External DAC Questions - Page 9 - AVS Forum
Forum Jump: 
Reply
 
Thread Tools
post #241 of 267 Old 01-30-2014, 03:01 PM
 
Heinrich S's Avatar
 
Join Date: Aug 2007
Posts: 974
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 29
Ratman, aren't you the same person who said that sound levels couldn't sum constructively? I'm sure you were. So please don't send me engineering degree links. I'm asking for help in understanding something and I'm sure it doesn't require an engineering degree.
Heinrich S is offline  
Sponsored Links
Advertisement
 
post #242 of 267 Old 01-30-2014, 03:35 PM
AVS Special Member
 
Glimmie's Avatar
 
Join Date: Sep 2000
Location: Los Angeles, CA
Posts: 7,785
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 36 Post(s)
Liked: 163
Quote:
Originally Posted by DonH50 View Post

Can you still get output devices for them?

Why does that matter? Transistors don't fail. That why we got rid of tubes. biggrin.gif

Glimmie's HT Page
Being redone - comming soon!

Glimmie is offline  
post #243 of 267 Old 01-30-2014, 03:35 PM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 17,651
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 375 Post(s)
Liked: 292
Trying to clear up my backlog of posts I have meant to reply smile.gif. Answering the question: "Are there universal laws as to what humans can and can't hear? If so, where and what are they?" Arny responded with:
Quote:
Originally Posted by arnyk View Post


You can find evidence related to their existence in Zwicker & Fastl's Psychoacoustics Facts and Models.

http://www.amazon.com/Psychoacoustics-Models-Springer-Information-Sciences/dp/3540650636

I don't know what "evidence" there is for any "universal laws" in Zwicker and Fastl's book. This is not easy reading material and I would not recommend it for anyone here to casually pick up and try to understand. The book is really a collection of their papers and studies and assumes a person in the field as the reader. The book does have extensive set of studies on our listening thresholds. This data is broad, complex and at times rough. Of note, "none are blind tests!" Let me repeat, they are not double or even single blind tests but clearly sighted. Here is the authors on set of tests:

"Method of Adjustment. In this method, the subject has control over the stimulus. For example, the subject can vary the level of a pure tone until it is just audible."

So clearly these are sighted tests. Interesting that Arny would be in favor of sighted tests here, but not elsewhere. I suspect he didn't know that these tests are sighted.

Related, the sample size can often be quite small. Here is an example below figure 7.1:

"The data indicated by dots connected with broken lines are for just-noticeable level and just-noticeable frequency differences. Medians and interquartile ranges for six subjects are given"

Just 6 subjects. So clearly not a broad sample size statistically speaking.

Note that my personal opinion is that such sighted tests are fine for this type of research. There is some degree of error but as first order approximation they are fine. Indeed 99% of the audio research in the world gets done sighted.

With this background, let's look at one of the easier to understand graphs, namely how small of a level difference we can detect:

i-5RdQr2z.png

As we see there, our detection threshold for this 1 Khz tone keeps going down/we become more accurate the louder the signal is. It is for this reason that I like to test for small differences in sources using headphones so that I can crank up the volume to hear all the low level detail. Now, let me ask you: how often have you seen someone mention at what SPL level they did their blind tests? Not often I bet. How easy do you think it is to get negative outcome? Pretty easy. Just lower the volume and detection threshold goes up and folk's can't tell the difference.

See how much insight and deep just evaluating on tiny bit of information is in this field/book? This chapter on JND goes on and on this way. The above graph for example changes if you change the duration of the gaps between tones! Changing the tone to noise also changes it. As would testing other frequencies (not documented in the book). There is no way I would want to convey the message that Arny is doing with respect to these being a set of cut and dry "laws" of how we hear. Every audio encoder (e.g. MP3 or AAC) has a perceptual model of our hearing system. Yet a human can still catch mistakes they make. And two encoders will sound different and hence, do not agree with each other what the model of our hearing system is.

Mind you, we are not at all blind with respect to how our hearing system works. The point being that as with many other technical arguments here, folks like to exaggerate what the science is, hoping the recipient doesn't have access to the research or knowledge to understand it. This is not ethical in my book.

