Bi-amping B&W CM10 - Page 13 - AVS Forum
Forum Jump: 
Reply
 
Thread Tools
post #361 of 651 Old 04-06-2014, 10:09 AM
Senior Member
 
Jon Middleton's Avatar
 
Join Date: Sep 2003
Location: SLC, UT
Posts: 343
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 77
It's too bad that speaker manufacturers feel obliged to put the extra set of terminals on their speakers. I think even the new Bryston Model T's have two sets, so it is the fashionable thing to do. It seems to me to be purely motivated by market forces rather than sonic considerations. I bought my 801's nearly 20 years ago at a high end store. They were a demo pair, so the saleeman and I put them in the back of my Landcruiser and took them to my house. The store sold Transparent, AQ, etc, all the usual suspects. I stopped by the nearest Radio Shack on the way home to pick up some speaker wire.wink.gif My WATT/Puppies are connected to the amp with some left over marine tinned 12 gauge wire. Just to one terminal, though.
Jon Middleton is online now  
Sponsored Links
Advertisement
 
post #362 of 651 Old 04-06-2014, 11:59 AM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 17,770
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 496 Post(s)
Liked: 326
Quote:
Originally Posted by arnyk View Post

However, as subjective and audiophiles and reviewers listen to amplifiers, they could all measure the identically same and the reviewers would still hear differences because their listening evaluations are so crudely performed.

The requirements are simple:

(1) Level matched
(2) Time synched
(3) Quick switched at listener discretion
(4) Effective bias controls.
(5) Listener training

They have been implemented by serious listeners for decades. The practical problem with them for journalists is that they don't always give the most expedient results.
If I may, let me present this double blind test:
i-nTpXsqF-L.png
This test was published in the Audio Amateur magazine in 1982. I trust it is acceptable to show as meeting your criteria since you were one of the three authors of that article.

Here are the amplifiers under test:
i-xjPp4w3-S.png

So pretty powerful amplifiers were used. The explanation goes to say :

i-3tbhPL7-M.png

Given the previous slide, and the 5 ohm impedance, we are pumping in the neighborhood of 180 watts into these speakers. How did the testing turn out? Here are the results:

i-zfdGz3C-XL.png

Notice the highlighted section in yellow. One of the clips generated "random" results. The others showed convincing difference in amplifiers. The text does not say why these tracks were selected. What if they had gotten "unlucky" and the other two tracks they selected were the same as the track "Lightfood?" That is, non-revealing? We would have had all "random" results and verdict would have been that the amplifiers sounded the same. Yet the problem would have been that of the people who created the test, not knowing what material is designed to bring out the difference we are trying to find.

Here is the bit from Conclusion of this article:
i-NVbTMcL-M.png

Is the message in yellow what we get in these discussions? How an amplifier sounds when stressed is our most plausible explanation of why they may sound different. We have a double blind test that shows exactly that. Yet, there is not one copy of this report anywhere online. Through thousands upon thousands of discussions and posts across countless forums, Arny did not think to post this. I had to buy a copy of the paper in an auction to get access to it. Why the reluctance? Why not present the data and let it speak for itself?

By the way in Arny's list and this test there is a critical protocol error: lack of control. A control is necessary to make sure the test fixture is working correct, that we are operating the test properly, and have listeners with the required discrimination. The authors got lucky in that they got an accidental control which was the different sounding amp. Had that not been there, and the outcome presented as "random," we would not have known about any protocol errors.

Amir
Founder, Madrona Digital
"Insist on Quality Engineering"

amirm is online now  
post #363 of 651 Old 04-06-2014, 12:08 PM
Senior Member
 
stereoeditor's Avatar
 
Join Date: Feb 2010
Posts: 323
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 111 Post(s)
Liked: 30
Quote:
Originally Posted by amirm View Post

If I may, let me present this double blind test:
i-nTpXsqF-L.png
This test was published in the Audio Amateur magazine in 1982.

I republished this article, with permission from Ed Dell at Audio Amateur, in Hi-Fi News & Record Review, of which I was the editor back then.

John Atkinson
Editor, Stereophile
stereoeditor is offline  
post #364 of 651 Old 04-06-2014, 12:26 PM
AVS Special Member
 
Frank Derks's Avatar
 
Join Date: Jul 2002
Location: Region A,B,C
Posts: 1,885
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 90 Post(s)
Liked: 86
Amplifiers can sound different.

1982 : State of the art amplifier design from that era had serious room for improvement.
Difficult 5 'Ohm' load from an inefficient electrostatic loudspeaker is mainly capacitive.
Amps driven into occasional clipping

What's the point to drag this into this thread? (Should be in 'All amplifiers sound the same' thread)

This article has little to do with blind testing a bi-amping setup.
Frank Derks is offline  
post #365 of 651 Old 04-06-2014, 01:11 PM
Advanced Member
 
andyc56's Avatar
 
Join Date: Feb 2013
Posts: 552
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 17 Post(s)
Liked: 180
Quote:
Originally Posted by amirm View Post

If I may, let me present this double blind test:

That troll post cost amirm $11 biggrin.gif
andyc56 is offline  
post #366 of 651 Old 04-06-2014, 01:17 PM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 13,854
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 405 Post(s)
Liked: 1050
Quote:
Originally Posted by amirm View Post

Quote:
Originally Posted by Frank Derks View Post

Both are clipped.
No they are not. You are confusing the limited (graphics) resolution of my digital scope in the tweeter amp. Even if it were, it is far less than the woofer amp.

