Multi-channel relies heavily on a center channel, this is not true stereophonic. I'd say here in the US surround sound is more common than stereo and being that it costs more to put together surround sound then people ARE willing to pay more for something that sounds more believable.
In the early days 2 channel beat out 3 channel purely because it was economically feasible back then. The 1930's-50's was a very different era compared to today. You have to remember LOTS of things had to happen with limited tech to double the number of channels, so 3 was out of the question.
In real life sound is not so controlled. Our hearing needs a certain amount of ambiance/space interaction to believe what we are hearing is normal instead of forced.
You can't be serious about cheaper. As I said that paper was about ten years old and we haven't seen a product. If there ever is one then I'll take a look.
I have heard many of those "soundbar speakers", you have got to be kidding to think this has any future except for novices who only care about audio for TV. Not one of them that I have heard (most) was worth having.
Yes, they are all too small, but you can't do that with bigger speakers, at least not inexpensively.
What I do is not a hypothetic "university research project". Its a real product with all the warts and practical compromises already built in. As I have said, vaporware is always the best design.
I don't think your responses here are realistic "stereo is dead", "soundbars are the answer". We are certainly not after the same things.
Last edited:
Maybe I need to clarify what I meant about the delay compensation thing. I am aware that you can on a traditional stereo adjust the volume and delay to make the stereo imaging remain stable if a listener is off axis. This is used in Meridian DSP loudspeakers. But what does this accomplish? You still have one sweet spot for one person and if you fix the imaging for the person on the right side then the center and left side has altered imaging. So I don't see this as much of a problem solver. I guess it's a nice gimic but good luck getting someone to where a head tracker so that it "works".
I am all about Surround Sound and Quad. But I think there is something about stereo. How easy it is to setup and how effective it is in it's simplicity. And ime a lot more real rooms - not dedicated listening rooms - are compatible with stereo than surround sound. BUT I tend to think a lot of the improvements audiophiles seem to be on a life quest to find are easily attainable with surround sound. If you are going for realism with stereo I am afraid you will never get what you are looking for. At least not without a healthy imagination/suspension of disbelief.
I am all about Surround Sound and Quad. But I think there is something about stereo. How easy it is to setup and how effective it is in it's simplicity. And ime a lot more real rooms - not dedicated listening rooms - are compatible with stereo than surround sound. BUT I tend to think a lot of the improvements audiophiles seem to be on a life quest to find are easily attainable with surround sound. If you are going for realism with stereo I am afraid you will never get what you are looking for. At least not without a healthy imagination/suspension of disbelief.
Last edited:
I know subjective descriptions are frowned on a bit around here, but I can't help but note that there's a audible difference when you sit too close to a speaker with a large center to center spacing. As I see it, there are two easy ways to sidestep the problem: Sit further back, or use a smaller spacing. In the drive for larger and larger waveguides, people seem to have neglected that they need a big ol' room to sound their best.
As for measurements, I wouldn't be comfortable doing that. My mic is a lot cheaper than what Geddes is using, and I measure outside, so we're going to get different results. While I am no slouch at this stuff, he has about a hundred IQ points, twenty patents and thousands of hours of work that I don't
But the theory, that's easy to discuss.
First, the equation:
nearfield to farfield transition = r
array length in meters = l
radiation frequency = f
r = ((l^2)*f)/700
The equation above is from a JBL white paper by Mark Ureda, named Line Arrays - Theory and Application.
You can also rearrange Ureda's formula so that we can focus on the length of the array:
l^2 * f = r * 700
l^2 = (r * 700)/f
l = sqrt ((r*700)/f)
l^2 = (r * 700)/f
l = sqrt ((r*700)/f)
Now let's crunch the numbers, first for 1000hz:
r = ((l^2)*f)/700
r = ((0.8^2)*1000)/700
r = 0.9meters, or 36 inches
(array is 31", both woofer and waveguide are playing here.)
r = ((0.8^2)*1000)/700
r = 0.9meters, or 36 inches
(array is 31", both woofer and waveguide are playing here.)
And now for 2khz:
r = ((l^2)*f)/700
r = 1.8meters, or 72 inches
r = 1.8meters, or 72 inches
And for 10khz:
r = ((l^2)*f)/700
r = ((0.38^2)*10000)/700
r = 2.1meters, or 82 inches
(This one assumes an array length of 15", the waveguide alone.)
r = ((0.38^2)*10000)/700
r = 2.1meters, or 82 inches
(This one assumes an array length of 15", the waveguide alone.)
Admittedly, the theory doesn't indicate that there will be a problem, as long as you stay at least seven feet back. To my ears, the distance to the array needs to be double or ideally triple of what the equation shows. I've also found that small arrays sound better than large arrays when you're seated closely.
Last edited:
Rotating the array does not change it's distance to the listener, and that's what we're trying to increase. Alternatively, we can simply reduce the size of the array.
Keep in mind, this isn't a knock against the Summas, if anything it's a plug for the Nathans. I don't see any good reason for an 18" Summa, and even 15" is more than I need.
The thing about directivity is that even a 3db or a 6dB reduction in off-axis energy makes a huge change in the power response, so a moderately sized waveguide makes a bigger difference in the response than the graphs show.
For instance, if you have a six inch loudspeaker with a perflectly omnipolar pattern, and it creates a reflection on a nearby wall, it will create a peak of as much as six dB. That's the equivalent of quadrupling the power at that frequency. As you can imagine, very audible. OTOH, if the off-axis response is down by 6dB, the peak can be no louder than 3dB, or equivalent to a 100% increase in power (at that frequency.) As I see it, how you get there doesn't matter. There are a number of ways to control directivity.
