IMO a growing concern is not even which is superior, but the motivations to pursue the difference between a good and a great speaker. I don't see it coming through as before. These days it is easier to make a speaker better than some have ever heard, with just a small fraction of the knowledge needed to build a great speaker system. Naturally people want to talk about it and they should, but some audiophile misconceptions and half-truths are getting into the mix along the shallower learning curves. The beginning of this century has been good so far for dispelling myths in general.
Ok, same speaker with passive xover,
with a regular dsp albeit one that has a small linear phase xover capability ,
and with dsp using FIR.
Speaker was a JTR 3TX.
Noesis 3TX
Uses a BMS 4593 coax on a 18Sound 1464 horn, with two 10" mids/bass below. So a 3-way TMM.
All made outdoors off an open deck at about 4m.
Here's with the factory passive xover.
No smoothing mag or phase.
Here's with a Linea Research dsp using IIR EQ's and what Linea calls LIR xovers, which are essentially linear phase LR's 24dB/oct.
No smoothing.
Here's with FIR dsp.
Again, no smoothing.
with a regular dsp albeit one that has a small linear phase xover capability ,
and with dsp using FIR.
Speaker was a JTR 3TX.
Noesis 3TX
Uses a BMS 4593 coax on a 18Sound 1464 horn, with two 10" mids/bass below. So a 3-way TMM.
All made outdoors off an open deck at about 4m.
Here's with the factory passive xover.
No smoothing mag or phase.
Here's with a Linea Research dsp using IIR EQ's and what Linea calls LIR xovers, which are essentially linear phase LR's 24dB/oct.
No smoothing.
Here's with FIR dsp.
Again, no smoothing.
Well, you are cheating a bit 
. The last one starts at 63 Hz the others are from below 30 Hz. The last one is +/- 30 dB, the others +/- 15 Nevertheless, the difference is quite clear. One would wonder what could have been done better in the passive crossover. I'm pretty sure there is room for improvement there as well.
. The last one starts at 63 Hz the others are from below 30 Hz. The last one is +/- 30 dB, the others +/- 15 Nevertheless, the difference is quite clear. One would wonder what could have been done better in the passive crossover. I'm pretty sure there is room for improvement there as well.Well, you are cheating a bit
Yeah, sorry about the scale difference...they were taken at different times for different purposes, and i had saved screen grabs i can find, but measurements i cannot (to re-scale)
Below 60Hz is immaterial on all.
In use, any of the three xover types would be passing to subs at about 100Hz, with at least 4th order.
I'm not qualified to assess the passives (and never hope to be
)But they seemed appearance wise to have had a lot of effort put into them.
Big suckers...
Attachments
Hey Scott, if you're still monitoring this thread, and as you are into tape, you might enjoy this:
https://worldradiohistory.com/Archive-All-Audio/Archive-Recording-Engineer/80s/Recording-1981-12.pdf
A nice gent recently posted it on one of the prosound forums.
The ads alone are worth looking at...
A real turn back in time
Thank-you ! (yes I'm still reading this thread)
1981 was the year I got married.
1974 was the year I built my first pair of speakers.
1979 was the year I discovered active crossovers, and their HUGE benefit.
Exactly
The mission of the test would be to see (hear) if the digital crossover sounds at least as good as the analogue, IN JUST DOING THE CROSSOVER FUNCTION.
If the answer is 'yes" then by all means proceed to take full advantage. If the answer is "no" then a decision would have to be made by the end user on how to proceed further.
I got that, but all I'm saying it won't be easy to do an honest compare.
There simply are too many choices/options to know or realise you've picked the right processor. (no, I don't believe all would be equally qualified to suit the test)
As Mark100 said earlier in this thread. To make it a proper test you should vary only one variable. With care you could mimic the passive results with active means. But how to go about it. What kind of DSP would be qualified to do the job. Even that choice would get a whole lot of different answers, even from those who run digital active.
This isn't an easy to solve comparison without running into an ever lasting debate.
There simply are too many choices/options to know or realise you've picked the right processor. (no, I don't believe all would be equally qualified to suit the test)
As Mark100 said earlier in this thread. To make it a proper test you should vary only one variable. With care you could mimic the passive results with active means. But how to go about it. What kind of DSP would be qualified to do the job. Even that choice would get a whole lot of different answers, even from those who run digital active.
This isn't an easy to solve comparison without running into an ever lasting debate.
