This fits to my experiences regarding horizontal wide vs. narrow directivity. And you are not unique. Many people like a good amount of lateral reflections ("spaciousness"). Imaging is important though, but not everything about good sound.
Not to mention the side lobes from 5 kHz upwards. From this critical frequency the CBT doesn't work anymore.
I need to address here that this comparison was also conducted with side wall treatment. So the difference I described isn't really related to lateral contribution.
Personally I only prefer side wall reflections if they arrive very late in time, though it's also dependent on the music material. But I use diffusion in the rear of the room which adds spaciousness and envelopment.
Were the whole sides treated or only the first reflection points?
When modeling the sound radiation as a sphere around the speaker there are not only reflections from the transversane plane which arrive to the ear, but also higher order reflections that arrive from different angles (but still "lateral"). Their amplitude highly depend on the directivity of the speakers. My point is that only treating the first reflection points doesn't eliminate all lateral reflections.
All arriving side wall reflections were attenuated. That's easy to measure.
I also did a comparison ones with slanting walls with almost no side wall reflections. Same result here.
I wouldn't have described the difference I heard regarding openness/clarity without attenuating high gain early reflections, since I would also describe something else than. But a major difference with them with no side wall treatment is that the CBT sounds tonality right and great, while Abbey doesn't. Abbey is actually much more dependent on treatment on side walls than CBT in a narrow room to sound good. And vertically the difference is larger. There's simply no comparison between these two speakers. You need a much larger horn speaker with a bigger voice coil to compare to the quality of CBT IMO.
Last edited:
I put this on my things-to-do list and will post when I have the measurements. My listening room is 14 x 23 with no side wall treatments and I have no problems with side wall reflections. It would be interesting to develop some type of metric for a proper comparison of the composite horizontal output at a number of vertical measurement heights.
Being that short it won't. I was perfectly aware of the smaller sources used in that theoretical example. I'm also perfectly clear and informed on my own design compromises.
I'm not burning down the CBT, as I said before. I just think the comparison is flawed. Yeah he had to choose something... I'll live
I hope you do get around to that! Would love to see that data (really!).
I have no hidden agenda, I'm just here to learn. Nothing to sell even.
Personally I think most, if not all speakers mentioned on this thread can be made to work. There isn't one among them I did not seriously consider. The most important factor for me was: getting away with it with permission from the missus.
Second one: what would it need/take to work with the room to let it perform as best as it could. I had to combine that with limited placement options and the general dislike for horn speakers from my girl. What can you do
. Personally I love that look!This brings up an interesting point about how to deal with directivity that changes with frequency. Both CD horns and CBTs have low frequency limits below which they become more and more omnidirectional. This makes for an uneven power response, which could color the sound as we perceive it.
I wonder if the problem largely goes away if we maintain directivity down to about 280 hz or so. At that frequency, the floor bounce will typically be less than 1/2 wavelength behind the direct sound, so it starts to correlate with the direct sound. At that point, the floor bounce is acting like reinforcement, and we can EQ the frequency response to be flat at the listening position, and the result might sound uncolored.
How do others suggest handling the loss of directivity at low frequencies?
It is my belief and this is held with some significant psychoacoustic support, that below about 500 Hz coloration of the sound because of room reflections is becoming minimal and vanishing at every lower frequencies because the ears processing time is getting so long that multiple reflections are occurring and the sound is become steady state. Thus a system that holds its DI down to 500 Hz will not be hampered when compared to a system that holds it lower than that. And a higher horizontal DI will always be preferred to a lower on in a small room because of very early reflections above 500 Hz have a negative effect on imaging. This belief is born out by all the opinions expressed here except Mr. Omholts. His are unique.
I agree with @Wesayso, basically each one of these speakers have their merits
The treble on the Summas also sounded 'dull' to me at first, but I think that perception was because of two things:
1) there's very little energy radiated into the back wall and the side wall, and due to that, you don't get the added reflections above 900Hz
2) constant beamwidth transducers simply sound different, and it takes some adjusting to get used to. After living with the Summas for a week or so, I found conventional dome tweeters to sound strange, and I still do.
But, again, different strokes for different folks. I think all of these speakers are great and if I had the room I'd have all three.
What would be the right slope for the HF response and from what freq onwards ?
Hi,
Sorry the comparison stops there. As owner of Summas and Abbeys- and now revised Summas- I can state that Summas are much better in resolution and soundstage and different animal compared to Abbeys.
I will provide more feedback after a month or so once I am settled in with my gear in my rental condo but seems like Pioneer/TAD S-1ex will be on sale shortly.
Regards,
Kishore
P.S>I will not dispute what you have heard since it is your room and your ears..'different strokes for different folks' as they say
That's good to heard. I have no experience with Summa.Hi,
Sorry the comparison stops there. As owner of Summas and Abbeys- and now revised Summas- I can state that Summas are much better in resolution and soundstage and different animal compared to Abbeys.
I will provide more feedback after a month or so once I am settled in with my gear in my rental condo but seems like Pioneer/TAD S-1ex will be on sale shortly.
Regards,
Kishore
P.S>I will not dispute what you have heard since it is your room and your ears..'different strokes for different folks' as they say
Something that's interesting, is that the CBT36 is much closer in clarity and details to my Klipsch K-410 horn that uses a 4" coil beryllium driver than what Abbey is. So I don't see this as being related to dome vs compression driver. Neither related to dispersion as early reflections are strongly attenuated with both. The Abbeys are simply lacking in this regard.
The coloration we perceive from a change in directivity is above the very lowest frequencies. Plus the room will dominate below the schroeder frequency. Meaning directivity here will be swamped be room modes, hence constant directivity here will basically not work much.
The important thing to have constant directivity down to the schroeder frequency, which is in most rooms around 200-300 Hz. Getting it down to 500 Hz is also a very decent compromise. But a shift in directivity higher in frequency becomes more of a problem and which I have described in this thread. Down to 1000 Hz is simply not good enough for a speaker to sound tonality correct, thus you become very dependent on strong treatment to minimize the change in reflected energy. Like I previously said; I'll much rather take a speaker with wide and uniform coverage low in frequency compared to a speaker with narrow coverage but collapsing/changing too early. This is very obvious when you can compare the speakers side to side.
I think that it is widely held that the response should fall from about 1 kHz to 10 kHz about 3 dB, so about 3 dB / decade.
This is the CBT-24 aka "epique"
An externally hosted image should be here but it was not working when we last tested it.
This is the new JBL EON 600 with a strange looking phase plug on the woofers
I'm willing to bet that you could improve the polars of the CBT with a phase plug on every single driver.
This is the ONE project I've really been itching to try, but I haven't been able to because my 3D printer kept sh itting the bed.
But...
I just bought a new 3D printer
....
I have to dissappoint you. But your approach won't work.
To eliminate side lobes you'll need to narrow the directivity of the single driver. JBL's funny looking coverage does the opposite.
What works are small fins (horns) between the drivers. They narrow their vertical directivity and attenuate side lobes. I know it, because I already tried that on my line array prototype (no, pink wasn't my choice!
). Full documentation
Being that short it won't. I was perfectly aware of the smaller sources used in that theoretical example. I'm also perfectly clear and informed on my own design compromises.
I'm not burning down the CBT, as I said before. I just think the comparison is flawed. Yeah he had to choose something... I'll live
I hope you do get around to that! Would love to see that data (really!).
I have no hidden agenda, I'm just here to learn. Nothing to sell even.
Personally I think most, if not all speakers mentioned on this thread can be made to work. There isn't one among them I did not seriously consider. The most important factor for me was: getting away with it with permission from the missus.
Second one: what would it need/take to work with the room to let it perform as best as it could. I had to combine that with limited placement options and the general dislike for horn speakers from my girl. What can you do
No problem - important thing is that what you have works for you and allows you to enjoy the hobby. When I talk to Don next I'll ask him to consider adding a floor-to-ceiling comparison to the CBT. I think the CBT offers wider appeal for DIY since many (myself included) don't have the option for a floor to ceiling array. Even if I did, I think the spatial presentation of a CBT is more realistic.