There is too much anecdotal testimony to entirely dismiss the BB's in my opinion.
The explanations of how they work are actually irrelevant to this discussion I reckon and may well be a deliberate red herring...doesn't matter how they work, the thing is that experienced ears say that they do something to the sound of audio systems.
Experienced ears state that other passive components make audible differences, so why not the BB, even if it is only the resistor within the BB.
Perhaps someone should get hold of some of those resistors and see if they alone make an audible difference when applied in the same way as BB's.
It does not take very much to alter the timbre of an audio system, and it may well be that subtle changes in levels of harmonics at sub -120dB are audible.
I am just trying to flesh out why experienced ears are saying BB's make an audible difference the same as everybody else who is sufficiently interested.
Perhaps sine wave testing is inappropriate in this situation ?.
If many experienced ears hear a difference there must be a reason other than psychological.
The real test is to devise the appropriate testing methodology.
Eric.
The explanations of how they work are actually irrelevant to this discussion I reckon and may well be a deliberate red herring...doesn't matter how they work, the thing is that experienced ears say that they do something to the sound of audio systems.
Experienced ears state that other passive components make audible differences, so why not the BB, even if it is only the resistor within the BB.
Perhaps someone should get hold of some of those resistors and see if they alone make an audible difference when applied in the same way as BB's.
It does not take very much to alter the timbre of an audio system, and it may well be that subtle changes in levels of harmonics at sub -120dB are audible.
I am just trying to flesh out why experienced ears are saying BB's make an audible difference the same as everybody else who is sufficiently interested.
Perhaps sine wave testing is inappropriate in this situation ?.
If many experienced ears hear a difference there must be a reason other than psychological.
The real test is to devise the appropriate testing methodology.
Eric.
Two logical fallacies here. Appeal to authority and appeal to popularity.
All those ears are connected to the brains of mortal human beings and as such no less susceptible to the same human weaknesses as anyone else.
"Experienced ears" and "experienced listeners" are some of the most meaningless terms out there.
se
No, the real test here is to see whether those experienced ears will allow themselves to be subjected to testing that may show they are imagining the difference they perceive.
The real test here is to see whether those who claim a difference exists in the face of data that shows it is the next best thing to impossible for an audible difference to exist can accept that the answer may well be psychological.
Unfortunately, experienced ears appear to be attached to closed minds, so the possibility of an error of perception is anathema to them - an insult to their experience and expertise, a slight on their golden knowledge.
Hi Eric,
No offense, but after reading the previous posts you recently made, two things come to mind - quite clearly.
1.) You do not have any idea how an experiment generated set of data contains "noise". This is normal BTW. By not using averaging, any random noise shows up very plainly. Not a surprise at all.
SY summed is all up nicely in his last post ....
And 2.) You are confused between what is observed, measured data, and reports generated from completely uncontrolled circumstances where the psychology has more effect on the outcome than what actually may have occurred.
The entire point of the word "anecdotal" boils down to uncontrolled observations that can not be repeated in any precise form. In other words, you can pretty much claim anything without the worry that you may be contradicted.
This can be a direct result from people repeating a claim often enough that it becomes fact for some of us. This form of misinformation is especially effective when a well known personality is used to deliver the message (JC in this case). Surely he wouldn't misrepresent the truth, or be so fundamentally mistaken - could / could he?
In short, the entire existence of some high end magazines have relied on anecdotal "reports" to survive. In other words, the "National Enquirer" of the audio industry. Notice how far they moved from measured performance once we understood what to start measuring.
You know, they never count the reports of people that didn't hear a difference. In fact, those people probably didn't write a report to begin with. The thing is, gathering "anecdotal testimony" together tends to pre-filter results that do not support the claim.
Listening tests on these have begun to the best of my knowledge, the planned second part of the investigation. From what I have seen, SY has done a marvelous job of measuring these things without being allowed to destroy them. Cal went out and paid for a pair of these to satisfy the questions that John would not answer with actual test data. He has the means, the equipment and I would think the motivation to make things clear. I see that you're not satisfied with the answers thus far. Therefore, may I suggest that you purchase a pair (you should be able to get an industry discount - as is the established custom in the audio world. At least dealer cost I would think). Then you can detail your test setup and procedure so that it will stand up to peer review. John may be able to help you get over the rough spots as he has had many years experience with this product. He should be able to design a proper test that will reveal these operating in their full glory.
