I had heard a very specific setup playing one specific jazz piece that sounded so convincing to me that I walked into the showroom because of it. The only reason I did not buy it was there were still some technical criteria that seemed could be improved on, and different people can be fooled in different ways at different times. Recording techniques is also important. I was watching TV, and one particular phone ring consistently fools me into thinking it was a phone in our home, when I look at my wife, she seems to just ignore the ring as though she is not fooled. My seating is about 30deg off axis from the TV, and I use the earlier versions of the Exaltations, which basically has the same passive circuit as the one that fooled me at the sculptor's house. Both of these designs are not ideal, but can still fool me. But when you start setting up a condition for testing, there are so many factors not related to the system itself that will cause such test to never fool the audience.
If you take into the psychological aspect of the audience, the mentality that they have to find the difference, the yes, you will eventually detect the difference. When you have learned to alway try and identify that specific characteristic difference, you can probably detect it all the time. But this does nor reflect how we normally listen.
Last edited:
Hello kindhornman
Why so much SPL loss from the throat compared to the other horn??
Rob
ha, I noticed that - yes it looks like somebody stuffed it with foam to make it sound like a dome tweeter,
Well, that's the rub. Linkwitz's dipoles start off with moderate-efficiency audiophile drivers (Scan-Speak et al), and then add 6 to 20 dB of equalization. (6 dB for the mid, and 20 dB for the bass drivers.) This has the effect of starting with 0.5% efficient drivers, and after electrically correcting for dipole loss, the effective driver efficiency ranges from 0.125% to 0.005%.
That's stupendously inefficient; 200-watt amplifiers are just barely adequate. If the drivers could tolerate it, a kilowatt might be more appropriate. The Orion is a loudspeaker intended for powerful, efficient Class AB or Class D amplifiers; tube amps need not apply, unless they are zany rack-sized amps with transmitting tubes and fan cooling. They exist: I've seen tube amps that run off 3-phase power on dedicated circuits, and do a pretty good job of heating the house in winter.
Equalizing horns is different; you're starting from 10 to 30% efficiency, so equalization losses don't matter that much. Digital equalization, powerful amplifiers, and constant-directivity horns are standard practice for sound reinforcement and movie theaters. Plenty of headroom in that situation.
I'm guessing that horn enthusiasts fall in two groups: the folks using fairly powerful Class AB transistor amplifiers (in the 70 to 300-watt range), and the vacuum-tube guys using amplifiers in the 3 to 60-watt range. The high-power guys are more comfortable borrowing from modern professional practice, using digital EQ, multi-amping, etc.
The vacuum-tube guys are usually more minimalist, avoid Linkwitz-style opamp-based equalization, and are less likely to use digital equalization. It's a different toolbox, and the sonic goals are different.
The esthetic preferences between the two groups can lead to wrangles over matters of taste. The big-watt guys are not going to understand the low-watt guys, and vice versa. In the crudest terms, it's heavy metal, rock-n-roll and action-movie soundtracks versus symphonic, opera, and chamber music.
Returning the Orion, yes, I've listened to them at some length at shows. I like the spaciousness and lack of box coloration of the Orions, but I can't get past the opamp equalization and the Class AB solid-state power amps.
Some people consider opamps and Class AB transistor amplifiers sonically transparent, the proverbial straight wire with gain. But I'm not in that group.
Last edited:
As Pano points out, it can be good enough at times, if sufficient factors are in place, to be fully convincing. Does it then sound 'real'? When it's good enough that you're not interested in asking that question, you're too involved with enjoying the music - well, that's good enough for me.
Once convincing music playback is achieved, then there are two dimensions to aim to advance in after that: first, to be able to play ever 'worse' recordings without losing that quality or integrity of sound; and secondly to achieve higher SPLs without compromise. The last barricades are recent pop recordings and mastering - the compression, studio processing and styles of production are such to test systems' abilities in both of those categories - worrying about full orchestral rendition is well behind you now as a goal ...
