Audible difference between analog interconnects

[plissken]

I find interesting these responses to my suggestion that perhaps Benchmark has async USB input on the DAC2 to check a marketing box.

 

Not that it isn't possible Benchmark added async USB for the reasons suggested by plissken or esldude. I don't know the answer.

 

But we do have a very plain illustration of Benchmark adding a capability to its DAC simply to check a marketing box with DSD, which you can read John Siau being very critical of repeatedly. Yet plissken and Dennis (esldude) felt they ought to raise technical (thus presumably more praiseworthy) reasons other than marketing for the DAC2's use of async USB input on top of its "perfect" jitter rejection through ASRC. While I don't have ESP and thus don't know plissken's and Dennis's reasons any more than I know for sure the reason Benchmark offers async USB input, I wonder if Benchmark marketing itself in ways that objectivists would likely favor could be causing people who feel they're on the objectivist "side" to root for and defend Benchmark.

 

I don't see a little mild exaggeration and cool sounding names ("perfect" jitter rejection, "UltraLock2" for ASRC) as a bad thing, particularly in service of a product that by all accounts is excellent. High end audio equipment isn't a necessity of life (however much we might feel differently when we're enjoying it); without effective marketing, there wouldn't be an industry on which to spend our disposable income. So I'm not trying to be pejorative when I wonder whether Benchmark is doing certain things for technical or marketing purposes. What I do want to do is point out that successful marketers are well aware of our predispositions, "subjectivists" and "objectivists" alike.

 

Jud you've been starring at your belly button a bit too long.

 

Audio gear is driven by inputs and outputs. Let's flip this on it's head: Why is Company X, Y, or, Z keep including TOSLink, S/PDIF, AES-EBU on their products.

 

Nice attempt at moving the goal posts. I wasn't taking about DSD. I was talking about Benchmarks ability to prove, via measurements, that their DAC is able to 100% extract all data and produce unmolested sound out of 1000 foot of CAT5 feeding the AES-EBU input.

 

So as not to worry about 3 - 5 foot of cable.

 

Remember John Siau doesn't recommend using high costs (esoteric, fancy, boutique, what have you cables) with THEIR DACS. On the other hand may be that your DAC sucks.

 

Only in your world does including a popular option equate to marketing and not meeting consumer demand.

[plissken]

BTW Async USB isn't a jitter reduction protocol / mechanism in and of itself. At least that is how I read what you are insinuating.

[Paul R]

Jud you've been starring at your belly button a bit too long.

 

Audio gear is driven by inputs and outputs. Let's flip this on it's head: Why is Company X, Y, or, Z keep including TOSLink, S/PDIF, AES-EBU on their products.

 

Nice attempt at moving the goal posts. I wasn't taking about DSD. I was talking about Benchmarks ability to prove, via measurements, that their DAC is able to 100% extract all data and produce unmolested sound out of 1000 foot of CAT5 feeding the AES-EBU input.

 

So as not to worry about 3 - 5 foot of cable.

 

Remember John Siau doesn't recommend using high costs (esoteric, fancy, boutique, what have you cables) with THEIR DACS. On the other hand may be that your DAC sucks.

 

Only in your world does including a popular option equate to marketing and not meeting consumer demand.

 

(*sigh*)

 

Okay, let's do flip this around. What do DACs that "sound better" than the Benchmark DAC2 do that makes them sound better? There are DACs that sound better, and there are people that - with some justification - call Benchmark DAC "ear rippers."

 

Your thinking does not seem to encompass the whole issue. you seem to be saying either than the Benchmark is as good as it gets, or that there are better engineers out there building better DACs.

 

Jud is, at least, putting forth an enormous effort to see and understand the entire issue. i read what he is saying more as there are a lot of good engineers making decisions about implementation details differently, and producing better or worse sounding DACs. A very reasonable position - but it does require one to acknowledge there are other factors out there that can make a difference.

 

-Paul

[Paul R]

BTW Async USB isn't a jitter reduction protocol / mechanism in and of itself. At least that is how I read what you are insinuating.

 

Seriously? What do you think "async" USB is for then? To the best of my understanding, the whole idea of synch USB and DACs that operate with it was to reduce jitter, and do it better than the ASRC based products that were available at the time.

[plissken]

(*sigh*)

 

Okay, let's do flip this around. What do DACs that "sound better" than the Benchmark DAC2 do that makes them sound better? There are DACs that sound better, and there are people that - with some justification - call Benchmark DAC "ear rippers."

 

Do you have a quantifiable measure of what is 'Better'. What happens when you think DAC A sounds better than DAC B and someone else thinks DAC B sounds better than DAC A?

 

you seem to be saying either than the Benchmark is as good as it gets, or that there are better engineers out there building better DACs.

 

I don't seem to be saying anything. What I said is in plain, English, text for all to read.

 

What I posted is that FFT plots show no difference with the output of a well designed DAC being fed with 1000 foot of CAT5e via AES-EBU and a standard AES-EBU cable.