Amir
Founder, Madrona Digital
"Insist on Quality Engineering"

amirm is online now  
post #244 of 267 Old 01-30-2014, 03:44 PM
AVS Addicted Member
 
Ratman's Avatar
 
Join Date: May 2002
Location: Collingswood, N.J.
Posts: 14,254
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 73 Post(s)
Liked: 235
Quote:
Originally Posted by Heinrich S View Post

Ratman, aren't you the same person who said that sound levels couldn't sum constructively? I'm sure you were. So please don't send me engineering degree links. I'm asking for help in understanding something and I'm sure it doesn't require an engineering degree.
If you need to ask, consider formal education as opposed to seeking a "degree" on a forum. wink.gif
Ratman is offline  
post #245 of 267 Old 01-30-2014, 05:06 PM
AVS Special Member
 
MarkHotchkiss's Avatar
 
Join Date: Mar 2007
Location: Long Beach, California
Posts: 1,476
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 82
Hi Heinrich,
Quote:
Originally Posted by Heinrich S View Post

So the input sensitivity has nothing to do with it? Or is it due to gain structure on the amp?
The input sensitivity should have nothing to do with it, as the input sensitivity should be almost the same for all amps (about 1 volt for an unbalanced input). The gain structure would effect it, but the gain structure is simply a byproduct of other design goals.
Quote:
So basically there is no spec that will explain it. If everything in audio can be explained technically, then I find it difficult to understand why this can't be supported by a spec.
It could be supported by a spec, but it would be a meaningless spec. What difference does it make where you might need to put the volume-control knob? What really matters is how loud the amp can get, and other specs, specifically the power rating, would tell you that.
MarkHotchkiss is offline  
post #246 of 267 Old 01-30-2014, 06:05 PM
AVS Special Member
 
mcnarus's Avatar
 
Join Date: Nov 2007
Posts: 5,997
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 260
Quote:
"Method of Adjustment. In this method, the subject has control over the stimulus. For example, the subject can vary the level of a pure tone until it is just audible."

So clearly these are sighted tests.
So clearly Amir does not understand what makes a test blind.
kbarnes701 and Jack D Ripper like this.

If you can't explain how it works, you can't say it doesn't.—The High-End Creed

mcnarus is offline  
post #247 of 267 Old 01-30-2014, 06:42 PM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 17,651
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 375 Post(s)
Liked: 292
Quote:
Originally Posted by mcnarus View Post

So clearly Amir does not understand what makes a test blind.
If you think those are blind tests, they must have put up these signs for you:

ahmar2-1782911.jpg

biggrin.gif

Amir
Founder, Madrona Digital
"Insist on Quality Engineering"

amirm is online now  
post #248 of 267 Old 01-30-2014, 07:54 PM
AVS Special Member
 
DonH50's Avatar
 
Join Date: Feb 2010
Location: Monument CO
Posts: 5,853
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 63 Post(s)
Liked: 224
Quote:
Originally Posted by andyc56 View Post

Quote:
Originally Posted by DonH50 View Post

I clearly remember the results, but now am struggling (30+ years later and with a whole lot more experience) to remember what I did that would have affected that, like changing feedback/coupling caps and such...
I'd guess that a change to the lower -3dB frequency was probably made at some point. Here's a SPICE simulation of a 200 Hz square wave with a simple HPF having -3 dB frequencies of 20 Hz and 2 Hz.

It looks like the -3 dB frequency needs to be about 1/100 of the square wave fundamental frequency for the tilt to be negligible.

Thanks Andy, that makes a lot more sense. It has been far too long for me to remember what all I did, and I was in college at the time so still pretty wet behind the ears. Wish I had that amp to figure it out!

"After silence, that which best expresses the inexpressible, is music" - Aldous Huxley
DonH50 is offline  
post #249 of 267 Old 01-30-2014, 11:27 PM
 
Heinrich S's Avatar
 
Join Date: Aug 2007
Posts: 974
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 29
Quote:
Originally Posted by MarkHotchkiss View Post

Hi Heinrich,
The input sensitivity should have nothing to do with it, as the input sensitivity should be almost the same for all amps (about 1 volt for an unbalanced input). The gain structure would effect it, but the gain structure is simply a byproduct of other design goals.
It could be supported by a spec, but it would be a meaningless spec. What difference does it make where you might need to put the volume-control knob? What really matters is how loud the amp can get, and other specs, specifically the power rating, would tell you that.