A good example why I haven't used 'scopes to judge clipping for decades. Last time I did use scopes to test amplifier one of the scope's inputs was connected across the output of an analog THD analyzer. I then had the option of a very sensitive view of any distortion.
Quote:
Quote:
The difference in spl by this advantage will be less than 1dB.
From what I recall, the difference was about 1.5 db or so. This sounds small until you consider that 1.5 dB improvement is the same as getting a 140 watt amp instead of 100.

That says more about the futility of looking at watts instead of dBs, than anything else. A classic newbie mistake if there ever was one!

Quote:
And this is with AVRs which have shared power supply so the differential is not as high as it could be with an amp with independent supplies.

Shared power supplies are only disadvantageous if one is naive enough to believe that all of the channels of a multichannel recording are identical to each other and perfectly correlated. If they are different then sharing the power supply enables dividing it up dynamically based on the needs of the music.
Quote:
Quote:
If listening conditions demand such amplifier output that it clips on regular occasions the user is better of buying a higher spec'd amplifier.
That's neither here, nor there.

Dismissing without a relevant explanation is just an attempt to distract.
arnyk is online now  
post #367 of 651 Old 04-06-2014, 02:08 PM
Senior Member
 
Jon Middleton's Avatar
 
Join Date: Sep 2003
Location: SLC, UT
Posts: 343
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 77
Quote:
Originally Posted by stereoeditor View Post

I republished this article, with permission from Ed Dell at Audio Amateur, in Hi-Fi News & Record Review, of which I was the editor back then.

John Atkinson
Editor, Stereophile

Nice, except that it was a couple of years ago. Now repeat the test using a digital source and compare an Adcom GFA 555 Mk II, a Bryston 4B SST2, and an Ayre VX-R. Don't clip them.
Jon Middleton is online now  
post #368 of 651 Old 04-06-2014, 02:51 PM
Senior Member
 
stereoeditor's Avatar
 
Join Date: Feb 2010
Posts: 323
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 111 Post(s)
Liked: 30
Quote:
Originally Posted by stereoeditor View Post

I republished this article, with permission from Ed Dell at Audio Amateur, in Hi-Fi News & Record Review, of which I was the editor back then.

So I followed the link to Amir's original mention of this article and there, to my surprise, was this posting by Arny Krueger:
Quote:
Originally Posted by arnyk View Post

Obviously you weren't that much into Audio ca. 1982, Amir. ;-) Everybody who was anybody read HFN&RR as it was known then.

Thank you for the compliment, Mr. Krueger. Although I didn't become the editor in chief of Hi-Fi News & Record Review until October 1982, I was deputy editor from April 1978 onward. Glad my efforts didn't go unappreciated :-)

John Atkinson
Editor, Stereophile
stereoeditor is offline  
post #369 of 651 Old 04-06-2014, 02:53 PM
Senior Member
 
hevi's Avatar
 
Join Date: Jan 2007
Posts: 360
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 31 Post(s)
Liked: 45
Quote:
Originally Posted by Jon Middleton View Post

Nice, except that it was a couple of years ago. Now repeat the test using a digital source and compare an Adcom GFA 555 Mk II, a Bryston 4B SST2, and an Ayre VX-R. Don't clip them.

Well, the irony that amirm and stereoeditor is pointing out with badly disguised enjoyment is that arnyk's *only* "semi-properly" performed, documented and published part in an ABX test seems to point in the direction that there are infact audible differences between amps, rather that there are "none", which is arnyk's general stance (or as he makes his stance come across here in the forum) when it comes to amplifiers...

Now, if I may ask what the heck the source has to do with it (AFAIK it stayed constant during the ABX tests, right?!?) -If you can detect an audible difference between amps with a such "inferior" source (as you are implying) as "analog", what makes you think you would *not* detect it using a superior source such as digital?!? In my world, it sure as hell isn't the source that is making the difference, because the source is constant, ritght?!?
amirm likes this.

hevi is offline  
post #370 of 651 Old 04-06-2014, 03:05 PM
Senior Member
 
Jon Middleton's Avatar
 
Join Date: Sep 2003
Location: SLC, UT
Posts: 343
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 77
Quote:
Originally Posted by hevi View Post

Well, the irony that amirm and stereoeditor is pointing out with badly disguised enjoyment is that arnyk's *only* "semi-properly" performed, documented and published part in an ABX test seems to point in the direction that there are infact audible differences between amps, rather that there are "none", which is arnyk's general stance (or as he makes his stance come across here in the forum) when it comes to amplifiers...

Now, if I may ask what the heck the source has to do with it (AFAIK it stayed constant during the ABX tests, right?!?) -If you can detect an audible difference between amps with a such "inferior" source (as you are implying) as "analog", what makes you think you would *not* detect it using a superior source such as digital?!? In my world, it sure as hell isn't the source that is making the difference, because the source is constant, ritght?!?

