as ideal as possible doesn't mean its ideal or near ideal
even high end dacs are still far from ideal
even high end dacs are still far from ideal
I see from your posts you misunderstand a lot of the discussion - your bias is likely causing this!
The newer Benchmark Media DACs are pretty good, IMHO. I can record and playback here, and it comes out pretty close. Certainly closer than a tape machine, although those are nice sometimes for certain things.
Seemed like a very apposite characterization from DF96 to me.
Since we are discussing credibility and bias, do you have any financial interest in the sale or design of any DAC or DACs?
Although biased like any other human, DF96 is pretty good at being logical. And it looks like you have been making claims that you aren't willing to back up when called out on it. Its like you claim research exists, then you tell other people to go find it. If you know for a fact research exists in support of your claims you should just point to it with clear links and stop playing games. Either that or stop making claims you can't back up. Otherwise, you risk continuing to dig yourself into a deeper hole in terms of losing credibility.
Last edited:
Wow, guys, you're producing a lot of posts per 24 hours, it's difficult to follow... 🙁
Of course, after witnessing 3 tests that leans toward the same kind of results, i forged some opinions. How wouldn't that be?
First, the low-bitrate MP3 audibility threshold combined with the non-audible CD v.s. HD files differences...
Then, a 10$ driver that can mimic a 1500$ one, once equalized, within 400hz-7.2khz
And now, 19,99$ DAC that is indistinguishable from 3-4000$ ones...
... what will it be, next?
The market, the hobby, is filled with illusions. We took for given audible differences when they were different only on paper (at best)...
Of course, after witnessing 3 tests that leans toward the same kind of results, i forged some opinions. How wouldn't that be?
First, the low-bitrate MP3 audibility threshold combined with the non-audible CD v.s. HD files differences...
Then, a 10$ driver that can mimic a 1500$ one, once equalized, within 400hz-7.2khz
And now, 19,99$ DAC that is indistinguishable from 3-4000$ ones...
... what will it be, next?
The market, the hobby, is filled with illusions. We took for given audible differences when they were different only on paper (at best)...
Last edited:
As far as me having an agenda of any kind, for whatever reason you might think, well: fair enough. It's your right.
Then, just duplicate my test(s) and show your own results, with your own participants in your own testing room.
If any info is missing in order to duplicate, just PM me.
Then, just duplicate my test(s) and show your own results, with your own participants in your own testing room.
If any info is missing in order to duplicate, just PM me.
Based on information given, mmerrill99 is indeed having (or had) financial interest related to converters.
It looks like it from the very beginning. But what does that change, anyway? With or without financial interest, some people just won't accept such thing (market/hobby filled by illusions) because that makes past purchases and decisions looks bad or even foolish.
Thanks Jon, I must have missed that upthread.
I don't disagree that there are any number of reasons for bias, but a belief (or rather, I suppose, the requirement that other people hold a belief) that pays the mortgage is a particularly tough one to let go.
Some fora (not this one, as far as I know) require industry folks to identify themselves as such to provide other members a suitable 'caveat lector'.
"It is difficult to get a man to understand something, when his salary depends on his not understanding it." (Upton Sinclair)
I don't disagree that there are any number of reasons for bias, but a belief (or rather, I suppose, the requirement that other people hold a belief) that pays the mortgage is a particularly tough one to let go.
Some fora (not this one, as far as I know) require industry folks to identify themselves as such to provide other members a suitable 'caveat lector'.
"It is difficult to get a man to understand something, when his salary depends on his not understanding it." (Upton Sinclair)
Jon, you may have discovered what many people already know. Most people are not good at hearing low level distortion, it's just that many audiophiles believe they are good at it when in reality they aren't. However, there is a small-ish number of non-audiophile people who are good at it. Like anything else, you can think of it as a bell curve with only a few people out on the tails. So, doing tests with audiophiles doesn't prove what is or is not audible to a few people who do hear distortion much better than average.
However, when you make a general claim that something is inaudible, you are including those few people out on the bell curve tail. By framing your claims that way you end up with claims that are often false. You haven't come close to measuring the best people for a particular skill.
However, when you make a general claim that something is inaudible, you are including those few people out on the bell curve tail. By framing your claims that way you end up with claims that are often false. You haven't come close to measuring the best people for a particular skill.
Last edited:
How do you know when your other equipment is good enough for testing DACs? Or, to put it another way, how do you know when a DAC is good enough for testing other equipment?kinsei said:
You seem to be trying to argue it both ways. Even low end DACs are so good that only the very best ancillary equipment can distinguish them from high end DACs (if true, this of course means that for almost everyone there is no need to buy a high end DAC). High end DACs are "far from ideal".
It would have been helpful if you did not feign surprise at a result which, given your other tests, perhaps was not such a surprise. In this controversial area credibility matters, because any result is likely to face scrutiny from those who believe that you should have found an opposite result if only you had done the test 'properly'.JonBocani said:
Last edited:
This is what I said that you & others are looking for research papers to back up
First off, is anybody denying that ABX testing can suffer from the "inattentional deafness"? I gave examples of it earlier in my original link to a series of ABX tests run over a year which failed to register real differences that existed. It wasn't until a final test when some new participant pointed out what to listen for that differences were now heard by all participants.
There are research papers on "inattentional deafness" if anyone was interested enough to search for the term. The failure of the people asking me to provide these papers shows their lack of real interest in this
Inattentional deafness in music. - PubMed - NCBI
" Inattentional deafness in music is defined as the inability to consciously perceive an unexpected musical stimulus when attention is focused on a certain facet of the piece."
It's a well known & well studied phenomena in vision (Inattentional blindness) & recent research is beginning to show the same in auditory perception.
The example music piece being mentioned in that research is here https://static-content.springer.com...52-x/MediaObjects/426_2014_552_MOESM1_ESM.mp3
It seems incredible but when the experimental group had to count the number of timpani beats a significant number of them missed the guitar solo
"Our findings clearly show that sustained inattentional deafness exists in the musical realm, in close correspondence to inattentional blindness with dynamic visual stimuli."
So let's look at ABX testing where on each trial the participant is asked to identify if X is the same as A or B - does inattentional deafness effect the results because of the cognitive load in doing the test?
Well if anybody followed up Jakob2's reference to Dave Moulton
"Dave Moulton for example stated having noticed that "double blind test novices" even had problems at the beginning to detect level differences of up to 6dB ." HERE
You will see this stated
Although it's badly worded, if untrained listeners are unsure that there are differences they fail to find as much as a 3dB difference (sorry Jakob, it's 3dB difference). There are a number of psychological reasons for this but one is our self-esteem - desire to be seen to "get it right" if we know a difference exists i.e motivation to keep trying until we succeed. Let's say that we were given one of those visual spot-the-difference tests but not told if there was a difference or not. think of how quickly we would give up Vs how long we would spend if we were told there was a definite difference.
Now a 3dB difference is a pretty obvious difference - far more gross than expected differences between DACs.
If such gross differences are unnoticed by ordinary listeners does this part of my statement make sense "isn't so available to consciousness which is what is being focused on in such forced choice blind testing." ?
Now, you say what about this part of the statement ".we perceive differences because we are physiologically affected by the devices" - where's the research to back this up?
With the advent of fMRI (functional magnetic resonance imaging) & other non-invasive measurements of brain activity, new techniques became available for investigating what, if any physiological effects resulted from sounds that are considered inaudible.
"Inaudible High-Frequency Sounds Affect Brain Activity: Hypersonic Effect"
"The activation of the thalamus may reflect its function as part of the limbic system, which also plays an important role in the control of emotions
(LeDoux 1993; Vogt and Gabriel 1993)."
"We have incorporated these features in the two-dimensional sound perception model: sound frequencies in the audible range function as a message carrier
and frequencies above the audible range, together with those in the audible range, function as a modulator of sound perception through the brain systems, including the reward-generating system"
Really, there is a wealth of research out there but one has to be motivate to find it & read it. My reluctance to spoon feed any papers here was based on the lack of demonstrated motivation i.e. the unwillingness to learn or consider any viewpoints/research which didn't match their own bias/agenda(LeDoux 1993; Vogt and Gabriel 1993)."
"We have incorporated these features in the two-dimensional sound perception model: sound frequencies in the audible range function as a message carrier
and frequencies above the audible range, together with those in the audible range, function as a modulator of sound perception through the brain systems, including the reward-generating system"
Now, prove me wrong by at least reading the papers before objecting to the research I presented for you!!
Last edited:
Just to make things very clear: i'm not coming on diyaudio to prove or defend my credibility. I'm coming here to share information and opinions, hoping that will lead to interesting discussions. I respect this website and i think there is a high quality of discussions and members, but it's still ''only'' a discussion forum board, nothing more nothing less.
Deuzio, ''feign surprise'' is the funniest thing i heard so far in here.. Damn guys, seriously? ''feign surprise''... how did you come up with that ? 🙂
Ain't no Al Pacino act here, i'm not surprised, i'm downright stunned! I've got tens of thousands in gears here that are now iron-marked ''illusion'' all over... I spent several thousands hours listening to what i thought was ''better'', when in fact it was just more of the same...
I didn't know about the CD sounding exactly the same as HD24/96.
and that didn't make me believe midrange drivers would sound the same...And all that didn't make me believe a 19,99$ DAC could sound the same as a 4k one... It's surprises over surprises.
Feign what? No, seriously, what do i miss here?
Unlike our crusader-in-chief here, i'm not afraid of such scrutiny or opposite views, mostly because i don't care much about it if not constructive. There is always people who will find bugs in other people's work, that's the way of life...
I'm extremely confident the test was made properly, with enough resources and dedication, and also far from the casual-basement-type of test most people could do normally.
Again, don't take my word for it: i welcome any visitor to come and see, and go through the test by himself. It's an eye-opener to experience such test.
Also, page 56... Do you all remember that test is NOT over yet? It's still ongoing... No conclusions as of now, only a general statement ''No day & night kind of difference possible''
We should at least wait until we have a bigger sample of participants, don't we ?
We should at least wait until we have a bigger sample of participants, don't we ?
The paper above describes a very interesting result, thanks for that - frequencies above the accepted limit of human hearing affecting perceived sound quality and (not surprisingly, if you accept that mental properties supervene on brain states) brain activity. The subjects were (again, not surprisingly) not able to consciously identify the high-frequency content, but they were - explicitly - able to state that preference in the context of a blind AB test of short (three minute, separated by ten seconds) snippet. The effect absolutely did not require the "longer term" to become apparent, unless by "longer term" you mean six minutes. Nothing in there discredits blind, short-term AB testing; rather, it depended upon it.
Inattentional deafness is not a new concept, but again, how does that support the existence of the "longer term" audibility effects you posit? It simply means that if you concentrate on aspect "X" of a recording, you become less sensitive to aspect "Y". The example given does not relate to sound quality (the subjects were asked to count timpani beats); do you have any examples that relate to evaluation audio quality?
Dave Moulton's blog post isn't peer-reviewed research, but it's notable that in that piece he too concludes that blind AB testing is the best available option for subtle audibility effects, quote "So, I recommend that you depend on blind (or better, double blind) testing to find out answers to questions about the audibility of effects like 96 kHz. sampling rates or 24-bit words.".
In general, these links are either highly tangential to the hypotheses that there exist "longer term" audibility effects in the evaluation of sound quality, or fail to support it at all. I really hoped you had something directly pertinent / less speculative when you said your claims were "verifiable" by "perceptual research". Since at least one of your "references" is linked direct from a google.ie search, one wonders if this is serious presentation of your evidence or just a 'Gish Gallop'.
To return to my earlier question, since you have again brought up "bias and agenda" - which DACs have you designed and / or sold?
Thank you, that's better than before. Your best points in one place. If we could reset and take up any questions from here forward that might be good for the discussion.
They don't sound the same unless the 24/96 is a copy of the CD at a higher sample rate. On the other hand, if the CD was made from a higher resolution 24/96 master, they should sound different. How different? It depends on the source material. If slammed to tape and distorted, CD and 24/96 may not sound all that different. If clean cymbals and other clean sounds, if a hi-res master doesn't sound better than a CD made from it then something is very wrong and needs to be investigated.
Little update on today's test:
We might have found an interesting music excerpts that seems to help participants (2 so far) getting better results: Agnes Obel - Smoke & Mirrors - 17 first seconds
YouTube
Not enough rounds or participants to prove anything, and i will need to re-check the SPL-matching, but that's the most ''positive'' results i got so far from the whole test.
We might have found an interesting music excerpts that seems to help participants (2 so far) getting better results: Agnes Obel - Smoke & Mirrors - 17 first seconds
YouTube
Not enough rounds or participants to prove anything, and i will need to re-check the SPL-matching, but that's the most ''positive'' results i got so far from the whole test.