Mcintosh integrated amp MA6600 for B&W 803D - Page 4 - AVS Forum
Forum Jump: 
 1Likes
Reply
 
Thread Tools
post #91 of 195 Old 07-23-2013, 12:02 PM
Member
 
sh25ep's Avatar
 
Join Date: Jul 2013
Posts: 30
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Brownstone322, thank you for posting the details of the ABX test. To respond on some particulars of your post:
Quote:
It doesn't matter whether or not they were actually guessing. They can in fact be guessing when they think they're not. It's semantics. With this data, there's no evidence that they were doing anything other than guessing.
If it doesn't matter and there was no evidence, then why do you end with they are all just guessing? With this data, there's no evidence that they were just guessing. It could be said either way.
Quote:
the valid methodology is to assume we can't hear differences and test to see if we can. The conclusion from a test like this would be that the data does not support the claim that the amplifiers had audible differences. That is what the data shows us
That assumption is geared toward the majority's desired conclusion. And the conclusion is derived from that assumption. The data does not show us anything other than half the people choose correctly and the other half didn't.
Quote:
We have no evidence that the users heard differences to begin with. Just because they claimed they could (in a single trial) doesn't mean anything.
If you thought that something sounded better than something else, what evidence would you possibly have on hand to give somebody to prove it? It would just be a case of if someone believed you or not.
sh25ep is offline  
Sponsored Links
Advertisement
 
post #92 of 195 Old 07-23-2013, 12:34 PM
Member
 
sh25ep's Avatar
 
Join Date: Jul 2013
Posts: 30
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
arnyk, thank you for blasting me and posting those numbers, however, my posting was addressed toward diomania as a sort of slap since he scratched me earlier.mad.gif The questions therein were rhetorical to all but diomania.
Quote:
Audio is both a science and an art.
I understand what you are saying. The aspect of audio that I was referring to is the sound tests. Thank you for the correction though.
Quote:
You appear to be tacitly admitting that you lack professional standing and/or relevant education.
I have zero professional standing in the audio field. I am a customer of the products.
I have some relevant education. I have years of end user experience.
I don't know what the word "tacitly" means.

I think that akhter was being sarcastic in half of his quotes you commented so seriously on.rolleyes.gif
sh25ep is offline  
post #93 of 195 Old 07-23-2013, 01:08 PM
Member
 
Brownstone322's Avatar
 
Join Date: Jan 2011
Location: The Old Dominion
Posts: 183
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 5 Post(s)
Liked: 62
Quote:
Originally Posted by sh25ep 
If it doesn't matter and there was no evidence, then why do you end with they are all just guessing? With this data, there's no evidence that they were just guessing. It could be said either way.

Again ... until the listeners can demonstrate a statistically significant level of correct answers, then the assumption (the correct assumption) is that the listeners cannot distinguish such differences and are, willfully or not, guessing. That's not my opinion. That's a statistical fact.

Quote:
Originally Posted by sh25ep 
That assumption is geared toward the majority's desired conclusion. And the conclusion is derived from that assumption.

It has nothing to do with "desired" anything. It is a reflection of empirical data.

Quote:
Originally Posted by sh25ep 
The data does not show us anything other than half the people choose correctly and the other half didn't.

Actually, that's not even what the Stereo Review data showed. It showed that no individual could make the correct choices more than 60% or so of the time (most more like 50/50). If you're suggesting that half the listeners were consistently right and half were consistently wrong, then you are mistaken. (Half right/half wrong is otherwise own as "guessing" or "flipping a coin"; it's of no further value.)

As for the System A vs. System B test, there was only one trial, so it's worthless. (Even then the data distribution was near random. Shocker.)
Quote:
Originally Posted by sh25ep 
If you thought that something sounded better than something else, what evidence would you possibly have on hand to give somebody to prove it? It would just be a case of if someone believed you or not.

Wanna prove it? Simple (and it's the third time I've pointed this out to you): You'd need to demonstrate that same preference through a repeated series of random tests. Anyone can claim anything in one instance. An isolated trial can't prove anything because (here we go again) you don't even need to be there to make a claim -- you could flip a coin and state your preference by e-mail. What is it about that point that you don't understand?

So you want someone to believe you? Then identify the same system as "better" five or 10 or 20 times in a row. That would be of statistical value. Should be easy, right? Show us the guy.

I don't care what you believe personally, because I recognize "audiophilia" for the religion that it is -- a non-scientific belief system that's largely made up. But it is unfortunate that you don't have an understanding of basic research methods and insist that you do.
Brownstone322 is offline  
post #94 of 195 Old 07-23-2013, 01:48 PM
Senior Member
 
syd123's Avatar
 
Join Date: Feb 2009
Location: mid-atlantic region of US
Posts: 449
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 34
Quote:
Originally Posted by syd123 View Post

Subjectivists such as yourself must find it quite deflating that not a single amplifier, cd player, or cable mfg. has ever cited in their advertisements tests in which blinded subjects picked their gear over a competitors? ..If their amp truly sounded better than a handful of competitors, then why not state this in your ads? What could possibly be more compelling?

And I still don't understand why you believe that because two different amps come from the same manufacturer (e.g., the McIntosh MC302 and MC452) they will automatically have precisely the same output levels for a given gain setting on the pre-amp. This may be true if they had the same wattage output and same input sensitivity, but the MC302 and 452 do NOT have the same specs!!! ..They differ by 150w/ ch. and have different input sensitivity ratings!! Please explain why you feel they're "volume-matched".