It may not be the source, digital is just easier. I'm not down on vinyl, I even have some records and a record player. It might be fun to use both sources. Maybe you'd find that analog is more "revealing" of the differences in amps. How many times have we read in "Stereophile" that this or that speaker is so good that it reveals different sonic characteristics of solid state components? Enough to make me want to throw up.
Jon Middleton is online now  
post #371 of 651 Old 04-06-2014, 03:16 PM
AVS Special Member
 
A9X-308's Avatar
 
Join Date: Mar 2008
Location: Australia; now run by adults.
Posts: 5,163
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 31 Post(s)
Liked: 42
Quote:
Originally Posted by arnyk View Post

I would say poorly designed.
I stand by my original description. As that poster has told us before he is electrically qualified and has many years in audio, and has no hesitation in typing multi thousand word posts nit picking the hell out of anyone else who disagrees with him, even to mining information posted years or decades ago (sometimes even in context) so it was either deliberately and intellectually dishonest or he's incompetent. Try writing it up as it was here and present it to the AES or similar body and see what sort of response it gets.
A9X-308 is offline  
post #372 of 651 Old 04-06-2014, 03:23 PM
Senior Member
 
hevi's Avatar
 
Join Date: Jan 2007
Posts: 360
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 31 Post(s)
Liked: 45
Quote:
Originally Posted by A9X-308 View Post

I stand by my original description. As that poster has told us before he is electrically qualified and has many years in audio, and has no hesitation in typing multi thousand word posts nit picking the hell out of anyone else who disagrees with him, even to mining information posted years or decades ago (sometimes even in context) so it was either deliberately and intellectually dishonest or he's incompetent. Try writing it up as it was here and present it to the AES or similar body and see what sort of response it gets.

are you talking about arnyk or amirm? *confused*

hevi is offline  
post #373 of 651 Old 04-06-2014, 03:43 PM
AVS Special Member
 
A9X-308's Avatar
 
Join Date: Mar 2008
Location: Australia; now run by adults.
Posts: 5,163
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 31 Post(s)
Liked: 42
I responded to Arny. I wasn't talking about him.
A9X-308 is offline  
post #374 of 651 Old 04-06-2014, 04:22 PM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 13,854
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 405 Post(s)
Liked: 1050
Quote:
Originally Posted by hevi View Post

Quote:
Originally Posted by Jon Middleton View Post

Nice, except that it was a couple of years ago. Now repeat the test using a digital source and compare an Adcom GFA 555 Mk II, a Bryston 4B SST2, and an Ayre VX-R. Don't clip them.

Well, the irony that amirm and stereoeditor is pointing out with badly disguised enjoyment is that arnyk's *only* "semi-properly" performed, documented and published part in an ABX test seems to point in the direction that there are infact audible differences between amps, rather that there are "none", which is arnyk's general stance (or as he makes his stance come across here in the forum) when it comes to amplifiers...

I think the above paragraph has just about everything wrong.

(1) AFAIK the HFN&RR article that I co-wrote has never been publicly criticized by any credible authority for any procedural problem. It is thus safe to say that Its procedures are just fine. BTW the article was published in 1982 - a mere 32 years ago. Talk about dredging the archives in a desperate effort to make trouble!

(2) It has always been clear to everybody officially involved with ABX that as the article was entited, "Some Amplifiers Do Sound Different". It is true that some people troll with the phrase "All amps sound the same". All generalities that global are false. Obtaining positive or negative results in ABX is as easy as knowing which equipment to choose to compare. That's just common sense. It is not my fault that some people get negative results when they compare equipment that measures to be so accurate that audible differences would be highly unlikely. If you check our archives of ABX test results at http://home.provide.net/~djcarlst/abx_data.htm you will find a nice mix of negative and positive results.
Quote:
Now, if I may ask what the heck the source has to do with it (AFAIK it stayed constant during the ABX tests, right?!?) -If you can detect an audible difference between amps with a such "inferior" source (as you are implying) as "analog", what makes you think you would *not* detect it using a superior source such as digital?!?

If you actually bothered to read the 1982 HFN article that you brought up, you would know that its positive results were based on playing back LPs. This fact makes the above question look pretty strange. Want to retract it? BTW, the actual tests were done in the spring of 1981 when CD players were as about as rare as hen's teeth.
CruelInventions likes this.
arnyk is online now  
post #375 of 651 Old 04-07-2014, 07:42 AM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 17,770
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 496 Post(s)
Liked: 326
Quote:
Originally Posted by arnyk View Post

That says more about the futility of looking at watts instead of dBs, than anything else. A classic newbie mistake if there ever was one!
Ah, the "you don't know jack" argument! biggrin.gif

Let's rewind the plot so that what we are discussing is not lost. OP in the very first post said this:
Quote:
Originally Posted by Mekail View Post

I have a Yamaha 5ch 1010 receiver and a Rotel RMB-1075 5ch power amp. The 1075 is 120 watts x 5. Bi-amping the front two channels will give me 240 watts.
That immediately triggered the alarm in the house of objectivity and folks piled on to shut the door down immediately with comments like this:
Quote:
Originally Posted by arnyk View Post

Essentially passive biamping cuts the load presented by the speaker into two pieces, and hooks each piece to a different identical amplifier. The maximum amount of power that can be delivered to each part is essentially the same whether the halves are connected to one amplifier or two.

You say we should look at things in db. OK, let's do that. OP says the power doubles. Doubling the power is 3 db. You say that it is "essentially" 0 db. My experiment showed the difference of 1.6 db or so. That is quite a ways up from 0 db that you said and represents genuine increase in total power delivered to the speaker.

You might say, "well, we don't hear watts and our hearing works close to log scale which db represents." I answered that in the other thread. Namely, the main benefit of any form of bi-amping in my opinion is the isolation of woofer from tweeter. By doing so, if the woofer channel distorts, and creates harmonic distortions, they won't bleed into tweeter and cause harshness and listener fatigue. That fact that we gain total power is icing on the cake.