I sure hope these things come with a serial number on each, and a certificate of authenticity. That's the least I'd expect. If not, the Franklin Mint should be able to show the distributors how to set this up at low cost. Serial numbers are the only way to ensure you have a matched pair. One has to be manufactured right next to it's mate.
-Chris
No offense, but after reading the previous posts you recently made, two things come to mind - quite clearly.
1.) You do not have any idea how an experiment generated set of data contains "noise". This is normal BTW. By not using averaging, any random noise shows up very plainly. Not a surprise at all.
SY summed is all up nicely in his last post ....
Important stuff in there. Ponder this for a while.
And 2.) You are confused between what is observed, measured data, and reports generated from completely uncontrolled circumstances where the psychology has more effect on the outcome than what actually may have occurred.
from : merriam-webster - dictionary
The entire point of the word "anecdotal" boils down to uncontrolled observations that can not be repeated in any precise form. In other words, you can pretty much claim anything without the worry that you may be contradicted.
This can be a direct result from people repeating a claim often enough that it becomes fact for some of us. This form of misinformation is especially effective when a well known personality is used to deliver the message (JC in this case). Surely he wouldn't misrepresent the truth, or be so fundamentally mistaken - could / could he?
In short, the entire existence of some high end magazines have relied on anecdotal "reports" to survive. In other words, the "National Enquirer" of the audio industry. Notice how far they moved from measured performance once we understood what to start measuring.
Well, that's simple - depending on how far you push things. Any changes I make in equipment that I can hear, I can also measure. Just different capacitors for instance. The trick is that the differences in measured performance can be mighty small ones. however, still measureable above any uncertainty or the noise floor of the test setup. Now that I've said that, take a look at the levels SY was measuring at.
No you don't. Look that ratio up and consider it again. Any differences "down there" can be caused by slight cable displacement, a small breeze, a static discharge somewhere in the area. The list goes on. In other words, the Bybee device was identical to a plain resistor in a test setup well below any threshold of hearing. That's below the noise floor of all audio gear I know of - and most test equipment too! Which brings us to the rest of your comment ...
So, to what standard would you be satisfied going to???
In other words, this is not the answer you wanted?
"I didn't get the answer I wanted, so let's throw out all that work and keep looking until I find the answer I like."
You know, they never count the reports of people that didn't hear a difference. In fact, those people probably didn't write a report to begin with. The thing is, gathering "anecdotal testimony" together tends to pre-filter results that do not support the claim.
Flawed reasoning. Besides, there are other factors at play beyond simple psychology. The entire basis of some reputations rest solely on the ability to hear what others can not. Now, there's a pretty strong motivator if you ask me. Self preservation.
Listening tests on these have begun to the best of my knowledge, the planned second part of the investigation. From what I have seen, SY has done a marvelous job of measuring these things without being allowed to destroy them. Cal went out and paid for a pair of these to satisfy the questions that John would not answer with actual test data. He has the means, the equipment and I would think the motivation to make things clear. I see that you're not satisfied with the answers thus far. Therefore, may I suggest that you purchase a pair (you should be able to get an industry discount - as is the established custom in the audio world. At least dealer cost I would think). Then you can detail your test setup and procedure so that it will stand up to peer review. John may be able to help you get over the rough spots as he has had many years experience with this product. He should be able to design a proper test that will reveal these operating in their full glory.
I sure hope these things come with a serial number on each, and a certificate of authenticity. That's the least I'd expect. If not, the Franklin Mint should be able to show the distributors how to set this up at low cost. Serial numbers are the only way to ensure you have a matched pair. One has to be manufactured right next to it's mate.
-Chris
Nope, zip lock bag, no branding, identification, or documentation. Cuts into the profits.
Hey, how do I get this industry discount stuff?
Industry accommodation sale. All reputable distributors know and respect this unwritten code. It's an "old boys thing". Actually, Dave should be able to take advantage, as should any respectable dealer in North America. I'm not sure how it translates to other areas of the world.
No way! Really???
How poor is that? Normally they want to build up the customer's confidence. They include at least a letter saying "Congratulations on your blah blah blah". I would expect the pair in Styrofoam fitted presentation style, cardboard constructed cases. The models up might even get that velvet stuff. After all, these amazing devices may possibly be damaged in shipping.
Do they need cyro treatment after shipping to restore proper functioning? That could possibly explain your null results. Were the leads demagnetized? We established that they are coated steel, so any residual magnetism may interfere with proper functioning as well. Do they need to be cleansed or charged using the proper crystals?
So many outs for them - so little time.
No way! Really???
How poor is that? Normally they want to build up the customer's confidence. They include at least a letter saying "Congratulations on your blah blah blah". I would expect the pair in Styrofoam fitted presentation style, cardboard constructed cases. The models up might even get that velvet stuff. After all, these amazing devices may possibly be damaged in shipping.
Do they need cyro treatment after shipping to restore proper functioning? That could possibly explain your null results. Were the leads demagnetized? We established that they are coated steel, so any residual magnetism may interfere with proper functioning as well. Do they need to be cleansed or charged using the proper crystals?
So many outs for them - so little time.
I can't quite believe there is a voice asking for exact repeatability between channels or tests at -145dBish below line level or better.
That's way down below the level of thermal noise in 50R resistors. Pick a supplier, and calculate the delta-T for even minor deviances from a nominal room temp and.. the whole thing dissappears into the noise floor (or the realms of audioasylum idiocy). Esp when CD delivers only c. -93dB S:N, and experience with really good (1/4" up to 2") analogue tape suggests that -70dB when actually delivered through the chain....sounds great
[size=-2]no, < -150dB matters. Obvious troll is obvious [/size]
That's way down below the level of thermal noise in 50R resistors. Pick a supplier, and calculate the delta-T for even minor deviances from a nominal room temp and.. the whole thing dissappears into the noise floor (or the realms of audioasylum idiocy). Esp when CD delivers only c. -93dB S:N, and experience with really good (1/4" up to 2") analogue tape suggests that -70dB when actually delivered through the chain....sounds great
[size=-2]no, < -150dB matters. Obvious troll is obvious
[/size]
Hi Martin,
Studer 820 using Dolby SR does better than -100 dB SNR. 320 nW for 0 as is the custom (a bit hot for me, by 3 dB). Admittedly, that's at 30 ips. At least Studer EQ's out the head bumps. Everyone else should go no higher than 15 ips least the head bumps move from the bass into the mids more.
Hi spind,
I'm glad you found something in that novel you liked!
-Chris
Studer 820 using Dolby SR does better than -100 dB SNR. 320 nW for 0 as is the custom (a bit hot for me, by 3 dB). Admittedly, that's at 30 ips. At least Studer EQ's out the head bumps. Everyone else should go no higher than 15 ips least the head bumps move from the bass into the mids more.
Hi spind,
I'm glad you found something in that novel you liked!
-Chris
Lets see if I got this correct.
If we take a few Bybees and throw them in a hat and then add a handful of Resistors maybe of different brands or at least different batches.
Selecting two at random, we rerun the test.
From the test results, we could not tell if we had:
BB or BR or RB or RR
Is this right?
If we take a few Bybees and throw them in a hat and then add a handful of Resistors maybe of different brands or at least different batches.
Selecting two at random, we rerun the test.
From the test results, we could not tell if we had:
BB or BR or RB or RR
Is this right?
Friends shouldn't let drunk friends post on forums....
Ok, ok, I came home from a long hot day in the sun and beach and waaaay too much beer and opened my big mouth.....my apologies for infuriating some of you, especially Stuart.
In review yes Stuarts testing is correct as are the comments following my drunken posts.
So back to the subject.....
Sine wave testing does not show any noticeable differences emerging out of the standard audio band noise spectrum.
How about testing with a music signal and see if differences are able to be discriminated ?.
Stuarts 192k/24 bit sound card is capable of resolving down to -145dB which I agree is noiseless enough to show up differences that are purported to be audible.
I figure one approach would be to play/record one channel of the same passage twice - ie one with BB in series, and the other without and then compare the two recordings.
Another approach might be to feed the same music signal into the two inputs of a differential line level amplifier stage (one leg with BB, one leg with RR) thereby resolving any differences in transfer functions that are above the noise of the rec/pb system.
I don't believe that all reviewers of BB are delusional and that there is something being changed by BB's that ought to be discernible, despite the babble on the BB web site.
I reckon different testing is required to establish more real world results than standard sine wave testing.
Eric.....sober now.
Ok, ok, I came home from a long hot day in the sun and beach and waaaay too much beer and opened my big mouth.....my apologies for infuriating some of you, especially Stuart.
In review yes Stuarts testing is correct as are the comments following my drunken posts.
So back to the subject.....
Sine wave testing does not show any noticeable differences emerging out of the standard audio band noise spectrum.
