Look at the directivity above 5K on the large screen array systems.
Good luck finding a decent horn/waveguide in a reasonable size that will get you there.
Rob
The Raptor did it down to 700Hz.
When Geddes posted response above 8KHz, I should be assuming polar response and not just the response?
I can not find the polar response on those larger screen arrays, it would be interesting to look at them.
OK but what's happening up top?? Here is the DI curve for the JBL Screen Arrays you mentioned. In a perfect world the directivity would be a straight line. You can see that the bass starts at a very low number and then the DI increases as you change over to the waveguide in the mids and top end. Take a close look above 2K or so. The number is getting higher and above 10K it starts an abrupt change, you can see the hook at the end.
A CD waveguide would have a straight curve like the older 2360 Bi radial horns
http://www.jblpro.com/pages/pub/components/23606566.pdf
That horn family are 2" throats and you can see above 10K they simple loose control.
The on axis response is another matter entirely and can be flat either by the increasing DI from the horn, Electrical compensation or a combination of both.
Rob
Attachments
The Raptor link is in this thread earlier. The measurements look pretty good on and off axis. Yes, there is a drop in off axis above 10KHz but the important question is does that really matter on the audiable side. All reviews of that design say it does not matter and its an awesome sounding speaker (Those reviews are as good as any Geddes speaker reviews too).
I agree 100% on the technical stuff and if having 100% directivity matters more to someone then how it sounds that is their choice. There is always compromises in audio, nothing is perfect, I would rather have low XOs and a drop off up high then higher XOs and Directivity to 20KHz.
Thanks for the JBL DI. I know those screen arrays are widly successful having heard them, I can safely conclude that there isnt anything ultra important about great DI in the upper frequency range. I also know some individuals with HT rooms that have never had a better HT setup then the JBLs.
I was trying to find out the real meaning of "response" POV from Geddes. I know understand that point to be really about the polars in the higher frequencies.
I agree 100% on the technical stuff and if having 100% directivity matters more to someone then how it sounds that is their choice. There is always compromises in audio, nothing is perfect, I would rather have low XOs and a drop off up high then higher XOs and Directivity to 20KHz.
Thanks for the JBL DI. I know those screen arrays are widly successful having heard them, I can safely conclude that there isnt anything ultra important about great DI in the upper frequency range. I also know some individuals with HT rooms that have never had a better HT setup then the JBLs.
I was trying to find out the real meaning of "response" POV from Geddes. I know understand that point to be really about the polars in the higher frequencies.
Last edited:
The XT1464 in Raptor is a narrow dispersion (60° x 40°) 1.4" throat horn that's only spec'd to 12.5 kHz. Check the polars and you'll see it collapses to 30° x 30° in the top octave, and might as well be tractrix or exponential in this respect:
http://www.eighteensound.com/renderPdf.aspx?pid=177
It's about power response in the end, and maintaining room-independence requires constant directivity through the top octave and EQing (rolling off) the VHF for the desired in-room response. The optimum curve shape is indeterminate, and likely varies with listening preferences, but it's more predictably controlled using EQ as in Geddes designs than via a directivity and room treatment crap-shoot.
For multi-channel HT with a center channel, it's all but moot; look at the off-axis angles involved. I provided an analysis of ScreenArray in another forum. It's a defined directivity waveguide intended for a specific application:
From the Screen Array patent:
A particular type of perforated screen is required, not typically used in HT, thankfully, as it reflects back into the waveguide. Bottom line: better than XT1464, but still not all that well suited to 2-channel except, perhaps with a particular deployment.
[Kinda huge, tho, if that's the thing.... ]
http://www.eighteensound.com/renderPdf.aspx?pid=177
It's about power response in the end, and maintaining room-independence requires constant directivity through the top octave and EQing (rolling off) the VHF for the desired in-room response. The optimum curve shape is indeterminate, and likely varies with listening preferences, but it's more predictably controlled using EQ as in Geddes designs than via a directivity and room treatment crap-shoot.
For multi-channel HT with a center channel, it's all but moot; look at the off-axis angles involved. I provided an analysis of ScreenArray in another forum. It's a defined directivity waveguide intended for a specific application:
From the Screen Array patent:
A particular type of perforated screen is required, not typically used in HT, thankfully, as it reflects back into the waveguide. Bottom line: better than XT1464, but still not all that well suited to 2-channel except, perhaps with a particular deployment.
[Kinda huge, tho, if that's the thing....
]
Last edited:
Bottom line: better than XT1464, but still not all that well suited to 2-channel except, perhaps with a particular deployment.