Courtesy of your blind testing, we know clipping even in small amounts is audible:

i-KCSpjj7.png

Clipping is audible due to harmonics it creates. The woofer is not capable of playing high frequencies so the high frequency harmonics of clipping get filtered in bi-amp configuration.

So while we can use a log function (i.e. db) to try to make the numbers smaller and quiet down the natives, in a transparent discussion among ourselves in the same camp, let's not go there. It looks like we are trying to sweep our mistakes under the rug.

Amir
Founder, Madrona Digital
"Insist on Quality Engineering"

amirm is online now  
post #376 of 651 Old 04-07-2014, 08:04 AM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 17,770
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 496 Post(s)
Liked: 326
Quote:
Originally Posted by arnyk View Post

I think the above paragraph has just about everything wrong.
There is nothing wrong with what he said Arny. Your test lacked any kind of control. If you knew you were testing clipping, you should have tried to have a sample that was highly clipped as to make sure the test, listeners, and scoring, all detect it appropriately. As it is, we have no idea if any of that happened. For all we know, as I showed with the AVS Forum blind audio test, that your scoring was wrong and no difference was found!

My wife used to be a lab technician in a hospital. At the start of every shift, they would run sugar water through their machine as a control to make sure it could detect it correctly. It mattered now how expensive the machine was, or that it had generated the right result countless days before. The protocol stipulated to always start with a known snapshot of the machine's performance. Likewise here, we need such a control in audio test and sadly that is what is always missed. The results always have a fog of inaccuracy as a result.
Quote:
(1) AFAIK the HFN&RR article that I co-wrote has never been publicly criticized by any credible authority for any procedural problem. It is thus safe to say that Its procedures are just fine. BTW the article was published in 1982 - a mere 32 years ago. Talk about dredging the archives in a desperate effort to make trouble!
Oh, the "as far as I know" argument. biggrin.gif We don't know what you know Arny. What we know is that this is the first time this test has ever been published online. I pointed out one potential issue with it above. If we had an independent observer audit the work, we may have found other problems. I know what you are going to say: you are beyond any mistakes. Here is a good test case to counter that. The Boston Audio Society blind testing is put forward as expert witness and evidence by our camp all the time. Here is one of their tests: http://www.bostonaudiosociety.org/bas_speaker/abx_testing2.htm

"I expressed my desire to try the test, and Remington went to cue up the record again, but I requested to be allowed to undertake the test with no signal passing though the system. Before realizing the import of what he was saying, Vanderkooy interjected: "Ah! You're going to listen to the sound of the relays." Yes, there is indeed a slight audible difference between the acoustic "click" made when the "A" and "B" relays pull in. This is due to the unavoidable differences in the mounting positions of the relay on the A/B/X box chassis and, although slight, it can be heard if one listens for it. I replied that I was going to listen to the difference in background hiss, and the subsequent series of blind trials showed conclusively that the two signal paths could be reliably distinguished on this basis alone."

I have highlighted the key section. Had professor Vanderkooy not been there, it is entirely possible the person being tested could have gamed the results and we would have an egg on our face making him a hero instead of zero. Fortunately we got lucky there that he did not try to listen for the clicks and were made aware of our testing protocol mistake.
Quote:
(2) It has always been clear to everybody officially involved with ABX that as the article was entited, "Some Amplifiers Do Sound Different". It is true that some people troll with the phrase "All amps sound the same". All generalities that global are false.
Amen to that! biggrin.gif Let's be absolutely clear that you are agreeing that two solid state amps can sound different. In all of these arguments you usually bring up low power tube amps as an example of amps sounding different. Have not heard you say two solid state transistor amps can sound different.
Quote:
Obtaining positive or negative results in ABX is as easy as knowing which equipment to choose to compare.
Say what? You mean you played a game in that article in order to get that headline of amps sounding different??? Tell us that is not true Arny.
Quote:
If you actually bothered to read the 1982 HFN article that you brought up, you would know that its positive results were based on playing back LPs. This fact makes the above question look pretty strange. Want to retract it? BTW, the actual tests were done in the spring of 1981 when CD players were as about as rare as hen's teeth.
Say what? smile.gif There is absolutely nothing wrong with what he said. He explained that if a difference between two amps was heard with LPs, that difference does not go away when you increase the fidelity of the source, i.e. use digital. What is your basis for disagreeing with that?

Amir
Founder, Madrona Digital
"Insist on Quality Engineering"

amirm is online now  
post #377 of 651 Old 04-07-2014, 08:06 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 13,854
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 405 Post(s)
Liked: 1050
Quote:
Originally Posted by amirm View Post

Quote:
Originally Posted by arnyk View Post

That says more about the futility of looking at watts instead of dBs, than anything else. A classic newbie mistake if there ever was one!
You say we should look at things in db. OK, let's do that. OP says the power doubles. Doubling the power is 3 db. You say that it is "essentially" 0 db. My experiment showed the difference of 1.6 db or so. That is quite a ways up from 0 db that you said and represents genuine increase in total power delivered to the speaker.

It seems to be very difficult to make an argument that can't be twisted and lost by someone who is determined to not follow it. ;-)

My actual argument is based on fact and experience. The fact it is based on is that so many dB more power means nothing unless you are clipping significantly at the lower power level. The experience I ask the reader to favor himself with is to move the volume control on his AVR by so many notches, leveraging the fact that just about every modern AVR's volume control works in steps of 0.5 dB.
Quote:
You might say, "well, we don't hear watts and our hearing works close to log scale which db represents." I answered that in the other thread.