While I stand corrected on my second point about amps w/ different wattages having similar output levels for a given input gain, I'm still waiting for a subjectivist to answer my first question:

So why don't gear companies cite DBT results to support their claim that their gear sounds better? ..You know, like drug companies do in supporting their claim that their drug performs better than placebo? Anyone?
syd123 is offline  
post #95 of 195 Old 07-23-2013, 02:22 PM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 14,382
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 754 Post(s)
Liked: 1167
Quote:
Originally Posted by diomania View Post

Quote:
Originally Posted by sh25ep View Post

The MC302 and MC452 will not have "different output levels for a given volume setting". The output is matched at Mcintosh.
How closely are they matched at McIntosh?

Hard to say because Mac specs the two amps slightly differently.

Might be a few 0.1' dBs off.
arnyk is offline  
post #96 of 195 Old 07-23-2013, 02:57 PM
Member
 
sh25ep's Avatar
 
Join Date: Jul 2013
Posts: 30
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Brownstone322
Thank you for trying to explain it further, but I know and understand exactly what you are saying. But there are more possibilities here than you are stating.

What would a "statistically significant level of correct answers" be? The assumption that they are guessing is guessing that they are guessing. Willfully or not....
It is not a statistical fact that they are guessing since the only numbers we have are that 54% were correct. Not that 54% were guessing.
And since half of the people believed that they wouldn't hear any difference, before the test even began, how does that skew the statistical result?
Quote:
You'd need to demonstrate that same preference through a repeated series of random tests. Anyone can claim anything in one instance.
Yes I see what you're saying and I too would've liked to see more tests.

But more possibilities to consider: If, during subsequent tests they (or some of them) decided to change their minds and liked the other amplifier better, then what would that mean? And if, some of them during subsequent tests had moved out of the camp of undecided and chose an amp, then what would that mean? And if, some of them during subsequent tests moved from liking one amplifier to undecided, then what would that mean?

All three of those things would have happened because as people listen over time and through subsequent tests, they start to notice things in the music and sound stage and differences in frequency, detail, clarity, and depth. All of which they don't always notice right off the bat. It depends on the individual and what part of the sound they are concentrating on. People will change their minds and won't pick the same one 20 times in a row. And it doesn't matter. They will eventually pick the better one with time. Or at least some people will. And the ones that don't....well maybe they can't hear the difference. But some people will.
The test was which one did they like. And the only way they would have known is through the differences in the two systems.
Quote:
An isolated trial can't prove anything because (here we go again) you don't even need to be there to make a claim -- you could flip a coin and state your preference by e-mail.
Are you saying that they were just picking one? Even when they had the option of choosing neither? They were all there listening and didn't flip a coin or send it by e-mail. They were there. Listening. If you do not believe the particulars of the test, then of course your conclusion will be based on what you think is questionable data. This is understandable but still comes down to rather or not you believe them.
Quote:
I don't care what you believe personally, because I recognize "audiophilia" for the religion that it is -- a non-scientific belief system that's largely made up.
If you told me that you can not hear a difference in amplifiers, then I would believe you.
If I told you that I can hear a difference in amplifiers, then you would say "where is your proof?"
It is a religion.rolleyes.gif I do pray and give thanks to the great Mcintosh god. And the beautiful goddess of speakers known as Paradigm.wink.gif
It may be a "non-scientific belief system" but it is not "largely made up". It is based on my own personal experiences and many many hours of testing over the years. My personal experiences and sound tests mean more than other peoples tests of which I can find flawed conclusions based on indecisive numbers and assumptions.
Quote:
But it is unfortunate that you don't have an understanding of basic research methods and insist that you do
I have?
sh25ep is offline  
post #97 of 195 Old 07-23-2013, 03:17 PM
AVS Special Member
 
kiwi2's Avatar
 
Join Date: Nov 2010
Location: New Zealand
Posts: 1,630
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 6 Post(s)
Liked: 144
Here's an interesting read from a few members that did some A/B testing a few days ago...

http://www.avsforum.com/t/1480939/audible-differences-between-2-channel-s-s-amps-and-avrs-when-operated-below-clipping/90#post_23550877
kiwi2 is offline  
post #98 of 195 Old 07-23-2013, 03:32 PM
 
diomania's Avatar
 
Join Date: Dec 2008
Posts: 1,389
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 49
Quote:
Originally Posted by sh25ep View Post

diomania, you said the the "Same 2 amps don't put out exact same voltage at the output either."
How do you know this? Did you have 2 of the same amps? How old were they? What kind were they? What did you use to measure the outputs? How old was the measuring device? Had it been recently calibrated? How far off were they?
Stalling, eh. rolleyes.gif I asked you a bunch of questions to your claims and it seems that you don't know the answers. That's typically the result of making things up as you go.
Quote:
How closely do you need them to match?
You said, "I did this with NAD and Rotel amplifiers and they matched." So I asked, how closely are they matched and what method did you use to figure that out?
http://www.avsforum.com/t/1415224/mcintosh-integrated-amp-ma6600-for-b-w-803d/60#post_23555371
Quote:
What method did you use to figure out that "Same 2 amps don't put out exact same voltage at the output either."?
Either you have a reading comprehension problem or short memory. Both are bad news for you when trying to communicate on internet forums. http://www.avsforum.com/t/1415224/mcintosh-integrated-amp-ma6600-for-b-w-803d/30#post_23550543
Quote:
How did you come up with that tolerance figure?
Now it's your turn to answer my questions in the links above. Once you do, I'll answer yours. By the way, it's ok to admit it if you don't know the answers. What say you?
diomania is offline  
post #99 of 195 Old 07-23-2013, 03:39 PM
Member
 