I am kind of taking a more cautious approach in terms do expanding SPL range. Basically, size always matters, and I like to first kind of limit expected physical in a design, then try to get the best balance out of that constraint. But sound is not the only goal, it makes sense to try and take a visually appealing design if it is going to be placed where it can be easily seen. Right now, I think the Exaltations are about the most convincing active speaker for it's size as I have heard, although there is some room for further improvement, it may not be possible without lots of investment in driver design, enclosure material and other electronics. The end price could balloon many times because of this. I think we would get to that level in the next round after more experience in larger systems which some of the technology will be developed.
In the race towards the 10.000th post, the discussion now seems to be focusing on personal preferences and taste. Having followed BTA for several years and sometimes questioned myself - why?, I think the essence of the BTA-thread for me is the way Lynn explains his philosophy and his choiches along the way. This in an analytical and entertaining way, often arguing towards convention and Hifi legend.
So, Lynns personal preferences and taste is, for me, the backbone of this thread. Also having some clues about the preferences of the major contributors of insight here is really valuable to be able to put this information in context. This has been addressed before.
But the others? People chiming in with reports of some personal preference, hardly within the (rather open) context of the thread. Its a bit tiresome. But its not that important and I guess its within the nature of this medium. This minor rant is also a post from someone not contributing much insight, just expressing personal preference for behaviour in the forum. So a quick look in the mirror then..
So, addressing the qoute:
Beeing a big-amp guy experimenting with horns and seeking a "you are there" sensation with both small ensemble classical, symphonical music, delta blues, "indie"-rock and whatever genre that triggers my curiousity, I find much valuable information in this thread. I even have some opamps in my DACs and crossovers and is not at all offended in beeing one of "the other guys". There is of course enormous differences between the different "big-amps", DSP-crossovers and DACs. Most will not do. At all.
Now, looking forward to learn how Lynns dual bass arrangement is working out. Horisontal vs vertical. This is a concern of mine as well and it looks to me that a horn covering 80-600Hz is the best solution (I have a few large horns, IWATA 300, LeCleach 200T, JBL 2360) as I plan on subwoofers this looks like a feasible option. Proper scaling and headroom is a priority of mine.
Anyones opinion on Parhams 9pi? PiSpeakers Forum - Re: 9Pi mid-bass horn dimensions??? - Wayne Parham, September 06, 2006 at 17:26:56 I have a few 2226 here and plan on making a quick mockup when work slows down in december.
I think if one is striving for most transparent reproduction, both listening impression and technical aspects need to track each other. If not means of technical data can be used to explain the listening impression, then I think more study needs to take place to find that relationship.
The more advanced electronics quite often rely on automatic layout. I have never seen a good automatic layout. Generally it is necessary to play with automatic and manual interactively to get a more ideal layout in short time. Low voltage solid state designs generally have higher current flow through the ground, this makes connecting components to ground a skill in the layout. Since solid state electronics generally have higher dynamic current requirements on the power supply compared with tube designs, it makes power supply matching with the circuit more important.
Solid state devices generally have higher bandwidth, this makes coupling between digital and analog sections.
The more advanced electronics quite often rely on automatic layout. I have never seen a good automatic layout. Generally it is necessary to play with automatic and manual interactively to get a more ideal layout in short time. Low voltage solid state designs generally have higher current flow through the ground, this makes connecting components to ground a skill in the layout. Since solid state electronics generally have higher dynamic current requirements on the power supply compared with tube designs, it makes power supply matching with the circuit more important.
Solid state devices generally have higher bandwidth, this makes coupling between digital and analog sections.
Last edited:
Lynn,
I was hoping you could elaborate more on your horn optimization process.
You said it took months to finalize. But given that T=.7 is *ideal*, the exit angle of an intended driver is a given, and the maximum size for shipping is a given, a spread sheet horn fitting those boundaries could be spit out in minutes.
