Stealing this for my thread lol! We must have the same driver....and we pretty much have the same horn...your comments have got me pumped up! What crossover do you have at 350hz? linear phase/minimum phase? slope? etc etc
Completely agree Art, i think we are saying the same thing.
My response to Jon was in part apology for wondering whether he had even tested......
His test simply narrowed the conditions down to where he can plausibly make the "sounds the same" assertion.
Fair enough I guess, but it certainly appears when narrowed down so, the assertion becomes contentious.
Folks appear to want to extrapolate the test beyond itself, to prove it wrong even on its narrowly defined test bed... imo.
Here's one thing that is interesting...we all know that there are differences between the different driver because, physics...one of the main iterations of this difference, that is removed from table, as pointed out by Art, the off axis energy. This is significant to me because in several of my threads, the topic has arised; Perception of the signal vs the driver/systems off axis energy....this experiment supports the idea that as long as you are in the sweet spot...beaming does not change the perception of direct signal...though an increase in off axis energy should be detectable vs the absence of..it appears in this test it was not a factor...maybe because the reflective energy was low in disproportion to the direct energy....ie they were deep into the critical zone....
1.65m listening distance, on-axis to the listener's head +/- 2 degrees,
Adjustable chair so the ears would be on-axis, even for short or tall participants.
Likely still in the nearfield...
This is what I would do in order to engage in critical listening while working in the studio...I'm just relating the experience, trying to speculate whether or not beaming is an issue for perception of the signal, in the case of reflective energy influence....it seems that room energy only gives cue of spatial perception/character....to be more diligent, the information regarding acoustic treatment and reflective energy needs to be addressed....unless it has been already.
Total volume: 147m³, build similarly as the University of Surrey's audio room dimensions & BBC R&D's ratio...
5.5 meters (width)
8.1 meters (lenght)
3.3 meters (height)
RT60 : 350ms
Noise floor : 29db (C-weighting)
I'm not sure what type of measurements could have been taken but some how the reflective energy would have to of been monitored and documented during the testing to provide some relevance to what has happened...
Last edited:
Imagine all the people
I say, argument would be worse! You still don't make any sense😉
No problem, Mark 🙂
Actually, we could't properly get the Radian CD down to 320hz. Lowest we could was 360hz.
We wouldnt fight against the driver's natural capacities with the EQ and having corrections +/- 15db was out of the question. Same for the 7200hz limit with other drivers.
Also, 360hz-7.2khz was comfortable for all listeners so we didnt have to push for a wider bandwith.
What kind of bandwith do you think you have from, let's say a little bluetooth speaker? A laptop? A mobile phone?
And, most probably, not a EQ's flat bandwith... 😉
Again, as explained numerous times: The bandwith chosen was determined by practical factors. That bandwith was found comfortable enough for all listeners to listen the music excerpts. Also, that's a very wide bandwith for any midrange in a 3-way loudspeaker... And the whole idea of this test was to start with the midrange, then move to the sub/woofers test and then tweeters test, with a common midrange, so it can be listenable...
But these sub/woofers and tweeters tests were never made. Partially because the midrange test results made it difficult to justify spending more time, money and energy on that.
Would be interesting but personnally I would do the opposite since the midrange is no more a mystery for me. I would be curious to have the same woofer-mid along with few tweeters that are EQ/SPL matched.
I might actually do that test.
But since I don't have my audio room anymore, I will do it at home, privately, hidden, safe from my detractors, with the results kept secrets.
Thanks JonBocani for sharing!
What I learn from this is
1: The importance of frequency response
2: The importance of a rigorous test setup, and to do it scientifically (so that your "own intuition" does not disturb the results by "adjusting" stuff, seen it too many times...)
3: There is always people that are fact resistant... but we have to live with them, and continue deliver fact and have patience!
