[mmerrill99]

Well, at the moment it seems cjl is struggling to understand IMD !

Oh, that's rich. I understand IMD perfectly. You do understand that the multitone test is still an IMD test, right?

I don't have any confusion over that - it's you that seems to have a problem understanding what InterModulation Distortion means & how if you increase the number of tones the distortion products increase.

It's arguably a better IMD test than the standard one, and I'm always a fan of increased testing and data on all devices.

Good, on that we both agree

That doesn't change the fact that all audible defects will be measurable though.

I would suspect they should be given a complete set of suitable tests.

 

[charleski]

Of course the other possibility is that the test is so ***** in every respect that it hides such an obvious difference!

 
No, that's not a possibility I'd seriously consider, because there's no basis for it. A 1.4dB variance in the frequency response is fairly subtle, and while I'd like to think I could spot it, I'm ready to be proved wrong. You, clearly, are not.
 
 

You know what IMD stands for, don't you? I wonder where those sideband distortions might be coming from?
Ah, yes difficult analysis - let's not do it then
You just chased your own tail in a logic circle blur - "This shouldn't be an issue for a home-audio system in which only one signal is present." - do you read back what you write - it's advisable ?

 
Right now I honestly don't think you really have a clue what IMD actually is. A signal composed of two frequencies will generate distortion products at the sum and difference of their frequencies: an 800Hz signal and a 1200Hz signal will generate products at 400Hz and 2kHz. This is different to harmonic distortion, in which the products appear at integer multiples of the frequency being tested, so a signal with an 800Hz component would generate products at 1600Hz, 2.4kHz, 3.2kHz, etc. Physical systems, especially loudspeakers, can generate subharmonic distortion products (rational fractions of the fundamental) as a result of interaction with their own intrinsic resonances.
 
Generally you want to test these two different types of distortion separately, as they may be caused by different defects in the design. This is why you need to take a little care in choosing the probe frequencies for an IMD test - if you fed in a combination of 1kHz and 3kHz you would get a distortion spike at 2kHz, but have no way of knowing how much of this was IMD from the 3-1kHz difference or 2nd harmonic of the 1kHz fundamental. Spraying a bunch of different frequencies at the system will produce a large range of distortion products that can be hard or impossible to disentangle, but, as I already said, it has its use as a 'ball-park' test that will cover a wide frequency range efficiently.
 
As for your last sentence: I only listen to one piece of music at a time. Maybe Vulcans find this illogical and listen to three or four on the trot to save time, but I'm not really interested in the listening habits of imaginary aliens. Alternatively ... you don't grasp the meaning of the word 'signal' in an engineering context.

 

[cjl]

   
Summary:
 
1. Schiit did 2 different types of IMD tests on a competing DAC.  They were surprisingly crappy in portions of the spectrum, enough to probably be audible.
 
2. Crappy IMD measurements that cross the threshold of audibility might correlate with subjective sound preferences.
 
That's it, right?
 
Seems pretty 'Well, duh' to me.

Yep, pretty much that.

 

[castleofargh]

want to know something about a DAC's chipset? you have god knows how many PDF you can download that tell you everything to the point where you know how the engineer's grand mother was making her apple pie. you even have all the stuff to tell you how to make a DAC using that chipset if you like.
now, you want to know something about the output signal of a DAC? here are some basic measurements not clearly named, often not even with a proper unit,  and made from undefined measurements. so we don't even know for example how the IMD test was done on most manufacturer's specs. oh but wait, I'm going too far already, we often don't even have IMD measurements on DACs specs...
 
so of course it's easy to argue that measurements might not tell the all story when we only get scraps of what the measurements should be. and when you argue about that with the manufacturers, the general consensus is that they don't provide more because specs are misunderstood by people...   so we can play that game for a while.
 
-"doc, my child doesn't know how to count"
-"did someone teach him?"
-"no because I'm afraid he might misunderstand".

