Did We Overhype USB Audio And Overlook Possible Pitfalls?

[lmitche]

There is an argument that great USB audio is still in its infancy and out of reach for the average consumer. You can spend $5k and do it yourself, but you need to do an awful lot of research to learn what to do, and have a decent background in digital hardware and software. It's still an enthusiasts game. To get the same performance in a turnkey system may cost $10k to $20k all in.

[esimms86]

I'm surprised there aren't more posts from AES users. The pros loves AES, as do many people with Genelecs at home. I have a Metrum Pavane that sounds phenomenal via AES though my Phasure NOS1A outdoes it with USB with or without the Intonation in play. The Pavane review posted on The Ear website led me to try out AES.

[gmgraves]

Hi George,

earlier in this thread you said

 

 

 

I may be misinterpreting this, but it sounds like you are saying that audio drivers are not necessary for superspeed. You specifically mention class drivers which I interpreted as meaning generic audio class drivers. If this is what you meant then what is superspeed using other than audio class drivers?

 

John S.

 

 

Hi John:

 

You didn't misinterpret. I misspoke. I meant the Isochronous firmware such as Gordon Rankin's "Wavestream" product. Don't know what made me conflate that with the drivers except perhaps the fact that I had been up all night when I wrote that. USB 3.1 has isochronous operation as part of the spec. While USB 1.0 through 2.0 use a "speak-when-spoken-to" protocol requiring something akin to "Wavestream" on order for isochronous or device initiated communications towards the host, USB 3.0 and 3.1 was designed to allow this without having to add third-party protocols. Sorry for the confusion. I'll go stand in the corner with a dunce cap on my head now....

[The Computer Audiophile]

Hi John:

 

You didn't misinterpret. I misspoke. I meant the Isochronous firmware such as Gordon Rankin's "Wavestream" product. Don't know what made me conflate that with the drivers except perhaps the fact that I had been up all night when I wrote that. USB 3.1 has isochronous operation as part of the spec. While USB 1.0 through 2.0 use a "speak-when-spoken-to" protocol requiring something akin to "Wavestream" on order for isochronous or device initiated communications towards the host, USB 3.0 and 3.1 was designed to allow this without having to add third-party protocols. Sorry for the confusion. I'll go stand in the corner with a dunce cap on my head now....

Somewhat related - I had a meeting with the USB Implementers Forum (USB-IF) at CES and they said Microsoft would be enabling native USB audio drivers for Class 2 components as soon as the USB stack is rewritten.

[davide256]

If you had asked me 6 months ago after I had just acquired the Regen, and was using with basic streamers, I would have said hell yes, USB is the way to go for digital connection solution. But now with the Aries mini, I much prefer the AQ Diamond coax for DAC connection as equal in depth of detail and simpler than asynch USB .

[One and a half]

Somewhat related - I had a meeting with the USB Implementers Forum (USB-IF) at CES and they said Microsoft would be enabling native USB audio drivers for Class 2 components as soon as the USB stack is rewritten.

 

What's the roadmap for that, or when?

[The Computer Audiophile]

What's the roadmap for that, or when?

They made it sound like it was happening in the not too distant future. Maybe six months is my guess.

[jabbr]

That said, my point that USB is a more robust interface than I2S still stands.

 

Two different purposes.

 

All DACs have input modules that convert external signals into something the DAC module itself accepts. I2S is a standard 'interboard' format that the actual DAC board itself accepts.

 

USB is an 'interbox' format. I2S over LVDS is being promoted as an interbox format.

 

 

Room treatments for headphone users

[Paul R]

Interesting you mention UDP. Most people think Ethernet means TCP/IP and all audio over Ethernet must use TCP. However, some home audio uses UDP. That said UDP is more than fine for home audio with a decent network.

 

Sure, but with UDP connections, you have to build information into the packet to provide for ordering and such. TCP handles all that at the media layers "under the covers."

