Cost no object dac/streamer

[RayDunzl]

Interesting. Should one read this as a suggestion for @amirm to broaden his set of measurements?

Would that lead to fewer or additional objections after the measurement and interpretation is posted?

 

[Blumlein 88]

Further to my previous post, here are the balanced output impedances I was able to find for the DACs tested (some from spec sheets, others measured by Stereophile). If the amp presents a load of 600 Ohm to the passive attenuator, we can assume that the amp + passive attenuator + cables present a significantly lower load than that to the DAC. Hence my suggestion that herein lies the most plausible source of audible differences between the units, particularly if the impedances of either DAC or amp vary significantly with frequency (which is for the most part admittedly unknown):

• Chord Dave: 66 Ohm
• Merlot: (unknown)
• Weiss DAC202: 44 Ohm
• DAC8 DSD: 114 Ohm
• Aune S16: (unknown)
• DCS Rossini: 2 Ohm
• DCS Delius: 1 Ohm

As to any audible difference between the Rossini and Delius, I have no plausible theory...

And we need to know which taper the Goldpoint was. 10 kohm or 250 kohm? Or which? As well as the type, length, and parallel capacitance of cable between the attenuator and amps. And with such a low input impedance combined with a passive attenuator you likely have the reason for the audible differences.

 

[svart-hvitt]

Would that lead to fewer or additional objections after the measurement and interpretation is posted?

Are absence of objections proof of truth?

 

[Wombat]

If someone knew, I am sure amir would measure them.
I ask you the inverse question: What is the conclusive evidence that the set of measurements usually presented (SINAD etc.) fully represent the behavior of a DAC and that there are no audible differences if these measurements are identical. The presented measurements are a good proxy, no doubt. But what makes them sufficient to be conclusive?
Scientific, statistically controlled proof please, not superficial logic.
I am not trying to be difficult. I am simply slightly taken aback by the reactions in the forum that anything that does not conform with the unwritten rules of the forum (identical measurements of a limited number of parameters = identical sound) gets disqualified as BS. I can well live with that in a audiofeelingsreview forum, but in an audiosciencereview forum it strikes me as inconsistent with the implicit mission of the forum. True science has always been curious and open to new insights.

Take the time and make the effort to grasp the knowledge that is available on the topic and see if you can come up with a valid alternative if you can't accept established knowledge.

P.S. Don't play that amateur argument of reversing the onus of proof. https://rationalwiki.org/wiki/Burden_of_proof

 

[hvbias]

I can't say I am that surprised with @jacobacci listening test. I read another DAC blind comparison test on another forum performed fairly well and people were able to distinguish differences under single blind comparisons.

A friend of mine was able to pass an ABX test with a Spectral SDR-2000 and an Assemblage DAC, also done under fairly stringent conditions. He told me it's less to do with the D/A chip, analog stage, etc and much more about the filters used. With the Spectral vs Assemblage he said the easiest way to pick them out was listen to the width of the stage on specific tracks. Once you find the thing you need to listen to just focus on that part. It is really no different than the MP3 vs lossless training.

Has anyone used a DAC with switchable filters tried them out? I have with a Lavry DAC, some of them can sound quite different.

Now imagine a DAC with no switchable filter but has something like "wide PLL", compare that to a DAC with a fairly run of the mill minimum phase filter. Someone that knows what to listen for will be able to tell them apart.

 

[stunta]

@amirm ? Streamers vs just ya DAC plugged into a computer, have we measured any streamers ?

The miniDsp SHD Amir measured is a streamer. Uses Volumio. Don't think it was measured as a streamer though.

 

[Wombat]

I can't say I am that surprised with @jacobacci listening test. I read another DAC blind comparison test on another forum performed fairly well and people were able to distinguish differences under single blind comparisons.

A friend of mine was able to pass an ABX test with a Spectral SDR-2000 and an Assemblage DAC, also done under fairly stringent conditions. He told me it's less to do with the D/A chip, analog stage, etc and much more about the filters used. With the Spectral vs Assemblage he said the easiest way to pick them out was listen to the width of the stage on specific tracks. Once you find the thing you need to listen to just focus on that part. It is really no different than the MP3 vs lossless training.

