Well, it's good thing you did Jan. How could you say that about JC?!!! 😉
I'm glad things have slowed down here. I hope someone is doing some good blind testing out there. And maybe this thread will actually accomplish something worthwhile. I'd do some blind testing but us life challenged people sometimes can't do what we want.
Wouldn't the optimum damping factor be speaker dependent, especially in those pre-Thiele-Small days? There were certainly a lot of designs in the '50s with variable damping factor for just that reason.
The designer of the loudspeaker stated in print that the optimum damping factor of the loudspeaker was 1. I just followed his advice, and he was right. It was a simple 2 way speaker, and the speaker itself had an optional 4 ohm resistor installed in the speaker cabinet for this reason.
I don't know if you yet understand, but adding a series resistor to the output of an amp is not really enough to make more than a superficial difference in the sound quality. Tubes still sound like tubes, and solid state will grate on you, if it is not really well designed.
This is not the LOGICAL conclusion offered previously. It might fool you in a double blind test, however. The question is: WHY?
This is not the LOGICAL conclusion offered previously. It might fool you in a double blind test, however. The question is: WHY?
Last edited:
Jan, I did not see your input, but I can easily guess it. It rankles you that I take credit where I can, having done many audio experiments, before most of you had left elementary school. Well, too bad, as it just so happens that 40 years ago, I was making more progress in audio design than most of you will ever achieve in a lifetime. I was very fortunate to have the drive to do audio design, forgoing better paid jobs, in order to do what I wanted to do. I have also been fortunate enough to know both Richard Heyser and Michael Gerzon, both real geniuses in audio theory, to learn about negative feedback and its ill effects, and the inherent quality of old 78 recordings, respectively. Alas, this was all back 30-40 years ago, and they both have been deceased for years, both unfortunately, dying early.
Where do you think this early understanding of what is not quite right about audio design comes from? ME? No, I only do circuit topologies, and I am darn good at it, as most of you copy my early work, whether you know it or not.
This is the true situation, either accept it, or tell me why you shouldn't.
Where do you think this early understanding of what is not quite right about audio design comes from? ME? No, I only do circuit topologies, and I am darn good at it, as most of you copy my early work, whether you know it or not.
This is the true situation, either accept it, or tell me why you shouldn't.
Sorry, i didn´t made it clear; what i really meant was, that the argument is based on the fact itself and can not be refuted due to the way the fact was established.
I understand that, but i gave some examples where people _have_ _to_ _trust_ their senses while otherwise they would not be able to do their work.
I didn´t mentioned musicians because i think they were generally be better in evaluating reproduction gear, but only because they show that humans have a very good aural memory (or could have) even for very small variations and second, that level variations will not swamp out even small differences in the presentation of music.
You will have noticed that in this thread often it will be stated as a fact, that aural memory is weak and that even small level differences will ultimatively lead to wrong results.
(Just to make it clear again, in a scientific experiment level differences were not acceptable, unless the experimentator will incorporate some counter measures against this confounder)
Our hearing sense works in a complex manner (just because our brain is involved) and if one doesn´t see it as a holistic system, one will most probably lose the general picture.
Unfortunately, discussion like in this thread tend to oversimplify certain aspects.
Good example. Let me try to say it this way; Toole & Olive have shown, that for loudspeakers rating and even ranking were strongly influenced in open tests. So you have the positive test result, that you _must_ _do_ loudspeaker evalution in a blind test.
Please compare that to your question above? 🙂
Would you still insist, that it isn´t possible to learn to deal with the various bias mechanism?
There is just a difference between scientific investigation and personal choice.
It seems to me that you´re choosing exactly the same option for loudspeaker evalution as Curly does for cable evaluation.
Yes, that is quite common. I remember a discussion with SY where we both cite Burstein´s comment on the stereophile dbt on amplifiers from memory; after rereading it, it became clear that we both were wrong, but each of us in a direction that supported our position. (And of course his version was slightly more wrong than mine 🙂 .....just kidding)
BTW, Burstein´s comments are still well worth a read.
You´re right, humans can be really bad observers, but it seems to depend strongly on confounders; obviously humans can be extremely sensitive observers if it is ensured that they were able to be in the right state of awareness.
You will remember the argument that a difference can´t be of practical relevance if not detected in a dbt by the participants.
I only posted the link to make clear that this conclusion (that seems to be reasonable at first) can be dead wrong.
I´m using the term "trained" usually with the notation "trained under dbt conditions" .
You (as most others) argue with the implicit understanding that everyone must be able to listen as sensitive in a DBT as usually under normal listening conditions.