This ignores masking, and phase, and all kinds of good stuff, but you get the general idea.
Last edited:
A loudspeaker can't be assessed as a real (natural) sound source. The whole reproduction chain from sound recording and mixing to the loudspeaker-room-system make up the illusion. The key is the indirect sound field (I stopped using the word "diffuse" because the sound field in acoustically small rooms is not diffuse). We still don't know enough to make the best out of stereo/multichannel, loudspeakers and rooms and there might be solutions that are mutually exclusive. All we know is that we're trapped in Audio's Circle of Confusion
Best, Markus
Some of that circle of confusion comes from second guessing the quality of the audiences playback system imo. I also don't think it's a great idea to master and mix an album in a room with an abundance of absorption when the target playback room most likely will just have everyday objects and not any specific absorption. By using absorption it also brings the cost of an accurate playback system way up. And is it really necessary?
And there's a multitude of good reasons for that!
So we still don't know where the Summa's far field begins. Maybe Earl has data on that? He already showed so much pretty colorful and detailed data in his paper that the other (bad) boys don't show
Why not? I don't consider $1200 inexpensive. It's not my fault your waveguides cost so much.
Exactly.
Why not? Relocating expensive electronics at their focal point or using acoustically transparent screens is not possible?
I don't think I said soundbars are the answer. While soundbars do also offer WAF factor, I merely stated people will choose what sounds more realistic (i.e. surround sound, binaural over headphones, etc etc..). If it takes different playback methods to work better with the room, there is no reason to simply label research vaperware. I thought you would respect this type of work instead of shunning it. You are correct, I am not trying to sell a product. As a hobby the more I learn the more I am interested.
Diffusion means absorption. Control rooms are "naked" so you do need to have some kind of absorption. Domestic listening spaces differ even more from each other than typical reflection free zone control rooms do. So what's the target? Anyway, a standardized mixing/mastering environment with a defined reflexion pattern (level, delay, angle, spectrum) would be most helpful to advance sound reproduction. At home you could follow the standard or not but at least you would know what goal you missed.
Best, Markus
For me it would be a speaker system which operates perceptually independent from the room. Or as close to that as I am going to get realistically. And to keep it simple, effective, and acce$$ible.
Smyth Research
Headzone PRO
DSPeaker-HeaDSPeaker
I'm selling my Headzone Pro - so if you're interested send me a PM
P.S. By the way, the headtracker is not necessary. There are NO front-back reversals when the HRTF is personalized.
Headzone PRO
DSPeaker-HeaDSPeaker
I'm selling my Headzone Pro - so if you're interested send me a PM
P.S. By the way, the headtracker is not necessary. There are NO front-back reversals when the HRTF is personalized.
Then i think you and Geddes are on quite different pages. $1200 for a high-performance waveguide or horn is not that expensive. I wouldn't even get one of my Compression drivers for that amount...... (JBL 2435hpl)
If you want expensive, check out the now discontinued TAD TH-4003.
I Have looked at Geddes speakers (not heard them) and too me they seem very reasonable priced. Considering the drivers and custom waveguide they incorporate.
Just my 2cents
-Tim
Have any theories on why it works? Just getting the delay times correct or something? Have you ever tried to just make the image flip with your mind?
It seems to me that particular artifact might be something like a 2-D artifact I noticed as a kid. With basic 2-D perspective illusions for the most part you perceive them the way they are intended by the artist. But you can sort of force it with your perception to invert. I remember a friend pointed this out to me when I was little and we were playing Excite Bike. The simple shapes instead of coming out then went inward and it was hard to ignore or go back to the way it should be perceived. Anyway blah blah I have a feeling this artifact is similar where your focus can determine the direction counter clockwise or clockwise. But when it switches directions the front and the back swap as well.
I have more of a problem with in-head-localization and a lack of bone resonance. But I am sure you got a shaker and some sort of reflection generator to fix that
Anyway I find it easier to just use 4 equal spaced speakers that don't give much of a perceived FR shift when you move your head forward and back within the sweet spots general area. It should also not shift very much if you are to turn around and walk away from the sweet spot.
Hello Patrick
Why are you using the dimensions of the waveguide in a line array equation? I always thought a line array was a group of closely spaced drivers. A Suma doesn't exactly fit that description. Is it correct to be using them to determine near vs farfield in this case?
Rob
Good question.
Let's talk about two scenarios:
In the first scenario, we have a seven inch full range in a box, with no diffraction treatment. In this scenario, we have a minimum of five drivers:
- The seven inch full range, which acts like an array of smaller transducers with a specific cone shape.
- Four duplicates of the original signal, created at the left, right and top edge of the cabinet.
- A duplicate at the floor.
You can totally model this in FRD tools, by using duplicates that are delayed in time and attenuated. This example is basically to demonstrate that even a single speaker in a box, with no tweeter at all, is an array.
In the second scenario, we have a fifteen inch covering 900hz and lower, along with a waveguide covering 900hz and higher. (Summa.) In this scenario, we have a nearly infinite number of drivers:
- The fifteen inch woofer, which acts like an array of smaller transducers forming one big cone
- The fifteen inch waveguide, which has a wavefront designed to match the woofer
- An infinite number of reflections created off the edges of the cabinet
I'm not a huge fan of one-way speakers, but I *do* believe that one of the reasons that one-ways can sound so compelling is that their minimum listening distance is modest.
Someone's bound to ask why headphones are listenable, and the answer is that the distance is fixed. So the frequency response is designed for that fixed distance. But in a room we don't have that luxury, and the frequency response in the nearfield is completely chaotic.
This is also the reason that loudspeaker arrays in small venues, which are quickly becoming ubiquitous, sound so terrible.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.