I'd agree with that. There's plenty of room for personal preference- though I'd suggest that this should be focused upon the speakers, not any other component.
What do you mean "no"? You've got this backwards. When you say "concentrating on hearing a difference", it means more focus on certain aspect. You should talk to Jakob2 regarding that "gorilla" video. In that video, when you focus on the ball movement, you will detect how many times that ball was tossed but miss the gorilla where as when you focus on the body movement, you will be able to detect the gorilla but miss the count of ball tossing.
This has what to do with audio DBT? Unless you are trying to reference McGurk Effect which was posted on this thread already.
Please let me know when you have gathered the info on those so that the debate can continue.
A mic/SPL meter reveals the reference output at your ear(s), unlike the signal going INTO your speaker boxes, albeit with a touch less precision.
B.
I suppose you´re referring to the famous swedish radio tests wrt transparency of audio codecs and Bart Locanthi identifies an idle tone that remained undetected in the swedish tests.
It depends.... 😉
As stated before, there are numerous examples and we have already discussed some of these thoroughly.
I´ve started with controlled listening tests back in the beginning of the 80s after reading some articles from Dan Shanefield.
In which way is stating a simple basic fact like presenting a straw man argument?
It is a simple fact, that _every_ experiment, which is used as a hypothesis test is prone to both error possibilies and it is good experimental practice against all error mechanisms, because we should be interested in _correct_ results.
False positives and false negatives are both incorrect results.
Audio is part of my business.
Last edited:
If you don't effect a change in any other part of the system (and you shouldn't!), the smarter and FAR MORE PRECISE/ACCURATE way of levels matching is to use the DVM, not a microphone. We're talking orders of magnitude more precise.
You seem to know more about these tests. So, can you please answer some questions:
When and where were these tests?
Were these double blind tests?
Which audio codec was used?
That wasn't the same kind of audio DBT (ABX) we are talking about. BTW, that report was touted by Robert Harley, the peddler of high-end audio electronics who has been trying very hard to discredit the value of audio DBT so that his business can flourish via deception. 🙄
See the following example of response which still doesn't cite actual example of audio DBT.
Dancing around the question... Yup, that is one of the traits among shill posters online.
Or are we talking about smart ways to ensure you aren't making mistakes in the way you test things.
For sure, a DVM will not tell you if the two DUT have matching colourations. Pink noise will tell you that in an instant.
Using a mic or your ears to judge loudness and your ears to judge colouration (in instantaneous A-B-B-A comparisons) is plenty accurate enough and sets the foundation for getting on with the next steps.
Ben
Last edited:
Robert Harley is one of the leading figures among high-end audio shills. You don't want to waste your attention on his marketing garbage. If you do, you will be filling your mind with garbage.
Mic closer, ears no way.
Pink noise to tell coloration in a DAC?! Instantly? What sorts of pathological effects are we talking about here?! 😱
Umm, looks like we need to start with Testing 101: you need to start this kind of equipment test by ensuring there are no irrelevant differences such as loudness (louder always sounds better) or in the tone compass (many folks just gush over rolled-off treble you get with tubes*).
Before you can start making wise comments about imaging and this and that mystery, you need settle that loudness and colouration as the same.
There's zero validity to listener judgments unless you first match for loudness and tone balance.... and then conduct quick-substitute comparisons.
Ben
*yes, soft-treble music can be very nice and there's no disputing taste; but you can make your solid-state gear behave like that with a touch of EQ and ditto for making your inexpensive DAC sound exactly like an expensive one
Last edited:
Err.... match voltage levels coming off your DAC (in this case it's a separate piece of equipment) with a DVM. Feed it a straight sinusoid (1k?) or something from a test disc (burn a disc with test tracks if you need to!).
That is testing 101.
That is testing 101.
We were talking about blind tests. ABX just happens to be a commonly used one (and one of the easiest to screw up IMO). Certainly the swedish one would have been much more rigorous than an ABX. And N=20k!!!
Everybody who reported was just passing on the original info published in the AES (at least that was where i 1st saw it).
And what does that have to do with the fact that it happened? A serious, well executed blind test and 20k people missed something so obvious.
dave