Did the BEM analysis show any differences that lead you to go with T=.8 instead of T=.7?
Did high frequency dispersion angle play in to the optimization in any way, and if so, what were the parameters/decisions that came into play?
I have read/heard somewhere that the -6db polar response angle of a horn (at what frequency?) could gleened by drawing a straight line from the x=0, y=0 point (axis at throat) to a tangent on the horn curve. Do you know if the angle of this straight line away from the x-axis is useful to make any decision regarding horn optimization to prevent undue beaming. I realize that CD is not a high priority for you, but avoiding undue beaming a la Tractrix is certainly a consideration.
I was hoping you could elaborate more on your horn optimization process.
You said it took months to finalize. But given that T=.7 is *ideal*, the exit angle of an intended driver is a given, and the maximum size for shipping is a given, a spread sheet horn fitting those boundaries could be spit out in minutes.
Did the BEM analysis show any differences that lead you to go with T=.8 instead of T=.7?
Did high frequency dispersion angle play in to the optimization in any way, and if so, what were the parameters/decisions that came into play?
I have read/heard somewhere that the -6db polar response angle of a horn (at what frequency?) could gleened by drawing a straight line from the x=0, y=0 point (axis at throat) to a tangent on the horn curve. Do you know if the angle of this straight line away from the x-axis is useful to make any decision regarding horn optimization to prevent undue beaming. I realize that CD is not a high priority for you, but avoiding undue beaming a la Tractrix is certainly a consideration.
There are of course sub camps.
I started with class AB Quad 303 33 but soon followed the J.E Sugden class A C51. All with the Quad ELS. Later over many years came my own amps all class A so I had unintentionally crept away from discrete SS Class AB to discrete Class A and recently into some in between world with DHT 20/20W amp. The solid state class A suited me better than a vintage Quad II
So as here I am now reaching out to the horn camp. If I had started on horns like you guys I would probably be in a different level in the DHT/Horn camp pushing the limits.
Technically, I think we all can see that the horn camps or sub camps where they are minded to stay or see the need to slip into a level of performance that is either with the top direct radiators or even ahead in term of accepted generally considered best, for their personal criteria. power, exquisite sound presentation, disarming? appearance. Horns can clearly out perform the visuals of a box in some creative art landscape, hiding themselves as some curvy form located in a designer studio creation adding to their appeal. Almost nouveau Art Deco.
The minimalist sub camp is my target with my intentional SS DHT 845, and it would be a bass reflex or other direct driver set up and mid/treble without supertweeter to start.
Just received a pair of alnico GPA 416's, so my next step will be to do some testing and see how it comes out against the TD15M's already in use.
A summary of the existing configuration, or at least what it was before I took things apart to prepare for more testing:
1. Aurum Cantus G3 ribbon, 2nd order high pass at 8 kHz.
2. Radian 745NEO/beryllium on Azurahorn AH-425, 3rd order high pass at 700 Hz.
3. TD15M in 3 cf sealed cabinet (-3 dB at 75 Hz), 2nd order low pass at 700 Hz.
4. TD15H in 5 cf cabinet tuned to 23 Hz (two 700 gram passive radiators) driven by plate amp.
The G3's will soon be replaced by Raal Lazy Ribbons.
Gary Dahl
A summary of the existing configuration, or at least what it was before I took things apart to prepare for more testing:
1. Aurum Cantus G3 ribbon, 2nd order high pass at 8 kHz.
2. Radian 745NEO/beryllium on Azurahorn AH-425, 3rd order high pass at 700 Hz.
3. TD15M in 3 cf sealed cabinet (-3 dB at 75 Hz), 2nd order low pass at 700 Hz.
4. TD15H in 5 cf cabinet tuned to 23 Hz (two 700 gram passive radiators) driven by plate amp.
The G3's will soon be replaced by Raal Lazy Ribbons.
