
- various USB cables
Of course I can’t say for sure why. It may be that I have wooden ears, or the fact that I use a pro audio DAC that has a very high quality clock built in that is also locked to a very high quality external 10MHz reference clock. The DAC is also a re-clocker that has outputs for both a high quality word clock and a de-jittered data stream if needed. .. but at the end of the day, just like you, my ears tell me what is important to me. As for jitter from servers and transports.. Asynchronous transfer is designed to eliminate that. The DAC asks for data packets at its pace and clocks them on down the line using its internal clocks. Any jitter from the source is therefore disregarded so if you hear a difference using devices as above, in theory it is not because they reduced jitter. That said, I don’t doubt you hear something. I’m just saying that attributing it to jitter flies in the face of everything we know about how this stuff all works. I’m also pretty sure we don’t really understand how all of this stuff works, nor do we really need to if we are getting the results we want. It would be interesting to use a source to feed an asynchronous DAC where you could vary the jitter and determine if you could hear it. |
one might usefully revisit more appropriate formats (optical, I2S, AES/EBU) to improve on what is at best an unacceptably wide range of outcomes with USB; I’m pretty sure that you will get an equally wide range of outcomes using these other formats. And reading your initial post, the problems you had with getting a reliable connection are not inherently USB problems since millions of others have absolutely no issues getting a reliable connection. The fact that you finally got it to work with a particular USB cable tells me something else was amiss. The best sounding connection we can debate, but not being able to get the source to communicate with the DAC via USB is highly unusual. I also have an Antelope DAC that I am extremely happy with although I do sometimes have to reboot my computer or the DAC to establish USB communication so maybe they do work better in a pro audio situation with that gear versus audiophile servers, and I sometimes have to toggle the internal word clock from lower rates to the maximum rate to get sound output, but a minor inconvenience for the wonderful sound at a reasonable cost. |
I just dropped my cartridge needle down onto my vinyl record or put my CD into my CD player and hit play and I get great music and nothing to worry about hey grandpa.. check the date on your flip phone, it is 2020 I have discovered such a tremendous amount of new to me music and new artists since going with Roon, Tidal, and Qobuz that I can’t imagine being locked into the extremely limited number of choices my personal collection allows, and that is in the thousands, I hear amazing things every day that I never knew existed. For less than the equivalent of what I paid for a single CD a month in 1985 I have access to literally millions of songs every day in CD or better quality... MILLIONS !!!! when I read about an album in a magazine or the web, chances are it is instantly available to me. When I hear an artist I like chances are I have instant access to their entire catalog along with recommendations for those that are similar. It is the golden age for music lovers that are willing to "worry about" how to connect their server to their DAC. I feel sorry for you if you are limited to physical media listening to the same stuff over and over and over. |
I have a ZEN Mk3 and an Aqua DAC, and the difference between USB cables is significant to say the least. I am not questioning what you hear. However, it is almost impossible to believe that an asynchronous USB DAC can't be designed to be immune from any nastiness coming in on the cable. The technology is too mature to accept that a multi-thousand $$ DAC can't effectively handle a USB input without expensive cables. but in real life buffering creates latency, Latency is completely a non issue in home audio. It matters not if the data stream starts now or a second from now. All that matters is proper timing once it starts. a clock is highly susceptible to any noise that is mixed with the signal ( and there is a lot of such noise created by the motherboard, processor, EMI etc), and this is just the beginning. a cable on the input will not have any effect on any noise generated inside the DAC. Any competent designer can filter out any noise coming in on the cable isolating the rest of the DAC from the noise. The idea that the USB cable can be an effective filter and the DAC input can't isn't reasonable. Again, I don't question what you hear, I question the design of a DAC that is affected by USB cables. |
This is not true, even Nuno Vitorino, the boss and designer of Innuos ZEN admits that if the transport is noisy, it is very difficult and sometimes impossible to filter the noise that is mixed with the audio signal. A testimonial from Innouos about the necessity of high end servers is like a testimonial from Elon Musk about electric vehicles. I agree once you mix noise into the audio it may be impossible to eliminate, but there are no audio signals in an Innouos Zen so what is he "admitting". As far as I can see they sell only servers.. no DACS, If I was selling expensive servers, I would also argue that DACs are incapable of filtering out noise so you must buy my server to get rid of it before the signal gets there. I have tried a pretty good variety of devices and cables listed previously to feed my DAC including a Roon Nucleus on a linear supply and Innouos Zenith MK3 with the Phoenix USB clocker as well as cables compared to my Mac Mini and $50 USB cable. I hear nothing "significantly" different or "vastly superior" about them. As I said earlier, it may be that I have wooden ears or my pro audio interface with reference, internal, re-clocking and jitter reduction handles it better than most. You can check my system and see it should be able to resolve any differences. In any case I feel very fortunate. I’ve done the experiment so I can sit back and listen to wonderful music without fretting about what I need to do about USB cables and servers and clockers , etc. A recent extensive review on Audiophile Style lauds the superiority of a $26K server that to be optimized needs expensive power conditioning, ethernet re-clockers and ethernet cables with expensive USB cables ....... Ethernet in - USB out does not require $30K+ to optimize. I stand by my position that if your DAC is significantly and vastly improved by a server and/or cable you have a DAC problem. I also understand the fun in trying it all and the joy (too strong?) of owning them. Have fun And a multi-thousand $$ DAC simply can not employ heavy filtering, simply because it kills the sound: adds modulations etc. I was very specific about where the filters are... I said Any competent designer can filter out any noise coming in on the cable isolating the rest of the DAC from the noise
so I’m talking about filtering and noise reduction of the incoming digital data stream, not the audio. Cleaning up the data stream will not kill the sound or all of these expensive servers and clockers would kill it too. Filtering noise from the incoming USB digital signal will not modulate anything. |