sh25ep's Avatar
 
Join Date: Jul 2013
Posts: 30
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
syd123
Quote:
So why don't gear companies cite DBT results to support their claim that their gear sounds better?
You mean like on their website? I don't really ever see any other advertisement for their products.
If they did such tests and posted them on their website, and it was in their favor, would you believe them? I wouldn't take their word for it.

When you're talking about this high dollar stuff, most people are going to go into a store and listen to it anyway and make up their own minds.
I mean it's an interesting thought (and possibly argument) that they could do it, but at the end of the day it wouldn't really matter.
sh25ep is offline  
post #100 of 195 Old 07-23-2013, 04:18 PM
Member
 
sh25ep's Avatar
 
Join Date: Jul 2013
Posts: 30
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
diomania, you're the one that stated first that they didn't match. Based on no further information from you.

I have told you of my experiences.
You're looking at me for a bunch of numbers. Go get them yourself.
Quote:
You don't believe me? Just put a voltmeter at the output of 2 of the same model amps and play a test tone and see it for yourself
You're the one making the claim. You go put a voltmeter on 2 amps of the same model and report back here.

The sun is cool to the touch. You don't believe me? Just go stand on it.

You're telling everybody that doesn't happen to have 2 amps of the same model on hand to either just believe you or prove you're claim for you. Why would they? Why would I?
Quote:
Either you have a reading comprehension problem or short memory. Both are bad news for you when trying to communicate on internet forums.
You just have totally missed the point of my post.

The questions I had for you are obviously ridiculous.
I worded them the way that I did, because your questions were not so obviously ridiculous.
sh25ep is offline  
post #101 of 195 Old 07-23-2013, 04:20 PM
Advanced Member
 
akhter's Avatar
 
Join Date: Aug 2005
Location: Hong Kong
Posts: 563
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 17
All this typing and the double-blind fanatics have convinced...no one (atleast none that have spoken up on this thread)...since the start of the thread till now.

I applaud the fanaticism to show us the light though...

As I said I am not here to convert anyone and hopefully some people will find my experience useful...

Agreeing to disagree would likely be too high of an expectations (especially since fanatics are involved) wink.gif
akhter is offline  
post #102 of 195 Old 07-23-2013, 04:46 PM
AVS Special Member
 
beaveav's Avatar
 
Join Date: Mar 2010
Location: SoCal
Posts: 1,830
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 61 Post(s)
Liked: 306
Quote:
Originally Posted by akhter View Post

All this typing and the double-blind fanatics have convinced...no one (atleast none that have spoken up on this thread)...since the start of the thread till now.

I applaud the fanaticism to show us the light though...

As I said I am not here to convert anyone and hopefully some people will find my experience useful...

Agreeing to disagree would likely be too high of an expectations (especially since fanatics are involved) wink.gif

Thank you for admitting you have your head in the sand and you like the view.

For every new thing I learn, I forget two things I used to know.
beaveav is offline  
post #103 of 195 Old 07-23-2013, 04:51 PM
 
diomania's Avatar
 
Join Date: Dec 2008
Posts: 1,389
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 49
Quote:
Originally Posted by sh25ep View Post

diomania, you're the one that stated first that they didn't match. Based on no further information from you.
Really? No further information from me? http://www.avsforum.com/t/1415224/mcintosh-integrated-amp-ma6600-for-b-w-803d/30#post_23550543
Quote:
I have told you of my experiences.
You sure did and I asked you related questions.
Quote:
You're looking at me for a bunch of numbers. Go get them yourself.
It is now clear that you have no idea what you are talking about. Thanks for admitting it, at least indirectly.
Quote:
You're the one making the claim. You go put a voltmeter on 2 amps of the same model and report back here.
So, after reading my reply "You don't believe me? Just put a voltmeter at the output of 2 of the same model amps and play a test tone and see it for yourself.", you came up with such reaction? Sorry to notice your problem. frown.gif
Quote:
You're telling everybody that doesn't happen to have 2 amps of the same model on hand to either just believe you or prove you're claim for you. Why would they? Why would I?
I see. So you weren't expecting people to believe you when you posted, "The output of 2 of the same model amps will be the same.", "The output of 2 of the same model amps will be the same. I have had 2 of the same model stereo amplifiers used in bi-amping and also bridged to mono and they matched. I did this with NAD and Rotel amplifiers and they matched.", right? Got it.
Quote:
You just have totally missed the point of my post.

The questions I had for you are obviously ridiculous.
I worded them the way that I did, because your questions were not so obviously ridiculous.
Says the one who doesn't know where the 20% value tolerance comes from. rolleyes.gif Of course they are ridiculous to you.
diomania is offline  
post #104 of 195 Old 07-23-2013, 05:01 PM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 14,382
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 754 Post(s)
Liked: 1167
Quote:
Originally Posted by sh25ep View Post

Brownstone322
Thank you for trying to explain it further, but I know and understand exactly what you are saying.