What I learn from this is
1: The importance of frequency response
2: The importance of a rigorous test setup, and to do it scientifically (so that your "own intuition" does not disturb the results by "adjusting" stuff, seen it too many times...)
3: There is always people that are fact resistant... but we have to live with them, and continue deliver fact and have patience!
Would be interesting but personnally I would do the opposite since the midrange is no more a mystery for me. I would be curious to have the same woofer-mid along with few tweeters that are EQ/SPL matched.
I might actually do that test.
But since I don't have my audio room anymore, I will do it at home, privately, hidden, safe from my detractors, with the results kept secrets.
An externally hosted image should be here but it was not working when we last tested it.
Thank you Art for your input.
Actually I agree with A), B) and C).
Obviously, testing the power response was not the primary goal of that specific test, nor was the max output (SPL) of each drivers.
Therefore, obviously, results would have been (probably) very different at 100db or 105db, when distortion levels would have made some drivers very easy to spot.
But what is the point doing that? None. The whole test was made to simulate ''normal Hi-Fi listening/room conditions''. And people usually don't sip their glass of Chambolle-Musigny, Talisker or diet Pepsi through an evening at home at 105db (108db in stereo)... 86-95db at listening position is much more realistic.
The endgame (which was never disclosed in this thread) was to select drivers in order to design an active 3-way speaker (commercial project). And that very test was just ONE test among many others, and the spirit was to question everything, from the DACs to the amplifiers, to the midrange, tweeter and woofer drivers. Voilà.
Obviously, again, if the project (or any project) needs drivers that can go up to 110db undistorted, it narrows down the choice of drivers. Subwoofer, midrange or tweeter.
Power response, also, would require a different room or at least positioning. A stereo set-up would also be mandatory for such test.
Bottomline, the results of this test simply highlighted the fact that the so-called hi-fi magic (often non-measurable, according to the followers) and decidedly esoteric conveyed by the audiophile market (shops, magazines, professional critics) is actually a deep and total illusion.
Also, and above all, this test, like the others before this one, highlights the undeniable fact that human hearing capacities are greatly overestimated, particularly by .... us, audiophiles / DIYers.
To conclude my response to your comment, I have the impression (subjectively speaking) that power response is another thing which, although real, has little influence on what ''normal'' people will consider to be really audible and/or important as to the overall sonic result of a sound system.
Last edited:
Thank you for your good words, esl 63. 🙂
Like a member said earlier in the thread: I am crapping on the Hi-Fi altar.
I shall be punished!
Interesting comparison.
I'm, among few other things, a sports car aficionado, and I do lapping on racing tracks on a regular basis.
The problem with your comparison is that NO car in the whole world could be hidden in a blind test.
The second you would sit in it, the second you would start the engine, would you recognize a Toyota Camry from a Volkswagen Passat. Especially if few minutes ago you would have sit in both, as the A and B presentation.
See what I mean?
You wouldnt even have to go on any road. Just sit in it and start the engine. In fact you wouldnt even have to BE the driver...
And that's for a Camry and a Passat. Two vehicles in the same category.
What happens if you compare a Lamborghini Avantador (RADIAN 950PB-beryllium) and a Hyundai Accent (VISATON FR10) ...?
Blindly. You're not even the driver. You're not even going on any road.
My guess is everyone, EVERYONE, would spot the Lambo... Even before sitting inside, because of the doors type differences and sitting level. Probably even because italian leather smells differently than south korean fabric...
Well, back to the blind test of the midrange, I basically took Lamborghini's, Hyundai's, Toyota's, Ford's, Porsche's, Fiat's.... And nobody could make the difference. 360-7200hz bandwith or not, power response or not, high SPL or not, the fact remains: the potential differences of these drivers are -at best- very slim, when listened within their capabilities.
That alone, is something to think about.
What kind of bandwidth? I dunno, but in those instances I'm not EQing everything flat from 360-7200 with a hipass and lopasss cutting off everything before and after.....