Thanks for replying! But correct me if I'm wrong, but a different gain structure and volume pot on one amp could result in the subjective terms described as "laid-back", "warm" etc etc when compared to another with a different gain structure?
Heinrich S is offline  
post #250 of 267 Old 01-31-2014, 12:47 AM
AVS Special Member
 
MarkHotchkiss's Avatar
 
Join Date: Mar 2007
Location: Long Beach, California
Posts: 1,476
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 82
Hi Heinrich,
Quote:
Originally Posted by Heinrich S View Post

Thanks for replying! But correct me if I'm wrong, but a different gain structure and volume pot on one amp could result in the subjective terms described as "laid-back", "warm" etc etc when compared to another with a different gain structure?
I can't say that I would recognize "warm" or "laid-back" since, as you said, those are pretty subjective terms. The gain structure should only affect the output voltage, resulting in a different volume. It should have no effect on the quality of the audio.

However, comparing two amps at different volumes would certainly sound different, and some people could then interpret the different volume in more subjective terms. Somebody might mistake a lower volume as "laid-back".
MarkHotchkiss is offline  
post #251 of 267 Old 01-31-2014, 03:28 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 13,654
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 288 Post(s)
Liked: 997
Quote:
Originally Posted by amirm View Post

Trying to clear up my backlog of posts I have meant to reply smile.gif. Answering the question: "Are there universal laws as to what humans can and can't hear? If so, where and what are they?" Arny responded with:
Quote:
Originally Posted by arnyk View Post


You can find evidence related to their existence in Zwicker & Fastl's Psychoacoustics Facts and Models.

http://www.amazon.com/Psychoacoustics-Models-Springer-Information-Sciences/dp/3540650636

I don't know what "evidence" there is for any "universal laws" in Zwicker and Fastl's book.

The idea that there are "universal laws" related to say Physics is hardly strange or unique to me. Here is just one of many examples:

http://en.wikipedia.org/wiki/Natural_law

"Natural law, or the law of nature (Latin: lex naturalis), is a system of law that is determined by nature, and so is universal."

I submit that people who study and understand Zwicker and Fastl's book may realize that wicker and Fastl are shedding light on universal natural law as related to psychoacoustics.
Quote:
This is not easy reading material and I would not recommend it for anyone here to casually pick up and try to understand. The book is really a collection of their papers and studies and assumes a person in the field as the reader. The book does have extensive set of studies on our listening thresholds. This data is broad, complex and at times rough. Of note, "none are blind tests!" Let me repeat, they are not double or even single blind tests but clearly sighted. Here is the authors on set of tests:

"Method of Adjustment. In this method, the subject has control over the stimulus. For example, the subject can vary the level of a pure tone until it is just audible."

So clearly these are sighted tests.

Amir, I seem to recall that we've had this disagreement before. I submit that it is possible to give the subject total control over the stimulus, and still have a DBT. In fact that is one of the tenets of ABX testing - we formulated a test where the subject controls all aspects of the test, and by following a few simple rules and guidelines, there is a perfectly valid DBT. I believe that I have a refereed JAES paper to back me up on that point. ;-)
arnyk is offline  
post #252 of 267 Old 01-31-2014, 03:38 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 13,654
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 288 Post(s)
Liked: 997
Quote:
Originally Posted by Heinrich S View Post


But correct me if I'm wrong, but a different gain structure and volume pot on one amp could result in the subjective terms described as "laid-back", "warm" etc etc when compared to another with a different gain structure?

The nature of human subjectivity, as I have experienced myself and observed it in my friends and acquaintances over the past 50 years, is that no actual physical change at all can result in the subjective terms described as "laid-back", "warm" etc etc when compared to another.

Since no change at all is required for those subjective terms, just about any actual change can also be described that way, level shifts included

As far as your amp comparison goes, I once did an experiment where I passed a tone with fixed frequency and level though an ordinary volume control and measured its level at the output. I simply twisted the knob to minimum and maximum and then adjusted the knob as closely to a predetermined amount such as 6 o'clock, 12 o'clock or 3 o'clock as I could. There were variations of up to several dB in what appeared to me to be the same setting of the knob.

And this my friend is why some people are so dogmatic about using good voltmeters to set levels. An axiom of audio: "Set levels by ear and hear differences in level by ear". Everybody who has attempted fine carpentry has practical experience in this area.
Rgb likes this.
arnyk is offline  
post #253 of 267 Old 01-31-2014, 06:10 AM
 
Heinrich S's Avatar
 
Join Date: Aug 2007
Posts: 974
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 29
Quote:
Originally Posted by arnyk View Post

The nature of human subjectivity, as I have experienced myself and observed it in my friends and acquaintances over the past 50 years, is that no actual physical change at all can result in the subjective terms described as "laid-back", "warm" etc etc when compared to another.