How about testing with a music signal and see if differences are able to be discriminated ?.
Stuarts 192k/24 bit sound card is capable of resolving down to -145dB which I agree is noiseless enough to show up differences that are purported to be audible.
I figure one approach would be to play/record one channel of the same passage twice - ie one with BB in series, and the other without and then compare the two recordings.
Another approach might be to feed the same music signal into the two inputs of a differential line level amplifier stage (one leg with BB, one leg with RR) thereby resolving any differences in transfer functions that are above the noise of the rec/pb system.
I don't believe that all reviewers of BB are delusional and that there is something being changed by BB's that ought to be discernible, despite the babble on the BB web site.
I reckon different testing is required to establish more real world results than standard sine wave testing.
Eric.....sober now.
I don't think they are delusional either - I think they are in denial and overestimate their ability to resolve differences and underestimate there own brain's ability to be suggested to. In this respect they are like every male driver on the planet.
Here's a real world result for ya. Minus 145db.
End of story.
Here's a real world result for ya. Minus 145db.
End of story.
Hi Eric,
That explains a lot then. I can hardly wait until Cal begins to post once he's three sheets to the wind!
I think I do know what you are thinking of though. Some components are sensitive to either the voltage across their terminals, or the current flowing through them. At times, a step function shows these dependencies more easily since there is no gradual change in the quantity of interest. Of course, that implies a rise time in the real world (nothing changes instantly when dealing with electrons). This then would imply a frequency limit, those pesky sine waves again! Still, a valid test. One that can be accomplished by using a square wave, or a pulse generator if you want to use a very small duty cycle, or large mark - space ratio.
Since Stuart tested these devices across the entire audio band, and there didn't seem to be any tendencies that would lead us to believe anything at all was happening beyond a simple resistance, I strongly suspect that any reasonable test for aberrations in the accepted audible bandwidth will fail to show us anything remotely interesting. Man, that was a run-on sentence if I ever saw one! Getting back to the subject now ...
I think there were a couple other people willing to run tests on these devices, we'll see. The beauty is that what tests are performed in a lab can easily be repeated and verified to a high degree of confidence. Of course, listening tests are also in the future. The pressure's on Cal I guess, for now anyway (until he posts his observations).
-Chris
That explains a lot then. I can hardly wait until Cal begins to post once he's three sheets to the wind!
That's a very familiar statement to see. However, have you ever heard of a dude named Fourier? He was into math in a big way, and he came up with the silly notion that all wave forms could be broken down into a series of sine waves (pesky things, those) at various relative amplitudes. Darned if no one has been able to prove him wrong. Remember that we're working in the world of math where there isn't any noise floor (which can also be modeled). So it would seem that testing with sine waves should be perfectly valid, as long as one tests up the the limit of the highest frequency of concern in the audio bandwidth. How about 100 KHz? That ought to cover things pretty well.
I think I do know what you are thinking of though. Some components are sensitive to either the voltage across their terminals, or the current flowing through them. At times, a step function shows these dependencies more easily since there is no gradual change in the quantity of interest. Of course, that implies a rise time in the real world (nothing changes instantly when dealing with electrons). This then would imply a frequency limit, those pesky sine waves again! Still, a valid test. One that can be accomplished by using a square wave, or a pulse generator if you want to use a very small duty cycle, or large mark - space ratio.
Since Stuart tested these devices across the entire audio band, and there didn't seem to be any tendencies that would lead us to believe anything at all was happening beyond a simple resistance, I strongly suspect that any reasonable test for aberrations in the accepted audible bandwidth will fail to show us anything remotely interesting. Man, that was a run-on sentence if I ever saw one! Getting back to the subject now ...
I think there were a couple other people willing to run tests on these devices, we'll see. The beauty is that what tests are performed in a lab can easily be repeated and verified to a high degree of confidence. Of course, listening tests are also in the future. The pressure's on Cal I guess, for now anyway (until he posts his observations).
-Chris
A minor complication with Fourier is that sine waves only model a periodic waveform. Music is not periodic, but it is sufficiently close that it is difficult to imagine a problem arising from this. It does mean that the LF limit has to be explored as well as the HF limit.
The objection that people often raise is fast transients, but Fourier can cope with them. It can even cope with discontinuities, and there are certainly none of them in an audio signal.
The objection that people often raise is fast transients, but Fourier can cope with them. It can even cope with discontinuities, and there are certainly none of them in an audio signal.
- Status
- Not open for further replies.