No one's denying that you can hear the EQing, just, that's all everyone's hearing in your, uh, "test". Aka, this is all semantics.
Y'all referred to testing by Floyd Toole several times. Well, here's a relevant posting on another website by Floyd Toole. A summary of sorts by Toole. Moderator, please remove this posting if not allowed.
A "Toole" summing up
Hello again. I have waited until you had a chance to vent your feelings before commenting. John has also asked a couple of specific questions of me that I think will be interesting to some of you at least....
...Link to original post by Floyd Toole...
Speaker Shootout - two of the most accurate and well reviewed speakers ever made - Page 19 - AVS Forum | Home Theater Discussions And Reviews
A "Toole" summing up
Hello again. I have waited until you had a chance to vent your feelings before commenting. John has also asked a couple of specific questions of me that I think will be interesting to some of you at least....
...Link to original post by Floyd Toole...
Speaker Shootout - two of the most accurate and well reviewed speakers ever made - Page 19 - AVS Forum | Home Theater Discussions And Reviews
Last edited by a moderator:
I can't say I'm too surprised by that result.
But here's something for everyone to consider:
I built a speaker that gives an entirely different character to the music when oriented vertically than it does when oriented horizontally. The difference is dispersion. Most waveguides and horns actually offer very little attenuation of first reflections (which is part of the reason they may be perceived similar to other drivers in the tests being discussed). For most people, the floor reflection happens about 35 degrees below the listening axis, and the ceiling reflection about 45 degrees above. Typical "directive" speakers are down maybe 3-4 dB at those angles. That's not too impactful.
I built a small line array speaker that's down more than 10 dB at 45 degrees off the vertical axis for all frequencies above 500 Hz. Much of the band is actually down 15 or even 20 dB. Admittedly, I focused on specifically attenuating the 30-50 degree off-axis range, and there may be some higher lobes at more extreme angles, but those correspond to multiple bounces to the listener unless diffusion is used.
Long story short, the vertical lobe is actually narrow enough to drastically reduce floor and ceiling reflections. In contrast, the horizontal dispersion was purposely kept as broad as possible given the budget and complexity constraints I imposed.
Part of the reason for building the speakers was to allow them to be listened to in both orientations to check the differences in perception. Since it's the same speaker with the same drivers, same crossovers, same distortions, etc, the test isolates the cause of the perceived differences. I hope to demo this to other people, but that's a little hard right now.
I performed this test in stereo, keeping the tweeter height consistent between orientations.
In many ways, the speakers sounded the same regardless of orientation.
But the soundstage was entirely different. It's not something anyone would have trouble identifying in an ABX setup, and preference to me was obvious. The vertical orientation with wide horizontal dispersion created a wide, engaging soundstage. The horizontal orientation with narrow horizontal dispersion created a narrow soundstage that seemed almost mono in comparison, and also sounded more dry, yet without any of the benefit to perceived detail you might expect from a more dry presentation.
So go figure: On one hand, we have different speaker drivers that can't be differentiated, and on the other hand, we have the same speaker in 2 different orientations that's perceived very differently.
Skylark Flying Towers: Nested Array Speakers in Denovo Cabinet -
Techtalk Speaker Building, Audio, Video Discussion Forum
Attachments
It's interesting. Good, solid background, and thanks for posting it. I'm not sure that it changes anything here.
But you have left out the single most important parameter. As I understand your testing, the FR, which you completely ignored in that statement, was EQed to point of being almost perfectly flat. And that is the main reason it was difficult to tell the differences among the various drivers.
Your conclusion is very misleading, however, for most of us who don't have the ability, or means, or interest to accomplish that very tight level of EQ. Most people here don't utilize DSP, which is essential to replicate your results.
What you haven't done is to compare the differences among these drivers without any EQ. That is where the real valuable information lies. And where significant performance differences are likely to show up based on price and quality. Not in what you have done in a very narrow test
- Status
- Not open for further replies.