You'll only have to prove this once in a single instance to make the case. Let's do it this way: take take the same DAC and amps as the original active analog system, only double up the DAC so every driver get's one. Next, use whatever DSP that gives you the most control. A linux based software system will probably be your best bet there. It's much more complex, but can be fully customised and can use the full potential of clock cycles and memory available to modern computers. Next make sure we feed both the same digital signal in some way.. file, SPDIF, streaming whatever..
When it comes to the filtering part: just measure the analog filters one by one, and create the corresponding filter in the DSP. This can even be done directly by converting the impulse response of the analog filter to the FIR coefficients. Obviously a high quality ADC should be used here. Another way would be to recreate the filters in software by just matching up slopes and phase as close as possible.
Then hook all of it up, and do a double blind test.
Thx for timeline. Here's mine, no doubt too verbose...
1971 - married
1967 - built first speaker, a plywood boombox with woofer, teeter, ratshack 8w battery driven amp, transistor radio as source. Screen door handle and secret padded compartment for bottle of booze and pot.
1976 - purchased first electrostats
till 2000 - remained enamored with electrostats and big planars & ribbons...classic audiophool shunning EQ, anything less than "pure signal path", always wanting class A, etc ..iow, an idiot... lol
2001 - ventured into live sound /DJ speakers and amps for parties. Learned the benefits of active xovers, EQs, and delays. Really liked the dynamics and bass, but missed the clarity, imaging, etc of hi-fi
2002 - discovered Meyersound, one of the makers of proaudio equipment built to meet or exceed hi-fi. Bought some full range 4-way mts4a's. Self-powerd, active analog, quad-amped .
Wow, true hi-fi that rocks. Learned the absolute superiority of outdoor listening with quality gear. Clarity, imaging, dynamics....i can have it all.
2003 or so - needed more bass when Meyers setup outdoors, built 4 of Tom Danleys Labhorns.
till 2015 - occasional visits from cops when cranking big Meyers (280lbs a piece) stacked on Labhorns
2015 - saw and built the Peter Morris DIY PM90&60...a low weight top that could run with the Labhorns, and save the hassle of setting up the heavy Meyers.
2016 - discovered FIR and linear phase tuning working on the PM90 & 60. But also continued working with traditional IIR xovers and EQ's due to providing the PM90s, labs, processing and amps for live gigs. (to avoid FIR latency) Made many comparisons between what I could achieve FIR vs IIR. No doubt in my mind FIR was better.
2020 - been on a DIY speaker craze since 2015...built 8 subs of three different types, 10 CD with mids trap boxes designed to array tight packed with interchangeable horns, a 15" prosound coax w plateamp, floor-to-ceiling line arrays with steering/shading, CBT line arrays, and lately 3 different size synergies.
All these builds have been very easy to tune with great results, really just due to the simplicity of tuning with dsp, especially with linear phase.
Imo, whether we can hear phase or not; whether passive, active analog, dsp IIR, or DSP FIR is superior...are all much less important than being able to consistently achieve excellent tuning....whatever design.
A varying freq or time response, even a little, dominates differences vs what method did you use, imho.
hmmm... interesting discussion but:
I find that things here are a bit confusing for me. As a hobby speaker builder I take advantage of DSP to achieve an objective, not to replicate an analogue filer or because it is my favourite technology.
Example: my latest SEAS 8" fullrange project had objectives to produce smooth on-axis slight tapered frequency reproduction. In addition, the bass had to be adjusted to suit my living room. To do this i need a lot of filters (yes fullranges have a ragged curve), and a Linkwitz Transform in the bass. DSP was therefore chosen based on these needs. I suppose it´s doable in the analogue domain, but why, it would complicate the project and make it costly?
Determine first your objectives, perform simulations and later on select your technology.
I find that things here are a bit confusing for me. As a hobby speaker builder I take advantage of DSP to achieve an objective, not to replicate an analogue filer or because it is my favourite technology.
Example: my latest SEAS 8" fullrange project had objectives to produce smooth on-axis slight tapered frequency reproduction. In addition, the bass had to be adjusted to suit my living room. To do this i need a lot of filters (yes fullranges have a ragged curve), and a Linkwitz Transform in the bass. DSP was therefore chosen based on these needs. I suppose it´s doable in the analogue domain, but why, it would complicate the project and make it costly?
Determine first your objectives, perform simulations and later on select your technology.
Must be a reason they chose Analogue-Active.
Hayward, CA, June 2020 – A summation of our no-holds-barred assault on the limits of dynamic loudspeaker design, the new Magico M9 establishes new benchmarks in musicality, transparency and fidelity. This four-way, six-driver floor standing system features the world’s first loudspeaker enclosure to combine inner and outer skins of carbon fiber with a revolutionary aluminum honeycomb core. Included with the M9 is a state-of-the-art analog outboard active crossover, the MXO. Designed in-house, this carefully crafted unit handles bass/midbass frequency separation. In addition, the M9 benefits from our latest generation of Nano-Tec speaker cones, featuring Aluminum honeycomb cores. The result is a revelation, a loudspeaker that can present intense crescendos with unconstrained power, yet reproduce the most delicate musical passages with transparency and stunning microdynamic detail.
The M9 - A Revelatory New Flagship Loudspeaker — Magico Loudspeakers
Hayward, CA, June 2020 – A summation of our no-holds-barred assault on the limits of dynamic loudspeaker design, the new Magico M9 establishes new benchmarks in musicality, transparency and fidelity. This four-way, six-driver floor standing system features the world’s first loudspeaker enclosure to combine inner and outer skins of carbon fiber with a revolutionary aluminum honeycomb core. Included with the M9 is a state-of-the-art analog outboard active crossover, the MXO. Designed in-house, this carefully crafted unit handles bass/midbass frequency separation. In addition, the M9 benefits from our latest generation of Nano-Tec speaker cones, featuring Aluminum honeycomb cores. The result is a revelation, a loudspeaker that can present intense crescendos with unconstrained power, yet reproduce the most delicate musical passages with transparency and stunning microdynamic detail.
The M9 - A Revelatory New Flagship Loudspeaker — Magico Loudspeakers
No doubt there was a reason here, why they went the DSP route...
It's easy to put up a counter offer...
B&O Tech: BeoLab 90 – Behind the scenes – earfluff and eyecandy
And it seems to work quite well:
We heard B&O's Beolab 90 speakers and we're still in disbelief
Or by our forum member:
It's easy to put up a counter offer...
B&O Tech: BeoLab 90 – Behind the scenes – earfluff and eyecandy
And it seems to work quite well:
We heard B&O's Beolab 90 speakers and we're still in disbelief
Or by our forum member:
Here's a brief review:
I've been going to audio shows for nearly twenty years now. In general, I've found that 80% of the speakers there are about as good as what you'd get at Best Buy, maybe a bit worse. I've noticed that you see a lot of names come and go.
The names that you see year-in-and-year-out tend to sell products that are consistently good. I've never heard a bad demo from Kef, Dynaudio, or Vandersteen. I'm not saying that they're GREAT, but they're consistently good, and once in a while, they're great.
One trend that I've noticed, particularly in the last five years, is that the overall quality of loudspeakers is getting very good. For instance, at the 2005 CES I'd say that most of the speakers I heard were average to mediocre. At the 2016 CES, most speakers sounded good.
I think there's a downside to this consistency though; a LOT of speakers sounded very very similar. For instance, both Kef and Wharfedale were demoing a modestly prices coincident speaker, and if I had a blindfold on I wouldn't be able to tell them apart.
I think what's happening here is that the quality of measurement software, and general knowledge of psychoacoustics is getting so good, even average and inexpensive speakers are light years beyond what we had twenty years ago.
Pioneer's $100 speakers come to mind; back in the 90s something like this would come in a nicer cabinet and it would cost $1000.
In summary:
I listened to dozens of speakers at CES 2016, and many of them were quite good. We're in a bit of a 'golden age' when it comes to loudspeakers, there's a lot of good stuff that doesn't cost a lot.
But I didn't see a lot that pushed any boundaries.
An externally hosted image should be here but it was not working when we last tested it.
The Nola Brio Trio really grabbed my attention. It has a lot in common with the LX Mini. I built myself an 'homage' to the LX Mini, and the Brio Trio reminded me a lot of my project, but the Nola sounds better. The speakers produce a soundstage which stretches well beyond their boundaries, and due to the tiny cabinets, they really disappear.
After a few hours of listening to various systems, listening fatigue was really setting in. For instance, I gave the Genesis line arrays a second listen, and though I loved them earlier in the day, I found that they weren't quite as good later in the day. I chalked it up to fatigue.
With less than ninety minutes left in the day, I made the trek over to Bang & Olufsen. My main goal wasn't to hear the Beolab 90s; I actually wanted to hear the Beolab 5s, as I've invested a lot of effort into acoustic lenses based on the Beolab.
