Evidence-based Speaker Designs

[digitalfrost]

Excellent design, but I'm more impressed by the ER18DXT (imo the best 2-way DXT kit design there is), which mates a larger driver (and its attendant advantages) to the DXT but has equally seamless directivity.

Interesting. I happen to own this speaker. I replaced it with KEF LS50, but now I'm keen to get ER18DXT out again and do another comparison.

I haven't used DRC back then, now I do.

 

[JJB70]

The Apple homepod is a superb piece of audio design, and appears to be very competitively priced against the other high quality wireless speakers from Sonos, Yamaha, Bluesound, Bose etc. I have friends who abandoned traditional hifi to jump into the Sonos or Yamaha Musiccast wireless systems and from what I can see they have no regrets or wish to return to the hifi world. I agree that there are limits to these multi-room systems and wireless speakers, and think that you can get some very good active monitors or passive speakers and a decent amp for a similar price to the better wireless speakers which will probably be ultimately more satisfying but I fully understand why the market made a transition to such speakers

 

[Ilkless]

Interesting. I happen to own this speaker. I replaced it with KEF LS50, but now I'm keen to get ER18DXT out again and do another comparison.

I haven't used DRC back then, now I do.

The documentation shows very well-designed on and off-axis that effectively leverages the DXT waveguide.

From the pdf of Mark K's documentation, note that: "The 67.5 axis is in white. You can see this, with difficulty. Also, the last curve in green is off axis 90 degrees - it doesn't show on the legend."

Averaged out to give a crude horizontal power response, you get:

A seamless crossover. The SEAS Idunn is crossed too high and too steeply to have such a good off-axis. It peaks (albeit lesser than many flat-baffle speakers) at around 2kHz. The guy I commissioned my Idunn build from changed the baffle shape (while keeping same C2C). My VituixCAD baffle diffraction models tell me this yields a 2kHz dip vs the original baffle that just about compensates, so the peak on mine might not be as prominent as in the stock design. I certainly couldn't hear any harshness attributable to it, but I'd like to have it measured sometime to confirm.

Here's vertical off-axis:

Mark argues this dip is much less steep than a LR4 design, especially one with a higher XO. However, he also concedes:

I would love a design using a modern 6.5-inch midwoofer (the ER18 is over 10 years old) that has a crossover designed along the same principles as the ER18DXT's - low and shallow, but still well within the DXT's ability, yielding good on and off-axis without compromising distortion.

 

[Juhazi]

Heissman has some models with DXT, this is with LR4 https://heissmann-acoustics.de/dxt-mon-rly/ and perhaps different voicing, here horizontals

 

[Ilkless]

Heissman has some models with DXT, this is with LR4 https://heissmann-acoustics.de/dxt-mon-rly/ and perhaps different voicing, here horizontals

I'm aware of the DXT-MON, but it uses a 5.5-inch (though newer and more advanced) midwoofer and is significantly smaller. Personally, I'd take the speaker with the larger midwoofer and pretty much just-as-good polars for the headroom and distortion. This reflects my own listening distance, room size, SPL and bass extension needs though, for which a 6.5-inch 2-way suffices with headroom to spare.


edit: realised this DXT-MON-RLY used the ER15. I was referring to the Wavecor DXT-MON. My comments still apply.

 

[Juhazi]

Oh yes, all models have small woofers. I have the MarkK speakers myself. LR2 xo sounds nice, previously I had Zaph L18 with LR4 xo. Grimm LS-1 has dxt and bigger woofer, copy that design with dsp and try different xo topologies?

Diy recipe with U16 woofer and DXT with LR2 http://www.audioexcite.com/?page_id=2386

 

[Mashcky]

Interesting. I happen to own this speaker. I replaced it with KEF LS50, but now I'm keen to get ER18DXT out again and do another comparison.

I’d love to hear your thoughts on the two speakers if you get the chance. I’m about to build some er18dxts myself.

 

[Ilkless]

Oh yes, all models have small woofers. I have the MarkK speakers myself. LR2 xo sounds nice, previously I had Zaph L18 with LR4 xo. Grimm LS-1 has dxt and bigger woofer, copy that design with dsp and try different xo topologies?

Diy recipe with U16 woofer and DXT with LR2 http://www.audioexcite.com/?page_id=2386

If I were to go DSP, I'd go with a Horbach-Keele quasi-coaxial. I think the DXT is an excellent candidate for a Horbach-Keele speaker. The expanded top-octave dispersion and dispersion control lower down makes a flat, moderate DI above baffle-step an attainable target. Plus the phased woofer array in that DIYAudio build lowers the directivity control to far below what the baffle step for that baffle geometry should be. I find it a very technically-proficient design.

 

[JohnBooty]

[Large pic of Apple HomePod removed]

Reminds me of the Danley Synergy Horn.

Kind of, but functionally I don't think they have anything in common if I'm understanding them correctly.