 

Jud is, at least, putting forth an enormous effort to see and understand the entire issue. i read what he is saying more as there are a lot of good engineers making decisions about implementation details differently, and producing better or worse sounding DACs. A very reasonable position - but it does require one to acknowledge there are other factors out there that can make a difference.

 

-Paul

 

What issue? Is there some aberration in the FFT plots by BenchMark that I'm missing?

 

Am I missing the point that while some manufacturers include USB input due to customer demand that BenchMark implements USB for marketing reasons instead?

 

Am I missing the point where someone is maintaining that Asynchronous USB is some how a jitter elimination schema instead of a no source clocking schema?

 

I acknowledge that if you are a DAC designer and USB cables make a difference then show that difference. If BenchMark can show no difference and, as some are insinuating, there are better engineering talents out there, then they should be able to, with Data, demonstrate that point.

[plissken]

Seriously? What do you think "async" USB is for then? To the best of my understanding, the whole idea of synch USB and DACs that operate with it was to reduce jitter, and do it better than the ASRC based products that were available at the time.

 

Sync= Use the computer to provide a clock

 

ASync= No clocking and use the DAC's clock

 

Async doesn't magically reduce jitter. It's about a choice in where you clock from.

 

If your supposition is correct then there would be no Asynch DAC's with Femto or other clocks. Just by virtue of Asynchronous USB you would be good enough.

[Jud]

Sync= Use the computer to provide a clock

 

ASync= No clocking and use the DAC's clock

 

Async doesn't magically reduce jitter. It's about a choice in where you clock from.

 

If your supposition is correct then there would be no Asynch DAC's with Femto or other clocks. Just by virtue of Asynchronous USB you would be good enough.

 

You really thought anyone was implying something different, clockless and "magical"?

[plissken]

Seriously? What do you think "async" USB is for then? To the best of my understanding, the whole idea of synch USB and DACs that operate with it was to reduce jitter, and do it better than the ASRC based products that were available at the time.

 

You really thought anyone was implying something different, clockless and "magical"?

 

I'm not implying anything.

 

I think people are misunderstanding why DAC manufacturers use Async vs Synchronous USB protocols.

 

Paul is stating that Asynch USB protocol has better jitter reduction mechanisms than BenchMarks in house ASRC

 

Feel free to point out my error.

[mansr]

I'm not implying anything.

 

I think people are misunderstanding why DAC manufacturers use Async vs Synchronous USB protocols.

 

Paul is stating that Asynch USB protocol has better jitter reduction mechanisms than BenchMarks in house ASRC

 

Feel free to point out my error.

 

Async USB removes at least one source of jitter (the clock recovery). How much better it performs as a result depends on the clock recovery mechanism and on how much jitter was in the source clock to begin with.

[Jud]

I think people are misunderstanding why DAC manufacturers use Async vs Synchronous USB protocols.

 

 

Async USB removes at least one source of jitter (the clock recovery). How much better it performs as a result depends on the clock recovery mechanism and on how much jitter was in the source clock to begin with.

 

Feel free to point out mansr's misunderstanding, since it is mine as well.

[mansr]

Feel free to point out mansr's misunderstanding, since it is mine as well.

 

Jitter reduction is one reason to use async USB, but it is not the only reason. For example, it allows using an external clock to synchronise multiple devices.

[BobSherman]

Feel free to point out mansr's misunderstanding, since it is mine as well.

Did not know how all this pertains to analogue interconnects... But... Oh well...

[plissken]

Async USB removes at least one source of jitter (the clock recovery). How much better it performs as a result depends on the clock recovery mechanism and on how much jitter was in the source clock to begin with.

 

I get that with Sync USB you have to discard the incoming clock data.

 

But Async USB in and of itself isn't an anti jitter source. It's just a clockless source.

[plissken]

Feel free to point out mansr's misunderstanding, since it is mine as well.

 

That Asynch USB is a jitter free protocol.

 

It's a clock free protocol and it's a fetched protocol. It makes it easier to Clock incoming data. Not that incoming data is jitter free.

 

There is a difference.

[Sal1950]

All this hate directed at John Siau has absolutely nothing to do with any techincal arguments over his designs. It is simply spawned from the fact that John won't talk the subjective party line on cables, DSD, etc. Even to the point of having Jud stoop to writing "people say" "call Benchmark DAC "ear rippers."

Does it so gaul you Jud that the premier subjective audiophile publication Stereophile still rates the DAC-2 as a Class A+ sounding DAC that you quote non-revealed sources as saying negative things?

Might also have something to do with the fact that he is friends with Mark Waldrep and that Mark uses the DAC-2 in his HDA recording studio to preview and also demo his recordings. Mark is another that must be silenced for not talking the same party lines.

You guys are worse than those writing the current political commercials.

Mason Proffit - Two Hangmen

 

It didn't take them long to try him in their court of law

He was guilty then of "Thinking", a crime much worse than all

They sentenced him to die, so his seed of thought can't spread

And infect the little children, that's what the law had said

So the hangin' day came 'round, and he walked up to the noose

I pulled the lever, but before he fell I cut him lose

They called it a conspiracy, and that I had to die

So to close our mouths and kill our minds, they hung us side-by-side

[mansr]

That Asynch USB is a jitter free protocol.