[Kinda huge, tho, if that's the thing....
Well, its a good thing Im as about as far away from 2 channel as you can get.
Everything I do is all about the HT design. Even my music is all digital and uses surround.
As for the Raptor, in the end we can agree that tech specs only go so far and its what listeners like. The Raptor has proven itself in listening tests against other designs so lets agree that a > 1" CD speaker can be a GREAT sounding speaker that lacks the full controlled directivity crowd sign of approval. It still has better directivity then dome design.
I think the advantages of a larger CD/horn overcome the issues of directivity that high up in the frequency. Again, speaker design is about choosing your compromises.
Last edited:
Yes, it's very much about choosing your compromise..
But what is "better" directivity?
Many designs with "domes" are often very uniform in their polar response. The exceptions are classic "two-ways" with a larger diameter midbass coupled to a 1" or smaller dome tweeter, and even then the usual loss in pressure off-axis between 1 kHz and the typical 2-3 kHz range can often achieve an increased sense of depth that often is more natural sounding (..and is found in better implementations that are essentially "flat" horizontally in the same freq. region).
Is *increased* directivity an advantage, and if so why?
Can you achieve that same advantage through other means?
Might there be *disadvantages* to a design that has an increase in directivity?
And particularly:
In the Raptor - isn't there the possibility that what people liked about had little if anything to do with it's polar response? In fact, perhaps people liked the design *despite* it's polar response.
Consensus appears to be that we do not know what comprises the optimum power response. Toole teaches that it is smooth and uniform, tracking the axial response. I'm suggesting that "ideal" may well be variable power response with controlled directivity. See Tannoy et al.
Another observation I consider highly relevant -- we go to DIY and other demo fests and compare various designs in medium to large listening spaces bearing little in common with our own smaller "venues" where the impacts of directivity and power response are far more significant. In many cases, other than the floor (or stage) boundary, it's virtually a freespace condition with long delays and typically uncontrolled reflectivity. Yes, after some time, we adapt and "hear through" the room, but advantage still goes to those designs better suited to the listening conditions characteristic of these larger spaces, and listener preferences must be considered in context of the auditioning environment....
Another observation I consider highly relevant -- we go to DIY and other demo fests and compare various designs in medium to large listening spaces bearing little in common with our own smaller "venues" where the impacts of directivity and power response are far more significant. In many cases, other than the floor (or stage) boundary, it's virtually a freespace condition with long delays and typically uncontrolled reflectivity. Yes, after some time, we adapt and "hear through" the room, but advantage still goes to those designs better suited to the listening conditions characteristic of these larger spaces, and listener preferences must be considered in context of the auditioning environment....
Last edited:
Seems to me that the opposite is really the case as all of the "experts" seem to be in complete consensus. There isn't a "consensus" on evolution either, but that doesn't make it any less a fact. Science is not a democracy.
Listening to any speaker is pretty far removed from evolution or even science.
Again, its fine to like accurate measurements (I do) but lets not think that the only good speaker is an accurate design based on ONE philosophy. There are many different designs for different applications, there isnt one design that is superior to all others since there is not a design that fits all the variables people could want.
Remember speakers are no more then furniture to most people there is no need for superior design. Just a design that fits the specific needs of that individual.
Last edited:
Yes, it's very much about choosing your compromise..
But what is "better" directivity?
Many designs with "domes" are often very uniform in their polar response. The exceptions are classic "two-ways" with a larger diameter midbass coupled to a 1" or smaller dome tweeter, and even then the usual loss in pressure off-axis between 1 kHz and the typical 2-3 kHz range can often achieve an increased sense of depth that often is more natural sounding (..and is found in better implementations that are essentially "flat" horizontally in the same freq. region).
Is *increased* directivity an advantage, and if so why?
Can you achieve that same advantage through other means?
Might there be *disadvantages* to a design that has an increase in directivity?
And particularly:
In the Raptor - isn't there the possibility that what people liked about had little if anything to do with it's polar response? In fact, perhaps people liked the design *despite* it's polar response.
Good questions, First, Domes absolutely suck for my applications. They have limited SPL and limited off axis response. I have no need for them whatsoever any more.
Waveguides/horns give the control needed and they also handle the dynamics better then any other tweeter choice so there is no other way to meet my application needs.
I have never read where directivity is a disadvantage, I think as time goes on more manufacturers will fix their off axis response because they are starting to realize how important it is. There will still be all types of designs and all those designs will have a fan base. Remember, people are not clones of each other and to think that there is only one superior design does not make sense at all in the real world.