If you understood what was being said, you wouldn't be claiming the following:
Quote:
But there are more possibilities here than you are stating.

or asking the following:
Quote:
What would a "statistically significant level of correct answers" be?

Or claiming the following:
Quote:
The assumption that they are guessing is guessing that they are guessing. Willfully or not....
It is not a statistical fact that they are guessing since the only numbers we have are that 54% were correct. Not that 54% were guessing.

We're talking Statistics 101 which was understood by the mathematicians, engineers and medical researchers who devised ABX back in the 1970s. This has been revisited from time to time since then by the members of that team and various independent authorities, but the practical meaning of much of this stuff remains the same.

It is possible to calculate the percentage of correct answers that give various levels of confidence that the observed results were not due to random guessing.

It is possible to ace a test with random guesses but that is highly improbable.

As you approach only 50% correct for a two alternative forced choice test (techy-speak for an AB or ABX test) the probability that the observed results were due to random guessing gets higher and higher. The actual requirements vary with the number of trials and the amount of confidence that is desired.

While it is possible to do a large number of trials and show mathematical support for less than 90% correct answers, common sense says that if you can only get the right answers say 60% or less of the time, whatever it is that you are hearing isn't very valuable to music lovers.
arnyk is offline  
post #105 of 195 Old 07-23-2013, 05:07 PM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 14,382
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 754 Post(s)
Liked: 1167
Quote:
Originally Posted by sh25ep View Post

diomania, you're the one that stated first that they didn't match. Based on no further information from you.

I have told you of my experiences.
You're looking at me for a bunch of numbers. Go get them yourself.
Quote:
You don't believe me? Just put a voltmeter at the output of 2 of the same model amps and play a test tone and see it for yourself
You're the one making the claim. You go put a voltmeter on 2 amps of the same model and report back here.

I've done this test dozens, even 100s of times. There are always slight channel imbalances. It seems like as many times as I've done this, by chance I should have found at least one amp where the channels matched within the resolution of the meter I was using. Of course when you use 4 or 5 digit meters which now cost like $25, exactly the same requires ridiculously close matching.
arnyk is offline  
post #106 of 195 Old 07-23-2013, 05:11 PM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 14,382
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 754 Post(s)
Liked: 1167
Quote:
Originally Posted by sh25ep View Post

syd123
Quote:
So why don't gear companies cite DBT results to support their claim that their gear sounds better?
You mean like on their website? I don't really ever see any other advertisement for their products.

Believe it or not, there are companies that still advertise in magazines and other places that are off-web.
Quote:
If they did such tests and posted them on their website, and it was in their favor, would you believe them?

Depends on how well they documented their claims. If you've done this kind of thing you know about small details that expose most fakers.
Quote:
I wouldn't take their word for it.

That's a choice you get to make.
Quote:
When you're talking about this high dollar stuff, most people are going to go into a store and listen to it anyway and make up their own minds.

Which ironically enough is about as reliable as flipping a coin.
Quote:
I mean it's an interesting thought (and possibly argument) that they could do it, but at the end of the day it wouldn't really matter.

History has shown that more better evidence does have some kind of positive effect on prospective purchasers.
arnyk is offline  
post #107 of 195 Old 07-23-2013, 05:12 PM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 14,382
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 754 Post(s)
Liked: 1167
Quote:
Originally Posted by sh25ep View Post

diomania, you're the one that stated first that they didn't match. Based on no further information from you.

I have told you of my experiences.
You're looking at me for a bunch of numbers. Go get them yourself.
Quote:
You don't believe me? Just put a voltmeter at the output of 2 of the same model amps and play a test tone and see it for yourself
You're the one making the claim. You go put a voltmeter on 2 amps of the same model and report back here.

Pretty much a tacit admission that the author doesn't even have a voltmeter to his name.
arnyk is offline  
post #108 of 195 Old 07-23-2013, 05:19 PM
Advanced Member
 
akhter's Avatar
 
Join Date: Aug 2005
Location: Hong Kong
Posts: 563
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 17
Quote:
Originally Posted by beaveav View Post

Thank you for admitting you have your head in the sand and you like the view.

Sure if that makes you feel better smile.gif
akhter is offline  
post #109 of 195 Old 07-23-2013, 07:15 PM
Member
 
sh25ep's Avatar
 
Join Date: Jul 2013
Posts: 30
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
diomania, Why do you keep linking back to your other post? What information is there that supports your claim?

I have answered many of your questions and you did not even answer the one (real) question that I had.

What is your definition of "matched"?
Quote:
It is now clear that you have no idea what you are talking about.
Your clarity is questionable and presumptuous.
Quote:
you came up with such reaction? Sorry to notice your problem.
What reaction? What problem? Except for the "report back here" part, I just told you to do what you told everybody else to do. And I figured you would be eager to report back here after you checked your amps and tested your numbers.
Quote:
So you weren't expecting people to believe you when you posted, "The output of 2 of the same model amps will be the same.", "The output of 2 of the same model amps will be the same. I have had 2 of the same model stereo amplifiers used in bi-amping and also bridged to mono and they matched. I did this with NAD and Rotel amplifiers and they matched.", right? Got it.
It's more important that they don't believe you.

