 
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
|
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
216.53.159.98
I've been trying to be more open-minded about the possibilites of Objectivists being correct about DBTs. In an effort to better understand Objectivists point-of-view, I went back and reread some of their posts about DBT's here on AA.I noticed often times Objectivists cite the requirements of SPL levels being closely matched and frequency response variations being ameliorated through equalizers. When this is done properly they claim that well-designed audio components should be virtually the same -- soundwise. I believe the validity of that statement "if" it's true, ONLY matters depending apon what one listens for when judging or trying to identify an audio component. Perhaps the reason for this difference in what Objectivists and Subjectivists claim is due to just how they listenand what they listen for?
Let me explain further. When I listen subjectively, frequency extremes and transparency is NOT what I am listening for. Instead I focus on dimensionality, soundstage width & depth or what some might call "air around the instruments". These things, at least from my experience, do not vary greatly with sound level changes, nor have I ever been ever to equalize these traits into a system via an equalizer if it doesn't already possess them.
I am NOT trying to "stir the mud," I am trying to discover if it's possible both sides are actually right at the same time! The example above would be IMHO one way that might be possible. I'm starting to believe that the DBT, ABX arguement, debate, discussion is a result, of the same arguement, debate, discussion presented by the 3 blindmen who after all touching an elephant pronounced:
1) An elephant is long and thin like a snake, after touching it's nose.
2) An elephant is thick and stout like a tree, after touching it's leg.
3) An elephant is large and wide like the side of a barn , after touching it's side.
All were correct, yet all were incorrect also! Perhaps these arguements, debates and discussions by Objectivists & Subjectivists are similar to those presented by the three blindmen?
Thetubeguy1954
Follow Ups:
When I listen subjectively, frequency extremes and transparency is NOT what I am listening for. Instead I focus on dimensionality, soundstage width & depth or what some might call "air around the instruments". These things, at least from my experience, do not vary greatly with sound level changes, nor have I ever been ever to equalize these traits into a system via an equalizer if it doesn't already possess them.You may already be aware of this, but just to make sure, I'll mention it anyway. In analog devices like amps, preamps and speakers, the frequency response defines more than just how much the device emphasizes or suppresses notes of different pitch. If you change the magnitude of the response vs frequency (say, in dB), the phase vs frequency must change also. Further, for devices having negligible distortion, there's a mathematical relationship between the frequency response (of which both magnitude and phase must be known) and the transient response of the device. So if you change the frequency response, the transient response is also affected. This means that the way the device responds to the attack and decay of a transient must change. It's not known exactly how this correlates to perception, but it's not at all unreasonable to expect that it could affect perception in some of the ways you mention.
Equalizers are mainly designed to "fix up" amplitude response, and they generally ignore phase. So there's a possibility that they could make the magnitude of the response better, but degrade the phase response. This in turn could possibly degrade the transient response.
Now there are some kinds of systems, such as the DRC package (digital room correction) in which both the amplitude and phase of the frequency response are measured and corrected (via a microphone at the listener's position). An example of the before measurements (in red) and after (in blue) are shown here . Notice how the "before" data doesn't really look like a step at all at the very start of the waveform. It goes up for an instant like it should, then goes negative, then back positive again with some ringing. The corrected trace below it in blue is still far from perfect, but you can see at the start that it behaves much more closely to an ideal step than the uncorrected data. It turns out that the DRC is actually correcting for the non-ideal phase characteristics of the loudspeaker crossover network.
So to sum up, frequency response necessarily includes both magnitude and phase, which are inseparable and together determine transient response. This relationship means that frequency response variations could cause other issues besides emphasizing or suppressing notes of a given pitch.
HiAS the real JJ pointed out ones ears do not always hear "one thing" for what it is.
An old hifi salesman’s trick to promote the brand De-jure was to raise the SPL of the speaker being pushed so that it is about 1 or 2 dB louder than the others.
The average potential customer will interpret the small level difference as that speaker being “clearer” not louder. Curiously, the same trick works on video, have a small treble and bass boost on the brand your pushing and people notice a better picture.An issue here is that since loudspeakers have so many flaws that they usually all sound quite different the issue is how to compare speakers.. Added to that is the fact that at normal listening distances, one is also getting great deal of room effects which are also often different speaker to speaker.
Compared to even cheap electronics however, loudspeakers have many flaws which because they do not help sell, are not generally mentioned in the magazines.
Of all the elements in the chain, loudspeakers, at least theoretically / by measurement, appear to be by far the weakest link in the re-production chain.Fortunate, they all have similar problems so we are pretty used to the way they sound vs real life.