Has anyone used a DAC with switchable filters tried them out? I have with a Lavry DAC, some of them can sound quite different.

Now imagine a DAC with no switchable filter but has something like "wide PLL", compare that to a DAC with a fairly run of the mill minimum phase filter. Someone that knows what to listen for will be able to tell them apart.

If there is a difference it can be isolated with signal analysis methods and equipment.


In terms of DIY double-blind audibility tests they are seldom properly documented in terms of environment, subjects, process, control, results, analysis and validation of conclusions. For a start the testers should be independent of the test subjects - seldom is the case with DIY tests.

'Take our word that the test was proper, and me and ? can hear a difference' is too vague.

 

[Sal1950]

It's not a reason to chuck out hi fi playback, but we have to keep in mind its importance in the scheme of things. We have to know we are reproducing something already severely flawed. The speakers and the microphones are usually directional units. And sound is typically omnidirectional in source. Even the omnidirectional microphones capture the sound down into mono or stereo.
I'm surprised sound reproudction sounds as great as it does!

You continue to miss the point of High Fidelity. When we discuss any component it's goal should be to reproduce as perfectly as possible the input without adding or subtracting any information. When we discuss the quality of any component, that is the detail it should be judged on. Too often components are judged to be "better or more musical" on the basis of added distortions, a completely wrong assuption and a direction that High End audio has wrongly taken.

 

[jacobacci]

P.S. Don't play that amateur argument of reversing the onus of proof

Why not?

 

[Wombat]

GIGO applies to HiFi, not just computer science and mathematics.

 

[Wombat]

Why not?

Plain English

 

[jacobacci]

Hence my suggestion that herein lies the most plausible source of audible differences between the units, particularly if the impedances of either DAC or amp vary significantly with frequency (which is for the most part admittedly unknown):

Interestingly the two units that ran through the attenuator were the Aune and the Merlot (not quite sure about the latter) because of fixed volume output. The others were adjusted using the DAC's volume. So for them the signal path was through a physical switch / cable directly to the brinkmann amps.
If output impedance was indeed a factor for the audible difference, then the high output impedance of some of these DACs would simply fall under the heading of bad design. The dCS seems to be good design (at least with respect to this parameter). Lucky I liked that one and not say the T+A, else I my credibility (if it exists at all) would take a serious hit.

Has anyone used a DAC with switchable filters tried them out?

Most modern DACs have switchable filters so I would suspect most of us have used one that has them. In our comparison we left all DACs at the default filters (which yes, is probably a fundamental flaw of the comparison).
But then comes the nasty question. Which of the filters (which are all compromises) is the most accurate?

 

[jacobacci]

Plain English

It is not me that is postulating that the set of measurements usually done by Amir is a full and complete representation of the audible (under all circumstances) characteristics of a DAC. All I am saying is that the set is a good proxy and MAY not be a complete representation.
In my view the burden of proof is clearly with the side claiming that the existing set of measurements is exhaustive and more than a proxy.

 

[Blumlein 88]

Interestingly the two units that ran through the attenuator were the Aune and the Merlot (not quite sure about the latter) because of fixed volume output. The others were adjusted using the DAC's volume. So for them the signal path was through a physical switch / cable directly to the brinkmann amps.
If output impedance was indeed a factor for the audible difference, then the high output impedance of some of these DACs would simply fall under the heading of bad design. The dCS seems to be good design (at least with respect to this parameter). Lucky I liked that one and not say the T+A, else I my credibility (if it exists at all) would take a serious hit.


Most modern DACs have switchable filters so I would suspect most of us have used one that has them. In our comparison we left all DACs at the default filters (which yes, is probably a fundamental flaw of the comparison).
But then comes the nasty question. Which of the filters (which are all compromises) is the most accurate?

Wait a minute. That is not what you described earlier. You listed the chain of devices and didn't mention this. So which is it?