 

[mmerrill99]

want to know something about a DAC's chipset? you have god knows how many PDF you can download that tell you everything to the point where you know how the engineer's grand mother was making her apple pie. you even have all the stuff to tell you how to make a DAC using that chipset if you like.
now, you want to know something about the output signal of a DAC? here are some basic measurements not clearly named, often not even with a proper unit,  and made from undefined measurements. so we don't even know for example how the IMD test was done on most manufacturer's specs. oh but wait, I'm going too far already, we often don't even have IMD measurements on DACs specs...

so of course it's easy to argue that measurements might not tell the all story when we only get scraps of what the measurements should be. and when you argue about that with the manufacturers, the general consensus is that they don't provide more because specs are misunderstood by people...   so we can play that game for a while.

-"doc, my child doesn't know how to count"
-"did someone teach him?"
-"no because I'm afraid he might misunderstand".

Yes, that's adequately demonstrated in recent posts

 

[charleski]

It's sometimes amusing to browse the AudioPorn sites, and the first DAC review on Stereophile currently is of the $5500 AudioNote DAC 2.1x.
 
This receives high praise from the enthusiastic reviewer. And yet, even as he's dishing out the plaudits he feels the need to qualify his statements: "the sound wasn't buzzy or fuzzy, but it skated right up to the border of same", "The harsh highs were, if anything, a little more grating through the Audio Notes than I heard from CD rips of the album played from my Apple iMac", "the DAC 2.1x Signature, in particular, may generate distortion products", "The question of whether the sound of the Audio Notes should be ... blamed for trading in pleasing distortions will never be answered to the satisfaction of all." But none of this diminishes his ecstatic conclusion: "I have never heard a CD player that beats this combination in the ability to involve me in the magic of notes and rhythms, or that presents lines of notes in such a musical and attention-grabbing manner. Vigorously recommended."
 
In the measurements section we learn that this device uses an old Analog Digital AD1865, a chip that's featured in a variety of odd-ball R2R designs (the large OBSOLETE stamp is particularly fetching). But, far more seriously, the designer has chosen to throw his textbook out the window (presuming that he actually owned one) and neglected to filter the output. As is entirely predictable, this wrecks the design, spewing out gobs of aliased image noise which then intermodulates down to the audible spectrum, with a 19kHz tone causing a -52dB spike at 6kHz. Harmonic distortion from low-frequency signals is particularly high, with the 2nd harmonic alone reaching -50dB. John Atkinson does, at least, have the integrity to declare in summary that, 'it is difficult to avoid the temptation to describe the Audio Note DAC 2.1x Signature as "broken" '. It clearly is.
 
So it measures like crap, but it 'sounds good'? Where have we heard that before?
Oh yeah ... vinyl.
 
Just as some people seem to enjoy the compression, noise and distorted frequency response of vinyl, could we be seeing a similar reaction to the nonlinearity of old R2R DACs and distortion produced by 'special' (i.e. wrong) filters? I suspect so. The human mind is incredibly plastic, and we can teach ourselves to enjoy all sorts of strange pleasures. I think the question really is: how far can this go? How long will it be before audiophiles discover the true source of digital audio nirvana and start paying $5k for DACs based on the big daddy of them all, the sublimely musical 14bit TDA 1540? Hold on, it's happening already: "when I heard the CD104 with just NOS and all capacitor mods - it was so good that I did not believe my ears. It is too good to be true." (yes, that's from Lampizator, makers of the masterpiece featured earlier in this thread, but they aren't the only ones).
 
... Actually, what am I doing wasting time on this thread. Bye guys, I'm off to buy up some ancient CD players so I can rip the chips out and put them in a fancy new box for the next HiFi Show.

 

 

[artur9]

   could we be seeing a similar reaction to the nonlinearity of old R2R DACs and distortion produced by 'special' (i.e. wrong) filters? I suspect so. 

That might possibly be true of some of these DACs but given the measurements @atomicbob took of the Schiit R2R DACs, that's not the case for them.  They measure fantastically well.

 

[watchnerd]

 
So it measures like crap, but it 'sounds good'? Where have we heard that before?
Oh yeah ... vinyl.
 

 
Or tubes.
 
(Full disclosure...I have some tube stuff I like. But I know it's distorted.)

 

[watchnerd]

  That might possibly be true of some of these DACs but given the measurements @atomicbob took of the Schiit R2R DACs, that's not the case for them.  They measure fantastically well.