Why people like the Raptor is not important to the directivity question, its important to the simple fact that its a 1.5" CD that measures well and sounds very good. It was using a specific design to counter Dr. Geddes too generalized "all > 1" CDs have bad responses".
For any given application there will be one and only one "optimal" solution. I, like most experts who have studied this situation, believe that accuracy IS the key. Subjective impressions are fleeting and unstable (especially when not blind), so to me the most accurate is the best. Any other position is hard to support with anything objective, only:
"Sounds good to me, so what does it matter?"
Fine, if thats how you want to view the situation - it's not my point of view that is for sure.
The question may be: do we want our reverberant field to sound flat, or should we make some timbreal adjustments.
I suspect we may be able to adjust the timbre of the reverberant field without too much effect on the perceived timbre on the whole, since we have clearly delineated what we want from a speaker, where there is (hopefully) very little in between the direct field and the late reverberant field.
I'm sure some of us know the sound of a small single driver on a conical waveguide, that beams at the top.
I'd like to try it the other way and increase the power response toward the top end. Short of designing for increasing dispersion, I can't see how this could be done unless on axis was in a partiall null.
Im 100% in the DBT camp, I have been doing the tests since University 22 years ago.
Have you done a controlled listening test with a 1.5" vs a 1" CD?? Again, I will post that its important to have accuracy and measurements but at the end of the day does anyone really know how much audible difference there is between the two in a proper listening test? If there isn't much difference they why all the stubborness to be chase perfect measurements or to post that "all responses are horrible" type blanket statements?
I always go back to think BOSE knowing they sound like crap but they also do a lot of R&D on what sound sells and they stopped fussing over the unimportant stuff a long time.
OK but what's happening up top?? Here is the DI curve for the JBL Screen Arrays you mentioned. In a perfect world the directivity would be a straight line. You can see that the bass starts at a very low number and then the DI increases as you change over to the waveguide in the mids and top end. Take a close look above 2K or so. The number is getting higher and above 10K it starts an abrupt change, you can see the hook at the end.
A CD waveguide would have a straight curve like the older 2360 Bi radial horns
http://www.jblpro.com/pages/pub/components/23606566.pdf
That horn family are 2" throats and you can see above 10K they simple loose control.
The on axis response is another matter entirely and can be flat either by the increasing DI from the horn, Electrical compensation or a combination of both.
Rob
Note that the 2360A has been replaced by the 2360B.
http://www.jblpro.com/catalog/support/getfile.aspx?docid=174&doctype=3
There is a slight improvement in the 16kHz polar but the main improvement is in response smoothness at 1kHz. I'd say that, outside of this forum, most of the users of 2" throat drivers use them for their very necessary increase in power output. 500 or 800Hz crossover points also work very well with 15 inch woofers in a theater environmnet. There are plenty of 2" throat drivers that work well to 15-16kHz so there is little compromise at the top end.
I've had good results with this system in Digital Cinema applications. I haven't tried the newer version screen array series but there performance parameters look pretyy good also. My impression is that the newer generation Optimized Aperature Waveguides are a different compromise with a little less emphasis on widest HF polars.
And the patented Screen Spreading compensation....marketing making lemonaid froma a bunch of sour lemons.
David
Thanks, The smoothness at 1KHz and 500Hz XO stuff is very important, more important then directivity up past 10KHz and its why I disagreed with the statement " > 1" CD have a bad response in the higher frequencies" Its not a general statement but its just one opinion.
If we can get directivity upto 10KHz, it can be damn good already, add to that what Zilch posted about Screens and directivity. Then > 1" CD choices can work pretty good in our HT environments.
If someone can give me a DIY build with XO at 600Hz, smoothness at 1KHz and directivity upto 15KHz then Im ready to buy the drivers
Last edited:
I think that this question has been well researched and the data points in the direction of a slightly falling direct response with a parallel falling power response. This is the way of nature and it only seems logical that it is what we perceive as natural.
An increase in HF power response does not seem to have any supporting evidence.
Does all this make a difference? Above 10 kHz most certainly not, and above 8 kHz its unlikely. As Dave points out, most people use larger format compression drivers for power handling and sometimes a lower crossover point. I don't need the power handling, and the crossover point is dictated by the woofer and a good 1" has no trouble doing this. So what is the advantage of a larger format in a home theater application? They are certainly far more expensive, larger and heavier. I suspect its mostly bragging rights - I'm just not into that kind of design goal. Perhaps some lower thermal distortion due to the larger voice coil? Maybe, but again to what extent is this a problem in a 1"? Too much audiophile audio design is done on the principle of "because it is there!".
Last edited: