I m musician and i think that there s people who are too much paid to make what is often rubbish music as there s writers that could write rubbish litterature as well..
To get back on topic of audibility of differences let say that from age 25 to 65 loss of sensitivity for men is 1dB/year at 8KHz, and 0.63dB/ year for women.
So so called golden ears are not even hearing the same thing from a year to the next one if not 5 months later since the threshold of level variation audibility is 0.5dB.
@ Jan.Didden,
just from the website it is indeed not so apparent, that the user will download an article about an experiment done with software and that the screenshot is probably just an illustration from that article... .
It´s always surprising what people might think although oneself would never consider something similar....
just from the website it is indeed not so apparent, that the user will download an article about an experiment done with software and that the screenshot is probably just an illustration from that article... .
It´s always surprising what people might think although oneself would never consider something similar....
Well, all right.
How you value this compared with an interview found in a HF magazine?
Sam Merril: Mason Hoffenberg Gets in a Few Licks
George
PMSL. You have been on the forum longer than me so you know more? Is that how it works? Chortle.
Actually it´s the other way round. If only one (or few) participants are true observers while all the others are only guessing (under the conditions of the test) then adding more and more guessing listeners will ensure that the null hypothesis will be retained.
That "trick" was used in Fremer´s amplifier challenge at the AES convention.
Increasing the number of trials will help if the true parameter of the underlying population (for the specific test conditions) is close to the number taken for the null hypothesis. Increasing the number of trials will then raise the power of the test.
Yes, this is valid only for a certain type of test. You are correct about the supposed GE crowd skewing the initial result, however I did not assume such a crowd was necessary part of the test. Just regular people randomly selected, result is 50-50 (null) then JC or bear or jakob2 are jumping in and claiming you ignored the 50% positives which may hide some truth (that SOME people can truly hear the desired outcome). You separate those 50%, repeat the test and you got a 50-50 null again. If so, you just quartered the probability that the original null was "biased toward null" or that the result was anything but a true statistical null.
Kind of a "positive control" that I am trying to find a practical example of. Reduces the probability that some statistically significant result is hidden by an (e.g.) odd distribution (non-Gaussian) property, in the population under test.
If there's a non-Gaussian property, it would be rather extraordinary for the addition of positive responders to turn the distribution normal. Of course, cherry picking one end of the distribution, adding results together, then claiming significance is a remarkable sort of dishonesty, but people do that anyway.
I just did a confirm on the Foobar ABX test.....it really is a total POS.
The difference between when a track is played in the player or the ABX utility is astounding.
A blind man would not be able to tell shite from clay when using this piece of smeg.
All Foobar ABX test results are invalid.
Dan.
The difference between when a track is played in the player or the ABX utility is astounding.
A blind man would not be able to tell shite from clay when using this piece of smeg.
All Foobar ABX test results are invalid.
Dan.
Last edited:
When are you going to make a contribution that is more than sniping?
Something with information or content?
You were preaching to me, not the other way around.
And, btw, perhaps you need to get away from your computer, and out more often...
I know, post an image of your "system"?
Or maybe an electronic/audio bit you've built?
Give urself some gravitas.
_-_-
I'm confused. The Foobar ABX test doesn't reroute the audio, just randomize the playback. Are you saying Foobar sounds bad? (personally I don't like the interface but to my knowledge the underlying playback software is sound.)
And... I looked at your stats, billshurv. 7,000 posts since you joined.
Average 7+ posts per day.
Most in the "Lounge".
Unless they are in this thread.
Most disparaging some thing or another, at least that is what a fast pass seemed to say.
You have Apogee Centaurs as your speakers - at least you did.
And, you spin vinyl. Many posts in that area.
Interesting...
In contrast I've made a bit more than 6,000 posts since 2002.
Average about 1.3 per day.
While I do own bigger and better Apogee speakers than you have, I consider them to be around for yuks and fun, since they're incredibly colored and flawed... someone else probably owns even BIGGER Apogees, I know. Someone will doubtless be insulted about these speakers, and also will be of the opinion that they are not colored or flawed...
<sigh>
Average 7+ posts per day.
Most in the "Lounge".
Unless they are in this thread.
Most disparaging some thing or another, at least that is what a fast pass seemed to say.
You have Apogee Centaurs as your speakers - at least you did.
And, you spin vinyl. Many posts in that area.
Interesting...
In contrast I've made a bit more than 6,000 posts since 2002.
Average about 1.3 per day.
While I do own bigger and better Apogee speakers than you have, I consider them to be around for yuks and fun, since they're incredibly colored and flawed... someone else probably owns even BIGGER Apogees, I know. Someone will doubtless be insulted about these speakers, and also will be of the opinion that they are not colored or flawed...
<sigh>
Bear it seems you are asking the impossible in my eyes. There is no perfect reproduction method that exists today, so to say you can set a standard by which perfection and a so called Blameless system can be assembled from individual components seems to be fairy dust. Since there are no perfect speakers that don't color in some sense the sound of every system how can you determine the blameless electronics specifications that will work with random speaker systems.
I don't work on the electronics side of the equation, I only have skills on the mechanical speaker side, the rest I have to leave to others judgement that I respect. I do understand on a general basis what is being done but can't comment on the minutia of the difference in one type of resistor vs another, I can't go there.
But I will say since we have no agreement on the simple fact of the best type of reproducer to use whether point source or omnidirectional or even planar how can you expect there to be a blameless standard for electronic components. There are in my eyes no golden ears, there are some who are better trained to hear specific things but to say someone has calibrated hearing is just to far from the truth.
I know far to many musicians who you would think would have the training and would have the pitch and tonal discrimination who could care less about how good a system sounds as long as they get a certain visceral feeling when listening to music.
Your questions are impossible to answer in the absolute, there is no there there.
Sorry Sy,
I took the bait.
I don't work on the electronics side of the equation, I only have skills on the mechanical speaker side, the rest I have to leave to others judgement that I respect. I do understand on a general basis what is being done but can't comment on the minutia of the difference in one type of resistor vs another, I can't go there.
But I will say since we have no agreement on the simple fact of the best type of reproducer to use whether point source or omnidirectional or even planar how can you expect there to be a blameless standard for electronic components. There are in my eyes no golden ears, there are some who are better trained to hear specific things but to say someone has calibrated hearing is just to far from the truth.
I know far to many musicians who you would think would have the training and would have the pitch and tonal discrimination who could care less about how good a system sounds as long as they get a certain visceral feeling when listening to music.
Your questions are impossible to answer in the absolute, there is no there there.
Sorry Sy,
I took the bait.
Me too, maybe get the developers to confirm there is only one path to the hardware? I would expect the ABX plug-in has no code involving the actual audio data.
Max - I don't see the "you are deaf or a liar" arguments doing any good for your cause. Does anyone else here share that extreme view?
I'm saying that I can load a track twice into the Foobar playlist.
I can then assign those two copies into the ABX plugin.
The tracks played directly from the playlist sound fine.
The tracks played from the ABX plugin sound bad, really bad in comparison.
I picked up on this a while back, and confirmed it just now.
Something is really wrong with the software.
Dan.
And you have picked up what has escaped everyone else! Remarkable work.
- Status
- Not open for further replies.