 

Layering something like RTSP (Real Time Streaming Protocol) over it using RTCP (Real Time Control Protocol) and RTP i(Realtime Transport Protocol) is easier done with TCP, and is commonly tunneled over HTTP connections.

 

With the speed and capacity of today's networks, UDP is not really needed as much to save on bandwidth and connections. IMO that is.

 

-Paul

[YashN]

Yes, but he mention there was no "additional" correction. Thus, implying (after first stating) there is correction to begin with.

 

Not necessarily, it could mean there's no correction in addition to the Cyclic Redundancy Check.

 

As an aside, does anybody know why when replying, the cursor would go to the first line right after suddenly in the middle of a sentence? Happens to me a few times. Quite odd.

[livelistenlearn]

Hi Guys - The title of this thread "Did We Overhype USB Audio And Overlook Possible Pitfalls?" is meant to be read as a question, not a conclusion or suggestion.

 

⤴︎ 'loaded', though

 

anyway...

 

back during post #1 (2014), USB was the thing. had a 2011 mac, bought a dac. music sounded distinctively better than cd player > (older) dac; and was more accessible… coming from existing speakers powered by the existing amps.

 

fast-forward to 2015: REGEN was born/delivered. and, (most) CA-ers rejoiced. invested in one, liked it, told superdad so. music from system with same mac + dac + amp + speakers = better (not enough time, + lack of expertise/equipment = ‘blind faith’ = take words for whatever worth).

 

then, mac died. grieved, bitched. bought new 2015 mac. with OPTICAL 24/192 output. remembered dac-maker used optical to ‘tune’. tried OPTICAL once, using it since.

 

A to your Q: YES/NO, imo.

 

think it’s not the connection/protocol. think it’s the implementation. always been.

 

agree that vega sounds good via USB. prefer OPTICAL for hugo. using HDMI for HT renderer > HT amp.

 

as pro mac user, have invested (significantly) in THUNDERBOLT. v. fast = video/photo editing + backing up files. no distinct audible benefit using TB for storing/accessing music (one man’s system)… but v. much less incidences of dropouts using Audi+/HQP.

 

observations:

 

1. most PCs and macs come with USB ports (for now). if I2S is better, the path leads inevitably to standalone streamers/renderers. one of USB’s main sells = plug + play. sans USB, CA = less fun, less accessible, less logical.

 

2. most (popular and relatively affordable) dacs do not have I2S/HDMI connectivity (yet). if these other options are better, many may need ‘better (connected)’ dacs = maybe good for industry, maybe not so good for user bank accounts.

 

3. odd/interesting that something that is good can sometimes precede a measure of negativity. REGEN seems to make most systems' USB outputs sound better, according to many CA-ers. yet now, it is being ‘used’ to present the case against USB.

 

naturally, goal in CA is closest proximity (accessible/affordable/practical) to perfection.

 

but… balance is key. as is, how revealing/nirvanic is one’s system?

 

in a decent 10-15k hi-fi system, how much of a difference do you think an alternative to USB connection will make… as compared to (a) software, (b) file format, © DSP, (d) amp, (e) subs and (f) speakers?

 

btw, Happy New Year, estimados señoras and señores.

may all ears/hearts stay young, and one’s hearing, ever discerning.

[YashN]

A good part of the problem here is that there are a whole bunch of different protocols for delivering music over Ethernet, and most of them are quite complex, dealing with metadata, cover art etc. This situation requires a fairly decent processor and a lot of software to deal with the plethora of what might get sent to the DAC. This is NOT good for the DAC.

 

What is needed is something like the NAA protocol which as simple as it can get for just sending the music and not dealing with all the other stuff.

 

John S.

 

Totally agree, in fact, not knowing the internals of Miska's NAA protocol, ideally what I'm looking for is just transporting a bitstream from A to B at very high frequencies and then output this on a pin. I used a high-frequency mode of PWM on the Arduino to control a Control Voltage point on an analogue synth, so something similar from a computer would be quite awesome.