Synergy Horn = multiple tweeters firing inward into a single horn

HomePod = tweeters (balanced mode radiators, I believe?) firing outward, not sharing a horn/waveguide/anything. (The "horn" at the bottom is the woofer's port and is not coupled to the tweeters in any way...)

 

[Cosmik]

People getting very excited about little speakers because of their graphs (a.k.a. "evidence"). But we know that audio is partly about power don't we? These little toys are going to run out of steam very quickly, and then they're going to rely on DSP to limit the bass and so on. Maybe they'll do that missing fundamental trick. The fact that the woofer is ported at a relatively high frequency means it gives its presence away.

These are the modern version of the 1970s music centre in comparison to a serious system. A serious system has guts. It doesn't do any compression or limiting. Its bass extends all the way down. It may even shock the listener with its power, and worry them over just how loud it's going to get.

I think people who go over to the Bose Wave radio and equivalents ("listen to that bass from such a little box!") are forgetting what it is to sit in front of some seriously meaty speakers and listen to a symphony at realistic levels, or The Who. Maybe they're fetishising certain regions of certain types of graph and not paying enough attention to the basics.

 

[svart-hvitt]

People getting very excited about little speakers because of their graphs (a.k.a. "evidence"). But we know that audio is partly about power don't we? These little toys are going to run out of steam very quickly, and then they're going to rely on DSP to limit the bass and so on. Maybe they'll do that missing fundamental trick. The fact that the woofer is ported at a relatively high frequency means it gives its presence away.

These are the modern version of the 1970s music centre in comparison to a serious system. A serious system has guts. It doesn't do any compression or limiting. Its bass extends all the way down. It may even shock the listener with its power, and worry them over just how loud it's going to get.

I think people who go over to the Bose Wave radio and equivalents ("listen to that bass from such a little box!") are forgetting what it is to sit in front of some seriously meaty speakers and listen to a symphony at realistic levels, or The Who. Maybe they're fetishising certain regions of certain types of graph and not paying enough attention to the basics.

Hah! That sounds like evidence based subjectivism to me!

(Not sure if you like that label, though).

 

[Juhazi]

Synergy horns have several mids (and woofers) firing in the horn that has one compression driver. Several iterations of course , but only with a single tweeter!
https://patents.google.com/patent/US6411718B1/en

Yes TV sound bars, Apple HomePd etc. are seriously limited in spl/distortion in low register. They rely on additional subwoofers. But there is serious knowhow and skill behind them. About 5 years ago I was at AES conference here in Finland and talked briefly with a guy from Bose... and Dr. Linkwitz and Dr. Klippel!

 

[JJB70]

People getting very excited about little speakers because of their graphs (a.k.a. "evidence"). But we know that audio is partly about power don't we? These little toys are going to run out of steam very quickly, and then they're going to rely on DSP to limit the bass and so on. Maybe they'll do that missing fundamental trick. The fact that the woofer is ported at a relatively high frequency means it gives its presence away.

These are the modern version of the 1970s music centre in comparison to a serious system. A serious system has guts. It doesn't do any compression or limiting. Its bass extends all the way down. It may even shock the listener with its power, and worry them over just how loud it's going to get.

I think people who go over to the Bose Wave radio and equivalents ("listen to that bass from such a little box!") are forgetting what it is to sit in front of some seriously meaty speakers and listen to a symphony at realistic levels, or The Who. Maybe they're fetishising certain regions of certain types of graph and not paying enough attention to the basics.

I think there is some truth in that, but it is also undeniable that there is a lot of clever engineering and science being pumped into wireless speakers and sound bars. There is a wide range of such speakers, some are clearly better than others, some of them are actually pretty refined and nicely balanced and can go up to a more than adequate volume without breaking down. Will they go to earthquake inducing volume? No (well, not without sounding pretty poor) but many (I suspect most) people neither want nor need such volume and for most people models like the Sonos 5 can fill a normal room with perfectly acceptable sound at sufficient volume. In the early days I did think most of these speakers were one trick ponies and that after the initial reaction of "wow, how can something so small make so much sound?!" wore off they were not very impressive. Now I find the good ones at least are very acceptable and I can enjoy music using them and thoroughly enjoy it.

 

[svart-hvitt]

People getting very excited about little speakers because of their graphs (a.k.a. "evidence"). But we know that audio is partly about power don't we? These little toys are going to run out of steam very quickly, and then they're going to rely on DSP to limit the bass and so on. Maybe they'll do that missing fundamental trick. The fact that the woofer is ported at a relatively high frequency means it gives its presence away.

These are the modern version of the 1970s music centre in comparison to a serious system. A serious system has guts. It doesn't do any compression or limiting. Its bass extends all the way down. It may even shock the listener with its power, and worry them over just how loud it's going to get.

I think people who go over to the Bose Wave radio and equivalents ("listen to that bass from such a little box!") are forgetting what it is to sit in front of some seriously meaty speakers and listen to a symphony at realistic levels, or The Who. Maybe they're fetishising certain regions of certain types of graph and not paying enough attention to the basics.