 

It's a clock free protocol and it's a fetched protocol. It makes it easier to Clock incoming data. Not that incoming data is jitter free.

 

There is a difference.

 

Data doesn't have jitter, clocks do. Remove the clock, remove the jitter.

[plissken]

Data doesn't have jitter, clocks do. Remove the clock, remove the jitter.

 

I thought even with Asynch you have to tell the computer to fetch and that the IRQ can introduce timing delay though.

[Paul R]

Do you have a quantifiable measure of what is 'Better'. What happens when you think DAC A sounds better than DAC B and someone else thinks DAC B sounds better than DAC A?

 

You may use any criteria you wish, though note I said "sound better" - which implies I was talking about subjective listening. I like this DAC better than that. It is also true that different DACs produce objectively different output. Using a benchmark DAC2 as a reference, how do you judge if a DAC "sounds better" or not?

 

 

I don't seem to be saying anything. What I said is in plain, English, text for all to read.

 

What I posted is that FFT plots show no difference with the output of a well designed DAC being fed with 1000 foot of CAT5e via AES-EBU and a standard AES-EBU cable.

 

I can see you believe that. You never answered if it made any difference in the sound however. Or exactly what kind of cable you are referring to.

 

 

What issue? Is there some aberration in the FFT plots by BenchMark that I'm missing?

 

Am I missing the point that while some manufacturers include USB input due to customer demand that BenchMark implements USB for marketing reasons instead?

 

Am I missing the point where someone is maintaining that Asynchronous USB is some how a jitter elimination schema instead of a no source clocking schema?

 

I acknowledge that if you are a DAC designer and USB cables make a difference then show that difference. If BenchMark can show no difference and, as some are insinuating, there are better engineering talents out there, then they should be able to, with Data, demonstrate that point.

 

Yep - you are either deliberately or not missing the point. Different high quality DACs sound different, based upon implementation choices by engineers. What do you see those choices are? Or do you contend that all DACs sound exactly the same or produce exactly the same output given the identical input.

 

-Paul

[plissken]

 

 

I can see you believe that. You never answered if it made any difference in the sound however. Or exactly what kind of cable you are referring to.

 

Uh 1000 foot of CAT5e and a standard (3-6 foot) AES-EBU cable as is outlined by John Siau.

 

John is stating very clearly there is no degradation in SQ and backs it with some FFT plots.

[mansr]

I thought even with Asynch you have to tell the computer to fetch and that the IRQ can introduce timing delay though.

 

Yes, but as long as the DAC-side FIFO doesn't underflow it's of no consequence. If the FIFO does underflow you get a dropout, not jitter.

[Jud]

All this hate directed at John Siau has absolutely nothing to do with any techincal arguments over his designs. It is simply spawned from the fact that John won't talk the subjective party line on cables, DSD, etc. Even to the point of having Jud stoop to writing "people say" "call Benchmark DAC "ear rippers."

 

Please do find at your very earliest convenience (right now would be really good) where I have ever said any such thing. The very worst I have said about the Benchmark DAC is that by all reports it is "excellent." Harsh critique, eh?

 

Now go ahead and find where I say anything different, particularly quoting anonymous sources as you allege, or STFU.

[plissken]

deliberately or not missing the point. Different high quality DACs sound different, based upon implementation choices by engineers. What do you see those choices are? Or do you contend that all DACs sound exactly the same or produce exactly the same output given the identical input.

 

-Paul

 

No, I'm deliberately choosing not to go off track with you.

 

My point by quoting John Siau is that their implementation has a very robust recovery system that for all intents and purposes with properly functioning cabling makes esoteric cabling a moot point.

[Jud]

That Asynch USB is a jitter free protocol.

 

It's a clock free protocol and it's a fetched protocol. It makes it easier to Clock incoming data. Not that incoming data is jitter free.

 

There is a difference.

 

Data doesn't have jitter, clocks do. Remove the clock, remove the jitter.

 

Exactly. Plissken, you're welcome to find where I've ever said data had jitter. In fact, I've used the example here a couple of times of writing down 1s and 0s by hand and typing them into your computer, and as long as the bits are clocked out of the buffer nicely, no timing differences on the input end matter.

[plissken]

Yes, but as long as the DAC-side FIFO doesn't underflow it's of no consequence. If the FIFO does underflow you get a dropout, not jitter.

 

I know. That's the point I'm trying to make. But the same is true of Synchronous also if the DAC is robust in it's reclocking. As a designer take the free ride that Async offers.

 

A-sync doesn't = jitter free. I get that it makes it easier to clock fetched data since you aren't on the computers time table.

 

I just want people to understand that Async has more subtlety than what they are thinking.

 

The fetch operation can have timing variance due to IRC and even bus contention. It's why I've made sure to always point out it's a clock free source of data and as you rightly pointed out a big problem that it solves.

[Paul R]

Async USB removes at least one source of jitter (the clock recovery). How much better it performs as a result depends on the clock recovery mechanism and on how much jitter was in the source clock to begin with.

 

+1 (Wow - something we agree 100% on! )