That's likely why you don't hear differences between DACs. You're listening for the wrong things.
Simply that you are implying that the same must apply in this test. It may be the case that he is biased (who alive isn't?) but he didn't start this thread with the same statement and has repeatedly denied it to be the case. You have stated your view a number of times, as has he, I don't see any value in the repetition
Unfortunately, most of the time, yes. I enjoy building loudspeakers and experiment with various transducers. But I will try next time.
I think that requires extended listening and then tricky not to find the change an improvement, although again, given enough time switching between the two it should be possible to eventually come to a preference
Great - doesn't hurt to try.
Have you ever not heard some anomaly in a speaker until someone pointed it out to you? Was it then very obvious & unavoidable to hear? Why did you not hear it yourself without it being pointed out to you?
This, if it ever happened to you, is called inattentional deafness - we are essentially deaf to differences which are there but we are focused on other aspect(s) of the sound.
Yes it's not a forced choice 6 second echoic memory test which a lot of people focus on in ABX.
The main factor is that the listener doesn't feel he has to perform & strains to hear a difference - it's just listening for pleasure as we do normally. I don't necessarily agree that the change will be perceived as an improvement & anyway A & B are presented in random order for each trial. it should also use many selected different tracks or selected parts of tracks so boredom doesn't become a factor.
Last edited:
When you say the Filo DAC is good, do you mean it sounds better than some other DAC?
tomahack said that I was missing the point even after I quoted the pre-judged test posting. I wanted to clarify that I was speaking of a past test, not this one.
It doesn’t appear that you understand how harmful bias can be when trying to conduct scientific investigation. These sorts of tests must be very carefully implemented in order to be valid. Do you understand that?
If the person conducting the test has a predisposition to a null result, he might push the results that way, even if he does it unconsciously, right?
If he believes he knows the outcome beforehand, if he believes that humans have weak hearing abilities, these prejudices might be reflected in how he conducts the test and/or interprets the result, don’t you think?
The reason the second “blind” is included in a “double blind” test is to try to prevent bias, unconscious or not, from tainting the results.
Repetition? On a forum? Are you daft?
Does it not bother you that the OP doesn't understand what a null test can and cannot demonstrate, when he is designing, implementing and interpreting a null test?
Last edited:
There's a lot of talk about blind testing eliminating biasing but a bias towards 'not hearing any difference' is one bias that blind testing (single or double) does not eliminate. If one believes that there is no difference to be heard between cables for instance & the participant knows that cables differences are being tested for, then the almost 100% outcome is that no difference will be heard - the motivation just isn't there - it's like the Hydrogen Audio forum moderator said when he claimed random pushing buttons to finish a blind test was perfectly acceptable - "Life's too short"
In that case I hope the preference being completely random and changing by the hour/day is an acceptable result. Let's also not forget the person with 2 TT's and 4 different cartridge/pre-amp combinations where each LP has a different prefered setup and all sound different in a way that careful DBT is not necessary.
Last edited:
Well if it was completely random then it would be a 50-50 preference i.e no preference, right?
Huh?
You really don't get that? People attach distinct preferences to highly colored components or combination of components where the simple difference test is a no brainer and none are conventionally "accurate" by a considerable margin.
If A and B are truly indiscernible, then you are correct, of course.
Assume, for a moment, that the test that led to the conclusion of indiscernibility was inherently flawed, that the very mechanism of the test itself tended to preclude the answering of the question that was posed, and/or that the particular implementation of the test was compromised in some way. In these cases the test hasn’t shown anything, so we cannot assume indiscernibility.
The preference test is instead of the ABX test, not a complement to it.
So, first off, you are saying that people's preferences don't randomly fluctuate but they actually do have a consistent preference based on sound? This is not what you said just a statement before this.
You are stating that this preference has nothing to do with "accuracy" - who said that any of the testing in this thread is about accuracy? The o/p stated that his test demonstrated no difference, nothing about accuracy. I suggested an A/B preference test, again nothing to do with accuracy but if a consistent preference is determined with the same two DACs the o/p used what does this tell you?
Answers on a postcard, please!
Last edited:
Are you saying that this person can hear differences, or is simply self-delusional?
A well designed and implemented test could tell you, either way.
In either case, this is about preferences. Since we cannot recreate, in the room, anything close to a replica of an actual concert performance, it comes down to preference.
Do you prefer the deviation from "reality" this way or that way?
Personally, I don't hold self delusion in disdain. Presumably one gets one's money's worth.
I recommend reading or watching Nelson Pass on the issue of preferences.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.