Science isn't a process of debate. Winning or losing a debate does not change whatever the science turns out to be.
Mark,
With all due respect, you haven't really disagreed at all.
It doesn't matter how products are evaluated, they are all designed using sound engineering principles, and if any problems are detected then solutions will be found using those same principles.
Sorry again, but isn't that kind of moving the goal post?
At first you claimed that the question wasn't answered - and further why my post was typical non - answer etc. etc. - and I provided a reasonable argument to show that indeed it was answered.
But now instead of admitting that, you are complaining about a non-answer for a question that you really haven't asked. 😉
As we all know the so-called "great debate" is going on at least for 50 years and we shouldn't pretend not knowing about it.
So a honest discussion is always appreciated, but then you should for example encourage Evenharmonics to bring up evidence for his claim(s) in this case citation of the "objective" listening tests of DACs where no difference was detected.
Maybe I missed it, but I can't remember that you objected to his refusal to post it.
Or should I/we accept something like "I posted the citation(s) in other threads and if you've missed it it is your problem" as sufficient?
Further, we all know, that humans are not infallable and that "sighted" listening comes with the risk of knowledge having an impact on sensory evalutions.
Otoh we all know (or should at least by now) that controlled listening tests (including "blinding") come with the risk of incorrect results too (for various reasons), especially when done in the often used variant of ABX-tests with small trial numbers and statistical analysis based solely on the proportion of correct results.
Last edited:
Science is about enquiring, asking questions, and hopefully finding meaningful answers.
Yes, there are fallacious arguements on both sides of this debate, and neither side can prove they are right by proving the other wrong.
Theoretically yes but in practice audio can be very tricky as JW has suggested.
I just sent a DAC out that came in for some work. It was an R2R / tube based unit (not of my design). It ended up with a JLsounds USB bridge
driving PCM1704's directly (owner will software upsample in PC). DAC's feed a simple triode OP stage.
Going on 'sound engineering principles' this DAC should sound like a pile of
rubbish, it would certainly measure pretty terribly.
It actually sounded really good, quite a shock! I kept thinking 'where is all
this distortion and mud I'm supposed to be hearing'? Big 3D soundstage,
amazing instrument separation and clarity. Really listenable.
So it goes, audio can be a frustrating game. 🙂
TCD
Naaling, in case you aren't aware, despite your logical questions, those who are arguing against you persistently, are either in audio business or affiliated with one. Their effort here is for the business, not the consumer's well being.
Don't want to crash your party, but I cannot stop wondering what are you, Joseph K, and quite a few others doing on this DIYAudio board?
You are all obviously accomplished designers, some of you have even a solid EE background, you have obviously a busy schedule and a big fat portfolio of projects; at the same time, it doesn't seem like you are eager to share much technical details (and discuss around) with your peers, anything that goes beyond your "sonic impressions", and that's understandable, since you need to put the bread and butter on your table. Instead, you are becoming defensive and/or passive-aggressive when challenged (which is to be expected due to the nature of any social media and the specifics of your discourse). Which pretty much excludes from your motivation the need for honest feedback. You also don't seem to seek technical feedback from other members.
I know of another successful audio designer (with more of an engineering approach) which when asked about sharing his knowledge replied "I don't feel an urge to educate my competition", a perfectly valid stance. But then he was also not wasting his precious time glossing about sonic impression and telling subjective stories about his products.
So then, once again, what are you doing here? Are you considering yourself as some kind of Missionaries of High End Audio, trying to educate the non believers? Advertising your work and/or products? Trying to contribute to creating a market for a new product (chip, equipment, feature, etc...)? Simply need to massage your ego (a very human need, agreed)? Searching/testing marketing cues? Anything else that I'm missing?
I've been reading this thread all the way through, just want to say that I am excited about the discussion on dacs and development.
But come on guys, I love nitpicking as much as the next person, but this is getting close to borderline politics, which goes against forum rules.
Don't know about the rest of you, but I've had some jobs myself where I've been forced to sign NDA's and the like. Even if you have any hard facts about a specific circuit or part, spilling the beans will get you flying out so fast you don't know what happened. And especially so if it's unreleased material, you can get in real trouble for that stuff. It's okay to say you where Impressed with something on a closed demo after the company has done a press release of the part, but that's more or less the limit on how much you can and should say.
Get real people!
Stop trolling!
But come on guys, I love nitpicking as much as the next person, but this is getting close to borderline politics, which goes against forum rules.
Don't know about the rest of you, but I've had some jobs myself where I've been forced to sign NDA's and the like. Even if you have any hard facts about a specific circuit or part, spilling the beans will get you flying out so fast you don't know what happened. And especially so if it's unreleased material, you can get in real trouble for that stuff. It's okay to say you where Impressed with something on a closed demo after the company has done a press release of the part, but that's more or less the limit on how much you can and should say.
Get real people!
Stop trolling!
