I can only explain from personal experience that if the recording mic location is at the ear, and the playback is also through earphones, the sensation provided is very surrounding and realistic even though the recorded background noise was higher than what I hear during live listening. People behind me actually sounded like they were behind me, and I had the tendency to turn my hear to confirm whether I was hearing the recording or a live person behind me. The same recording when played back through speakers cannot have the same effect, and I have never experienced any speaker playback that provided such sensation. I would atribute it to the interaural effect because that seems the only explainable means that I can comprehend.
George - I don't even remotely see the connection to waveguides, even if your assumptions were somehow correct.
That said, it is well know that cross-talk cancellers work best with directional speakers. But this has nothing to do with "preference".
My point is that just pure waveguides cannot achieve a narrowly focussed beam over a sufficient bandwidth to even evaluate whether the narrow beam is prefered or not. The only way that might even have a chance is combining ultrasonic modulation with a wave guide. Cross-talk cancellers only work a specific positions. Not really effective if we consider all the contraints in listening positions.
George - If you believe that then you believe that directivity makes no difference in sound quality. I don't think that is true at all. Prefered or not there IS a difference in the sound quality.
Can you explain how you come to that conclusion "If you believe that then you believe that directivity makes no difference in sound quality" ? The effects of beam width, pattern effects on ITD and IID can, and have to some extend been analyzed. We know that there are many issues that effect how playback is perceived, there just isn't a clearly sorted how data as to what effects what perception to what degree. Let's face it, any in any field of technology, there are things that would require further investigation. Why, even health related research come up with different conclusions every few years. Audio will not be different.
Hello soongsc
I remember a while back I asked you if you had any horns and the answer was no. The reason I am bringing this up is you don't need waveguides to experience the effect CD has. Horns will do quite nicely and once you have compared a CD horn/waveguide set-up with smooth of axis performance to a conventional system with poor off axis performance it's a no brainer. Overall the CD system simply sounds better through out the room.
To hear the difference all you have to do is move off axis and listen too the upper octave drop out. That's enough for me right there. Simulations are fun and all but nothing beats hands on trying to make horns and waveguides work for you in your own room.
Rob
Hi Rob,
There are many aspects in design tradeoffs. I have listened to many speaker systems, reviewed whatever data is available on those. So it really does not matter whether I own one or not. Sure, there are things that I like about horns, and certain aspects that I do not like. If you look through the thread, sims are only assisting to decide what to build and test. The question at hand is how narrow a beam is best, and how (or rather whether) a narrower CD beam can be achieved for actual comparison. To date, there is not such information available.
There are many aspects in design tradeoffs. I have listened to many speaker systems, reviewed whatever data is available on those. So it really does not matter whether I own one or not. Sure, there are things that I like about horns, and certain aspects that I do not like. If you look through the thread, sims are only assisting to decide what to build and test. The question at hand is how narrow a beam is best, and how (or rather whether) a narrower CD beam can be achieved for actual comparison. To date, there is not such information available.
Perhaps you need to try a bit harder.
Go to google scholar, advanced search
at exact phrase enter: directionality
at at least one enter: "er geddes"
you will get this
"er geddes" "directionality" - Google Scholar
the Huddersfield guys are interesting so hit related articles
related:OE7TLPH_Ks0J:scholar.google.com/ - Google Scholar
and you can find all sorts of stuff. For example:
Characterisation of small room acoustics for audio production - University of Huddersfield Repository
Characterisation of the directionality of reflections in small room acoustics. - University of Huddersfield Repository
http://eprints.hud.ac.uk/3580/1/IOA_Proceedings_Vol30-6_RS24-2008.pdf
two minutes
I'm sure you can come up with other interesting search terms. You are likely to find the situation is about the way Geddes described it to you.
Thanks much.
Edit, update: the papers on ducts and low freq don't seem relevant. The Toole ref is 1990 and content prolly familiar. The Huddersfield people have a measurement tool but no relevant conclusions that I've seen (and nothing like human tests there). Some papers for sale for $30. Reading... reading.... and I hope others are doing so too and with better grasp than I have.