 

Because of the layers of the OSI stack, Ethernet deals with way more than just a bitstream in it packetised nature.

 

What would be needed is a much, much simpler protocol, with no error correction, no packets, just the ability of rapidly varying a voltage repesenting a digital '0' to a digital '1' and being able to send that over some distance.

 

At the reception pin, we already know how we'd deal with it: a good filter for DSD and we'd be smoking.

 

This would do with a much simpler interface at the sending device and at the receiving device (I think).

 

Now, one often overlooked issue (not by some of us) is all the electrical complications that arise when you're sending data over a normal cable, thereby damaging signal integrity.

 

So far, it looks like some people here have had great results with Ethernet-to-Optical and using Optical to do the transportation of data to point B. So perhaps getting from A to B can only be done well with optical but ideally, I'd like just the bitstream to be sent to the minimally effective DAC.

 

At the DAC, I only want the D/A process to occur (with buffering as necessary, so a minimally noisy memory system).

 

Are there any technical difficulties in just sending a pure bitstream from a computer to a receiving device? I've been thinking of that for a good while.

[YashN]

But now with the Aries mini, I much prefer the AQ Diamond coax for DAC connection as equal in depth of detail and simpler than asynch USB .

 

Do you know if this coax connection can be used for DSD?

[YashN]

then, mac died. grieved, bitched. bought new 2015 mac. with OPTICAL 24/192 output. remembered dac-maker used optical to ‘tune’. tried OPTICAL once, using it since.

 

Even some older iMacs have optical outputs

 

Mine is a 24" which was released in 2009 (dark ages in computer terms), and hidden within the normal mini-jacks is an optical connection.

 

For computer audio for a video playback with VLC in surround and Netflix audio, we use that Digital Out to my amp, and the soundis really, really, good: solid sound, clear and we have a wonderful soundscape in surround.

[Rexp]

Right. Back up to $4000.

No one going to chime in with a USB DAC suggestion that doesn't cost a fortune and doesn't need all the add ons?

[wgscott]

Not necessarily, it could mean there's no correction in addition to the Cyclic Redundancy Check.

 

As an aside, does anybody know why when replying, the cursor would go to the first line right after [ /QUOTE] suddenly in the middle of a sentence? Happens to me a few times. Quite odd.

 

Have you tried using a linear power supply and a regen on your USB keyboard?

[wgscott]

No one going to chime in with a USB DAC suggestion that doesn't cost a fortune and doesn't need all the add ons?

 

The Dragonfly comes to mind.

[firedog]

No one going to chime in with a USB DAC suggestion that doesn't cost a fortune and doesn't need all the add ons?

 

Haven't heard it, but guessing the Ayre Codex (under $1800) has very good USB implemtation like its' big brother. I don't know if you consider it a fortune or not. It's considered a mid priced DAC around here.

[Paul R]

No one going to chime in with a USB DAC suggestion that doesn't cost a fortune and doesn't need all the add ons?

 

The iFi iDSD Nano or Micro come to mind, though the sound better with an iFi iUSB power and a Gemini cable. Really are hard to beat. $500 for the Nano config, about $850-$900 for the Micro.

 

Dragonfly, Wavelength anything, etc.

[Norton]

I'm surprised there aren't more posts from AES users. The pros loves AES, as do many people with Genelecs at home. I have a Metrum Pavane that sounds phenomenal via AES though my Phasure NOS1A outdoes it with USB with or without the Intonation in play. The Pavane review posted on The Ear website led me to try out AES.

 

Yes from my experience I'd rate AES first, Spdif second and USB or toslink joint third in terms of SQ. The Bryston BDP into Chord DAC64 via AES, for example, was superb.

 

With both Hugo and Esoteric DACs I never use USB, as Spdif is superior. Presumably USB is used simply as it's the default interface for commodity PCs - which doesn't seem an auspicious starting point for audiophile SQ.