No one asked about any specifics of the part that would be covered by said NDA. All that was asked was to explain why or by what mechanism the part would sound better when it measures worse, still has high speed digital I/O pins, and feeds the MCLK through both ICs.
So, get real. Stop trolling.
Naaling:
I. Am. Not. Going. There.
Not mentioning any names, just watching, and hoping for actual contributions to thread topic.
I. Am. Not. Going. There.
Not mentioning any names, just watching, and hoping for actual contributions to thread topic.
Too late. You already have. Now its time to back up your statement, or admit you are wrong.
I rest my case.
This is completely out of character for me, but I'm just so disappointed this very nice thread got ruined.
I really hope we can keep this to topic, because if it keeps up the way it has for the last 10 pages or so I really hope some OP closes it instead.
Anyone that took offense at my request to keep to thread topic, kindly fill out this form and do whatever you want with it afterwards, because I really don't care.
This is completely out of character for me, but I'm just so disappointed this very nice thread got ruined.
I really hope we can keep this to topic, because if it keeps up the way it has for the last 10 pages or so I really hope some OP closes it instead.
Anyone that took offense at my request to keep to thread topic, kindly fill out this form and do whatever you want with it afterwards, because I really don't care.
Worse has happened. I still live.
I merely requested that we keep to the forum rules. If that is not acceptable then do something else.
I merely requested that we keep to the forum rules. If that is not acceptable then do something else.
I have no doubt that a device that measures poorly can sound good. I have owned and listened to tube amps. I've listened to CD players from the early 90s. I think the bar to clear is lower than many here will admit. I just don't believe there is any evidence that there are properties that affect sound quality that are not measurable in the analog output.
Understood.
Hypothetically, if such properties did exist what would you accept as evidence of it?
Also, in the context of responding to the above question please clarify what you mean by the term 'measurable.' Measurable by an AP, theoretically measurable, something else?
Last edited:
When discussing anything like this I feel that everything should be kept solely to technical details and that any subjective impressions be left out.
Regardless of anyone's reputation subjective impressions are pretty much meaningless. Especially for devices which have already pushed the nonlinearities so far below the threshold of audibility. For loudspeakers it's slightly different as the way they react with the room is of vital importance and a subjective impression can be useful, but not for DACs.
It's enough to say I heard both and preferred A over B, that's fine. And then leave it at that. But when you say A was significantly better than B when both are essentially transparent...there's no way that A was significantly better than B. Was A playing undistorted music with an excellent level of fidelity? Yes? Was B? Yes? Then they were at worst very similar or at best actually identical.
For there to be an actual significant difference one would have to be broken. Night and day differences require one to be playing the song backwards or for one to be twice as loud as the other. When comparing devices as good as this there is no significantly better than the other. Hell the two might not have even been level matched. This sounds like a very casual setting so they probably were not.
There's significantly better in my head mind you. I prefer the sound of B because it's built in the way I like the idea of.
When discussing something in a thread like this we should keep the discussion technical because these are merits that are common to all devices, for direct comparison, and common to us in that we all understand what specifications mean.
If one device has oversampling of 256 instead of 128 then great. Explain why this has the potential to sound better.
I for one appreciate the fact that the digital filters in the AK4191 actually have as much stopband attenuation as the noise floor so that absolutely nothing will alias through. I don't consider one with only 100dB sufficient in a product with a noise floor way below this, such as in the AK4499. Now of course I doubt I'll hear it but it is technically superior in one respect.
Regardless of anyone's reputation subjective impressions are pretty much meaningless. Especially for devices which have already pushed the nonlinearities so far below the threshold of audibility. For loudspeakers it's slightly different as the way they react with the room is of vital importance and a subjective impression can be useful, but not for DACs.
It's enough to say I heard both and preferred A over B, that's fine. And then leave it at that. But when you say A was significantly better than B when both are essentially transparent...there's no way that A was significantly better than B. Was A playing undistorted music with an excellent level of fidelity? Yes? Was B? Yes? Then they were at worst very similar or at best actually identical.
For there to be an actual significant difference one would have to be broken. Night and day differences require one to be playing the song backwards or for one to be twice as loud as the other. When comparing devices as good as this there is no significantly better than the other. Hell the two might not have even been level matched. This sounds like a very casual setting so they probably were not.
There's significantly better in my head mind you. I prefer the sound of B because it's built in the way I like the idea of.
When discussing something in a thread like this we should keep the discussion technical because these are merits that are common to all devices, for direct comparison, and common to us in that we all understand what specifications mean.
If one device has oversampling of 256 instead of 128 then great. Explain why this has the potential to sound better.
I for one appreciate the fact that the digital filters in the AK4191 actually have as much stopband attenuation as the noise floor so that absolutely nothing will alias through. I don't consider one with only 100dB sufficient in a product with a noise floor way below this, such as in the AK4499. Now of course I doubt I'll hear it but it is technically superior in one respect.