If someone is reading this and is thinking that they may want to buy 2 identical amplifiers and are wondering "do they match?" the answer is yes they will match.
The answer is not, no go check the numbers.
Quote:
Says the one who doesn't know where the 20% value tolerance comes from.
Once again, the questions I had for you are obviously ridiculous. You are referring to one of those questions out of context.

You have accused me of stalling, not knowing the answers, making things up, not having any idea what I'm talking about, and having some "problem".
I have not done any of these things to you. Nor will I.
I have answered your questions with the information I have had on hand at the time of answering.
I have the means to get the voltage numbers that I guess you are requesting. But why would I? The voltage numbers are only half the story. That is why I asked you what your definition of "matched" was. And also, how matched do you need them to be before they are considered "matched"?

I'm gathering, from your post, that your way is to use a test tone and a voltmeter. What frequency of test tone and at what listening level? Through speakers or without speakers? Is this the best way to do it?

arnyk suggested (while he was answering my obviously ridiculous questions) that matched is "0.1 to 0.3 dB 20-20 KHz". But he did not say at what listening level. Is this the best way to do it? Is this all frequencies at once or some kind of tone sweep? He also mentioned the voltage thing. Are we doing both here?

Speaker manufacturers consider their speakers matched if they are within 1db of each other. This is a tolerance and most likely not the actual value of any said speaker system. So if the end result was within 1db, even while using different amplifiers, is that matched?

Do you see the problem(s) with you questions now?
And it's still all a little more complicated than that.
sh25ep is offline  
post #110 of 195 Old 07-23-2013, 07:28 PM
Member
 
sh25ep's Avatar
 
Join Date: Jul 2013
Posts: 30
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
What the hell arny?

Do you have any questions or just tacit slams, blasts, and mud dragging?
sh25ep is offline  
post #111 of 195 Old 07-23-2013, 07:57 PM
Member
 
Brownstone322's Avatar
 
Join Date: Jan 2011
Location: The Old Dominion
Posts: 183
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 5 Post(s)
Liked: 62
Quote:
Originally Posted by sh25ep 
But more possibilities to consider: If, during subsequent tests they (or some of them) decided to change their minds and liked the other amplifier better, then what would that mean? And if, some of them during subsequent tests had moved out of the camp of undecided and chose an amp, then what would that mean? And if, some of them during subsequent tests moved from liking one amplifier to undecided, then what would that mean?

All three of those things would have happened because as people listen over time and through subsequent tests, they start to notice things in the music and sound stage and differences in frequency, detail, clarity, and depth. All of which they don't always notice right off the bat. It depends on the individual and what part of the sound they are concentrating on. People will change their minds and won't pick the same one 20 times in a row. And it doesn't matter. They will eventually pick the better one with time. Or at least some people will. And the ones that don't....well maybe they can't hear the difference. But some people will.

It's very simple, and I already pointed it out: A/B/X methodology, which is what Stereo Review used. It won't matter if they hear (or imagine) differences in "sound stage" or "frequency, detail, clarity, and depth." It won''t matter if their perceptions change with time. Either they match A/B to X or they don't. All else is cheap talk.

Quote:
Originally Posted by sh25ep 
The test was which one did they like.

Which, you may recall, is why the methodology was bogus to begin with. The authors even referred to their methodology as "A/B/X," but apparently they didn't understand what it A/B/X is. Again, we don't care which system they "liked better" until they proved that they can hear differences to begin with, which they never did.

Would you stop defending those test results already? 'Cause they're worthless.

Quote:
Originally Posted by sh25ep 
And the only way they would have known is through the differences in the two systems.

Who says they "knew" anything? The data does not tell us that.

Quote:
Originally Posted by sh25ep 
Are you saying that they were just picking one? Even when they had the option of choosing neither? They were all there listening and didn't flip a coin or send it by e-mail. They were there. Listening.

Get this: You're missing the point. Of course they were there, but because of the way the test was conducted, the results were no more reliable than if they had not been there at all. Can you not understand that? (See "review" at the bottom.)

Quote:
Originally Posted by sh25ep 
If you do not believe the particulars of the test, then of course your conclusion will be based on what you think is questionable data. This is understandable but still comes down to rather or not you believe them.

Actually, I do believe in the particulars of the test. That's why I recognize that the Matrix HiFi results tell us nothing, not what you think it does, and not what the authors apparently think it does. I was trained in research as part of my master of science degree (trained against my will, but trained nonetheless). Were you? I'm not a professional researcher, but I learned enough to know when a data set proves something (and when it doesn't).

Quote:
Originally Posted by sh25ep 
If I told you that I can hear a difference in amplifiers, then you would say "where is your proof?"

To be entirely honest, I wouldn't care what you told me.


Let's review:

In the Stereo Review experiment, no listener could consistently pick among amplifiers using an A/B/X methodology (despite vast differences in amplifier topology and cost) beyond the threshold of guessing. That's not to say they were guessing; it simply means that no one did any better than some guy who was guessing. Conclusion: The data does not support the hypothesis that the users could distinguish among amplifiers.

In the Matrix HiFi experiment, users were asked to state their preference among two systems (in a single trial) without first establishing that they could in fact hear differences among the the systems. Conclusion: The entire data set can be disregarded. Do you still not see why? Consider this hypothetical example...