Loudspeakers compress dynamically, ALL voice coil drivers begin to show the effects of “power compression” and the shift in parameters that accompanies it beginning at 1 / 10 to 1 / 8 rated power.
Loudspeakers spread out a complex signal in time, some frequencies produced before others, most multiple driver loudspeakers cannot preserve the waveshape of the input signal, there is too much time dispersion..
Loudspeakers produce both even and odd harmonically related and non-harmonically related “free sound” which is not part of the input signal and the degree is also level dependant. Make no mistake, the speaker can’t have any idea what the musicians wanted, its job to to produce the drive signal only, any additional sound is coloration.
Loudspeakers have directivity, if this is not “constant” then the spectrum of the on axis sound is different than the off axis spectrum.
Loudspeakers with more than one driver normally also have self interference in the crossover region which is visible in polar or spherical measurements of the speaker (again, as this does not “help” sell, hifi magazines don’t discuss any of this).
The self interference produces lobes at particular frequencies, pointed in undesired /unexpected directions. Which raises the level of reverberant sound at that frequency.On the other hand, as one increases the level, the size of the audience, acoustic power and room size all of the problems become critical show stoppers, hence are well known outside of hifi.
ON DBT’s.
Studies have shown peoples acoustic memory is very short, in this kind of testing it was necessary to be able to switch back and forth as quickly as possible.
On the other hand, if one were testing speaker wire, it would be logical to use a pair of heavy duty Relays (one at the amp end, one at the speaker end). These (high current types) have self wiping contacts and have very low R etc and would allow one to rapidly (nearly instantly) switch between one set of wires and another.
This allows one to sit back as long as they want to see if anything pops out later on.
Even when you control which is which through the relays, if there is a difference you will hear it switching from one to the other. Once you are sure (if so) have a second person switch back and forth out of your control and knowledge of which is which and then see if that effects your ability to hear some difference.
If you can reliably hear a difference form one to the other, then you have an audible difference.
You don’t need to be blind, just being open minded to the results (what ever that is) and being able to switch back and forth rapidly gets you 90% of the pie.
If you can hear a difference this way in cables or what ever, it is worth doing, if not, think hard about drawing that wallet out.The last BT I was in was for commercial power amplifiers, for reference I brought in a Threshold Stasis as well. Using very revealing speakers, it was interesting that the best sounding Pro amp (QSC PL-236) sounded slightly detectable as different than the Crowns and Chevin and about the same as the Threshold.
The differences were in the (what it sounded like to me) low level, decay side of transients if that makes sense.
Curiously too, with some program material, the Threshold began to sound “different” at a modest level. An oscilloscope revealed even at a modest level, the amp was instantaneously clipping due to the very dynamic signal.
One couldn’t hear this (I never noticed anyway) unless you could compare to the “without” condition, until then, the Threshold was “plenty of power” for me.
From that day on, I have used what are “Pro” amplifiers in my stereo and use a PL-236 (max of 1300W /ch) on a pair of speakers with 100dB sensitivity now and never clip (headroom is dynamic ranges friend).
Tubes (depending on the topology) have an edge here as the “max” outout is a more gradual limit as opposed to a Solid State brick wall.
One can “see” why too, if the music had a peak to average level of only 30dB and the amp a max power of 200W, then the average level without clipping peaks is only . 2 Watt. Fortunately for the industry, most “modern” pop music, in an effort to be “loud”, has a P/A ratio of about 10dB (or less from FM).Anyway, what you want for one channel is 2pc of a DPDT relay or 4pc of SPDT relay.
A DC coil is better here, you need a supply for that too.
For instance, ALL ELECTRONICS (800-826-5432) has cat# RLY-351 a SPDT relay for $2.40 each.What ever you compare or listen for, it will be easy to hear “if” there is a difference or not when you can switch quickly. Like you suggest, it may take time to be able to hear some feature so being able to switch at any time would be useful here.
A relay switcher like this would also allow you to switch between amplifiers too by only using one end of it.
I think you will find both extreme positions on this issue (as is normally the case) is not the place “where reality dwells”. There are real mysteries and room for improvement in audio, there is an awful lot of BS too.
Have fun, do try this, best listening.Tom Danley
Danley Sound labsWhat (tube I presume) amp do you have?
Hello Tom,If you are asking me: "What (tube I presume) amp do you have?" I own a Mastersound Reference 845 from Italy. It's as beautiful to look at as it is to listen to. I love the way Italian's voice their audio components. IMHO they are making some of the most realistic audio components there are, at least the ones I've heard sound that way.
