What about the low-frequency directivity? SL has theorised that maintaining uniform response is important across the spectrum?
The proposal seems to be direct only at upper frequencies, while being omni in the lower region, which is the traditional problem with typical box speakers.
SL seems to be gearing to proving or disproving his theory:
ORION/PLUTO Users Group • View topic - LinkwitzLab website up for 10 years!
Thanks.
The proposal seems to be direct only at upper frequencies, while being omni in the lower region, which is the traditional problem with typical box speakers.
SL seems to be gearing to proving or disproving his theory:
ORION/PLUTO Users Group • View topic - LinkwitzLab website up for 10 years!
Thanks.
Actually, this technique has been exploited for years in car stereo installations to provide a reasonable sound stage for driver/passenger positions. Imagine speakers installed in front doors/kick panels, pointed to ear height of the opposite seat.
The driver will hear on axis sound from the passenger side speaker, and off axis sound from the driver side speaker, and vice-versa, causing a more or less equal left/right levels at each position, due to the fact that the local speaker's off axis level will be less than the on axis level from the opposite speaker. If done right, the sound stage can be quite convincing.
Interestingly enough, under dash waveguides have even been available since the early/mid 90's, though maybe not as scientifically implemented as with Earls' techniques, I'm sure. Here's a link to a tutorial from the company that pioneered these waveguides for car stereo:
Tutorials at SpeakerWorks/USD Audio
Lukas
Correct you cannot compensate for the delay, azimuth is azimuth and is directly related to the ear spacing and source angle. Not to mention moving off-axis and moving closer to room boundaries also compounds this problem. Stereo attributes are strongest in the time sensitive region while the upper region dominated by levels is soley a monophonic area in our hearing. Therefore upper frequencies do not add to the stereo effect and can actually be more confusing and degrading to stereo. This artifact was well known even from the early days and by the inventor.
This "white paper" seems more like a marketing brochure. How many times is Summa mentioned? No official references either. I do applaud the work in supporting a position for good power response.
This is not really anything ground breaking either and if you must discuss stereo any further with directivity there are better concepts. If you want to discuss a system with level controlling directivity when moving about the room:
http://www.extra.research.philips.com/hera/people/aarts/RMA_papers/aar01p.pdf
Last edited:
Hi Lukas
Yes, this is mostly correct. The amplitude versus time tradeoff is not perfect, but its a whole lot better than most common speaker setups for Hi-end speakers. Read the reviews on my web site, the technique does work and it is significant.
Durwood
Regarding Ron Aarts paper, its basically the same thing that my speakers do. The difference is that mine are practical cost effective products and the Aarts paper is just a theoretical discussion. Put into practice with real transducers, etc. and his technique won't work out all that well, or at least it must not have since we don't see any products of this type.
Yes, the paper is more of a marketing brochure than an engineering report. I wrote it because of the number of people who do not understand directivity and what it means. I also wanted to show some real responses since its extremely rare to see this type of data for commercial products. Given the results it is no wonder.
Someone mentioned SL and the Orions. I'd love to see data on those since I highly doubt that they are all that great as regards polar response. As far as LF directivity is concerned, the paper discusses why this is not important at LF. Since Dave Clark is an old friend and lives very close by, he told me of his study (I saw the Orions setup at his home on New Years Eve.) I don't think that it is going to come out quite like Mr Linkwitz is hoping. It won't be blind so it won't convince any sceptics no matter how it comes out.
The effect is real and is called "trading" or "equivalence" in psychoacoustics. The displacement of a phantom source caused by an interchannel time difference can be fully compensated by an interchannel level difference and vice versa (see Blauert).
Problem is that this effect varies from person to person and is signal dependent. While the configuration suggested by Earl helps avoiding the complete breakdown of the stereophonic illusion when the listener leaves the sweet spot, it is not suitable to replace a center channel loudspeaker.
Best, Markus
Psychoacoustics knows a couple of parameters that describe spaciousness. The problem is that there are no studies that ever qualified those parameters for stereo reproduction in acoustically small rooms.
Within that context not only the loudspeaker is of interest but also the format (multichannel vs. stereo) and last but certainly not least the room. Linkwitz seems to forget about that just like Earl seems to forget that loudspeaker directivity down to 200 Hz may play a major role.
Best, Markus
"Major role", I don't think so, "significant", maybe, "insignificant", maybe as well. I think for a "good design" one has to be careful of the trade-offs. This extension to lower frequencies is extremely expensive in terms of cost and size, so if its just a noticable improvement, but at an extreme cost, that is not going to be very appealling. Fine for a very rich man. in a very big room who wants to pay the excessive expense involved in making a "one-off" - when you find him let me know and send him my way! I need more clients like that.
Markus
Or, did you mean a "paper" design? You know those Uber-designs that get written about around all the time, but somehow never seem to get made, so then no one can criticize them either way - its just "hypothetical" and "you never know". Yea let's ramble on for days about the kinds of things "that could be done"! One thing I learned being a productive engineer was that as soon as a project starts to look like that, its time to stop because your just wasting your time - "Blowing smoke", as they say.
Or, did you mean a "paper" design? You know those Uber-designs that get written about around all the time, but somehow never seem to get made, so then no one can criticize them either way - its just "hypothetical" and "you never know". Yea let's ramble on for days about the kinds of things "that could be done"! One thing I learned being a productive engineer was that as soon as a project starts to look like that, its time to stop because your just wasting your time - "Blowing smoke", as they say.
Thanks for the link! Finally found the title of the Bauer paper I was searching for. Interesting to see that Bauer used dipoles and toe-in 50 years ago
Best, Markus
Markus
Quite the contrary. I did a paper design of a system using dipoles and looked into it. In the end the added directivity was not as great as you might think when one is using very large diameter drivers like I do. Then there is the required added electronics. By my calcs, the system price jumped by almost 50%. Others will probably quote all kinds of different numbers, but that was my conclusion.
So let's build a waveguide
I don't care about the "how". All I care about is performance.
Just send the cash!!
I think I inspired a couple of your projects years ago, and I've really been enjoying your posts on mp3car. One thing that you haven't discussed is that your projects work very nicely in a small environment. It doesn't necessarily have to go in the car.
For instance, my home office is around five hundred square feet, and it's really a bit small for my Summas. There's a gap of nearly two feet from the two diaphragms in the speaker, and due to this huge gap, you need to sit REALLY far away to get the full benefit of the speaker's directivity. If you sit too close to the speaker, the performance of the waveguide degrades, particularly from 900 to 1800hz, which is right smack in the middle of the midrange. Subjectively I've noticed that the sound is smoother and the image is more faithful to the recording when listened at the proper distance. The biggest improvement is a big step up in articulation.
As soon as you back up, everything is roses, and the speaker sounds fantastic. But they really need a big room. At one point I'd devoted an entire floor of my house, fully 900sf, and they really shined there.
I have a friend-of-a-friend who built a BARN to house his audio equipment! And seriously, spending $5,000 to build a room around your loudspeakers makes a heck of a lot more sense than spending $5,000 on amplifiers and DACs. Particularly if you can DIY.
Now one of the cool things about using arrays to control directivity is that you can sit much closer to the speaker than if you used a large waveguide.
Anyways, not saying one solution is better than the other, just saying that a lot of people are clamoring for larger and larger waveguides without considering that it requires a larger room.
gedlee said:
I don't agree it's the same. Your dispersion pattern is very simple and uniform and does not alter the directivity shape. Using DSP the dispersion pattern is specifically shaped between different frequency bands. You can find this type of practice in the pro audio world with array steering. While your method might be simpler, you can do the other method for about the same price (or cheaper) and maybe get better results.
Perhaps you haven't noticed the market increase in "soundbar speakers" from yamaha, philips, marrantz, etc etc? While these may not be exactly like what is proposed, it is still being investigated/researched in many universities too. Stereo is and will die a slow death. It is the reason multichannel has overtaken the market and preferred over 2 channel. The problem with the current soundbar speakers is the same as with many vanilla commerical offerings...the speakers are too small to deliver the dynamics neccesary for realism.
Thanks for the link! Finally found the title of the Bauer paper I was searching for. Interesting to see that Bauer used dipoles and toe-in 50 years ago
Best, Markus
I also have the more indepth paper that seems to have been lost a while ago when they were doing maintenance on the website. If you would like I can send it to you via email.
Last edited:
Hi Patrick
You stated that a couple of times in the past and I'm curious what the problem here might be. Are there any measurements that would show where the Summa's far field begins? Words like "smoother", "faithful" and "articulation" don't mean much to me when talking about loudspeaker performance.
In small rooms the big Summas should perform even better because of their higher directivity down to lower frequencies.
Best, Markus
Yes I tried the bessel array but went back to a regular array. Bessel could work in the midrange if the crossover was out of the bass region but I was put off by the loss of lower frequencies due to the one inverted driver, otherwise it worked as described. That was a few years ago already.
I have a few more tricks up my sleeve but you can only do so much in a tiny space.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.