Before I start this writeup, let me say thank you to all that participated in this audibility test, and to Doug P for allowing it to happen at the Iowa event. It was highly appreciated.
For a lot of the details on the test I conducted, you can find a lot of info in the Attitudes thread:
For the tracks utilized:
I did measure the caps used in the xovers, and they reflected a very tight measurement between the types actually used in this test. The capacitance value should therefore be negligible. ESR was not balanced, and therefore should be seen as differences between caps and their manufactured properties.
There were 3 tests over the course of a half hour. All were of different circuit positions, caps, track timings and orders, and musical selections. Therefore each of these tests could have a different affect on the outcome in terms of results. I am not revealing the caps used in this test at this time. Xovers were not visible, nor was what I was doing in the case to the attendees. They were also not crowd visible or viewable in their entirety until after the test. A few audible cues were used initially, and then removed for the X portion.
Lights were dimmed to remove some of the visibility of the test, and allow the ears to function more independantly of visual cues. The crowd would not let me leave off the lights completely.
Crowd verbal opinion, post-test:
I did get a lot of positive feedback from participants, saying they heard differences, even if they could not identify the X in the test. Being this is subjective, discount what you feel, but they said this was the case. Most of the people that volunteered info or those I asked of it, came to the conclusion that the first was actually the easiest to tell, the second was less so but also distinguishable, and some could not tell a thing about the last test.
That all said, I will now reveal a lot more detail....
The first test involved switching of alike cap makes/models in both the midrange and tweeter circuits with another set from another make/model. These were placed in the last series cap position for both the mid and tweeter, from a 3rd order electrical highpass. (IE- one cap in the midrange, one cap in the tweeter, both in the same electrical position; switched at the same time.) Switches were heard between A and B initially, and not heard when going to X. This was to see if the difference was easier to detect over more than one driver, and to allow the harmonics of the adjacent driver to also be affected simultaneously over the course of the test.
I received a few subjective written comments on the ballots that B was 'clearer/more dynamic', while A was 'edgier'.
Tracks used were 1,2,3 in the link above^. The order of tracks was the 3 tracks for A, then 3 tracks for B, then 3 tracks for X; Answer was B.
Of the 26 ballots, I had one ommission, 8 answered correctly, and 17 answered incorrectly. This apparently was not as easy as the groups' reflected opinion stated. Still to be noted is that the crowd as a majority thought they heard a difference.
This was a single cap swap in the woofer's shunt path. You can probably guess what I used of certain type and be correct, but I'm still not revealing the actual ones involved. Switches were heard between A and B, and rapid-fired to X in a large number of clicks to avoid detection, over 3 switches in the case. Unless you are a robot or supercomputer with AI, there is not a feasible way to decipher the stopping position. This was to see if the normally accepted woofer cap is really okay to the attendees.
Subjective written comments consisted of B being 'less detailed' and 'recessed vocally', while the A was 'better on the Norah and RyG tracks'.
Tracks used were 2,5,6 linked in the list above^. Order was 5-A, 5-B, 2-A, 2-B, 6-A, 6-B; 5/2/6-X. Answer was A.
Of the 26 ballots; 6 omissions, 13 correct, 7 incorrect. This one was apprently fairly easy.
This test was a single cap swap between 2 differingly constructed caps. The position was the first cap in the 3rd order high pass after the attenuation resistor. This was to see if upstream components made a bigger difference.
Subjective written comments included: 'B is clearer on drums', 'A is whiney, B is warm', and 'Violin better on A, while bass better on B', and 'B is an attenuated top-end'. (I can't see the last one being true, as the tweeter was not affected.)
Only track 7 listed above was used, and 2.5 minutes for A, B, then X. Answer was B.
Of 26; 11 were correct, 8 were incorrect, 3 omissions.
This means that while 3 people answered all 3 correctly, 3 other people had omitted answers but were correct in what they did answer. So, 6 people scored 100%.
3 people answered all incorrectly, or 0%.
9 people got 66.66% correct, and 8 people got 33.33% correct.
Until I update with plots for everything, this will remain how the results actually stand.
Thanks for reading...
Login or Sign Up
- Log in with