Without commenting specifically on the multiple claims of improvement with no factual basis in the evidence, only more conjecture, one problem in the ARTA presentation is the shift in the reference level between any two plots for CSD and varying scale for impulse. The auto-level and auto-scale are poor methods for comparison presentation. Software such as the older LAUD and currently Praxis that provide full control of the presentation would be far better for this purpose.
That said, there are no conclusions that can be reached in comparing various drivers without a before/after set of measurements. I could show you sets of "comparison" measurements I have of two drivers that would seem to indicate some sort of benefit from a treatment if I showed only one without and another with. When one looks at the before/after set of the individual treated driver, the only valid way to analyze any treatment, many of those cases would show a detrimental effect, despite the apparent improvement of the case with no before/after. It's quite possible to have an untreated driver that is better than another one that has been treated. Also, if one treats the "better" driver of a pair (not known beforehand, of course) and measures afterwards, then one would conclude, erroneously, that it was the treatment than resulted in the better response.
The methodology of comparing drivers with no before/after measurements of treated drivers is simply fatally flawed if one is attempting to examine the effects of some treatment. It looks interesting, but there is nothing that can be concluded from them, though there seems to be no lack of effort to do so in a positive spin despite the lack of the necessary measurements.
Consider that both treated drivers are ringing dramatically well beyond 3ms in a scale of only 25db as shown in the CSDs. That ringing is on the order of 10db! Then there's that monstrous "valley" in the treated drivers (one of several) that is so much more than 5db as it starts that you can't even see it in the CSD angle. BTW, LAUD and Praxis (and others) allow you to change the angle of the CSD to view areas occluded in any particular angle. The tools being used are either lacking in presentation controls or options are not being used judiciously.
Claim what you will, none of it is accurately based on the data presented. The required data is conspicuously lacking. This is the technical discussion, after all. Conjecture based on insufficient evidence is still the norm.
Dave
That said, there are no conclusions that can be reached in comparing various drivers without a before/after set of measurements. I could show you sets of "comparison" measurements I have of two drivers that would seem to indicate some sort of benefit from a treatment if I showed only one without and another with. When one looks at the before/after set of the individual treated driver, the only valid way to analyze any treatment, many of those cases would show a detrimental effect, despite the apparent improvement of the case with no before/after. It's quite possible to have an untreated driver that is better than another one that has been treated. Also, if one treats the "better" driver of a pair (not known beforehand, of course) and measures afterwards, then one would conclude, erroneously, that it was the treatment than resulted in the better response.
The methodology of comparing drivers with no before/after measurements of treated drivers is simply fatally flawed if one is attempting to examine the effects of some treatment. It looks interesting, but there is nothing that can be concluded from them, though there seems to be no lack of effort to do so in a positive spin despite the lack of the necessary measurements.
Consider that both treated drivers are ringing dramatically well beyond 3ms in a scale of only 25db as shown in the CSDs. That ringing is on the order of 10db! Then there's that monstrous "valley" in the treated drivers (one of several) that is so much more than 5db as it starts that you can't even see it in the CSD angle. BTW, LAUD and Praxis (and others) allow you to change the angle of the CSD to view areas occluded in any particular angle. The tools being used are either lacking in presentation controls or options are not being used judiciously.
Claim what you will, none of it is accurately based on the data presented. The required data is conspicuously lacking. This is the technical discussion, after all. Conjecture based on insufficient evidence is still the norm.
Dave
Last edited:
I'm sure there would be an explanation from a psychoacoustics point if audible improvements are perceived, however, I have not researched this yet.
Last edited:
I would guess that the real point is that none of these drivers has any useful response above 500 Hz so what ever changes at 2k and above is pretty much irrelevant. The response of the drivers has been presented in a number of ways and all show it is not very accurate.
On the other hand, accuracy is a measurement of how well the output matches the input. If you want to argue that inaccurate reproduction sounds better than accurate reproduction, so be it.
On the other hand, accuracy is a measurement of how well the output matches the input. If you want to argue that inaccurate reproduction sounds better than accurate reproduction, so be it.
not convincing at all
Seems to be lots of pseudo-rigorous data manipulation by proponents w/o attention to details such as driver to driver variance, appropriate driver selection, inappropriate comparisons purported to support claims, it goes on and on....
no effort whatsoever to correlate pretty pictures with lots of colors to what is actually heard in any quantitative fashion, so what's the point?? It's like comparing apples to aardvarks and Rembrandts to radishes... all personal preference, little technical relevance...
nothing new...
Seems to be lots of pseudo-rigorous data manipulation by proponents w/o attention to details such as driver to driver variance, appropriate driver selection, inappropriate comparisons purported to support claims, it goes on and on....
no effort whatsoever to correlate pretty pictures with lots of colors to what is actually heard in any quantitative fashion, so what's the point?? It's like comparing apples to aardvarks and Rembrandts to radishes... all personal preference, little technical relevance...
nothing new...
Last edited:
Certainly can't argue with that. I'm just thinking if there are different aspects of inaccuracy, and one is preferred over the other, then there might be an explanation for it.