Since no change at all is required for those subjective terms, just about any actual change can also be described that way, level shifts included

As far as your amp comparison goes, I once did an experiment where I passed a tone with fixed frequency and level though an ordinary volume control and measured its level at the output. I simply twisted the knob to minimum and maximum and then adjusted the knob as closely to a predetermined amount such as 6 o'clock, 12 o'clock or 3 o'clock as I could. There were variations of up to several dB in what appeared to me to be the same setting of the knob.

And this my friend is why some people are so dogmatic about using good voltmeters to set levels. An axiom of audio: "Set levels by ear and hear differences in level by ear". Everybody who has attempted fine carpentry has practical experience in this area.

Thank you for this!!!!

I suspected there may have been a reasonable explanation for what I experienced. Whether it was the actual reason or not I can't say for sure, after all, it may have been my imagination, but the reason you provided is a GOOD reason and I accept it. I wish there was a spec on the amp sheet that could offer insight into this, it would make life easier knowing if the cause was related to a spec.

Thanks again for the reply.
Heinrich S is offline  
post #254 of 267 Old 02-01-2014, 09:46 AM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 17,651
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 375 Post(s)
Liked: 292
Quote:
Originally Posted by arnyk View Post

Amir, I seem to recall that we've had this disagreement before. I submit that it is possible to give the subject total control over the stimulus, and still have a DBT. In fact that is one of the tenets of ABX testing - we formulated a test where the subject controls all aspects of the test, and by following a few simple rules and guidelines, there is a perfectly valid DBT. I believe that I have a refereed JAES paper to back me up on that point. ;-)
Trying to portray an obviously sighted test as blind is like a preacher justifying having sex with a prostitute as "God told me to do it!" biggrin.gif Yes, we had this discussion before in the context of jitter and you citing the dolby listening tests, not realizing until I pointed out that the test was completely sighted. Backpedaling started to no effect. When the author of a test does not use the word blind, you don't get to put that their mouth with some voodoo logic.

Let me share a personal experience. Years ago I got a PS Audio Power Plant. This is what it looks like:

1100psa.1.jpg

This is a power "regeneration" device. It takes the AC power, converts it to stable DC, then using a Microprocessor, creates a precise AC frequency, e.g. 60 Hz, and outputs it at the rated voltage and current. The upshot is that the output waveform is much cleaner than input. They went one step further: since they have a DSP, they made the frequency programmable. The "common wisdom" is that these devices are rip offs and the notion of changing the frequency just adds insult to injury. That the power supply in our audio devices is already removing all the noise from AC so adding such upstream devices does nothing. That is how I went into this evaluation.

I connected the Power Plant to my Mark Levinson DAC and ran some of my difficult clips though it while I changed the frequency of the AC voltage generated by it. As I went up from 60 Hz, I could detect improved quality. Surprised, I assumed it was just placebo that "higher is better." More surprised was when I kept increasing the frequency and all of a sudden, the improvements stopped and from there on, the sound actually degraded! In other words, there was an optimal frequency to drive my DAC. My memory is very hazy as to what that number was but I seem to recall it was in 85 Hz or so. I went looking for a picture of the power plant for this reply and by chance, landed on the review of it on Stereophile. I was super surprised to read this section:

"The Power Plant has a feature that, as far as I know, is unique among equipment offered for audiophile use: variable AC frequency. This so-called Power Factor allows AC frequency to be varied from the normal 60Hz to a high of 120Hz. (The 50Hz setting is for countries where this is the normal AC frequency.) PS Audio claims that increasing the Power Factor makes the equipment's power supply more efficient, increasing the effective value of capacitors and decreasing the transformer's radiated magnetic field.

In my system, increasing the Power Factor produced a startling sonic improvement. Dynamics—already enhanced compared to the raw AC condition—were even more dramatic, to the extent that, after comparing the same CD track with the Power Factor at 60Hz and at 90Hz, one listener accused me of having turned up the volume. (I didn't—honest!)

There was a significant improvement in overall clarity, and voices sounded more rounded, less synthetic. My wife, a fan of the late John Denver, said that with the Power Factor at 90 his voice "sounded more like John." The most beneficial effect seemed to be at the setting of 90Hz or thereabouts, the 120Hz setting resulting in some thinning of the sound."


