[Pharmaboy]

There's a longstanding pattern, on Head-Fi and other sites, where discussions of "which digital input works best" degenerate into quasi-physics & electronics debates. Usually somebody asserts some variation on, "Bits are bits. No sound differences are possible," then it's off to the races.

I prefer to talk about the admittedly subjective experience of how different digital inputs sound to me. Here are my experiences:

• After running computer speakers off the output of PC sound cards for ~10 years and putting up with indifferent, noisy sound, I finally get my 1st standalone DAC about 10 years ago (Stello DA100). It was a big step up in overall sound, though far short of the quality I'd heard in the past w/vinyl and/or tape (analog) sources.
• Somewhat unusually for the that time, Stello admitted that the USB input (and associated circuitry) were not optimal for best sound. They advised users to use the toslink/optical or digital coax inputs for best sound. That prompted me to buy my 1st S/PDIF convertor, the 24/96 Musical Fidelity V-Link. And damned if Stello wasn't right--toslink/digital definitely sounded better than USB (I compared them at some length). The differences were subtle, but all in the direction I wanted to go (ie, less glare, less obviously "digital" repro of upper midrange & treble, and a bit more realism in bass).
• Eventually I replaced the Stello with a nice sounding, inexpensive unit from Peachtree Audio (DAC iTx). It's USB input was vastly better than the Stello's, but again, toslink/optical sounded slightly better.
  
• I started reading that coax sounded even better than optical, at least in some cases. So I got an inexpensive BJ coax cable and checked. They were right--coax did sound better to me. Slight gains in warmth, dimensional quality/soundstaging, etc.
• ~2 years ago I got my first pair of headphones (Fidelio X2's), and was able to hear these small input differences even more clearly by HP.
• 1 year ago I got my first R2 R (non delta-sigma) DAC, the Audio GD NOS 19, and that was a sonic revelation to me...easily the least "digital" digital sound I ever heard. This 2016 unit had the latest/greated Audio GD USB input card, so of course I compared USB to toslilnk/optical and coax. Again, coax sounded somewhat better.
• About 9 months ago I upgraded the V-Link to the 24/192 version, which only outputs coax, no optical. Around that time I got a (for me) pricey coax cable that some in Head-Fi strongly recommended, the Oyaide R510. Wow--another revelation. The already good coax sound took a distinct jump in quality.
• Since then I got the Audio GD DAC-19, their oversampling-capable R2 R design. It has an even newer USB input implementation. Again I tested USB vs coax inputs' sound, and it was no contest.

I won't bore you with microscopic audiophile details & adjectives about the sound of coax vs USB. It really comes down to a slightly improved quality of relaxation, unforced sound, organic sound (things these 2 Audio GD DACs do in spades to begin with).

True, I'm limited to 24/96 by using the S/PDIF converter, but that doesn't matter to me (at least, not yet).

So in conclusion, I have extablished to my own satisfaction, empirically, that coax out of an S/PDIF converter sounds better than USB (this has held true across 4 different DACs).

That's enough for me...coax is all I use now.

 

[taxman2]

Well, if you are really up to SQ best way is to get a DDC (direct digital converter) and convert from usb or spdif to I2S. Believe me, there is no better sound.

 

[Pharmaboy]

Well, if you are really up to SQ best way is to get a DDC (direct digital converter) and convert from usb or spdif to I2S. Believe me, there is no better sound.

I'm not familiar with "DDC." Can you explain? Are there any DDC products being used in desktop audio?

For the last year or so, I planned to get a bigger/badder S/PDIF converter (Singxer SU-1)...but purchase of other amps/headphones came before this. Also, I've seen rumors that Singxer will soon come out with a new version of the SU-1 that incorporates a more robust power supply (one of the most common "mods" to the SU-1 was to wire it into a big power supply).

However, I don't think the Singxer box is this DDC you mention...

 

[winders]

DDC = Digital to Digital Converter

The Singxer SU-1 and Singxer F-1 are examples of DDCs. They convert one digital format to another.

 

[taxman2]

That's exactly what I am talking about. And yes, I've removed the internal power board and connected it directly to LPS.

 

[castleofargh]

Well, if you are really up to SQ best way is to get a DDC (direct digital converter) and convert from usb or spdif to I2S. Believe me, there is no better sound.

"believe me", the least convincing way to make a technical statement since

^_^



those who care about a preferred sound should obviously try themselves and make their own subjective decision for the specific playback system they own. while those who care about objective fidelity should also try themselves and measure the output of the DAC. pretending to be all about fidelity while never measuring anything is kind of a paradox IMO.
and of course for those who don't want to bother with testing and purchasing redundant stuff(very understandable), shouldn't it be obvious that if you guys don't care enough to test, then it wasn't all that important in the first place?

each interconnect protocol is focused on specific pros&cons, which makes switching from one to another a great option to try and minimize a specific problem. like going optical to get rid of ground loop. but each solution also has specific weaknesses and optical ends up being kind of the less preferred solution when we have no ground loop or local RF interference. so IMO use what's most convenient, and if some specific problem kicks in, then will be the time to try other options.
I understand that people are paranoid about noise, jitter and whatever, it's in our nature and marketing does everything to keep us focused on small stuff as significant improvement have become something rare. but if I was really paranoid, I wouldn't leave it to some dude I don't know on the internet. so back to trying ourselves. ^_^

 

[Roseval]

Archimago did some measurements with the Oppo Sonica, comparing USB, Toslink, Coax and DLNA
Makes one wonder if there is any difference at all in a modern well designed DAC
See VI : http://archimago.blogspot.nl/2017/06/measurements-oppo-sonica-dac-ess-sabre.html#more

 

[taxman2]

Thank's for showing it. There might be a subtle difference but USB bested them all.