It's like room modes, one or two outstanding modes might not be prefered against evenly excited multiple modes that create a good SPL curve.
Last edited:
...so what's the point?? It's like comparing apples to aardvarks and Rembrandts to radishes...
..
Why not conduct those rigor double blind tests you are calling for by yourself and post e'm here ???
..:yaaaaaaaaaaawn:
Last edited:
Why not conduct those rigor double blind tests you are calling for by yourself and post e'm here ???
..:yaaaaaaaaaaawn:
why would he do that? It would be a complete waste of time no?
Would YOU accept the 'word' of a skeptic if he reported back that he did a DBT and heard no difference?
No, so the answer is (and has always been) that the CLAIMANTS are the ones who be tested...after all there is nothing stopping me reporting 'no difference' no matter what I heard, if I were dishonest enough to do that.
WHY would I do a test on cable differences (do they make any difference BTW? haha) if I felt cables made no difference? If I did, would anyone believe me? So no, it is the guy who is adamant he can pick cables apart that does the test, same deal here, surely.
Actually, there are some out here who WOULD like to hear the results of a DBT by anyone skeptical. Besides, by your own logic, would the skeptics believe the believers if THEY reported hearing a difference in a DBT? That's why I and others have proposed a gathering of BOTH parties for a DBT.
Unfortunately, that isn't likely to happen, so this entire debate seems silly.
Keep in mind that we "skeptics" have never said that the sound of a diaphragm (as opposed to immobile surfaces) won't be changed. It may be, it may not be, that's not the question. The issue arises from the claims about what changes do occur, the "interpretation" of those changes (when) measured and the impact (perception) of those changes. I have to agree with John, those drivers just aren't useful above 500Hz given the nature of the response before and after treatment if one desires accurate response.
Have no doubt, there will never be a DBT done of sufficient rigor in any case, IMO. It's not trivial for the audibility claims. To start it requires that the study be done on the same set of drivers, first untreated, then treated. That doesn't happen with the simple sets of measurements made, it's not likely to happen for any DBT. You can't do quick A-B testing of untreated vs. treated, so the test protocol that would be required is going to be well beyond anything likely to be attempted here.
It is easy, however, to measure/treat/measure, but we don't see that happening much, either, except in John's case.
Dave
I'll repeat for the Nth time that I was part of a blind test of EnABL'd drivers at RMAF. Lynn Olson was sitting right beside me. Lowther drivers, Nelson Pass personal system. The EnABL pattern was not visible. Not a double blind test - but "somewhat blind."
I did NOT like the treated drivers, much to my disappointment. In may ways they were identical, expect at certain points in the music were the driver would "shout" or break up. Also in some complex tonal passages the untreated driver was more reveling and natural. A let down for sure. Was thinking of how to break it gently to Bud.
Except that I was Wrong. The drivers I thought were untreated were EnABL'd, and vice-versa. So much for assuming I knew which were playing -and for prejudice. Thought I knew the order, but I was wrong. So everything I did not like about the EnABL'd drivers actually turned out to be the untreated drivers.
My basic impression was that the EnABL'd drivers just did not have the "rough edges" that the stock drivers had. Not noticeable much of the time, but during numerous passages, it was obvious. The choice was easy.
I did NOT like the treated drivers, much to my disappointment. In may ways they were identical, expect at certain points in the music were the driver would "shout" or break up. Also in some complex tonal passages the untreated driver was more reveling and natural. A let down for sure. Was thinking of how to break it gently to Bud.
Except that I was Wrong. The drivers I thought were untreated were EnABL'd, and vice-versa. So much for assuming I knew which were playing -and for prejudice. Thought I knew the order, but I was wrong. So everything I did not like about the EnABL'd drivers actually turned out to be the untreated drivers.
My basic impression was that the EnABL'd drivers just did not have the "rough edges" that the stock drivers had. Not noticeable much of the time, but during numerous passages, it was obvious. The choice was easy.
So how does anything we've been saying conflict with that? Is it possible to hear a difference? I don't see why not, never said otherwise. Will someone prefer treated over untreated? Possibly.
Is the driver an accurate reproducer before treatment (does the output reasonably match the input)? No. Is it accurate after treatment? No. Might someone still prefer the sound? Yes.
Dave
Is the driver an accurate reproducer before treatment (does the output reasonably match the input)? No. Is it accurate after treatment? No. Might someone still prefer the sound? Yes.
Dave
It is, of course, IMPOSSIBLE to do a DBT before and after with the same driver (the order would give it away, and the time delay would raise questions about accuracy of listeners' impressions). It IS, however, possible to DBT with multiple driver sets and derive a statistically significant difference (or not). Or it could be done with a few sets in order to get a pretty good idea of the results without statistics. My offer still stands (and I think I'm not too far away from you, Dave, though we'd need to find a "believer").
Edit: Oh, and the offer is still to host and provide the beer and/or single malt...
Hmmm. So now I'm confused. There's the possibility of a difference and some may like it over untreated? What's the arguing about? I'd say on that basis the debate is done. Only I'm still interested in seeing what those who are measuring are finding regardless of the rigor of their tests.