[vortecjr]

Yes from my experience I'd rate AES first, Spdif second and USB or toslink joint third in terms of SQ. The Bryston BDP into Chord DAC64 via AES, for example, was superb.

 

With both Hugo and Esoteric DACs I never use USB, as Spdif is superior. Presumably USB is used simply as it's the default interface for commodity PCs - which doesn't seem an auspicious starting point for audiophile SQ.

 

The Bryston BDP uses an off the shelf motherboard and it doesn't output AES/EBU on it's own. The AES/EBU signal appears to come from an XMOS based USB to AES/EBU converter. Not sure which BDP version. They also use a Julia PCI card which is now discontinued by ESI. That Julia PCI card is single ended so the AES/EBU signal is derived from SPDIF.

 

Jesus R

[livelistenlearn]

Even some older iMacs have optical outputs

 

Mine is a 24" which was released in 2009 (dark ages in computer terms), and hidden within the normal mini-jacks is an optical connection.

 

For computer audio for a video playback with VLC in surround and Netflix audio, we use that Digital Out to my amp, and the sound is really, really, good: solid sound, clear and we have a wonderful soundscape in surround.

 

cool.

 

still (occasionally) fire up a 27” 2009 core 2 duo 3gHz imac running osx lion for work on legacy fcp7 files.

 

but, only late-2013 (or newer) mbps and mac pros do 24/192 output via optical/toslink. imacs = mid-2014 onwards, minis late-2014 onwards.

 

pre-2013 apple machines were/are limited to 24/96 (or lower) via optical/toslink.

[livelistenlearn]

Yes from my experience I'd rate AES first, Spdif second and USB or toslink joint third in terms of SQ. The Bryston BDP into Chord DAC64 via AES, for example, was superb.

 

With both Hugo and Esoteric DACs I never use USB, as Spdif is superior. Presumably USB is used simply as it's the default interface for commodity PCs - which doesn't seem an auspicious starting point for audiophile SQ.

 

imo, 2014/2015 mbp-to-hugo =

(24/192) OPT > USB (REGEN) > COAX (SPDIF).

(mbp running customised audi+ filters with/without dirac plug-in).

 

have (retired) exposure dac that benefitted from BNC connection to cd player, but limited to 24/96 input.

 

have one balanced XLR cable, for use only with zoom H4N/H6 portable recorders and mics. no argument about nice ‘pro’ feel and ‘click-in’ connection + voice only recordings come across as PDG.

 

thanks for sharing re: dac64. tried dave? he’s got XLR-in on the derriere as well… but will have to wait until one enjoys some kind of spectacular windfall

[Jud]

cool.

 

still (occasionally) fire up a 27” 2009 core 2 duo 3gHz imac running osx lion for work on legacy fcp7 files.

 

but, only late-2013 (or newer) mbps and mac pros do 24/192 output via optical/toslink. imacs = mid-2014 onwards, minis late-2014 onwards.

 

pre-2013 apple machines were/are limited to 24/96 (or lower) via optical/toslink.

 

...which illustrates a reason for USB ubiquity. Output any resolution you like, including DSD, from your chosen software or a downloaded file using a Mac laptop or mini you already own. Nothing else allows you to do this with less fuss and learning than a USB connection between computer and DAC. Spending some extra money on cables or boxen between the computer and DAC is less daunting for many (most?) than spending rare leisure time figuring out optical Ethernet or SDIF-1 on purpose-built servers.

[livelistenlearn]

agreed, jud.

 

USB convenience/ubiquity is, by and large, indisputable… and USB progression (with backward compatibility) is all good, for the benefit of CAs, esp. neophytes who already have music players = PC/tab/phone (wireless connectivity notwithstanding).

 

however, we seem to be venturing into SQ… and the ‘superiority’ of alternative/optional connections.

 

btw, that ‘vintage’ imac is hooked up to 2 x 2TB FW drives