 
I'm sure the hardest core R2R guys would slam Schiit for having a filter, too.

 

[Vkamicht]

Hi all, really interested in this topic and I've read a number of pages but I had a question I haven't seen brought up (maybe missed it) - are there any level matched analog recordings showcasing the difference between these two types of DACs? I am going through inner turmoil because I'm a skeptic but I "want to believe." I would like the ability to do an ABX on foobar or something without having to switch equipment (plus I can't really afford much), I'm wondering if the difference would be audible via line out recording. My thought being if people are truly hearing the "digital harshness" sound of the DS DACs, then even if my playback system is DS, I would still be able to pick out which audio file was run through the DS DAC, right? (essentially twice after playback) I mean it would only sound worse making multiple passes... and the R2R recording wouldn't have this problem. On the flip side if the argument is that anything coming out of the DS is going to sound awful no matter what, then I would propose to do the same test via the R2R. The "DS sound" is then recorded into the file itself and an ABX should reveal this. Is there a flaw in this thinking?

 

[charleski]

  That might possibly be true of some of these DACs but given the measurements @atomicbob took of the Schiit R2R DACs, that's not the case for them.  They measure fantastically well.

Oh I'd absolutely agree. The flagship Burr-Brown PCM63-K was capable of extremely good results when implemented properly ... at a price. And current darlings like the PCM1704 are almost as good (though it costs six times as much as a top-flight ΔΣ chip like the CS4398). If you want to spend money just for the sake of spending it, then it's perfectly possible to get a decent result with R2R.
 
[As an aside, looking at the money people are spending on old PCM63 chips makes me wish I hadn't junked the old DAC I built back in the '90s.]
 

  Hi all, really interested in this topic and I've read a number of pages but I had a question I haven't seen brought up (maybe missed it) - are there any level matched analog recordings showcasing the difference between these two types of DACs? I am going through inner turmoil because I'm a skeptic but I "want to believe." I would like the ability to do an ABX on foobar or something without having to switch equipment (plus I can't really afford much), I'm wondering if the difference would be audible via line out recording. My thought being if people are truly hearing the "digital harshness" sound of the DS DACs, then even if my playback system is DS, I would still be able to pick out which audio file was run through the DS DAC, right? (essentially twice after playback) I mean it would only sound worse making multiple passes... and the R2R recording wouldn't have this problem. On the flip side if the argument is that anything coming out of the DS is going to sound awful no matter what, then I would propose to do the same test via the R2R. The "DS sound" is then recorded into the file itself and an ABX should reveal this. Is there a flaw in this thinking?

I wouldn't be at all surprised to learn that a DAC like the AudioNote 2.1x could be distinguished from a proper design based on a ΔΣ chip in an ABX test. That doesn't mean it sounds better, unless you have a preference for coloured sound.

 

[prot]

It's sometimes amusing to browse the AudioPorn sites, and the first DAC review on Stereophile currently is of the $5500 AudioNote DAC 2.1x.

This receives high praise from the enthusiastic reviewer. And yet, even as he's dishing out the plaudits he feels the need to qualify his statements: "the sound wasn't buzzy or fuzzy, but it skated right up to the border of same", "The harsh highs were, if anything, a little more grating through the Audio Notes than I heard from CD rips of the album played from my Apple iMac", "the DAC 2.1x Signature, in particular, may generate distortion products", "The question of whether the sound of the Audio Notes should be ... blamed for trading in pleasing distortions will never be answered to the satisfaction of all." But none of this diminishes his ecstatic conclusion: "I have never heard a CD player that beats this combination in the ability to involve me in the magic of notes and rhythms, or that presents lines of notes in such a musical and attention-grabbing manner. Vigorously recommended."