 

[Wombat]

S

It is not me that is postulating that the set of measurements usually done by Amir is a full and complete representation of the audible (under all circumstances) characteristics of a DAC. All I am saying is that the set is a good proxy and MAY not be a complete representation.
In my view the burden of proof is clearly with the side claiming that the existing set of measurements is exhaustive and more than a proxy.

Show me where it is stated that the existing set of measurements is exhaustive. I am sure @Amir has not made such a claim. Anyway, show me.

 

[jacobacci]

So which is it?

When not needed, we set the Goldpoint to "maximum volume", i.e. to not attenuate. If I am not mistaken, in that setting, the Goldpoint is just a switch.

 

[jacobacci]

Show me where it is stated that the existing set of measurements is exhaustive. I am sure @Amir has not made such a claim. Anyway, show me.

I am not saying Amir made that claim. The argument is usually: There are no audible differences between DACs that measure well (using the said set of measurements). Any listening test that demonstrated audible differences is flawed, as there cannot be any.
I have repeatedly seen terms such as "complete waste of money, they all sound the same". But this time around I really cannot be bothered to go find all the quotes. I have patiently answered all the challenges thrown in my direction without resorting to qualifications such as "amateurish", but this is really getting a bit long in the tooth.

 

[andreasmaaan]

Interestingly the two units that ran through the attenuator were the Aune and the Merlot (not quite sure about the latter) because of fixed volume output. The others were adjusted using the DAC's volume. So for them the signal path was through a physical switch / cable directly to the brinkmann amps.
If output impedance was indeed a factor for the audible difference, then the high output impedance of some of these DACs would simply fall under the heading of bad design. The dCS seems to be good design (at least with respect to this parameter). Lucky I liked that one and not say the T+A, else I my credibility (if it exists at all) would take a serious hit.

Don't worry, no threat to your credibility from me You've clearly stated what your goals were in doing the test, and it seems to me those goals were met. Plus you've been generous with the depth you're willing to go into describing your methods.

Still, your results are major outliers when compared to more rigorous tests that have been conducted, so I can't help trying my hardest to explain them.

I'm not sure the "high" output impedance of the DACs would come under the heading of bad design so much as the weirdly low input impedance of the amp, or the impedance structure of the setup as a whole.

When not needed, we set the Goldpoint to "maximum volume", i.e. to not attenuate. If I am not mistaken, in that setting, the Goldpoint is just a switch.

Not quite: the loads of the Goldpoint + amp + cables would still be in parallel to the DAC if I'm not mistaken.

And even without the attenuator in the chain at all, the load presented by the amp itself would be quite low in comparison to many of the DACs.

 

[Wombat]

I am not saying Amir made that claim. The argument is usually: There are no audible differences between DACs that measure well (using the said set of measurements). Any listening test that demonstrated audible differences is flawed, as there cannot be any.
I have repeatedly seen terms such as "complete waste of money, they all sound the same". But this time around I really cannot be bothered to go find all the quotes. I have patiently answered all the challenges thrown in my direction without resorting to qualifications such as "amateurish", but this is really getting a bit long in the tooth.

A weasel out post. Those posts you allude to will not be hard to find. They may even be inacurrate - that should be checked before referencing them. You have not provided solid support to your opinions in technical matters or reasoning. Having calibrated ears would be a good start.
This has been yet another 'but I can hear it and you don't measure it' type of thread diversion. I guess it won't be the last.

 

[Frank Dernie]

Yes, but accurate to what, that signal has passed through a ton of equalizers and compressors already, by some mastering engineer, through his speakers, good for his hearing.
Some guys here talk like we are discussing for pacemaker devices, these things should reproduce music.
Have anyone here listened to live music? No dac or amplifier can reproduce it,
Accurate yes, but first is the listening..

Accurate to the recording. Yes, a mastering engineer has messed with it, yes it depends on the chosen microphone and its location in how convincingly it picks up the sound but still, once there is a digital file of whatever the artists are satisfied with and have released, accuracy to that is what I mean, and that is entirely measurable and achievable at modest cost.
Except the listening room and speakers.