 

[triodesteve]

Archimago did some measurements with the Oppo Sonica, comparing USB, Toslink, Coax and DLNA
Makes one wonder if there is any difference at all in a modern well designed DAC
See VI : http://archimago.blogspot.nl/2017/06/measurements-oppo-sonica-dac-ess-sabre.html#more

Years ago when usb was relatively new, Gordon Rankin was trying to understand why it sounded so different (and not good). He realized that he didn't own an instrument
that could measure well enough to see the problems.
As often seems to be the case people measure for things that aren't important (or not known to be important yet). Think distortion in the 70's. How did that turn out?

Just listen...your ears all tell you which is best in your system.

 

[bfreedma]

Years ago when usb was relatively new, Gordon Rankin was trying to understand why it sounded so different (and not good). He realized that he didn't own an instrument
that could measure well enough to see the problems.
As often seems to be the case people measure for things that aren't important (or not known to be important yet). Think distortion in the 70's. How did that turn out?

Just listen...your ears all tell you which is best in your system.

What problems do you believe to exist in interfaces that you don't believe we have instruments capable of measuring? Particularly those that fall into the range of human audibility?

The article is well thought out and specific. Simply throwing out a general comment about lack measurement capabilities without specifying what they are is patently unfair to Archimego and his hard work.

 

[winders]

Archimago did some measurements with the Oppo Sonica, comparing USB, Toslink, Coax and DLNA
Makes one wonder if there is any difference at all in a modern well designed DAC
See VI : http://archimago.blogspot.nl/2017/06/measurements-oppo-sonica-dac-ess-sabre.html#more

That's one device.....most DAC manufacturers will tell you that you get better sound using S/PDIF, AES/EBU, or I2S....

 

[triodesteve]

What problems do you believe to exist in interfaces that you don't believe we have instruments capable of measuring? Particularly those that fall into the range of human audibility?

The article is well thought out and specific. Simply throwing out a general comment about lack measurement capabilities without specifying what they are is patently unfair to Archimego and his hard work.

How would I know, I'm not an engineer. I'm a winemaker. I also don't pay much attention to wine reviews or wine judging. You know who can be a wine reviewer? Anyone with a keyboard.

It reminds me of Navi-guy. So many folks on head fi treated him like a god. Good for him for designing something affordable but its not like the thing was a dragon slayer. But boy did I get fried on reddit for saying my 12 year old dac was much better ( it should be...it was many times the most). My intent was not to slam the person doing the measuring. It was to say to everyone "Why are you taking his work at face value"?

I don't understand why people don't trust their own hearing. My other dac is a Chord Mojo. I listened to all three inputs in the context of MY system. Why did I not like coax? I have no clue...but it was pretty easy to hear.

 

[bfreedma]

How would I know, I'm not an engineer. I'm a winemaker. I also don't pay much attention to wine reviews or wine judging. You know who can be a wine reviewer? Anyone with a keyboard.

It reminds me of Navi-guy. So many folks on head fi treated him like a god. Good for him for designing something affordable but its not like the thing was a dragon slayer. But boy did I get fried on reddit for saying my 12 year old dac was much better ( it should be...it was many times the most). My intent was not to slam the person doing the measuring. It was to say to everyone "Why are you taking his work at face value"?

I don't understand why people don't trust their own hearing. My other dac is a Chord Mojo. I listened to all three inputs in the context of MY system. Why did I not like coax? I have no clue...but it was pretty easy to hear.

If you don't understand why people can trust their hearing for personal preference but not in any way to draw technical conclusions, read about psychoacoustics and expectation bias. There isn't a human living or ever has lived who isn't influenced by them.

When you say you don't understand what Archimego posted and that you don't need to understand the science to draw a conclusion, all I hear is "my opinion outweighs someone else's knowledgeable and detailed research". Particularly with the assumption that your "by ear" testing had no controls such as level matching, let alone proper isolation of the tester from the source being used.

I always find it interesting how many people feel comfortable saying that regarding audio, but would never say the same about, for example, the medical profession. How often do you hear something like "my blood test showed high cholesterol but since, in my opinion, I don't have high cholesterol , no need to listen to the expert" or "our current cholesterol testing method may not be perfect, so I'll ignore it until everyone who eats fast food says it is".....

 

[triodesteve]

A) You are assuming he's an expert.
B) Way more research done in medicine than in audio. But as in both, many dead ends and things we don't understand. If measurements were all that, CD's would have really been perfect.

I actually mentioned an expert in my first response. Gordon Rankin developed the usb protocol that almost everyone now uses. He found a problem, researched it, wrote the code to remedy it. He has the chops to back up his findings. When he talks, I listen. Too many designers just read tech sheets and slap things together.

Anyway there is plenty of room in the world for both of us. And neither will die from bad measurements like we would from ignoring cholesterol screenings.
PS: My doctor announced at my last visit that Brietbart was the only news worth following....so now I doubt his ability to understand my cholesterol numbers!

 

[castleofargh]

again a clear question will usually solve that issue.
- what do I prefer? it's a question about my subjective preference, like asking which singer or food I prefer. I won't start measuring stuff to find an answer to that.
- which output has the highest fidelity? this is a question about objective fidelity, you answer it with an objective method, not by asking how someone feels about it.

it's that simple.

That's one device.....most DAC manufacturers will tell you that you get better sound using S/PDIF, AES/EBU, or I2S....

if a manufacturer has an advice about a device, it seems like a good idea to listen. he might not know everything, but let's hope he knows at least a little about the stuff he built ^_^.