TEST SUBJECTS:

User 1 is a self-described "audiophile"; he believes his hearing and perception to be superior to those of others.

User 2 is deaf; he communicates with sign language.

User 3 doesn't care about any of this.

TESTING PROCEDURE:

User 1 listens intently and thoughtfully, takes a few notes, and selects "System A."

User 2, because he can't hear dick, flips a coin and selects "System B."

User 3 skips the trials and goes to a bar and gets drunk on Bass Ale, but he sends in his vote for "System B" by text. (He picks "B" because "Bass" comes before "Ale"; it wasn't an easy decision.)

There are no further trials and no attempt at A/B/X matching.

ANALYSIS:

As the test was conducted, Audiophile Guy's selection is no more reliable than those of Deaf Guy or Drunk Guy. While we know that Deaf Guy and Drunk Guy were guessing, we have no way of knowing that Audiophile Guy wasn't guessing.

Why don't we know? Because we can't know, not with that testing procedure.

You can bang your fists and argue until your keyboard breaks, but those are facts, not opinions, and you can't make them go away.
Brownstone322 is offline  
post #112 of 195 Old 07-23-2013, 10:05 PM
Member
 
sh25ep's Avatar
 
Join Date: Jul 2013
Posts: 30
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Quote:
It won''t matter if their perceptions change with time. Either they match A/B to X or they don't. All else is cheap talk.
That's too simple of an assessment. Since people's perceptions change with time, it's more complicated than that.
Quote:
Again, we don't care which system they "liked better" until they proved that they can hear differences to begin with, which they never did.
Because they "liked better", they heard differences. That's what liking anything better means.
Quote:
Who says they "knew" anything? The data does not tell us that.
Well please excuse me and let me rephrase. And the only way they would have known what they liked is through the differences that they at least thought they heard in the two systems.
Quote:
the results were no more reliable than if they had not been there at all.
That's only true if you don't believe them. They did have the neither option. I keep pointing that out because that gives credibility to their claims of being able to hear a difference.
Quote:
I was trained in research as part of my master of science degree
Well hello Mr. Fancy Pants!
Quote:
I learned enough to know when a data set proves something (and when it doesn't)
Maybe you learned to much.
Quote:
To be entirely honest, I wouldn't care what you told me.
I didn't say you would care. What I meant was that you wouldn't believe me.
Quote:
That's not to say they were guessing; it simply means that no one did any better than some guy who was guessing. Conclusion: The data does not support the hypothesis that the users could distinguish among amplifiers.
Yes I agree. The data also does not support the hypothesis that all amplifiers sound alike to everybody.
Quote:
The entire data set can be disregarded. Do you still not see why?
No I still do not see why. I do not have my Fancy Pants master of science degree. Is this what is needed to understand that weird hypothetical example?
Quote:
we have no way of knowing that Audiophile Guy wasn't guessing
Yes we do. If you can grasp the concept of the "neither" option and believe him when he claims that one sounds better than the other.
Quote:
You can bang your fists and argue until your keyboard breaks, but those are facts, not opinions, and you can't make them go away.
You can bang your fists and argue until your keyboard breaks, but stating your assessment of the situation as fact does not make it so.
akhter likes this.
sh25ep is offline  
post #113 of 195 Old 07-23-2013, 10:28 PM
Senior Member
 
kraut's Avatar
 
Join Date: Sep 2012
Posts: 379
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 60
Quote:
Yes I agree. The data also does not support the hypothesis that all amplifiers sound alike to everybody.

no, the data says the null hypothesis that amplifiers driven within their parameters have no discernible sound differences is not disproved.
That is all, there is no evidence that would confirm the alternate hypothesis: amplifiers driven within their parameters can be distinguished.



sheesh, after all this crap about some electronics...back to the music and some North Mississippi Allstars..really loud. "Do it like we used to"...that is the "Spirit"...
kraut is offline  
post #114 of 195 Old 07-24-2013, 04:02 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 14,382
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 754 Post(s)
Liked: 1167
Quote:
Originally Posted by sh25ep View Post

What the hell arny?

Do you have any questions or just tacit slams, blasts, and mud dragging?

So you really don't have a voltmeter, eh?

Thanks for sharing! ;-)
arnyk is offline  
post #115 of 195 Old 07-24-2013, 04:05 AM
Member
 
sh25ep's Avatar
 
Join Date: Jul 2013
Posts: 30
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
The test should not be considered useless because it is statistically inaccurate, "statistically meaningless", "not statistically significant", or "not following the scientific process".
Are the only people that can post useful results on tests supposed to be scientists locked in a lab room doing constant scientific testing?
Quote:
then the valid methodology is to assume we can't hear differences and test to see if we can
That is not valid methodology. It's not objective to assume one way and test for the other.
It is objective to assume neither way and test for both ways. If neither way can be confirmed, then there is no assumption to be made.
This is not to say you can't assume for your own personal beliefs, sound tests, and general adventures through life. It just means you can not accurately and automatically apply the assumption to the said conclusion of said test.
Quote:
In contrast, the Stereo Review test used an A/B/X methodology, which works like this (assuming level matching):

1. User hears one source at random (described as A).

2. User hears second source at random (described as B).

3. User hears source X, which is A or B repeated at random.

4. User switches among A, B and X and attempts to match X to A or B.

5. Test repeats, although the user won't know whether the sources are the same or have swapped between A and B.

That methodology sets a high bar, and it will expose random guessing every time.
1. Do they hear it through a switch box? See my posts on page 1 and 2 for correction on this.
2. Do they hear it through a switch box? See my posts on page 1 and 2 for correction on this.
3. Do they hear it through a switch box? See my posts on page 1 and 2 for correction on this.
4. They switch it at a switch box? It's very reasonable to say that they did, for the essence of time, since they had so many people go through and do this. See my posts on page 1 and 2 for correction on this.
5. It did not say that the test repeated. This throws off the validity of the test.