Since I'm not in the habit of falling down rabbit holes, you're going to have to do better than that Amir. If you have something to say, say it here!

BTW the convention of using dB to describe power is not my innovation. I believe I first saw it used in High Fidelity Magazine many, many decades ago. An archive of that magazine is here: http://www.americanradiohistory.com/High-Fidelity-Magazine.htm
Quote:
Namely, the main benefit of any form of bi-amping in my opinion is the isolation of woofer from tweeter.

The above opinion is grossly flawed by its conflation of active biamping and passive biamping, which are vastly different things. Amir, if you think they are the same or even comparable, then you might be the only person who is knowlegable about audio engineering who thinks so! This is such a gross mistake that IMO it raises significant questions about someone who repeatedly makes it (I've ticketed you for this infraction within the week).
Quote:
By doing so, if the woofer channel distorts, and creates harmonic distortions, they won't bleed into tweeter and cause harshness and listener fatigue. That fact that we gain total power is icing on the cake.

IME woofers do a pretty good job of reproducing the nonlinear distortion including IM that a clipping amplifier creates.
Quote:
Courtesy of your blind testing, we know clipping even in small amounts is audible:

i-KCSpjj7.png

"Small amounts" of clipping? Could you be more vague? If I get to pick what a "Small Amount" is, it will always be inaudible! ;-) Furthermore, there are at least two vastly different but reasonable ways to quantify clipping. One is the percentage of the time that the music is clipping, and the other is the amount of the music that is clipped off the top. I don't think that either is sufficient by itself.
Quote:
Clipping is audible due to harmonics it creates. The woofer is not capable of playing high frequencies so the high frequency harmonics of clipping get filtered in bi-amp configuration.

Yet another newbie mistake. Clipping creates both harmonics and IM products. In a band-limited system (e.g. a woofer) the IM products are often far more audible. It is a law of science that EVERY nonlinearity that creates THD also creates IM when reproducing music, because music is generally not just a single pure tone.
arnyk is online now  
post #378 of 651 Old 04-07-2014, 08:11 AM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 17,770
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 496 Post(s)
Liked: 326
Quote:
Originally Posted by A9X-308 View Post

I stand by my original description. As that poster has told us before he is electrically qualified and has many years in audio, and has no hesitation in typing multi thousand word posts nit picking the hell out of anyone else who disagrees with him, even to mining information posted years or decades ago (sometimes even in context) so it was either deliberately and intellectually dishonest or he's incompetent. Try writing it up as it was here and present it to the AES or similar body and see what sort of response it gets.
Oh, the "you are an idiot" argument. biggrin.gif I read your original post. Other than making an accusation, you did not explain what you didn't like about the test. Regardless, every objection was put forward in the other thread and addressed. I even went as far as doing exactly what Arny said the load should be (even though it made no sense) and repeated the experiment with the same results: http://www.avsforum.com/t/1492314/question-on-bi-amping/690

So again, let's not rehash the same topic over and over again. Please read the other thread to make sure your objections are not addressed.

Amir
Founder, Madrona Digital
"Insist on Quality Engineering"

amirm is online now  
post #379 of 651 Old 04-07-2014, 08:15 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 13,854
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 405 Post(s)
Liked: 1050
Quote:
Originally Posted by amirm View Post

Quote:
Originally Posted by arnyk View Post

I think the above paragraph has just about everything wrong.
There is nothing wrong with what he said Arny. Your test lacked any kind of control. If you knew you were testing clipping, you should have tried to have a sample that was highly clipped as to make sure the test, listeners, and scoring, all detect it appropriately. As it is, we have no idea if any of that happened.

We had a very specific sample that demonstrated the audible problem that we heard quite quickly and clearly. I don't have the article at my disposal so I can't quote from it how we characterized it. We could repeatedly drop the needle in a certain groove more or less, and right there was an example of the problem. A year later we were doing the same thing with the "phrase repeat" feature of our new CD players. These days we edit the file and play the snippet of a .wav file.

If memory serves the LP was by the Eagles and "Hotel California" comes to mind. But this was over 3 decades ago, so I could have those facts all wrong.

We have prima facaie evidence that we could hear the difference - the test scores which were overwhelmingly positive for audible differences. Did we even miss a trial? Maybe one. Well you have the article!
Quote:
For all we know, as I showed with the AVS Forum blind audio test, that your scoring was wrong and no difference was found!

I am unaware of that. Please cite a post with a URL.
arnyk is online now  
post #380 of 651 Old 04-07-2014, 08:26 AM
Senior Member
 
UndersAVS's Avatar
 
Join Date: Mar 2008
Posts: 259
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 18 Post(s)
Liked: 35
Amir,
You've demonstrated that a Yamaha amp clipped when the impedance was low. Nothing more. The rest can be chalked up as gross speculation concerning insignificant details, or off topic and petty bickering. Not something one would expect from someone so highly critical of other comments.

Have a good day.
UndersAVS is offline  
post #381 of 651 Old 04-07-2014, 08:41 AM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 17,770
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 496 Post(s)
Liked: 326
Quote:
Originally Posted by UndersAVS View Post

Amir,
You've demonstrated that a Yamaha amp clipped when the impedance was low. Nothing more.
I am sorry but seems like the context of the test was lost. Here is the start of my explanation of the testing: http://www.avsforum.com/t/1492314/question-on-bi-amping/660#post_24002262 with Arny saying:

Good point but all you are doing is addressing is a poor choice of words. The power amp is still generating the voltage at the frequencies that the crossover blocks. If either amplifier is generating audible distortion because it is being overdriven, then they both are generating audible distortion. If either amplifer is not generating audible distortion, then neither will be generating distortion.