I think there may be some truth to the claim that New Small can sound big at a High Price, as much as I speculate that Old Big at Low Price can sound even better.

So if you have the space and the taste, why not? I guess this is what drives many DIY people. They can make big bulky boxes with great Man Acceptance Factor that sound good, often better than New Small High Price.

 

[Wombat]

People getting very excited about little speakers because of their graphs (a.k.a. "evidence"). But we know that audio is partly about power don't we? These little toys are going to run out of steam very quickly, and then they're going to rely on DSP to limit the bass and so on. Maybe they'll do that missing fundamental trick. The fact that the woofer is ported at a relatively high frequency means it gives its presence away.

These are the modern version of the 1970s music centre in comparison to a serious system. A serious system has guts. It doesn't do any compression or limiting. Its bass extends all the way down. It may even shock the listener with its power, and worry them over just how loud it's going to get.

I think people who go over to the Bose Wave radio and equivalents ("listen to that bass from such a little box!") are forgetting what it is to sit in front of some seriously meaty speakers and listen to a symphony at realistic levels, or The Who. Maybe they're fetishising certain regions of certain types of graph and not paying enough attention to the basics.

The graphs show levels of ~80dB or less. This OK for casual use but how do they go if pushed or for transients above that level? General question for the forum.

 

[Juhazi]

Most of them utilize power limiting of some kind, easy with dsp.

 

[restorer-john]

This OK for casual use but how do they go if pushed or for transients above that level?

One alcohol-fuelled party and they'll have a one way ticket to the local landfill.

 

[Thomas_A]

It seems that quite many want a perfectly linear response. This is normally perfectly fine and should be done if you work with multichannel and especially with a centre dialog speaker (also for upmixing stereo). As I wrote elsewhere on the forum, there are errors introduced in the timbral spectrum for the central phantom image in stereo due to physics, as has been published by Shirley, Toole and others.. I doubt that a recording engineers apply standardised corrections according to those errors. In my ears I voice my DIY stereo speakers with more energy in the 1-2 kHz region compared to the 2-4 kHz o. The speaker has some flaws but this is a DIY speaker with a €20 cone tweeter and €40 5 inch midwoofer. It sounds fine though.

 

[JohnBooty]

People getting very excited about little speakers because of their graphs (a.k.a. "evidence"). But we know that audio is partly about power don't we? These little toys are going to run out of steam very quickly, and then they're going to rely on DSP to limit the bass and so on. Maybe they'll do that missing fundamental trick.

[...]

I think people who go over to the Bose Wave radio and equivalents ("listen to that bass from such a little box!") are forgetting what it is to sit in front of some seriously meaty speakers and listen to a symphony at realistic levels, or The Who. Maybe they're fetishising certain regions of certain types of graph and not paying enough attention to the basics.

Yeah. The grins that come from properly massive speakers absolutely cannot be replicated by smaller units. I own some pretty accurate smaller speakers, but nothing in my collection approaches the smiles that these guys produce: http://www.bicamerica.com/showpage.php?brand=3&type=15&spkrID=120

Sometimes available for $400/pr, shipped to store, and look at these stats. Based on my listening I've no reason to doubt them. The horn tweeter is not the smoothest, but it's not awful. It's very low-tech speaker, and a lot of folks will scoff, but the fact is that you don't need KEF LS50 levels of engineering when you have adequate displacement in the first darn place.

• Frequency Response: 24Hz-20kHz +/- 3dB
• Sensitivity: 95dB @ 2.83V/1 watt, 1 meter
• Recommended Power: 430 Watts Peak, 225 Watts RMS

The appeal, for me, is not the massive potential max output. A lot of people erroneously think that's the sole appeal of big speakers and powerful amps.

I typically listen at 80dB average or less because I value my hearing. The appeal is the absolutely huge (and effortless) dynamic range these speakers are capable of. In my judgement, this sort of capability is what's required to truly capture the emotion of live music.

I have a number of smaller bookshelves of typical ~85dB efficiency, and they're certainly fun (especially crossed over to capable powered subwoofers) but they are still a joke compared to the BICs. I have some MTM bookshelves of medium-high 91dB efficiency, and they come a little closer in terms of effortless dynamic range.

And that dynamic range is the area where small speakers, especially portable bluetooth speakers, just totally fall down. They're fine if you have no alternative (gotta do what you gotta do) but they can't really stir emotions.

Please understand, though, that I'm not arguing against an evidence-based approach to audio! A reasonable FR is indeed a prerequisite for good listening. And dynamic range is not some ineffable magic notion either. We can and do measure that.

(attached photo is not my living room, but it shows that these things look somewhat respectable with the grilles on....)

 

[FrantzM]

I think we are getting carried away. No one is trying to tell you, the audiophiles that Soundbar , Sonos or Bose Acoustimass, etc sound better than your favorite brand or model of speakers. We are simply saying that some mundane speakers and components designs are based on serious Scientific principles. They are not for most of us IOW we are not their market target but have to admit that sometimes they fill our space and lives with what many of us are after: Music.