Gary Dahl
I took those polars both used in the advertising brochures and in the AES paper. They were made with 1/3rd Octave pseudo random noise stimulus. The noise had a repeat cycle, so that is the cause of the slight "sawtooth" repeating pattern that you see. Obviously the pen speed was sufficient to catch the fine detail of that pattern, and certainly any detail of the polar curves themselves.
The biradial polars are that smooth and there is no fine detail to see in the front of the dispersion pattern (the back half has some cancelation lobing that varies with frequency).
Its okay to be sceptical but you shouldn't accuse people of deception when you have no facts to back it up.
David S.
Dave,
I don't think I accused anyone of being deceptive. but I do believe that most manufacturers do smooth out the response curves for marketing reasons. Now comparing the two different approaches would be rather difficult as they are just such different approaches to the end result. I am sure that I would more than likely prefer the JBL with the bi-radial horn over the UREI coaxial as I am very familiar with the old Altec designs. The horn is just to small on the UREI, there is always problems with the transition from the cone driver to the compression driver. Either the horn is to small to get low enough or you can say the cone driver is hard pressed to smoothly reach the cutoff of the small horn. It always seemed rather silly to try and correct these coaxial designs with an added midrange driver to fix the midrange vocals.
I don't think I accused anyone of being deceptive. but I do believe that most manufacturers do smooth out the response curves for marketing reasons. Now comparing the two different approaches would be rather difficult as they are just such different approaches to the end result. I am sure that I would more than likely prefer the JBL with the bi-radial horn over the UREI coaxial as I am very familiar with the old Altec designs. The horn is just to small on the UREI, there is always problems with the transition from the cone driver to the compression driver. Either the horn is to small to get low enough or you can say the cone driver is hard pressed to smoothly reach the cutoff of the small horn. It always seemed rather silly to try and correct these coaxial designs with an added midrange driver to fix the midrange vocals.
Gary
May I ask a question about the crossover frequency that you and Mr. Olson chose. I suppose you aim to use the AH-425 as low as possible. JMLC's rule of thumb is that two times the fc (2 * 425 = 850) is the lowest recommended cut off frequency. Have you settled on 700 Hz after some tests? Is 700 Hz the -3 or -6 dB point? I lack experience with horns, hence my interest. I plan on using a 3d or 4th order high pass with the same Radian driver but haven't settled on the crossover frequency yet. It is obvious that horn gain must drop around and below fc, but it is not clear how catastrophic or gradual the degradation is. Many thanks.
Pierre
Last edited:
Nice tweeter option there Gary . I hadn't seen those till now . The 105dB is going to be handy ! There's one more thing to be envious of, as I try to save up for some Radians !
Ps. On the padding-down issues for the CD ( which will be more significant on the 745 Neo/Be ) I was talking to CV ( Shinebox ) at the weekend and apparently I've come to the same conclusion as a couple of the London group - that resistors of this type, used in series :
Buy Through Hole Fixed Resistors POWER THICK FILM 10W 4R7 BI Technologies BPC104R7J LF online from RS for next day delivery.
are actually more transparent than the autoformers .
OK, I suppose I'm operating more in a 10-15R series resistance game myself, but worth considering if it fits into anything you're playing with , crossover-wise .
I used 6.8R and 10R & could hear NO difference at all on the 288H , and a slight softening of dynamics ( or was it a taming of the HF resonances ??! ) with 22R . I guess it's heading more towards a current-sourced operation at that stage, but not unlike how Jean Michel runs the TAD2001 .
Ps. On the padding-down issues for the CD ( which will be more significant on the 745 Neo/Be ) I was talking to CV ( Shinebox ) at the weekend and apparently I've come to the same conclusion as a couple of the London group - that resistors of this type, used in series :
Buy Through Hole Fixed Resistors POWER THICK FILM 10W 4R7 BI Technologies BPC104R7J LF online from RS for next day delivery.
are actually more transparent than the autoformers .
OK, I suppose I'm operating more in a 10-15R series resistance game myself, but worth considering if it fits into anything you're playing with , crossover-wise .
I used 6.8R and 10R & could hear NO difference at all on the 288H , and a slight softening of dynamics ( or was it a taming of the HF resonances ??! ) with 22R . I guess it's heading more towards a current-sourced operation at that stage, but not unlike how Jean Michel runs the TAD2001 .
I'm not sure if you and I are talking about the same thing.
I do know that there are reproduction systems which sound convincing.
However, in my eyes, there is difference between 'a convincing sound' and 'listening to it blindfolded, an experienced listener would think one is listening to live performance'.
Indeed, very much so.
I'm not sure what you are saying here.
Are you saying that in normal listening, one would think/feel one is attending live performance?
I don't know - didn't try that trick.
I mean that of the several 1000s that attended the demos, most were amazed at how much it sounded "like the real thing". Many were surprised and perplexed that speakers could do that. And across many recordings and genres.
I also remember walking into a demo room in Vegas at The Show and hearing a symphonic recording that was very, very lifelike. It strikes you right away, how lifelike it is. Is it perfect? No. But is it close enough that you don't care? Yes.
There are some systems and rooms up in Seattle that also have that reputation, tho I haven't heard them personally.
I've had that same "illusion" with all direct radiator and electrostat systems with the right music. Audiophile classical recordings are great for sound quality, proper mic'ing, etc. But in my experience the best performances are usually found on old recordings. Same with jazz. Rudy Van Gelder recorded some of the best jazz the world has ever known, but his recording techniques are far from natural. Roy DuNann knew how to really capture the sound of an instrument. Like wise some of the greatest American music ever made pre dates WW2; swing, bebop, delta blues, etc. It's never going to sound real on any system, but the music is what matters. Or how about the multi tracking, mixing and processing Eddie Kramer did on Jimi Hendrix's albums. Never going to sound real, but he is one of my favorite rock musicians.
I have a direct to disk LP of some jazz guitarist that sounds so real on my ESL63s. But I couldn't tell you the recording or artist name since the music is so bland.
Last edited:
Recordings
My opinion is still that John Wood, at 'Sound Techniques' in London ( Chelsea ) did the best recordings I'm aware of . The best examples of these are on the early 'Transatlantic' LP's featuring Pentangle and John Renbourn ; but obviously a lot of the early Island work is very close in quality , and close to my heart, as you can tell from my username . His ability to render tone and natural dynamics set him apart from the rest at that time in those genres of music .
My opinion is still that John Wood, at 'Sound Techniques' in London ( Chelsea ) did the best recordings I'm aware of . The best examples of these are on the early 'Transatlantic' LP's featuring Pentangle and John Renbourn ; but obviously a lot of the early Island work is very close in quality , and close to my heart, as you can tell from my username . His ability to render tone and natural dynamics set him apart from the rest at that time in those genres of music .
Of interest in this paper is the variation of *optimum* directional patterns with distance from the speakers, both for nominal directional pattern AND relative directional patterns between high and low frequencies. (See Table 1 on page 791, but note that the ratio D1/Y should read Y/D1.)
The general trend is that the farther you go back from the speakers, the wider the directional pattern should be for optimum, but note the RELATIVE beamwidths between high and low frequencies also changes. However, for the ubiquitous equilateral triangle setup, high and low beamwidths (defined as the total angle between 3db down points) are the same at only 49 degrees. (Note that beamwidth is defined as 3db down vs 6db down that is usually considered for constant directivity criteria.) Therefore, flat off-axis FR at 25 degrees either side appears sufficient for optimum localization. Closer in, the recommended beamwidth angle for high frequencies should be less than for low frequencies, albeit with VERY narrow directivity.
Because most loudspeakers would not meet these strict criteria, and in-real-room reflections would likely alter these guidelines, experimental toe-in is obvious. But the bottom line is these criteria may be a useful consideration for optimization or selection of speakers based upon a user's defined layout, whether by choice or as dictated by the room or seating conditions.
Last edited: