Certainly a larger room is better, and physics enters into things.
In this thread, I guess Lynn is the one to decide if his listening space is large enough for certain designs. There are a lot of people with HT rooms, so clearly its not an insurmountable issue for many. DIY Audio is an international site, not a USA site. It was started in Australia and has members all over the world. My impression is that generally the amount of space available to our typical member is smaller than that discussed.
BUT:
Let's not continue discussions of the "morality" of too big an audio room, and how many big rooms are in what areas of the world. We really aren't able to know all the facts, but clearly some are able to have them and for some it is a matter of priorities.
SO:
We should freely discuss optimal room sizes for various speaker types, and how smaller rooms can compromise them, but please: not the immediately above issues
I live in a flat in San Francisco, and I won't ever have the greatest audio due to space constraints, but still am fascinated by audio generally, even if I can't take advantage of all I read here, and it's good to know which speakers work the best in which spaces. And I still enjoy music on my sub-optimal, but quite satisfying setup.
Variac, Chief Moderator
You know, I'd like for Lynn to weigh in on this issue.
Lynn, how much influence, on the system you're designing, does the size of your listening room impact how you're accomplishing your goals?
I ask because this thing is gonna be pretty big! So, based upon the drivers, the efficiency and physical size, I'm guessing you must have a pretty large room you're filling with sound?
Lynn, how much influence, on the system you're designing, does the size of your listening room impact how you're accomplishing your goals?
I ask because this thing is gonna be pretty big! So, based upon the drivers, the efficiency and physical size, I'm guessing you must have a pretty large room you're filling with sound?
Directionality
Ken Kantor poularized the philosophy of removing early room reflections, and IME, placing a speaker along the long wall performed best, set up to Ken's old yard sticks. It maintained the most acurate tonality from recording to recording, and minimized imaging vagueness. It also provided the best presentation of HRTF/crossfeed encoded material that I ever heard, with sharp images moving fluidly and contiguously from 20' left of left, through the speaker plane, to 20' right of right, for recording encoded as such. Speakers were 6' from each side wall. Not to advocate HRTF crossfeed, but it illustrated to me that pushing out the side wall maintains extremely cohesive and wide imaging, while favouring tonal accuracy.
This room was a rare treat. For most of us, we need to deal with that first reflection, and thats my current situation. I'm surprised the discussion hasn't included use a good side wall absorber, at the first reflection points. Given the absorption these offer down to 500 Hz or better, I would think they largely alleviate the need to go to a highly directional speaker.
The difficulty with these sorts of comparisons is that its impossible to A/B this scenario, a highly directional speaker vs a more common wide dispersion design+side wall absorber, and then make any conclusions based on directionality alone.
Ken Kantor poularized the philosophy of removing early room reflections, and IME, placing a speaker along the long wall performed best, set up to Ken's old yard sticks. It maintained the most acurate tonality from recording to recording, and minimized imaging vagueness. It also provided the best presentation of HRTF/crossfeed encoded material that I ever heard, with sharp images moving fluidly and contiguously from 20' left of left, through the speaker plane, to 20' right of right, for recording encoded as such. Speakers were 6' from each side wall. Not to advocate HRTF crossfeed, but it illustrated to me that pushing out the side wall maintains extremely cohesive and wide imaging, while favouring tonal accuracy.
This room was a rare treat. For most of us, we need to deal with that first reflection, and thats my current situation. I'm surprised the discussion hasn't included use a good side wall absorber, at the first reflection points. Given the absorption these offer down to 500 Hz or better, I would think they largely alleviate the need to go to a highly directional speaker.
The difficulty with these sorts of comparisons is that its impossible to A/B this scenario, a highly directional speaker vs a more common wide dispersion design+side wall absorber, and then make any conclusions based on directionality alone.
Having followed this entire crazy thread I can say "too big!" For me - that is. The sketches and ideas that Lynn and others have come up with are much too big for my tiny cottage by the sea.
But the designs are still fascinating and full of great information. Just because I'll never build Lynn's "Beyonds" for my living room doesn't mean that the way they turn out isn't interesting.
Shootz, I'll never own a jumbo jet, but getting the chance to brainstorm with the experts on the design of one would still be fun. Even though it won't fit in my driveway.
But the designs are still fascinating and full of great information. Just because I'll never build Lynn's "Beyonds" for my living room doesn't mean that the way they turn out isn't interesting.
Shootz, I'll never own a jumbo jet, but getting the chance to brainstorm with the experts on the design of one would still be fun. Even though it won't fit in my driveway.
I'm still convinced that Lynn's work will pay off in less ambitious (smaller) versions. Just changing his max output requirement would make a huge difference in the final size I think.
It is obvious to me that my partially open baffle speakers are in too small a room, but they still sound very good. Actually the room's volume isn't small, it is just too narrow- no room for space behind the couch
Lynn must be at a doctor's appointment....
It is obvious to me that my partially open baffle speakers are in too small a room, but they still sound very good. Actually the room's volume isn't small, it is just too narrow- no room for space behind the couch
Lynn must be at a doctor's appointment....
Re: Directionality
I use a sort of side absorber/diffuser, but I prefer not to use any more absorption than necessary - people tend to over do it with absorbers. And it is diffciult to get good absorption down to 500 Hz. or so, A simple thin absorber isn't going to do that. The really difficult reflections are the floor and ceiling. I use ceiling baffles to break up that reflection and a 3 in thick rug right at the floor bounce. But to the sides and behind the listener I use no absorption at all - I want as much reverb here as possible.
But I still don't think that side absorbers changes the need for directionality as it is more than just the first reflections that are at issue. Its all the reflection in the first 10 ms, and, if you measure it out, thats about the first 20 or 30 in a small room. The first 2-3 ms. is critical for imaging - I even make sure that there is nothing that can diffract (different than reflect) the sound within several feet of the speakers. But the first 15 ms. is critical for coloration and directional sources makes a very big difference in the levels of the reflections in this time frame.
DDF said:
I use a sort of side absorber/diffuser, but I prefer not to use any more absorption than necessary - people tend to over do it with absorbers. And it is diffciult to get good absorption down to 500 Hz. or so, A simple thin absorber isn't going to do that. The really difficult reflections are the floor and ceiling. I use ceiling baffles to break up that reflection and a 3 in thick rug right at the floor bounce. But to the sides and behind the listener I use no absorption at all - I want as much reverb here as possible.
But I still don't think that side absorbers changes the need for directionality as it is more than just the first reflections that are at issue. Its all the reflection in the first 10 ms, and, if you measure it out, thats about the first 20 or 30 in a small room. The first 2-3 ms. is critical for imaging - I even make sure that there is nothing that can diffract (different than reflect) the sound within several feet of the speakers. But the first 15 ms. is critical for coloration and directional sources makes a very big difference in the levels of the reflections in this time frame.
At Ai we are really enthusiastic about our smaller ESP12 and ESP 10. The ESP 10 is about 1/10 the size of the ESP15 (Summa) but with a very comparable performance. Very directional, efficient and high output - now that should be appealing to you guys with small spaces. All three system have the same waveguide, plug and compression driver, only the waveguide mouth is a bit scaled down on the smaller models - but not that much.
Re: Re: Directionality
The absorbers I have in mind are the resonating dissipative style, using peg board, frame and lossy material to provide a good broad band absorption. They're definitely DIY, but simple enough to build, unless flush with cash. They really work well. The absorption only needs to be 7 to 10 dB to not impact image shift (from Queen & Olive).
I can't visualize having 10 to 20 discrete reflections within 10 ms (drawing it out in a 16x13 room), with a broad absorber at the sides, with the following caveats.
I'm neglecting the first vertical refelction here of course. I really agree with controlling the first ceiling reflection. In that wide room, the ceiling was acoustic tile. Not the prettiest looking room, but it sounded wonderful, the room colouration was very low. Early vertical reflections affect timbre, but vertical reflections delayed by even 10 ms can affect image height (from an old JASA paper). Break them up and pad them down a bit.
I'm also neglecting the front wall which I feel is an absolute necessity to providing adequate depth cues, to reinforce frequency dependant and level ones (ie further back = less high end and quieter).
On the floor reflection, I'm philosophically more on the fence. If you want the effect to be one where the performers are in your room, I believe that the floor bounce is an aid in that mirage. Floor reflections also only affect image shift if the envelope is greater than several seconds (Kantor) so I think they aren't a terrible problem. The old Snells on the other hand were tonally wonderful by eradicating the floor from the equation. Pick your favoured illussion and design accordingly.
With the side absorption and some ceiling control, its still fairly easy to maintain a healthy RT60 (no carpet),
I focus on tonality when dealing with the room, but I know there are many that experiment at length with room boundaries and imaging. Hopefully they can share their practical experiences here.
gedlee said:
The absorbers I have in mind are the resonating dissipative style, using peg board, frame and lossy material to provide a good broad band absorption. They're definitely DIY, but simple enough to build, unless flush with cash. They really work well. The absorption only needs to be 7 to 10 dB to not impact image shift (from Queen & Olive).
I can't visualize having 10 to 20 discrete reflections within 10 ms (drawing it out in a 16x13 room), with a broad absorber at the sides, with the following caveats.
I'm neglecting the first vertical refelction here of course. I really agree with controlling the first ceiling reflection. In that wide room, the ceiling was acoustic tile. Not the prettiest looking room, but it sounded wonderful, the room colouration was very low. Early vertical reflections affect timbre, but vertical reflections delayed by even 10 ms can affect image height (from an old JASA paper). Break them up and pad them down a bit.
I'm also neglecting the front wall which I feel is an absolute necessity to providing adequate depth cues, to reinforce frequency dependant and level ones (ie further back = less high end and quieter).
On the floor reflection, I'm philosophically more on the fence. If you want the effect to be one where the performers are in your room, I believe that the floor bounce is an aid in that mirage. Floor reflections also only affect image shift if the envelope is greater than several seconds (Kantor) so I think they aren't a terrible problem. The old Snells on the other hand were tonally wonderful by eradicating the floor from the equation. Pick your favoured illussion and design accordingly.
With the side absorption and some ceiling control, its still fairly easy to maintain a healthy RT60 (no carpet),
I focus on tonality when dealing with the room, but I know there are many that experiment at length with room boundaries and imaging. Hopefully they can share their practical experiences here.
Re: Re: Re: Freq and imaging
Pardon my ignorance, but how is this possible ? "Much better" in what respect, exactly ?
I have never heard an "audio system" anywhere NEAR the live music experience. And I've heard most of the world's "best" (admittedly, not your friend's selenium coax).
I mean, live sound is by definition a reference. ANY reproduction is just that, a reproduction. Now, "stereo" can get pretty close to some types of music (remeber Dunlavy's "music professor couldn't tell them apart" demo), but on any type of orchestral material domestic audio falls woefuly short of target.
NOT EVEN CLOSE.
johninCR said:
Pardon my ignorance, but how is this possible ? "Much better" in what respect, exactly ?
I have never heard an "audio system" anywhere NEAR the live music experience. And I've heard most of the world's "best" (admittedly, not your friend's selenium coax).
I mean, live sound is by definition a reference. ANY reproduction is just that, a reproduction. Now, "stereo" can get pretty close to some types of music (remeber Dunlavy's "music professor couldn't tell them apart" demo), but on any type of orchestral material domestic audio falls woefuly short of target.
NOT EVEN CLOSE.
Variac said:
Uh, not this time ... just a lot of wrangles getting my big PowerMac Dual G5 repaired. Applecare has dropped the ball on this one, and it's been 19 days and two service calls so far. The liquid-cooled dual processor module has been replaced (twice), the power supply (twice), and tomorrow morning the tech comes back and installs the second (new) motherboard. I suspect strongly the overclocked, liquid-cooled Dual 2.7 GHz G5 has reliability issues in several parts of the design. I've been pestering Apple for a replacement machine - this is getting a little out of hand. (Yes, I have Dual Athlon PC, but I really dislike using WinXP any more than the absolute minimum.)
I'm posting this on the partly-repaired machine - the fan/temp sensing system needs to be calibrated, of all things, and the cal procedure failed today, thus the return visit tomorrow. So I type away and listen to fans whirring away at what sounds like 75~80 dB. I'll give up pretty soon and return to the laptop downstairs with its flaky WiFi connection.
Directivity is an area, like amplifiers, where Earl and I must respectfully disagree. Narrow-directivity speakers don't do it for me - I've always liked the spatial presentation (and resolution) of electrostats, going all the way back to first hearing the Quad ESL57 at Radio People in Hong Kong in 1962.
In 1991, the Ariels were designed for Mike Spurlock in Portland, who owned a set of stacked ESL57's, and wanted a speaker that had similar qualities but was also a good match for triode amplifiers. To my surprise, a number of Quad owners in Hawaii and the UK have built Ariels, and consider them reasonably close sonically. I guess this isn't too amazing, since I follow the BBC/DEL Shorter and KEF/Laurie Fincham design philosophy.
If I was in the NRC/Floyd Toole camp, I'm sure my speakers would sound rather different. The "school" of the design usually has a strong influence on the resulting sound - all those tradeoffs, and different schools make different choices.
I see myself as a technical artisan, and a writer. I've been thinking about this a bit, and the role of technical artisans in Germany and Japan are very different than the US and the UK, which are cultures that celebrate inventors and capitalist heroes (think John Galt).
In the US & UK, artisans are given derogatory names like "dilletante" or "tinkerer". In Germany, though, artisanship goes back a millenium - this is the culture than invented the clock, after all, along with many musical instruments. There are similar traditions in Japan, and I surmise that this was a subtle influence on me as I grew up in the Kobe/Osaka/Kyoto area. Like many artisans, I design for my own pleasure - I don't have to satisfy a corporate client, nor follow the dictates of a marketing department. This flows out a quite conscious decision when I left Audionics in 1979 to never work for an audio company again - or if I did, it would be my own company, not an entity controlled by someone else.
Uhh, not so much. I heard some pretty exotic systems back in 1958~59, in the quite generous-sized homes of some of the Western community in the little town of Shukugawa.
Other influences were the Shadow Vector quadraphonic project in 1973, and hearing the BBC Soundfield mastertape recordings at the BBC Research Labs in 1975. Meeting the BBC team and Laurie Fincham in 1975 left a very strong impression. I was very lucky to visit the BBC in the glory days, before the cuts of the 1980's.
Other influences were the Shadow Vector quadraphonic project in 1973, and hearing the BBC Soundfield mastertape recordings at the BBC Research Labs in 1975. Meeting the BBC team and Laurie Fincham in 1975 left a very strong impression. I was very lucky to visit the BBC in the glory days, before the cuts of the 1980's.
Re: Re: Re: Re: Freq and imaging
I don't go to the symphony, so the concerts I've attended we're amplified. Admittedly the sound quality is often poor, whether it's due to the acoustics of the venue, bad seating location, or an inadequate setup, however, to sound better than "any I've attended" is saying something. That's also just from a sound quality standpoint, excluding the energy and visual impact of a live event. I realize it's an apples and oranges comparison, but I've never heard recorded music played back with such a "live" sound indoors or out either....more precise imaging and accuracy in detail, yes, but not the sense of space to add realism that I've found only open alignments to deliver. A significant advantage is due to a pair of point sources playing a well recorded stereo signal, however, I've heard boxed speakers (probably with better components using better equipment) used outdoors numerous times and none have come close in realism. Get a pair of decent OB's outdoors with some kind of structure behind them to put the rear wave in play, then you'll understand what I mean.
For orchestral music, you'll need speakers tailored to that because classical music places specific demands on a speaker that are difficult to reproduce. Even though it's not my cup of tea, I always throw a couple of classical pieces into the mix when I'm testing my latest speaker creation to help identify its weaknesses. Also, the fault of recorded classical music never sounding even close to live most likely lies in the recording itself. How can a full orchestra possibly be mic'd properly? I'll bring a few classical pieces with me next time to check it out.
Bratislav said:
I don't go to the symphony, so the concerts I've attended we're amplified. Admittedly the sound quality is often poor, whether it's due to the acoustics of the venue, bad seating location, or an inadequate setup, however, to sound better than "any I've attended" is saying something. That's also just from a sound quality standpoint, excluding the energy and visual impact of a live event. I realize it's an apples and oranges comparison, but I've never heard recorded music played back with such a "live" sound indoors or out either....more precise imaging and accuracy in detail, yes, but not the sense of space to add realism that I've found only open alignments to deliver. A significant advantage is due to a pair of point sources playing a well recorded stereo signal, however, I've heard boxed speakers (probably with better components using better equipment) used outdoors numerous times and none have come close in realism. Get a pair of decent OB's outdoors with some kind of structure behind them to put the rear wave in play, then you'll understand what I mean.
For orchestral music, you'll need speakers tailored to that because classical music places specific demands on a speaker that are difficult to reproduce. Even though it's not my cup of tea, I always throw a couple of classical pieces into the mix when I'm testing my latest speaker creation to help identify its weaknesses. Also, the fault of recorded classical music never sounding even close to live most likely lies in the recording itself. How can a full orchestra possibly be mic'd properly? I'll bring a few classical pieces with me next time to check it out.
