That's not exactly what I was saying. However, I would agree there is no proof behind the claim that some people make about most dacs being audibly perfect today for 100% of the human population. No dac of today is audibly perfect for 100% of the human population, IMHO. Can I prove it, or can the opposing claimants prove their opinions to be fact? No, IMHO, I don't think there is funding for research today to sufficiently delve into such matters.
Last edited:
Are you still saying that there is proof "$200 or $500 dacs that measure pretty well sound as good as any dac can ever sound, they don't."?
You are free to express your opinion. I'm just curious what your opinion is based on.
Except for very small number (old model), every report of level matched DBT of DACs show that regardless of their price, the listeners couldn't tell them apart. Do you know of test result that supports your claim "$200 or $500 dacs that measure pretty well sound as good as any dac can ever sound, they don't."?
Sure. Is there enough data on how the tests were conducted to replicate them? Where listeners given any training on audibly differentiating dacs? What kind of power amps, headphones, speakers where used? What volume levels were used? Did they use expert listeners?
Look, I could set up a test of dacs so that nobody could tell them apart or at least some people could tell them apart. If I could select skilled listener participants, everyone could probably tell any dacs apart. You tell me what you would like the outcome to be and I can design the dac listening experiment to show what you want, and without any peeking either. Since most people who do that kind of testing are only interested in the bulk of the population under what they consider average listening conditions, and not interested in expert listeners under critical listening conditions, they get what they expect. If it were otherwise, the outcome would be different. Since I am only talking about expert listeners under critical listening conditions, I say there is no proof one way or the other about that demographic, they haven't been tested with any recent dacs if they have ever been tested at all.
Last edited:
But these must be forum tests (or even worse the dreaded "forum ABX") and need to be held to a different standard, maybe the participants should just post as individuals on what works for them.
No, Scott. We need new research. We have agreed on that before. There is no funding for it we both know that. Nothing has changed. In the meantime, we can describe opinions and personal experience or just shut up. Should I take your satire as a vote for shutting up?
No not at all, everyone speaks to their own experience which applies only to them until someone takes up the challenge to go further. If there is going to be no examination or judgement applied to how anyone conducts their personal tests then they all stand on equal ground. If someone says they made a good faith effort to compare the Benchmark to their favorite $300 DAC and heard no difference that has equal weight to any other opinion.
There isn't much to say John. A well engineered Sabre dac can sound pretty good with DSD512 if the best algorithms we have today for doing it are used. By well engineered, there is no magic to it, just figuring out how much imperfection is tolerable in each area, the various power supplies, clocking, output stages, case shielding, etc., for picky listeners and doing the best you can for them. The secret sauce is in the i7 DSP, for now at least.
I am expecting about the same to be true for AK4499, except we know less about what it can do and how much care and attention each of the areas will need for our picky listeners. Maybe there will be opportunities to design discrete output stages and maybe a couple of discrete power supplies for AVCC. If so, you know who I know that is interested in a piece of the analog stuff.
There isn't much more than that I can say that might help keep you up to date on developments.
With that I would like to end here for today, and hopefully for longer. Despite days like this when I talk more than I should, I don't like all the arguing. Big waste of time.
All the foregoing, IMHO only, of course.
I will post link/s after you address the question on the status of proof that "$200 or $500 dacs that measure pretty well sound as good as any dac can ever sound, they don't.".
I can set up a test to make one DAC sound different even to novice listeners. How? Adjust volume so that there is sufficient discrepancy. What does a test like that provide? Oh, I don't know, maybe an opportunity to kill time?
How about setting up a test that controls visual bias and matches levels so that the participants can use their ears only?
Easily.
I have had a test comparing vinyl rip with wav issued to this vinyl as a bonus download. Same mastering. Clapton, 45rpm 2LP "I still do". People were unable to tell the difference in ABX, except for one, who concentrated to a very small click at the beginning of the vinyl rip.
Another nice example is
Listening Test Part II. Can you tell which......
So people can easily tell the difference between SOTA DACs, however the same people are unable to tell the difference in ABX between vinyl rip and data file and between original file or tube loop. Or they do not participate (based on their previous failure) the test. Or are finding excuses on foobar changing the sound. Do not wonder then that the stories about the same people hearing differences between good DACs in a sighted test are considered anecdotal. However, I believe these people are self-convinced they do hear a difference and try to construct fantastic pseudo theories explaining the "difference".
Yes. These same people latch on to any kind of theory to explain what they think they hear and haven't tested properly. One week the problem is jitter, the next week it's power supply noise, then VREF 1/f noise, DS noise modulation, the oversampling filter, and on and on. The magnitude of the perceived issue is never considered. It's just the next "issue" with digital for the gurus to fix and proclaim an analog-like sound. Of course, those who are already wary of the digital bogeyman and know nothing about what is being discussed eat it up.
Last edited:
It may be stretching your beliefs, but this is actually something that can be done. Measuring all relevant aspects of an electrical signal, for example, and drawing conclusions about how it will sound from that. Much progress has been made thusly by the scientifically inclined.
Mapping of measured differences to perceptions is still a demandig and ongoing task (I didn't read your post as if you'd think it is already over and done) especially as different sound events can lead to the same percepted impression and considering the high variability between humans.
Which matters even more if multidimensional stimuli (and judgements) are examined.
Which matters even more if multidimensional stimuli (and judgements) are examined.
actually no version of Pro Tools has any ADCs or DACs. Pro Tools is a software program that can use anyone's converters. Most upper tier people do not use Avid (the owners of Pro Tools) converters, even though the latest incarnation is much better than previous ones, still not as good, and indeed the previous 2 generations were dreck.
Alan
Of course,
Ashihara, K.: Audibility of complex tones above 20kHz.
The 29th international congress and exhibition on noise control engineering, August 2000, Nice, France
You reassure me. An instant, I had the impression that things had changed completely for the last 5 years that I retired ;-)
- Status
- Not open for further replies.