In the measurements section we learn that this device uses an old Analog Digital AD1865, a chip that's featured in a variety of odd-ball R2R designs (the large OBSOLETE stamp is particularly fetching). But, far more seriously, the designer has chosen to throw his textbook out the window (presuming that he actually owned one) and neglected to filter the output. As is entirely predictable, this wrecks the design, spewing out gobs of aliased image noise which then intermodulates down to the audible spectrum, with a 19kHz tone causing a -52dB spike at 6kHz. Harmonic distortion from low-frequency signals is particularly high, with the 2nd harmonic alone reaching -50dB. John Atkinson does, at least, have the integrity to declare in summary that, 'it is difficult to avoid the temptation to describe the Audio Note DAC 2.1x Signature as "broken" '. It clearly is.

So it measures like crap, but it 'sounds good'? Where have we heard that before?
Oh yeah ... vinyl.
...

After reading such stuff it is extremely difficult to avoid the temptation to describe Stereophile as a bunch of incompetent liars .. or maybe I should say 'vigurously competent liars' .

 

[castleofargh]

they have that kind of unique situation where measurements and subjective opinions can mix and oppose each other. if anything, compared to most other guys, I would say that at least they have some measurements and I greatly appreciate that.
 
now when the graphs show a problem and the feedback is "it sounds amazing", I have to say that I'm always wondering if it's one of those cases where distortions sound nice, or if it's marketing and mutual interests that are typing on the keyboard at that moment. 
 
 
@Vkamicht I honestly don't know. some would probably argue that the ADC ruined the very thing that was making the R2R DAC genuine sound. even if that's kind of silly when you know that the record was most likely made on a delta sigma if it's not an old one. I often do what you say and record what I want to test so that I can just abx it instead of having to prepare a tedious blind test. but I really don't know how much the ADC impacts the sound of both sources. maybe if I had a better ADC I would have more confidence in that kind of test.
but at least we would need both DACs to be recorded on the same system with the same settings, just asking for people to send a recording of a song or a sweep, could then make us test more than the 2 DACs and would be unfair. but it could at least be more than us talking in the wind ^_^.

 

[nick_charles]

  they have that kind of unique situation where measurements and subjective opinions can mix and oppose each other. if anything, compared to most other guys, I would say that at least they have some measurements and I greatly appreciate that.
 
now when the graphs show a problem and the feedback is "it sounds amazing", I have to say that I'm always wondering if it's one of those cases where distortions sound nice, or if it's marketing and mutual interests that are typing on the keyboard at that moment. 

 
 
There is another possible explanation, that even what looks like impressive distortion and noise and FR deviations are *sometimes* still below our detection ability in normal listening and that the supposed huge differences between components can sometimes be attributed to the nature of uncontrolled tests where little or no effort has been expended to remove the array of biases and that even without the biases spawned by foreknowledge of what the reviewer is listening to we have a test method so flawed that any number of imagined differences and/or audio characteristics are discovered.
 
Be that as it may removing the knowledge of what they are listening to would seem to be an utterly logical first step. The argument that these expert listeners have such superior hearing and judgment that they can remember in clear and total detail what something sounded like three weeks ago in a different room with different speakers and different tracks at different volume levels and thus make 100% judgments off differences between A and B and are not in any way (conscious or not) swayed by appearance or price stretches credulity.
 
The late Tom Nousaine did some interesting tests. he gave listeners a box that had a circuit in it that was either a transparent pass through or introduced 2.5% distortion. Long term listeners scored randomly (50/50) in their ability to determine if the box had distortion or not. Then the distortion was plugged into a signal using a ABX box listeners able to switch between two signals (distorted/undistorted) quickly performed far better

 

[icebear]

  they have that kind of unique situation where measurements and subjective opinions can mix and oppose each other. if anything, compared to most other guys, I would say that at least they have some measurements and I greatly appreciate that.
 
now when the graphs show a problem and the feedback is "it sounds amazing", I have to say that I'm always wondering if it's one of those cases where distortions sound nice, or if it's marketing and mutual interests that are typing on the keyboard at that moment. 
...

There is for sure a chance that measurements show some kinds of odd behavior but the equipment sound subjectively good.
But in general there can be no doubt that advertising and reviews go hand in hand. There isn't a single product review in the you know what hifi magazine (in any country) of a companie's product and the related full ad page of that manufacturer is either before the review or directly behind. Sometime next issue announcements are followed by back page high gloss ads. So much for objective reviews