By the way, I want to apologize for the length of this post. I wanted to put my experience with the Beolab into perspective. I don't want people thinking that it was the only speaker that I listened to at CES, or that I've only listened to a handful of world-class speakers. I've listened to hundreds, and I even bought a pair based on demos at audio shows. (I bought my Gedlee Summas based on a demo that Earl did at the 2005 RMAF.)
Let's cut to the chase here. The Beolab 90 is an industry-changing speaker. In my life I can count on one hand the speakers that are able to make the room 'disappear.' The first time I experienced it was with Quad Electrostatics. The second time was with Danley SH50s. There are plenty of speakers which throw a nice soundstage, but those two speakers make the boundaries of the room disappear. The listening room in my home is tiny, barely 150', but close your eyes with an SH50 and you're transported to where the recording was made.
I need to stress that this is rather unique, as there ARE speakers which throw a huge stage, no matter what's on the recording. I've heard a lot of line arrays that do that, and you can create that effect electronically using crosstalk cancellation.
But the Quad and the SH50 were different, when the recording was small, the sound was small, and when the recording was great, it was like a window onto the recording venue.
The Beolab 90 does this.
Listening to the Summas back-to-back with the SH50s, I could tell that Geddes is correct when he says that HOMs are obnoxious. The SH50s image like crazy, but the treble isn't silky smooth like it is with the Summas. (And don't take my word for it, you can see it in the measurements of both.)
The Beolab didn't have that problem; there is no waveguide and no horn, and there are no HOMs.
Listening to the Quads, I knew I could never live with them; I'm a horn guy and the Quads lacked the dynamics I'm accustomed to.
The Beolab 90 is a BEAST; I'm surprised Bang & Olufsen wasn't kicked out of CES. I never even got close to the SPL limits of my Summas, but if the Beolab 90 had a limit, I couldn't tell what it was. B&O cranked it up for a couple of tracks, and you could FEEL the bass. Each cabinet basically has four high excursion subs in it, and thousands of watts.
The easiest way for me to describe the demo is this:
I have a pair of really nice headphones, Sennheiser HD380s. The Beolab 90 sounds like my headphones. Except the soundstage is out there in the room, instead of in my head. The sound is clean and dynamic. The soundstage is whatever you give it; give it a good recording and the soundstage is pinpoint and huge, give it a crummy recording and it's crummy.
Another thing that I really appreciated about the Bang & Olufsen demo was their transparency. For instance, the Dynaudio folks ran a demo using a track that appeared to have crosstalk cancellation on the recording. I've messed around a lot with crosstalk cancellation, so I know it when I hear it; it makes the stage insanely wide. The B&O folks weren't doing what most do, they weren't playing a series of "audiophile-approved" recordings carefully curated to wow the crowd. They were basically picking tracks at random from their music server. (80% of the tracks in the demo were tracks I owned, so I feel fairly confident that I wasn't snowed.)
So... Any questions?
If you couldn't tell yet, this is the best speaker I've ever heard, bar none. In fact, I feel a little silly lumping it in with other loudspeakers. It's like comparing a tube amp to a solid state amp. They both amplify the music, but they're completely different animals.
Attachments
Last edited:
Interesting
Interesting. Each one of us has pleaded our case, based on the choice of another.
Peace. Joy. Live long and Prosper.
Interesting. Each one of us has pleaded our case, based on the choice of another.
Peace. Joy. Live long and Prosper.
As typically spec and design oriented diy-ers (and that's what makes diy much more interesting than off the shelf stuff), the marketing and sales channels aspect seems a bit overlooked here. Manufacturers such as Magico, Wilson and B&W cannot afford commercially to come up with an all out active digital speaker such as B&O or Kii &co.
The M, W etc sales is through traditional hi end sellers that must sell expensive amps and waterhose cables at 100$ per foot. Integrated solutions would be the end the business model in their chain of distribution&sales.
It is only partially a tech thing thing: it is whom you want to sell to. A potential Magico or Wilson buyer would never be interested in a Kii, D&D (or even a Quad ESL 63 to name a very different animal) simply because is is not expensive enough to show off with in your home. It is essentially the Rolex/Ferrari thing.
The M, W etc sales is through traditional hi end sellers that must sell expensive amps and waterhose cables at 100$ per foot. Integrated solutions would be the end the business model in their chain of distribution&sales.
It is only partially a tech thing thing: it is whom you want to sell to. A potential Magico or Wilson buyer would never be interested in a Kii, D&D (or even a Quad ESL 63 to name a very different animal) simply because is is not expensive enough to show off with in your home. It is essentially the Rolex/Ferrari thing.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.