If they used a switch box, then that is not valid "methodology" that "sets a high bar".
That is bad methodology that sets a low bar.
sh25ep is offline  
post #116 of 195 Old 07-24-2013, 04:09 AM
Member
 
sh25ep's Avatar
 
Join Date: Jul 2013
Posts: 30
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
You may have missed a post arny.eek.gif Go back and read it for further slam material.
sh25ep is offline  
post #117 of 195 Old 07-24-2013, 04:27 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 14,382
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 754 Post(s)
Liked: 1167
Quote:
Originally Posted by syd123 View Post


So why don't gear companies cite DBT results to support their claim that their gear sounds better? ..You know, like drug companies do in supporting their claim that their drug performs better than placebo? Anyone?

We've been asking ourselves this question since we did some of the first ABX tests back in the late 1970s. Here are some answers that we have come up with based on over 35 years of observation:

Most audio industry insiders know about the results of doing good listening tests but are in denial. The common denial strategies are:

(1) Double Deaf Denial. These individuals believe that even though tens of thousands of people of all ages and levels of experience with audio obtain the same null results, they are all hearing-challenged.

(2) Magic System Denial. These individuals believe that even though audio systems of just about any cost, opulence, or performance level in existence have been used during ABX tests and obtained similar outcomes, people would start hearing differences if they stopped using such cursed equipment in their DBTs and used the right magic boxes instead.

(3) Procedural Denial. These individuals believe that the various added controls used during DBTs change the listening experience in subtle but profound ways that even what they think are obvious differences are totally masked. A sales guy helicoptering over and buzzing around the listeners like you find at your typical high end audio salon would greatly improve listener performance.

(4)"It is a business", Denial. ABX, schmabx. ABX has been around for almost 40 years and there has been a great business to be done while ignoring it. Why change?

(5) Philosophical Denial. Only Objectivists do ABX tests, and only Subjectivists can set up good sounding audio systems and know how to listen.
arnyk is offline  
post #118 of 195 Old 07-24-2013, 04:34 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 14,382
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 754 Post(s)
Liked: 1167
Quote:
Originally Posted by sh25ep View Post

You may have missed a post arny.eek.gif Go back and read it for further slam material.

You' seem to have run away from every debunking of your many highly questionable posts that I've posted so far. So why should I work harder when I've already got you totally paralyzed? ;-)
arnyk is offline  
post #119 of 195 Old 07-24-2013, 05:07 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 14,382
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 754 Post(s)
Liked: 1167
Quote:
Originally Posted by sh25ep View Post

If you are in camp "C".... well you might not even know it.

First, the wire. If you're using lamp shade wire from the hardware store for your speaker wire, then this is not going to be sufficient to test out different decent quality amplifiers with. The reason is that low quality wire can drag down the performance of decent quality amplifiers (as well as the entire system) and as such, make different amplifiers sound alike.

Contrary to your apparent belief, they sell high end speaker wire and interconnects to objectivists and allow them to hook them up to their switchboxes. Doing that results in no thermonuclear explosions. Been there, done that and no joy.
Quote:
Second, the switching. If you are using a switch box of some kind to switch the amplifiers on the fly, then this is not going to be sufficient to test with either.

Here we see a "new truth". It appears that you want people to believe that it s impossible to build a switch box that doesn't totally mask differences. Even though cable swapping is slow and therefore ruins listener accuracy and sensitivity, we've tried that. We've also shown that the random unmatched listening levels that almost all audiophiles use for their tests produce reliably discernible differences, even with a switch box in place. Been there, done that and no joy.
Quote:
Much like the wire, the switch box can drag down the performance and make the amplifiers sound alike.

Please provide an actual technical critique of one of our switch boxes. Here is what they look like since its pretty obvious you've never actually seen one for yourself:


Quote:
The only good way of switching to different amplifiers is by physically disconnecting the speaker wires and moving them from one amp to the next.

Spoken like someone who has zero experience with well-made switch boxes...

Quote:
Lastly is the point of level matching. The only level matching you need to do is to match the perceived volume of one amplifier to the next. Perceived volume. This is the volume that YOU think that the amplifiers match at. No volt meter, no watt meter, no db meter is needed. The reason is this...At what frequency are you supposed to volt / watt/ db match?

By even just asking this question you show that you probably have no clue about the relationship between measured and perceived listening levels. Believe it or not there is this area of science called Psychoacoustics that explains all this. After you finish your introductory statistics class, try a good basic uinversity class in psychoacoustics. You will find it to be quite a revelation.
Quote:
Even if you volt / watt / db match over the entire frequency range (using pink noise), different people still hear different frequencies in different ways.