I think this is the 432'd time (more or less) this particular issue has been re-hashed ;-)

He is right in the last sentence. Up to that point, he and others kept repeating that line that both amps in a passive bi-amp situation clip at the same time. I have bolded that above in Arny's statement. No real measurements were provided to back that other than stating it as "everybody knows." So I set out to test that hypothesis. I used a 4 ohm load for the "woofer" amp. I used a high value resistor for the tweeter as that is the impedance that amp would see, when driving a 20 Hz tone to the tweeter. This allowed me to do the test without a crossover. I then turned up the volume until I observed clipping. In all *three* AVRs that I tested, the woofer amp clipped before the tweeter one. So the statement that both amps clip at the same time was shown to not be true.
Quote:
The rest can be chalked up as gross speculation concerning insignificant details, or off topic and petty bickering. Not something one would expect from someone so highly critical of other comments.
I am unclear what the "rest" is. Again, the test was designed to show the talking point regarding bi-amping being useless due to both amps working identically is not right. That would only be right in the case of ideal, paper amplifiers. No one in that thread, or here has provided measurements to back that talking point. It was pure speculation as you are stating. They had not verified their hypothesis.

The test was not designed to do anything else than above. Its purpose was narrow and specific to invalidating a claim.
Quote:
Have a good day.
You too smile.gif.

Amir
Founder, Madrona Digital
"Insist on Quality Engineering"

amirm is online now  
post #382 of 651 Old 04-07-2014, 08:44 AM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 17,770
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 496 Post(s)
Liked: 326
Quote:
Originally Posted by arnyk View Post

I am unaware of that. Please cite a post with a URL.
Here it is, a few pages back in this thread: http://www.avsforum.com/t/1507920/bi-amping-b-w-cm10/270#post_24566328

Amir
Founder, Madrona Digital
"Insist on Quality Engineering"

amirm is online now  
post #383 of 651 Old 04-07-2014, 09:46 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 13,854
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 405 Post(s)
Liked: 1050
Quote:
Originally Posted by amirm View Post

Quote:
(1) AFAIK the HFN&RR article that I co-wrote has never been publicly criticized by any credible authority for any procedural problem. It is thus safe to say that Its procedures are just fine. BTW the article was published in 1982 - a mere 32 years ago. Talk about dredging the archives in a desperate effort to make trouble!

Oh, the "as far as I know" argument. biggrin.gif We don't know what you know Arny. What we know is that this is the first time this test has ever been published online. I pointed out one potential issue with it above. If we had an independent observer audit the work, we may have found other problems. I know what you are going to say: you are beyond any mistakes. Here is a good test case to counter that. The Boston Audio Society blind testing is put forward as expert witness and evidence by our camp all the time. Here is one of their tests: http://www.bostonaudiosociety.org/bas_speaker/abx_testing2.htm

"I expressed my desire to try the test, and Remington went to cue up the record again, but I requested to be allowed to undertake the test with no signal passing though the system. Before realizing the import of what he was saying, Vanderkooy interjected: "Ah! You're going to listen to the sound of the relays." Yes, there is indeed a slight audible difference between the acoustic "click" made when the "A" and "B" relays pull in. This is due to the unavoidable differences in the mounting positions of the relay on the A/B/X box chassis and, although slight, it can be heard if one listens for it. I replied that I was going to listen to the difference in background hiss, and the subsequent series of blind trials showed conclusively that the two signal paths could be reliably distinguished on this basis alone."

I have highlighted the key section. Had professor Vanderkooy not been there, it is entirely possible the person being tested could have gamed the results and we would have an egg on our face making him a hero instead of zero. Fortunately we got lucky there that he did not try to listen for the clicks and were made aware of our testing protocol mistake.

Vanderkooy and Lipshitz independently developed their own ABX box. I am not in any way responsible for its design or operational characteristics. They were commenting on their own handiwork, not mine.

The above shows how far some will go to impute mistakes I never made to me.
arnyk is online now  
post #384 of 651 Old 04-07-2014, 11:14 AM
AVS Special Member
 
Frank Derks's Avatar
 
Join Date: Jul 2002
Location: Region A,B,C
Posts: 1,885
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 90 Post(s)
Liked: 86
Amir, your test is flawed. It doesn't proof your point at all.


"Good point but all you are doing is addressing is a poor choice of words. The power amp is still generating the voltage at the frequencies that the crossover blocks. If either amplifier is generating audible distortion because it is being overdriven, then they both are generating audible distortion. If either amplifer is not generating audible distortion, then neither will be generating distortion.

I think this is the 432'd time (more or less) this particular issue has been re-hashed ;-)
"
Arny is right in both sentences.

"....You say we should look at things in db. OK, let's do that. OP says the power doubles. Doubling the power is 3 db. You say that it is "essentially" 0 db. My experiment showed the difference of 1.6 db or so. That is quite a ways up from 0 db that you said and represents genuine increase in total power delivered to the speaker. ..."

Despite the availability of twice the power it simple doesn't double with passive bi amping. There is no 3dB gain. Your claimed 1.6dB difference is a wrong conclusion. (And incorrectly calculated)

Redo the test as suggested, watch and hopefully learn.
Frank Derks is offline  
post #385 of 651 Old 04-07-2014, 12:00 PM
AVS Special Member
 
J_Palmer_Cass's Avatar
 
Join Date: May 2005
Posts: 6,290
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 118 Post(s)
Liked: 83
Quote:
Originally Posted by amirm View Post

Quote:
Originally Posted by UndersAVS View Post

Amir,
You've demonstrated that a Yamaha amp clipped when the impedance was low. Nothing more.
I am sorry but seems like the context of the test was lost. Here is the start of my explanation of the testing: http://www.avsforum.com/t/1492314/question-on-bi-amping/660#post_24002262 with Arny saying:

Good point but all you are doing is addressing is a poor choice of words. The power amp is still generating the voltage at the frequencies that the crossover blocks. If either amplifier is generating audible distortion because it is being overdriven, then they both are generating audible distortion. If either amplifer is not generating audible distortion, then neither will be generating distortion.

I think this is the 432'd time (more or less) this particular issue has been re-hashed ;-)

He is right in the last sentence. Up to that point, he and others kept repeating that line that both amps in a passive bi-amp situation clip at the same time. I have bolded that above in Arny's statement. No real measurements were provided to back that other than stating it as "everybody knows." So I set out to test that hypothesis. I used a 4 ohm load for the "woofer" amp. I used a high value resistor for the tweeter as that is the impedance that amp would see, when driving a 20 Hz tone to the tweeter. This allowed me to do the test without a crossover. I then turned up the volume until I observed clipping. In all *three* AVRs that I tested, the woofer amp clipped before the tweeter one. So the statement that both amps clip at the same time was shown to not be true.
Quote:
The rest can be chalked up as gross speculation concerning insignificant details, or off topic and petty bickering. Not something one would expect from someone so highly critical of other comments.
I am unclear what the "rest" is. Again, the test was designed to show the talking point regarding bi-amping being useless due to both amps working identically is not right. That would only be right in the case of ideal, paper amplifiers. No one in that thread, or here has provided measurements to back that talking point. It was pure speculation as you are stating. They had not verified their hypothesis.

The test was not designed to do anything else than above. Its purpose was narrow and specific to invalidating a claim.
Quote:
Have a good day.
You too smile.gif.



In the past I have used a passive bi-amp setup on a pair of speakers, and the difference is easily heard (meaning the system can play clean at a much higher SPL level). As a disclaimer, the speakers that I used in this test were 4-way speakers. For this test, the three way speaker section was driver by a Sony AVR, and the 10" subwoofer driver was driven by an external subwoofer amplifier.

I don't think I would waste my time on a passive bi-amp with a typical 2 way speaker.
J_Palmer_Cass is offline  
post #386 of 651 Old 04-07-2014, 12:29 PM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 13,854
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 405 Post(s)
Liked: 1050
Quote:
Originally Posted by amirm View Post

Quote:
Originally Posted by arnyk View Post

I am unaware of that. Please cite a post with a URL.
Here it is, a few pages back in this thread: http://www.avsforum.com/t/1507920/bi-amping-b-w-cm10/270#post_24566328

I guess this is what you are talking about:

"OK, it is your turn Arny. Do you have one of these on AVS Forum? Or any other forum? If not, I think I have won the game."

You can't get many bragging rights by claiming win a game when you're playing against just yourself, Amir.

When Clark's ABX paper passed JAES peer review, they said nothing about requiring things be posted on Internet audio forums. Come to think about it, there was no Internet way back then, or even for the next decade or so. Hmmm.

Do try to be relevant! ;-)
arnyk is online now  
post #387 of 651 Old 04-07-2014, 12:33 PM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 17,770
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 496 Post(s)
Liked: 326
Quote:
Originally Posted by Frank Derks View Post

Amir, your test is flawed. It doesn't proof your point at all.
Unless you can explain why, there is nothing i can say about that.

Quote:
"Good point but all you are doing is addressing is a poor choice of words. The power amp is still generating the voltage at the frequencies that the crossover blocks. If either amplifier is generating audible distortion because it is being overdriven, then they both are generating audible distortion. If either amplifer is not generating audible distortion, then neither will be generating distortion.

I think this is the 432'd time (more or less) this particular issue has been re-hashed ;-)
"
Arny is right in both sentences.
For that to be right, both of these waveforms would have equal audible distortion:

i-FR7FvFX-XL.png

Or these two:

i-9Jjsjrb-XL.png

Is this what you saying? Because if you are, it is not correct.
Quote:
"....You say we should look at things in db. OK, let's do that. OP says the power doubles. Doubling the power is 3 db. You say that it is "essentially" 0 db. My experiment showed the difference of 1.6 db or so. That is quite a ways up from 0 db that you said and represents genuine increase in total power delivered to the speaker. ..."

Despite the availability of twice the power it simple doesn't double with passive bi amping. There is no 3dB gain. Your claimed 1.6dB difference is a wrong conclusion. (And incorrectly calculated)
And I didn't say it did. I said the key benefit is that when one channel clips, it doesn't bleed into the other. The counter to that was that they both clip simultaneously. I ran the experiments to show that was not the case and no data is shown otherwise.
Quote:
Redo the test as suggested, watch and hopefully learn.
I know about your test but it is irrelevant to this discussion. The most powerful signal that we play is typically low frequencies. It is those that tax the amplifier the most. In a single amp situation, any distortion created as a result of low frequencies creates harmonics that travel to the tweeter as I have repeatedly explained. It was said that splitting the speaker into two halves does nothing to alleviate this. To prove that is wrong, I used a 20 Hz tone and drove both channels: one clipped (woofer) and the other did not (tweeter). This is precisely the scenario that brings out the differences between single and bi-amp and disproves the theory of simultaneous clipping.

Amir
Founder, Madrona Digital
"Insist on Quality Engineering"

amirm is online now  
post #388 of 651 Old 04-07-2014, 01:06 PM
AVS Special Member
 
Frank Derks's Avatar
 
Join Date: Jul 2002
Location: Region A,B,C
Posts: 1,885
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 90 Post(s)
Liked: 86
Music or sound effects contains more than just a single frequency.

Look at the two tone wave in the link below. It's the hf tone that clips first long before the lf tone clips.
http://www.st-andrews.ac.uk/~www_pa/Scots_Guide/audio/clipping/page3.html

The 'hf' amp clips anyway. Even if it's power delivery is far below it's rated power.
Frank Derks is offline  
post #389 of 651 Old 04-07-2014, 01:08 PM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 17,770
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 496 Post(s)
Liked: 326
Quote:
Originally Posted by arnyk View Post

I guess this is what you are talking about:

"OK, it is your turn Arny. Do you have one of these on AVS Forum? Or any other forum? If not, I think I have won the game."
No I am not. 99% of the post was something else related to the question you asked me here. And you went for that line? Again, I said that tests need to have a measure to detect errors of all kind, including scorekeeping. In that post I showed how a blind test run on this very forum had a major protocol error that I happen to catch. Had I not done that, they would have claimed that I and a number of others didn't have as good a hearing as others who got it "right." Turned out they got it wrong and we were the one in the right. I am not saying your test had protocol errors but if it did, we would not know it. Your tests lack controls which are standard in many blind tests. Here is ITU BS1116 recommendation for testing for small differences which you say you believe in:

"A major consideration is the inclusion of appropriate control conditions. Typically, control conditions include the
presentation of unimpaired audio materials, introduced in ways that are unpredictable to the subjects. It is the differences
between judgement of these control stimuli and the potentially impaired ones that allows one to conclude that the grades
are actual assessments of the impairments."


See? A "major consideration" is the control. Now, what they prescribe above is specific to audio compression but the notion is just as valid.
Quote:
When Clark's ABX paper passed JAES peer review, they said nothing about requiring things be posted on Internet audio forums. Come to think about it, there was no Internet way back then, or even for the next decade or so. Hmmm.
You are confusing the Web for the Internet. The Internet certainly existed in 1982 and so did Usenet newsgroups where you post in all the time. I know because I implemented the modem driver on our Unix systems then and was surprised by the avalanche of newsgroup messages. I just searched for my posts and could not go back before 1986. Here is a post from that era:
Quote:
Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 6/24/83; site houligan.UUCP
Path: utzoo!watmath!clyde!cbosgd!ihnp4!houxm!whuxl!whuxlm!akgua!akguc!codas!peora!ucf-cs!novavax!houligan!amir
From: amir@houligan.UUCP (Amir Majidimehr)
Newsgroups: net.rec.photo
Subject: Re: Vivitar 450mm f4.5
Message-ID: <297@houligan.UUCP>
Date: Fri, 14-Mar-86 14:57:53 EST
Article-I.D.: houligan.297
Posted: Fri Mar 14 14:57:53 1986
Date-Received: Sun, 16-Mar-86 08:57:10 EST
References: <5126@stolaf.UUCP>
Organization: Gould Electronics, Ft. Lauderdale, Florida.
Lines: 17

I too liked the reviews of the 450mm. I called a mail-order place in
Portland, and they gave a me similar story. That is, Vivitar had serious
reliability problems with the lens. And of the three that they sold, two of
the came back because of defects. They claimed that they couldn't
get them anymore.

This is a real shame. Where else can you get a 450mm, f4.5 for $350 or so?
My only other alternative was the Canon 450mm f4.5. But I still can't justify
its $730.00 price. I hope Vivitar solves their production problems soon.
The only thing that I can't figure out is why they are advertising so much
for this lens???
--
Amir H. Majidimehr
Unix Development Group
Gould Inc, Computer Systems Division
{sun,pur-ee,brl-bmd}!gould!amir

Good to know that I made grammar errors even then and it is not a sign of aging. biggrin.gif

Amir
Founder, Madrona Digital
"Insist on Quality Engineering"

amirm is online now  
post #390 of 651 Old 04-07-2014, 01:21 PM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 17,770
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 496 Post(s)
Liked: 326
Quote:
Originally Posted by Frank Derks View Post

Music or sound effects contains more than just a single frequency.

Look at the two tone wave in the link below. It's the hf tone that clips first long before the lf tone clips.
http://www.st-andrews.ac.uk/~www_pa/Scots_Guide/audio/clipping/page3.html

The 'hf' amp clips anyway. Even if it's power delivery is far below it's rated power.
The article is not relevant to our discussion as it has nothing to do with bi-amping. It talks about the ramifications of clipping (and the old argument on what damages the tweeters) but we are not discussing that in the specific. We take clipping as something that can occur with music or tones, and show whether both amps clip in bi-amp configuration or one. My measurements show the latter.. Your article even if it were on this topic, has no measurements. I have already said that paper/ideal assumptions would generate different results.

Amir
Founder, Madrona Digital
"Insist on Quality Engineering"

amirm is online now  
Reply Audio theory, Setup and Chat

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