The question is, why that should be the case?
The reason for doing a DBT is that a specific person can´t control the various bias mechanism; the same person (just a minute before a marionette in hand of his bias) transmutes to a perfect observer, being in a zen state of awareness to detect _any_ audible difference presented. 🙂
Does that seem reasonable? Really no need for this person to get used to the (sometimes very) strange/unusual conditions of a specific test protocol?
Really no need to see, if some training under this specific test conditions does improve the listeners ability to detect differences?
That is all, what is meant by me with the term "training".
That´s why i wrote, that both sides should admit, that the other side _might_ be wright. (even if they think it is unlikely)
Just to quote jneutron on this point "if you bring science to the table than it has to be real science" 🙂 (i really like that one, because it in short enlightens the much higher responsibility for the objectionistic side of the discussion; if i am an objectivist i can´t abandon the rules of science just because i _believe_ that i am right)
That might be the case; that´s why a test should be as ´relaxed´as possible, but as strict/controlled as needed, but in any case it has to be shown, that a participant in this test is reaching a sufficient level of sensitivity for the task -> that is the reason why positive controls are mandatory.
Negative controls are needed to show that a difference detected was established by the EUT and not anything else.
Just to repeat the point from above; because it is not guaranteed that a specific test protocol is not such a confounder that the results are falsified.
Wishes
Last edited:
The truth is: What is, is. It doesn't take double blind tests to PROVE that no differences in wires exist to non-believers. They just have to keep that opinion. The strict double blind tests are configured to give null results. And if for some reason, they don't give null results, the test will be worked over, until it does give null results. This is because the tests are designed with a null conclusion in mind, from the start.
This is made by taking out type 1 errors, and increasing type 2 errors, and therefore confusing the test participant. Prove me wrong!
This is made by taking out type 1 errors, and increasing type 2 errors, and therefore confusing the test participant. Prove me wrong!
Not sure what I am trying to prove wrong here. But one thing you seem to be wrong about is that there are a lot of people that think different cables make absolutely no difference. I don't think there is a single person in this thread that would deny you can engineer a cable that will certainly produce an audible difference. Minimizing variables like LCR will minimize type 1 errors yes but the reason for doing this isn't to somehow fool you and increase type 2 errors. It's to make sure you are actually testing for what you want be testing for - for example standed vs solid, this dielectric vs that dielectric etc....
Well, let's run the experiment. Maybe he can. John? (the Leventhal paper, BTW, makes no such claim, just in case that's what you were thinking)
Well, nothing PEER REVIEWED that's for sure. But isn't it obvious? Read Les Leventhal (sp) and his statistics evaluation of ABX testing. That's a start. I had dinner with Les once, and he explained it to me.
Well I am surprised to find that cables do something I had not guessed at.
They get a bunch of smart guys in to long, silly arguments! Powerful mojo, have they.
They get a bunch of smart guys in to long, silly arguments! Powerful mojo, have they.
Well I am surprised to find that cables do something I had not guessed at.
They get a bunch of smart guys in to long, silly arguments! Powerful mojo, have they.
😛
Ok, next. Rod Rees' articles in 'The Audio Amateur' in the late 70's and early '80's.
For example: Try 1,1982 'Auditory Test Methods and Science'.
For example: Try 1,1982 'Auditory Test Methods and Science'.
0 hits on google for 'Auditory Test Methods and Science'. I have tried to look some others you have referenced and only found fragmented copies and just gave up in frustration.
Bo
This is called setting up a strawman. You are probably aware of the folks who find CAT5 cable improves their sound system. Cat5 is spec'd at .188 ohms/m. Some folks also make sure both speaker cables are the same length. So in a not unrealistic example 8 meters of cable could be used per loudspeaker with all four conductor pairs in parallel for a total resistance of .376 ohms. With a single 8 ohm loudspeaker rising to 16 and dipping to 4 the difference would be (-.2017 - -.7803) or .5786db. You can certainly argue that some would not hear that. But I have been present when such adjustments are made and they can be heard. (Do not confuse detecting level differences with tonal balance level differences.) Of course there also are people who use longer cables, have larger differences in speaker impedance or smaller cables. In a concert system it is unlikely any of the cables are 8m or less and the gauge used is usually 13ga. (That is what Neutrik speakons take.)
As to the inductance and capacitance these factors may have some influence on the loudspeaker connection, but in my OPINION you should be looking at the interaction of the cable soakage and time energy dispersion to the amplifiers feedback path and associated errors.
- Status
- Not open for further replies.