I am so surprised to see them hearing the same effect and around the same frequency. Anyway, let's review your litmus test for a test bing bind. My testing of the threshold of audibility or lack thereof of this frequency effect is an analog of what went on the research by Zwicker and Faslt. As you say, the testers had control of the stimulus and so did I. I didn't know of the outcome prior to the test and neither did they. By the way, the machine was free to me so I had no motivation to use it. And like the researchers, I have another tester at Stereophile arriving at same conclusions/thresholds as I have.

So please explain if my test is double blind and hence, authoritative. And if it is not, what separates it from the Zwicker and Fastl tests.

Amir
Founder, Madrona Digital
"Insist on Quality Engineering"

amirm is online now  
post #255 of 267 Old 02-01-2014, 10:52 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 13,654
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 288 Post(s)
Liked: 997
Quote:
Originally Posted by amirm View Post

Quote:
Originally Posted by arnyk View Post

Amir, I seem to recall that we've had this disagreement before. I submit that it is possible to give the subject total control over the stimulus, and still have a DBT. In fact that is one of the tenets of ABX testing - we formulated a test where the subject controls all aspects of the test, and by following a few simple rules and guidelines, there is a perfectly valid DBT. I believe that I have a refereed JAES paper to back me up on that point. ;-)

Trying to portray an obviously sighted test as blind is like a preacher justifying having sex with a prostitute as "God told me to do it!"

Amir, according to your logic, it wold appear that every ABX test ever done was a sighted test because in ABX tests the listener can control all aspects of the test. Since The AES seems to disagree with you, I suggest that you take it up with them. ;-)

Obviously, the issue is not whether or not the listener controls the test, but whether or not the controls he exercises can be reasonably be expected to significantly affect the outcome of the test in such a way that his biases are represented in the test's outcome.
arnyk is offline  
post #256 of 267 Old 02-01-2014, 11:59 AM
AVS Addicted Member
 
kbarnes701's Avatar
 
Join Date: Jan 2008
Location: Main Listening Positon
Posts: 16,225
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 303 Post(s)
Liked: 1143
Quote:
Originally Posted by arnyk View Post
 
Quote:
Originally Posted by amirm View Post
 
Quote:
Originally Posted by arnyk View Post

Amir, I seem to recall that we've had this disagreement before. I submit that it is possible to give the subject total control over the stimulus, and still have a DBT. In fact that is one of the tenets of ABX testing - we formulated a test where the subject controls all aspects of the test, and by following a few simple rules and guidelines, there is a perfectly valid DBT. I believe that I have a refereed JAES paper to back me up on that point. ;-)

Trying to portray an obviously sighted test as blind is like a preacher justifying having sex with a prostitute as "God told me to do it!"

Amir, according to your logic, it wold appear that every ABX test ever done was a sighted test because in ABX tests the listener can control all aspects of the test. Since The AES seems to disagree with you, I suggest that you take it up with them. ;-)

Obviously, the issue is not whether or not the listener controls the test, but whether or not the controls he exercises can be reasonably be expected to significantly affect the outcome of the test in such a way that his biases are represented in the test's outcome.

 

You mean that people conflate a) having control of the stimulus and b) knowing which stimulus is being used at any given moment?  They think that is the same thing? Amazing.

kbarnes701 is offline  
post #257 of 267 Old 02-01-2014, 12:25 PM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 17,651
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 375 Post(s)
Liked: 292
Quote:
Originally Posted by arnyk View Post

Amir, according to your logic, it wold appear that every ABX test ever done was a sighted test because in ABX tests the listener can control all aspects of the test. Since The AES seems to disagree with you, I suggest that you take it up with them. ;-)

Obviously, the issue is not whether or not the listener controls the test, but whether or not the controls he exercises can be reasonably be expected to significantly affect the outcome of the test in such a way that his biases are represented in the test's outcome.
You didn't answer my question Arny. I gave you a specific scenario. Is it double blind or not?

Amir
Founder, Madrona Digital
"Insist on Quality Engineering"

amirm is online now  
post #258 of 267 Old 02-01-2014, 12:38 PM
AVS Special Member
 
mcnarus's Avatar
 
Join Date: Nov 2007
Posts: 5,997
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 260
Quote:
So please explain if my test is double blind and hence, authoritative
First of all, being double blind would not by itself make it authoritative. Second, it was not double blind because you were getting a non-sonic stimulus, i.e., the frequency readout.