One never does level matching with pink noise since it contains all frequencies mixed together and the essence of the procedure is matching at specific audible frequencies one at a time. I guess we need to add some course work in practical electronics to your course load. You wouldn't be a hifi salesman, would you? ;-)
Quote:
And different amplifiers put out different frequencies in different amounts in different ways. For example, one amplifier might be a little hotter in the mid-range while another might be a little bumpier in the bass. All the the bass and treble controls set to flat and with both amplifiers working properly.

Actually modern amplfiers (at least the good ones) have very flat response coming right out of the box. If you ever read and understood your basic Stereophile amplifier review you'd already know that:



Quote:
My thanks to the old man and his insight comparing the 6600, 6900, and Classe. It helps to know that somebody has heard a difference between the the Mcintosh units and also his thoughts on the Classe.

Actually the Classe and McIntosh amps I've found reviewed with frequency response curves are kind of on the marginal side. They are bad enough that they might even sound different from each other. I provided a frequency response of a good power amp above for your future reference... And below I provide a frequency response curve of a kinda marginal amplfiier.




While I don't have a fancy picture here is the text from a typical mainstream AVR review from http://www.hometheater.com:

"This graph shows that the AVR-2313CI’s left channel, from CD input to speaker output with two channels driving 8-ohm loads, reaches 0.1 percent distortion at 126.8 watts and 1 percent distortion at 152.1 watts. Into 4 ohms, the amplifier reaches 0.1 percent distortion at 154.4 watts and 1 percent distortion at 192.5 watts.

There was no multichannel input to measure. THD+N from the CD input to the speaker output was less than 0.005 percent at 1 kilohertz when driving 2.83 volts into an 8-ohm load. The signal-to-noise ratio with an 8-ohm load from 10 hertz to 24 kHz with “A” weighting was –113.19 dBrA.

From the Dolby Digital input to the loudspeaker output, the left channel measures –0.10 dB at 20 Hz and –0.27 dB at 20 kHz. The center channel measures –0.09 dB at 20 Hz and –0.15 dB at 20 kHz, and the left surround channel measures –0.09 dB at 20 Hz and –0.29 dB at 20 kHz. From the Dolby Digital input to the line-level output, the LFE channel is –0.01 dB at 20 Hz when referenced to the level at 40 Hz and reaches the upper 3-dB down point at 118 Hz and the upper 6-dB down point at 121 Hz.—MJP"
]

Looks to me like it would pass level matching right out of the box and outperforms the Classe when it comes to that critical function.
arnyk is offline  
post #120 of 195 Old 07-24-2013, 05:53 AM
Senior Member
 
syd123's Avatar
 
Join Date: Feb 2009
Location: mid-atlantic region of US
Posts: 449
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 34
Quote:
Originally Posted by arnyk View Post

We've been asking ourselves this question since we did some of the first ABX tests back in the late 1970s. Here are some answers that we have come up with based on over 35 years of observation:

Most audio industry insiders know about the results of doing good listening tests but are in denial. The common denial strategies are:

(1) Double Deaf Denial. These individuals believe that even though tens of thousands of people of all ages and levels of experience with audio obtain the same null results, they are all hearing-challenged.

(2) Magic System Denial. These individuals believe that even though audio systems of just about any cost, opulence, or performance level in existence have been used during ABX tests and obtained similar outcomes, people would start hearing differences if they stopped using such cursed equipment in their DBTs and used the right magic boxes instead.

(3) Procedural Denial. These individuals believe that the various added controls used during DBTs change the listening experience in subtle but profound ways that even what they think are obvious differences are totally masked. A sales guy helicoptering over and buzzing around the listeners like you find at your typical high end audio salon would greatly improve listener performance.

(4)"It is a business", Denial. ABX, schmabx. ABX has been around for almost 40 years and there has been a great business to be done while ignoring it. Why change?

(5) Philosophical Denial. Only Objectivists do ABX tests, and only Subjectivists can set up good sounding audio systems and know how to listen.

Thanks for the thoughtful reply Arnyk... The only other denial strategy I might add (though I suppose it's a variant of #3) is the "Testing Axiety" denial. ...The silly notion that differences which are sooo apparent during normal listening suddenly disappear when subjects are put under the duress of testing. ..Ugh.

So if no mfg. uses DBT to demonstrate that their gear sounds better, how about the idea of using DBT data to prove it sounds the same? For example, I've often been tempted to do the following:

Line up a supplier of cheap but cool looking speaker cables (with blueish annondized connectors and maybe even a wood box for packaging) then using DBT's compare them with the most expensive Stereophile class "A" recommended speaker cables, then use the results in my marketing. ..Something like, "In carefully controlled blinded trials, subjects found my $85 Excalibur cables to be absolutely indistinguishable from the $2500 AudioQuest Celestials (..or whatever). And to make it even more compelling, my cohort of listeners would include Univ. of Penn. music dept. students plus as many area audiophiles as I could line up. ....Of course, I'd happily provide details of the testing protocol and equipment used in the "fine print" of my ad. ..I wonder if I could be sued for doing this by AudioQuest (or whomever)? ..Don't see how I could if I adequately substantiate my claims. ..Alas, I don't have the stomach engage in such shenanigans.
syd123 is offline  
Reply 2 Channel Audio

User Tag List

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off