Sorry Dave, I'm not getting your point. There was a lot of recent talk about blind listening tests of the process. I've done at least one. Nothing more than a data point - but blind tests have been done. Maybe I misunderstood the recent discussion. I preferred the treated drivers, as did most who attended the test. The difference was better, not just different. Just wanted to go on record - again - to say that at least one good test was done and the results were in favor of EnABL.
Thru the years there has been much debate over whether EnABL does anything at all, or anything more than a tiny added mass would do. Since I heard the differences I'm very glad to see measurements that show differences, subtle as they are. The measurements don't really show me what I heard, but they do point that way.
So why all the posts?
Ah, but that's not what's going on here. Proponents state (in various obtuse ways) that there is more going on than just "they hear a difference".
The implication is that there is scientific justification proven by (among other things) the nonsensical "blink test" ramblings... mostly exposed by skeptics as technical dishonesty disguised as proof. Not a shred of rigour anywhere...
As for semiblind testing at RMAF... anecdotal and subjectively valid for those who heard it... certainly not a controlled evaluation of any kind... maybe an expose' of personal bias against proper testing protocols, though.
Why am I required to disprove someone else's opinion stated as factual? It's already self-evident in all the ramblings and quasi science babble
Ah, but that's not what's going on here. Proponents state (in various obtuse ways) that there is more going on than just "they hear a difference".
The implication is that there is scientific justification proven by (among other things) the nonsensical "blink test" ramblings... mostly exposed by skeptics as technical dishonesty disguised as proof. Not a shred of rigour anywhere...
As for semiblind testing at RMAF... anecdotal and subjectively valid for those who heard it... certainly not a controlled evaluation of any kind... maybe an expose' of personal bias against proper testing protocols, though.
Why am I required to disprove someone else's opinion stated as factual? It's already self-evident in all the ramblings and quasi science babble
My post was confusing now that I re-read it. I should have left out references to DBTs I think. The claims to which I referred are those elaborate ones that indicate what I would call "near miraculous" changes such as total elimination of breakup and similar. Empirical evidence disproves that rather dramatically, when it is provided, that is.
It's perfectly reasonable that an induced change to a driver might be perceptible and be considered an improvement since that is based on preference. No argument with that.
Dave
It sort of *is* the question - actually the only question for most - no ?
Only for the minority of primarily technically interested and for the "passionate thinkers" its not in the middle of interest.
As a side note:
Those changes in measurements, I and John have shown simply *must* change presentation - no doubt about it - you *only* can ask about how much and if its possibly in your preference (and if its "systemic" of course).
As long as John is not disclosing his treatment procedure - which I have asked him to do several times now - I'd say his treatment should not *automatically* be considered to be EnABL treatment.
Look at the plots and you will see that with his treatment there seems to be dampening involved first hand - whereas with the drivers I had a chance to measure there seems to be "mode spreading" involved.
@Soongsc
Possibly do not compare this "mode spreading" with exciting as many room modes as possible.
Technically seen its quite a different perception thing happening IMO.
*If* my measurements are not a "single case" (but can be confirmed by others) we possibly should look in a second step into how this behaviour folds back into operating range.
This possibly could be done by simply doing some distortion measurenents first hand.
If, for example, 3rd order distortion changes significantly with before / after measurements - well it would not convince anybody nor would it be "the importent point", but at least it would possibly allow for some further understanding of the mechanisms involved and also allow you "EnABL'ers" to develop on your method.
Michael
Last edited:
exactly my point...alas.
One side will simply NOT accept the thoughts/proofs/results of the opposing side (deliberate choice of words).
"I heard a difference for the better with enabled drivers in a DBT"....then the dbt had some flaw in it.
"I was and remained a skeptic after doing a dbt with enabled rivers".....then you simply proved your existing bias and is further proof that dbt's are useless in audio.
Now, substitute enable with cables, caps, isolation cones....
I wish there were more true 'searching for the truth' rather than simply defending already held beliefs. Once we had that, then we would make sure the tests themselves were up to scratch.
Good luck with your proposed group tests.
My ONLY contact with enable was the supposed benefits with enabling the BOX (ie the cones were not touched) and only one box was done. That, I think, is supposed to somehow skew the soundstage and show that enable is doing something. (??)
No-one picked up the slightest hint that it was doing anything. That was extended listening by many people (ten or so at least) over the course of a weekend.
Why you chose to reply to this earlier post rather than the later ones that clarified the position of myself and I believe many other "skeptics" leaves most of the above straw-man arguments curious. Though I cannot speak for all, my position, as clarified more than once and recently above, has nothing to do with your comments. That is, we generally do not dispute changes in perceived response of treated drivers, however we take issue with a number of other claims, especially those that are easily disputed with measurements.
Your last point seems to be right in line with most of us "skeptics", so we are in the same camp, so-to-speak.
Dave
Your last point seems to be right in line with most of us "skeptics", so we are in the same camp, so-to-speak.
Dave
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.