The Mastersound Reference 845 can be seen at the link provided. It's a monster at 15.7"W x 25.2"D x 10.2"H & 135LBS. It produces 40W/ch of the most convincing illusion of live music I've ever heard.
The other company listed (Ayon) is to the best of my knowledge a collaboration between Viac & Mastersound. I've heard some say Ayon is better and I've heard others say they are too close to pick a winner. As I've never heard Ayon, I don't know either way.
Thetubeguy1954
I considered this idea a long time ago but never got around to building one. I considered using make-before-break relays to avoid momentarily unloading and then reloading the amplifier. It occurred to me that one drawback of this test is that it compares two potentially flawed pieces of hardware. A better test IMO would be testin each set of wires individually against a shunt. The shunt is the idealized substitute for one of the A/B test elements. In the real world, the shunt can be just a bypass relay or a switch. You already have something very close (close enough) to a shunt for testing interconnect cables in your preamplifier if you have a tape monitor switch and a tape monitor output which is not buffered by an isolation amplifier stage. By connecting the interconnects between the tape output and monitor input and switching the monitor switch between source and monitor, you can alternately compare the interconnect to the shunt. I have yet to test an inexpensive interconnect which did not pass this test flawlessly, it being impossible to tell under any circumstance whether or not the interconnect is in or out of the circuit. This is entirely consistant with both theoretical and measured performance for these wires. IMO, if for any reason, an interconnect were to produce an audible loss of signal such that it was below the level of the shunt, it is automatically so flawed as to be unacceptable. Jon Risch has a problem with this test. Probably his real objection is that it makes sense, works, is easy to perform, and results in the correct answer, the answer he doesn't want to hear.
Isn’t it amazing that the claims made for improvements in the sound are supposed to be so obvious and so easily heard, until there is a DBT done.
Actually, I didn't need a DBT. Just being honest with myself was enough. Here was another surprising result. Last year I bought a Toshiba DVD player. I decided to try it as a cd player. My main system uses a 1991 vintage JVC 1 bit 8x oversampling player (which sounded only very slightly different from the 20 bit 1989 Denon player it replaced.) With several duplicate discs which were easy to synchronize, I could not hear any difference between the JVC and Toshiba players in direct A/B comparison under any circumstances. Even more surprising and remarkable to me was that using the JVC's fixed output, the two units had identical signal levels which didn't need any correcting for level matching in the comparison. This convinced me of what I suspected all along, both players operate flawlessly performing their function interchangably and making no audible difference. About the only objection I have to the Toshiba player is that it does not perform some of the functions I would like it to or if it does, not nearly as conveniently or intuitively. As for so called audiophile type equipment which costs far more and sounds different, as with wires which might sound different from a shunt, to the extent of their difference, I consider them defective and cost being no object, I would not choose them. That their characteristic sound may mitigate other problems with some peoples' other components does not justify buying them in preference IMO. How nice that at least for me, there are some areas in life where one of the cheapest options is also the best option.
jneutron may come along and speak for himself. I understand less than 99.9% of what he has to say. However, as a scientist it seems to me that he delves far deeper into wave theory and aspects of wires than most deal with at the engineering level.He has some interesting things to say about localization and such. Hopefully, he'll comment.
Having said that, I want to respond more directly to your post. I have often thought it would be interesting to devise a DBT that really did test the ability to differentiate over long periods of trial listening in the home environment.
Say if you could take two cables such as Cardas and Wireworld that many subjectivists would believe are much different sounding cables, disguise them so they could not be differentiated physically, and then place them in various homes for periods up to two or three weeks. In some instances the two sets of cables would be the same, in other instances one would be the Cardas and the other the Wireworld. None of the participants would know which two sets they had. At the end of the test period they would be asked to choose which one they preferred. One choice they could make it to say they could not distinguish one from the other.
I once proposed a long-term listening DBT such as that over at the old AR cable forum and the naysayers scoffed, ridiculed and laughed at the suggestion, saying that such long term testing was wholly unnecessary as the short term DBTs had already pretty much established that most cables of similar length and gauge cannot be distinguished under blind conditions.
____________________"To dance beneath the diamond sky with one hand waving free."
Those who would say, " that such long term testing was wholly unnecessary as the short term DBTs had already pretty much established that most cables of similar length and gauge cannot be distinguished under blind conditions," are being inconsistent with the scientific method. Your question about whether long versus short term listening is a testable hypothesis, but meeting the double blind requirements may be difficult. But I think a single blind test would be useful, if disguising the ics does not compromise their performance.
First, it's tentatively.And the answer in science is that hypotheses are not accepted UNTIL SOMEBODY SHOWS SOME EVIDENCE.
Falsifiable evidence. Not evidence from a sighted test where one can't even establish potential falsification, too.
Anyone knowing anything about logic knows you cannot prove a hypothesis, you can only disprove one. This is why we have null hypotheses which we hope to reject. Again I don't know why anyone should pay attention to someone who knows so little. I suggest you seek a refund on your graduate education.
First, since I've said, on this board, in the recent past, that there is no absolute proof of a negative (which is the same thing you're saying), you're lecturing to your better on that one.Now, simply put, you claimed that "all hypotheses are tentatively accepted". Bull.
Until there's an experiment, an hypotheses is just that. Not accepted, not rejected, just a hypotheses.
One does not have to ACCEPT a hypotheses to test it.
Your original statement was abject, utter crap, and your illict straw man an attempt to defame.
GET OUT, QUACK!
"you're lecturing to your better on that one." What a laugh; I have been teaching research methods for 40 years. I doubt if you could even pass such a course given your existing misinformation and gross ego.You test hypotheses with data on your operational hypotheses. But you cannot logically accept a hypothesis, you only can hope to reject the null hypothesis. The findings are only tentatively accepted with replication being an all important additional support.
You notice that I seldom have the needs to inflate my importance and to claim to be an expert or authority. Such claims are the bastion of the charlatan, which sir you are.
I do intend to not monitor this page, but please know that any misrepresentation of what science knows or can claim will get my response that you are a charlatan.
It is plain as day that you're nothing more than a willfully malicious stalker. Your position on hypotheses in THIS article (as opposed to your first, which was another position entirely) agrees with what you can find me saying on this very page.
Your original claim was not that all acceptance was tentative, which is very true, but rather, given the phrasing, that one must accept all hypotheses tentatively AT THE START. Your phrasing suggested that one must accept something in order to test it.If you meant something else, then your original expression was not clear, to say the least.
One does not have to accept anything in order to test it.
Having an open mind does not require one's brain to fall out.
a
And then you accuse me of not understanding.Sure. Uh huh. Right.
Let's see:
Let's look at two statements:
All hypotheses are tenatively accepted. No amount of short DBTesting of ic proves anything definitively.That's yours.
Now, let's consider: All hypothesis are only ever tentatively accepted. No ....
Your statement suggests positive acceptance. The second does not, and makes the situation clear. Your choice of the word "all" at the beginning makes your statement universal, and suggests that the statement "the moon is made of limburger cheese" has to be tenatively accepted until such time as it's disproven. That, of course, is absurd.
I am not responsible for your poor writing. I don't claim to be the best writer in the world, either, but it would be a lot smarter for you to admit that you wrote something poorly than it would be to claim that somebody else doesn't understand.
What is also telling is your utter lack of diligence, since I've said, clearly, what eventually amounted to your intended position, on this page and the one that preceeds it, and done so repeatedly.
None the less, you insisted that you read carefully, and claimed that I was a "charlatan", even though you FINALLY got around to saying what I'd already said.
Your lack of diligence, your apparent negligence, etc, look very bad. Do you also claim to teach scientific ethics? Enquiring minds want to know.
a
Listen, creep, you wrote something poorly and then picked a fight over it.
You have, because of your own poorly phrased statement, claimed that I am incompetent in the most very basics of my present and prior jobs. You WILL retract that allegation.I think, frankly, that it was all malicious bait. I can not, however, prove that, I simply point out the repeated misconduct on your part as possible corroboration.
This is all that I said, whereupon you chose to pontificate on that being wrong. I am sorry that you are threatened when someone calls you, but you chose to make the misstatement not me.You must also retract your charge that I baited you and that you are more competent than I. Of course I expect no such retraction, nor should you. If you are so competent, why are you so truculent and threatened?
You and probably I have irritated many here. I mainly post to tell others of my experiences with cables, fuses, etc. Once again, I will try to limit myself from posting here as this page is like WWI with neither side able to move against the other, or as someone said, a train wreck.
Your statement "now accepted" is again either a result of your lack of dilligence or intentional deception.I have stated that over, and over, and over. It is elsewhere on this page, and if not there, on the part of the next page that just got folded over.
Your insistance that you have taught me ANYTHING except that you are deceptive, either negligent or dishonest, and malicious is quite wrong.
I accept that your phrasing was weak and I read it in a different form than you intended. How's that?
a
Well?
It will be a local observation, and the findings can not be transfered to me or others as valid data ( only audiophile hints ). Two different frame of referance. No global value.
You may be correct, but then what value would any DBT ever be? I don't see how a DBT comparing two different cables or components could ever be globally applicable. The significance of the results would always be limited to the particular components or cables under test under the particular circumstances with the particular participants.What is the difference, as far as applicablility, between the DBT I propose and all the others that have ever been reported? Mine simply goes to the "field" rather than remains in the "lab".
____________________"To dance beneath the diamond sky with one hand waving free."
I'm equally sure that we told you that long term DBTs have been done.
____________________________________________________________
"Nature loves to hide."
---Heraclitus of Ephesus (trans. Wheelwright)
> I'm equally sure that we told you that long term DBTs have been done. <Do you perchance have any citations on long term DBT's?
I don't have copies of the articles, but I know Daniel Shanefield did some. I think one of the articles by Shanefield shown in the references on Eyespy's site or on the ABX site describes one. Mtrycrafts at Audioholics.com could probably give you the precise reference. Sorry, I didn't make a copy of it when I visited him.Arny Krueger thinks they have been done and he would know, and I have linked below to a post of his at Stereophile.
E. Brad Meyer pointed this out at a meeting of the Boston Audio Society, as well.
http://www.bostonaudiosociety.org/bas_speaker/abx_testing.htm
____________________________________________________________
"Nature loves to hide."
---Heraclitus of Ephesus (trans. Wheelwright)
- http://forum.stereophile.com/forum/showthreaded.php?Cat=0&Number=298&page=&vc=1 (Open in New Window)
Thanks.What speakers did you get to replace the Stratus Mini's in your main rig? I still have my pair that are waiting for me to get off my butt and set up that HT system. :) Someday....
Paradigm Signature S2 speakers.
____________________________________________________________
"Nature loves to hide."
---Heraclitus of Ephesus (trans. Wheelwright)
I'm pretty impressed with those. A friend of mine got a pair for his adult son. We ended up with a very slight toe-out and they worked beautifully.Enjoy!
That's why I bought them!Because of the set up (WAF), I have to toe them in quite a bit. With a subwoofer, one can obtain full range sound.
____________________________________________________________
"Nature loves to hide."
---Heraclitus of Ephesus (trans. Wheelwright)
"I'm equally sure that we told you that long term DBTs have been done."You probably did. I'm over 60. Does that entitle me to selective memory yet?
____________________"To dance beneath the diamond sky with one hand waving free."
Over 60, eh? Don't ask for too much sympathy from me!
____________________________________________________________
"Nature loves to hide."
---Heraclitus of Ephesus (trans. Wheelwright)
Small differences in level do not sound like differences in level, they sound like differences in "air" or "fullness" or "soundstage" or any number of other kinds of perceptual description.Small differences in frequency response also do not necessarily sound like small differences in frequency response, rather they sound more like small differences in timbre, openness, soundstage, etc...
My point is not that there are no audible differences in things, there certainly are quite substantial differences between at least some kinds of equipment, but rather that things that are detectable in a DBT are also (and usually easily) detected as "being measurably different".
Relating the measurements to the perception is NOT a finished art, in fact for small differences it's barely a "started" art.
There are some differences that can be argued as "way too small". This, however, does not manage to discount quite a bit of the measured differences actually observed in equipment.
If you have an SNR of something like 110dB in an average system, then there will be no audible distortions from that. If it's flat to .1dB, you're likely to hear no audible differences from THAT. The combination of the two would suffice to say "differences are extremely unlikely to be audible".
Very few things give that kind of performance, though.
Consider. If I take narrowband noise (inside of one critical bandwidth) and add another (independent) narrowband noise to it, the lower power(probe) noise is inaudible when it's about 6dB below the louder noise.
If, however, I put in a high-energy sine wave at 19kHz, and a 1kHz sine wave 90dB down from that, you'll hear the 1kHz sine wave plain as day in any quiet listening room with good equipment.
So, SNR (that includes THD as a subset of SNR) is "not very useful". I liken it to the "mostly harmless" description of the Earth in HHGTTG.
Knowing the SNR does not tell you much until you know both the spectrum (short-term) of the signal *and* of the noise. And you can not reasonably reduce this to anything like "1 number".
Very informative post. It proves that most of the most prominent posters on this board have something to say when they're not catfighting. Thanks for this info... although it blows a few theories of mine! But that's not a bad thing.
This post is made possible by the generous support of people like you and our sponsors:
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2025, Audio Asylum, all rights reserved ]