Re: Re: Re: Re: Re: Freq and imaging
Oh, that explains it ... Sorry for misunderstanding.
But I'm sure unamplified voice, trumpet, clarinet or piano can be heard quite often. Even for these any speaker I've heard falls well short (this is again not 100% fair as we are talking the whole recording/reproduction chain, but let's asume speakers are the most guilty party in that chain). I have never been fooled by reproduced sound of a piano, for example. Some do get closer, but ... no even cigarette butt
This is probably true. How can orchestra and the venue be miked properly ? How does one record nuances and reflections inside a concert hall or a large church for example ? I'm not recording engineer, but I imagine these are really unsolvable problems with today's technology.
I'm afraid we are on the very, very long journey as far as domestic reproduction of classical music is concerned.
johninCR said:
Oh, that explains it ... Sorry for misunderstanding.
But I'm sure unamplified voice, trumpet, clarinet or piano can be heard quite often. Even for these any speaker I've heard falls well short (this is again not 100% fair as we are talking the whole recording/reproduction chain, but let's asume speakers are the most guilty party in that chain). I have never been fooled by reproduced sound of a piano, for example. Some do get closer, but ... no even cigarette butt
This is probably true. How can orchestra and the venue be miked properly ? How does one record nuances and reflections inside a concert hall or a large church for example ? I'm not recording engineer, but I imagine these are really unsolvable problems with today's technology.
I'm afraid we are on the very, very long journey as far as domestic reproduction of classical music is concerned.
Lynn Olson said:
The Toole School has been very vocal popularizing their philosophy, of very even response over wide dispersion angles, with the top end tailing off as you move off axis. Idea is flat long term integrated power in room with a gentle low pass shelf characteristic.
How would you describe the BBC and old KEF school?
The BBC/old-KEF constants I've seen over the years are a softer top end, and crossovers that aren't designed to be in phase at xover. Many years ago, the various KEF papers and the BBC designs seemed to sport mainly odd order acoustic xovers with their well known pros and cons: no reverse null here. These "uncorrelated" xovers certainly have a place in the trick bag and I've been getting superior results with them in some applications, especially ones where drivers are extremely close together.
Re: Re: Re: Directionality
In reverse order - T60 in small rooms is a meaningless concept. It is even loosing its following for larger rooms.
I disagree on the floor reflection, and any reflection in the median plane cannot affect imaging as its effect is equal in the two ears. But these reflections all affect coloration - ceiling or floor - the ears don't know the difference.
I also disagree about the front walls as no early reflections add anything of value to a playback system - no matter where they come from. I believe that the area behind the speakers should be fairly dead - the only place where sound absorption is advised.
DDF said:
In reverse order - T60 in small rooms is a meaningless concept. It is even loosing its following for larger rooms.
I disagree on the floor reflection, and any reflection in the median plane cannot affect imaging as its effect is equal in the two ears. But these reflections all affect coloration - ceiling or floor - the ears don't know the difference.
I also disagree about the front walls as no early reflections add anything of value to a playback system - no matter where they come from. I believe that the area behind the speakers should be fairly dead - the only place where sound absorption is advised.
Re: Re: Re: Re: Directionality
T60: perhaps this is picking nits now, but you are correct that I should use ultra precise language. RT60 was meant to be short hand for reverberation. Insert more precise language at will: for example intelligibility indices, C30, etc.
The floor reflection one surprised me. However, read it for yourself before you form conclusions:
"Effect of early reflections from upside on auditory envelopment", Journal of the Acoustical Society of Japan, 16, 2 1995
This paper shows vertical images are spread if a vertical reflection is delayed by at least 10 ms, and the reflected power is at least 1/2 the incident.
We'll have to respectfully agree to disagree on the front wall.
gedlee said:
T60: perhaps this is picking nits now, but you are correct that I should use ultra precise language. RT60 was meant to be short hand for reverberation. Insert more precise language at will: for example intelligibility indices, C30, etc.
The floor reflection one surprised me. However, read it for yourself before you form conclusions:
"Effect of early reflections from upside on auditory envelopment", Journal of the Acoustical Society of Japan, 16, 2 1995
This paper shows vertical images are spread if a vertical reflection is delayed by at least 10 ms, and the reflected power is at least 1/2 the incident.
We'll have to respectfully agree to disagree on the front wall.