Edit, update: the papers on ducts and low freq don't seem relevant. The Toole ref is 1990 and content prolly familiar. The Huddersfield people have a measurement tool but no relevant conclusions that I've seen (and nothing like human tests there). Some papers for sale for $30. Reading... reading.... and I hope others are doing so too and with better grasp than I have.
Last edited:
Horns and waveguide "families" are commonly available with 60° through 120° beamwidths, generally covering the range of interest here. I earlier showed a pair of $15 JBL PT waveguides which approximate the patterns and response characteristics of Nathan vs. Orion.
As Rob states, at some point, it's necessary to set the sims aside, build, measure, and listen.
[OR, we can just continue blathering about it ad infinitum....
]http://www.diyaudio.com/forums/multi-way/103872-geddes-waveguides-460.html#post2237477
Last edited:
I certainly would not have been able to do the number of builds in the amount of time and budget. Like over 100. Howmany horn s have you build (not bought) and posted in this thread. I have at least four. Whom has done more than I? Most of the people posting here only buy and put together. I sim the contours, get them custom made, and measured, and compared against sims. I'd like to see what others post here, but I mostly see bought and put together and measured. You can do better than that, can you?Horns and waveguide "families" are commonly available with 60° through 120° beamwidths, generally covering the range of interest here. I earlier showed a pair of $15 JBL PT waveguides which approximate the patterns and response characteristics of Nathan vs. Orion.
As Rob states, at some point, it's necessary to set the sims aside, build, measure, and listen.
[OR, we can just continue blathering about it ad infinitum....
I would be interested in seeing comparison between 90deg included and 5deg included if it can be done.
Last edited:
If I had a lathe I would consider trying simple conical contours but I don't and could not justify the expense as this is a hobby not my livelihood. I spend enough time planning designing and building my cabinets, taking measurements and then doing the crossovers in a sim. Then temporarily building the speakers up so you can verify the sims. Then you have you have to get the rough cabinets finished and then reinstall everything. So no time for me to make my own horns especially with so many readily available.
I think it's fine that you do. I have a question? Did you build up a speaker system using any of those contours? If you did how did it come out? I don't remember you posting about any completed systems. I would be curious how they worked out. Did you get the results you expected?
Rob
Last edited:
I think you probably have a more simpler goal. Good for you.I am saying there are easier means to get at answers to the fundamental directivity question(s) at issue here with resources readily available to anyone having interest in doing so without making a life-long production number out of the pursuit....
If the results are as good as I can like them for 3 months, this thread will be the first to hear about it. Let's also not forget that Earl did not find anything that satisfied him, and thus the Summas.If I had a lathe I would consider trying simple conical contours but I don't and could not justify the expense as this is a hobby not my livelihood. I spend enough time planning designing and building my cabinets, taking measurements and then doing the crossovers in a sim. Then temporarily building the speakers up so you can verify the sims. Then you have you have to get the rough cabinets finished and then reinstall everything. So no time for me to make my own horns especially with so many readily available.
I think it's fine that you do. I have a question? Did you build up a speaker system using any of those contours? If you did how did it come out? I don't remember you posting about any completed systems. I would be curious how they worked out. Did you get the results you expected?
Rob
From the work I've done, it's easy to get the normally published data to look good, I'm sure that with some of the on and off axis measurements that I have posted are probably better than lots of commercial products our there. But other measurement data that are normally not published, some of which I have raised in the "measurements" thread, really do not show good results. The most simple thing is CSD.
Soren Bech has done the fullest studies along those lines. He simulated multiple reflections via secondary sources in an anechoic chamber. Later studies used the exact off-axis response that a simulated reflection would need, i.e. the response from the angle prescribed by the image model.
If the issue is "what frequency response should the reflections be?" I'm not sure Toole answers that. Recently they have done much simulation of the in-room response of systems, he now emphasises the response angles that typically generate the early reflections, however most of his early studies show that sytems with very nonflat power response will be prefered over systems with better power response and worse direct response.
I like the Lipshitz and Vanderkoy paper on that. They used a forward firing KEF 104.2 with a sideways firing ESL 63 on top. Since the side firing ESL had the listener sitting in the null, they could perturb the sound power radiated into the room independently of the direct response. Amongst their conclusions was that holes in the power response were pretty benign. Peaks in the power response might be audible, and flat sound power was usually a problem.
D Smith