If you can't explain how it works, you can't say it doesn't.—The High-End Creed

mcnarus is offline  
post #259 of 267 Old 02-01-2014, 01:19 PM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 17,651
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 375 Post(s)
Liked: 292
Quote:
Originally Posted by mcnarus View Post

First of all, being double blind would not by itself make it authoritative. [Second, it was not double blind because you were getting a non-sonic stimulus, i.e., the frequency readout.
Oh I have done the test without looking with the same results. But let's not pile on more on your plate smile.gif. Let me move the argument forward with a sentence from the authors:

"All the psychophysical methods discussed so far have the common feature that a final value of the threshold or of ratio can be deduced from a single trial. In the first two methods described, the subject is actively involved in the task by controlling the stimulus. sometimes such an activity may produce bias, as for example in loudness comparison."

So clearly the tests involve the tester knowing everything as I did. And as a result, the authors, as me, state potential for bias. Doesn't leave much wiggle room for you all's argument, does it?

BTW, the sentence after the above offers a mitigation. I leave it as an exercise for the reader to determine if that makes it "double blind."

I am curious if you all have the book and have read this, why it has not been noted and arguments made in absence of it. If you have not read it, then you are just speculating and we could do without that.

Amir
Founder, Madrona Digital
"Insist on Quality Engineering"

amirm is online now  
post #260 of 267 Old 02-02-2014, 06:28 AM
Super Moderator
 
DrDon's Avatar
 
Join Date: Mar 2002
Location: Metro Detroit
Posts: 12,498
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 25 Post(s)
Liked: 218
Thread has been rolled back to remove off-topic, insulting and bickering posts.

Further such behavior will result in the loss of posting privileges in this thread without further warning.
gnk1 likes this.

Walking the fine line between jaw-dropping and a plain ol' yawn.
DrDon is offline  
post #261 of 267 Old 02-02-2014, 09:19 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 13,654
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 288 Post(s)
Liked: 997
Quote:
Originally Posted by amirm View Post

Quote:
Originally Posted by mcnarus View Post

First of all, being double blind would not by itself make it authoritative. [Second, it was not double blind because you were getting a non-sonic stimulus, i.e., the frequency readout.
Oh I have done the test without looking with the same results. But let's not pile on more on your plate smile.gif. Let me move the argument forward with a sentence from the authors:

"All the psychophysical methods discussed so far have the common feature that a final value of the threshold or of ratio can be deduced from a single trial. In the first two methods described, the subject is actively involved in the task by controlling the stimulus. sometimes such an activity may produce bias, as for example in loudness comparison."

So clearly the tests involve the tester knowing everything as I did. And as a result, the authors, as me, state potential for bias. Doesn't leave much wiggle room for you all's argument, does it?

BTW, the sentence after the above offers a mitigation. I leave it as an exercise for the reader to determine if that makes it "double blind."

I am curious if you all have the book and have read this, why it has not been noted and arguments made in absence of it. If you have not read it, then you are just speculating and we could do without that.

I am unaware of "sometimes" being the same as "certainly". "Potential" does not usually always mean the same as "Actual". "May" does not usually mean the same thing as "Does".

I'm willing to admit that due to human error or even malfeasance any particular listening test allegedly using any protocol might produce invalid or irrelevant results. So what?
arnyk is offline  
post #262 of 267 Old 02-02-2014, 10:50 AM
AVS Special Member
 
Glimmie's Avatar
 
Join Date: Sep 2000
Location: Los Angeles, CA
Posts: 7,785
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 36 Post(s)
Liked: 163
Quote:
Originally Posted by amirm View Post

Trying to portray an obviously sighted test as blind is like a preacher justifying having sex with a prostitute as "God told me to do it!" biggrin.gif Yes, we had this discussion before in the context of jitter and you citing the dolby listening tests, not realizing until I pointed out that the test was completely sighted. Backpedaling started to no effect. When the author of a test does not use the word blind, you don't get to put that their mouth with some voodoo logic.

Let me share a personal experience. Years ago I got a PS Audio Power Plant. This is what it looks like:

1100psa.1.jpg

This is a power "regeneration" device. It takes the AC power, converts it to stable DC, then using a Microprocessor, creates a precise AC frequency, e.g. 60 Hz, and outputs it at the rated voltage and current. The upshot is that the output waveform is much cleaner than input. They went one step further: since they have a DSP, they made the frequency programmable. The "common wisdom" is that these devices are rip offs and the notion of changing the frequency just adds insult to injury. That the power supply in our audio devices is already removing all the noise from AC so adding such upstream devices does nothing. That is how I went into this evaluation.

I connected the Power Plant to my Mark Levinson DAC and ran some of my difficult clips though it while I changed the frequency of the AC voltage generated by it. As I went up from 60 Hz, I could detect improved quality. Surprised, I assumed it was just placebo that "higher is better." More surprised was when I kept increasing the frequency and all of a sudden, the improvements stopped and from there on, the sound actually degraded! In other words, there was an optimal frequency to drive my DAC. My memory is very hazy as to what that number was but I seem to recall it was in 85 Hz or so. I went looking for a picture of the power plant for this reply and by chance, landed on the review of it on Stereophile. I was super surprised to read this section:

"The Power Plant has a feature that, as far as I know, is unique among equipment offered for audiophile use: variable AC frequency. This so-called Power Factor allows AC frequency to be varied from the normal 60Hz to a high of 120Hz. (The 50Hz setting is for countries where this is the normal AC frequency.) PS Audio claims that increasing the Power Factor makes the equipment's power supply more efficient, increasing the effective value of capacitors and decreasing the transformer's radiated magnetic field.

In my system, increasing the Power Factor produced a startling sonic improvement. Dynamics—already enhanced compared to the raw AC condition—were even more dramatic, to the extent that, after comparing the same CD track with the Power Factor at 60Hz and at 90Hz, one listener accused me of having turned up the volume. (I didn't—honest!)

There was a significant improvement in overall clarity, and voices sounded more rounded, less synthetic. My wife, a fan of the late John Denver, said that with the Power Factor at 90 his voice "sounded more like John." The most beneficial effect seemed to be at the setting of 90Hz or thereabouts, the 120Hz setting resulting in some thinning of the sound."


I am so surprised to see them hearing the same effect and around the same frequency. Anyway, let's review your litmus test for a test bing bind. My testing of the threshold of audibility or lack thereof of this frequency effect is an analog of what went on the research by Zwicker and Faslt. As you say, the testers had control of the stimulus and so did I. I didn't know of the outcome prior to the test and neither did they. By the way, the machine was free to me so I had no motivation to use it. And like the researchers, I have another tester at Stereophile arriving at same conclusions/thresholds as I have.

So please explain if my test is double blind and hence, authoritative. And if it is not, what separates it from the Zwicker and Fastl tests.

That was a very good audiophile report of the device.

Now lets see some scope images of the input AC waveform versus the DC rails of the DAC power supply. I'm quite sure your DAC uses regulated power supplies hence my doubt how this device could alter the sonic characteristics.

I am not trying to be confrontational here but if you honestly heard something and coupled with your engineering background, don't you want to investigate why? Because it doesn't make sense on the surface.

I do acknowledge a cleaner AC input will possibly lower the noise floor. And also how this deliberate harmonic addition could effect an unregulated power supply, I don't see how it has an effect on a linear regulated supply. Even less so on a switch mode power supply.

P.S. I know you already know better than this but make sure we can see the time and voltage scales on the scope display. None of this popular audiophile magazine crap of showing convincing waveforms with the scales unknown please.

Glimmie's HT Page
Being redone - comming soon!

Glimmie is offline  
post #263 of 267 Old 02-02-2014, 10:52 AM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 17,651
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 375 Post(s)
Liked: 292
Quote:
Originally Posted by arnyk View Post

I am unaware of "sometimes" being the same as "certainly". "Potential" does not usually always mean the same as "Actual". "May" does not usually mean the same thing as "Does".

I'm willing to admit that due to human error or even malfeasance any particular listening test allegedly using any protocol might produce invalid or irrelevant results. So what?
Well, are you aware of how often such disclaimers are provided by the researchers for double blind/ABX tests? Have you for example for results that match your arguments?

How often a single trial is used? How often do you see an double blind/ABX test with no statistical analysis? What potential for subject bias exists in double blind/ABX tests?

If the disclaimer is due to "human error or malfeasance" why do the authors use the phrase: "may introduce *bias*?" Isn't that actually characterizing the problem with the test, i.e. the listener bias getting in the way of achieving correct results as opposed to any other human error or unknown problem? And why did they propose a mitigation for that problem? We can't fix the potential for human error or malfeasance, right?

Should we mistrust any double blind tests you put forward for the same excuses you are providing? And if mistrust exists in this situation, why is it that you stated the results of their research as "laws" of how we hear?

Do you think the authors don't know what double blind testing is and therefore, did not know to state it as the protocol for these tests?

Is it really that uncomfortable to say these are not double blind tests? What damage comes from that Arny? I am a strong believer in double blind tests yet, as with countless researchers, have no problems at all with also accepting these tests. Very little audio research is done with double blind tests. If we excluded all the rest, we would make zero progress!

Amir
Founder, Madrona Digital
"Insist on Quality Engineering"

amirm is online now  
post #264 of 267 Old 02-02-2014, 10:59 AM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 17,651
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 375 Post(s)
Liked: 292
Quote:
Originally Posted by Glimmie View Post

That was a very good audiophile report of the device.

Now lets see some scope images of the input AC waveform versus the DC rails of the DAC power supply. I'm quite sure your DAC uses regulated power supplies hence my doubt haw this device could alter the sonic characteristics.

I am not trying to be confrontational here but if you honestly heard something and couples with yoour engineering background, don't you want to investigate why. Because it doesn't make sense on the surface.

P.S. I know you already know better than this but make sure we can see the time and voltage scales on the scope display. None of this popular audiophile magazine crap of showing convincing waveforms without the scaled unknown please.
We are on the same page Glimmie. smile.gif To be clear, my test did not in any way, shape or form, comply with single or double blind standards. As I noted though, I did close my eyes, randomized the frequency by pushing the up/down buttons many times and I could still land in the same region of frequencies as far as improvements. I wanted to perform one final test that I could not: Switch instantly between 60 Hz and 85 Hz to see if I could tell them apart blind. But there is no such button. You have to sequence through the frequencies and that creates a gradual result. Without this, I am not here to declare my results as something valid.

I like your suggestion. Time permitting I will go ahead and make measurements and report back. Here is a puzzling bit for now: if I kept increasing the frequency at some point, the Mark Levinson DAC would display an error that said something about the power. Unfortunately the code is not documented. I don't know what it is checking. Maybe overvoltage??? Do you have any ideas? Maybe I can tease this out using instrumentation.

Amir
Founder, Madrona Digital
"Insist on Quality Engineering"

amirm is online now  
post #265 of 267 Old 02-03-2014, 07:14 AM
 
bralas's Avatar
 
Join Date: Jan 2014
Location: Lafayette, LA
Posts: 117
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 20
Quote:
Originally Posted by arnyk View Post


You've lost me again!

I wonder what post you are referring to. Is it even on this thread? A link or post number might clarify things.

http://www.avsforum.com/t/1512853/monitor-audio-rx6-distortion/30#post_24260405

Here: :rolleyes:

 

Inside 'this thread', you've associate me with some unknown forum member that posted within the last year... yet you seem unable to recall multiple exchanges with me ~1 week ago? My reply toward you was illustrating how low freq vs hi freq energy is distributed within complex audio. When one observes audio over the longer time-base,  low freq impulse(s) WILL "null" giving opportunity for the higher freq average power to rise, thus matching the lower frequency power. The Rane technical note #128 is spot and I've explained it to you IMO quit succinctly three times now.

 

Remember?   

bralas is offline  
post #266 of 267 Old 04-14-2014, 10:05 PM
Member
 
ron hawaii's Avatar
 
Join Date: Apr 2014
Posts: 29
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
I have pretty good system but terrible source.

Cambridge 751r receiver gallo fronts strada center min21 surrounds dual subs

Unfortunately most of my music library is mp3s.

Do you think a DAC would help? Or use internal cambridge dac with apple tv as media player is ok?

If there is a setup convenient and not too expensive that would help please give input
ron hawaii is offline  
post #267 of 267 Old 04-15-2014, 02:47 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 13,654
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 288 Post(s)
Liked: 997
Quote:
Originally Posted by ron hawaii View Post

I have pretty good system but terrible source.

Cambridge 751r receiver gallo fronts strada center min21 surrounds dual subs

Unfortunately most of my music library is mp3s.

Do you think a DAC would help?

Surely you jest!
Quote:
Or use internal cambridge dac with apple tv as media player is ok?

It would sound good with a good source. Fix the source!
Quote:
If there is a setup convenient and not too expensive that would help please give input

My strategy is to obtain the original CDs by various means and rip and encode them myself. To that end I built an 8-core 64 bit PC with 3 DVD drives that can rip and convert 3 CD's at a time at top speed. As soon as I find a 4th DVD drive...
arnyk is offline  
Reply Audio theory, Setup and Chat

Tags
Marantz Sr5005 Audio Video Receiver Black , Paradigm Reference Studio 20
Gear in this thread

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