These ain't your daddy's speakers...

[DaveH]

Very cool, WIP.

 

 

[Kerry Marchman]

Very cool. Technology is an amazing thing. But, can they be bought for three fiddy?

[ARM OF HAMER]

                                           Yeah you can spend a fortune on Stereo Equipment just like you can on many things. [Like guitars and amplifiers] Hard to get a real feel for what these speaker do from a video but they sure look cool and obviously a lot of time and money were put into producing them. Be curious to know the price as there are speakers that are over a million dollars. But for plain old audiophileness I don't believe anyone has beat this guy, and sadly he passed not long after this system was completed and, in the end, the whole system was sold for a fraction of what it cost to build it. Meet Ken Fritz. 

 

[burningyen]

[JGale]

Here’s what I’m not understanding about how this speaker works:

My first impression is this is an open-baffle, 20way crossover, implemented in DSP, that uses the DSP to also massage out room and other effects on the signal reception by the listener. He places an omnidirectional mic at the seated position like a normal room correction software. More speakers would give more options over smaller frequency ranges= more precise control. Also means you could vary the crossover point among the entire stack and optimize for whatever variable you choose within those frequency ranges. That’s a standard crossover with RoomEQ albeit with 20 crossover points to determine.

Done every day. Sometimes twice a day if you’re a real audiophile. The only real Mania is the number of speakers he has chosen to use. 

He harps on the problem of overshoot of the speaker cone stop/start due to inertia of the speaker’s moving parts. Ok, what does Fourier math have to do with relieving that? Does the selected speaker size follow some Fourier generated curve? Comb effects are very audible, so he has invented a huge problem that requires more DSP involvement to solve.

If he generates a sine wave on the individual speakers and superimposes the signal on that, you would hear the carrier wave.

If you generate the carrier wave over a frequency range in response to the incoming signal and superimpose the rest of signal on it, what is different from a standard crossover?

And fuckman, at least make the magnet covers a cone like a good omnidirectional speaker reflector.

To quote a deer fren. Buy a pair of Genelecs and be done with it.

*** one other thought.

Is not the very nature of a Fourier solution the summation of a series. Not a closed form solution, but one arrived at by summing a bunch of little solutions until you have enough to pass some critical error threshold. More DSP just to avoid errors induced by the math.

Genelec, B&W, Wilson and other manufacturers have done this for you at well under a $100K.

Edited April 19 by JGale

[JGale]

Ah! Shazam! I was approaching it like a Neanderthal.

They're doing the signal manipulation in the Time Domain. Grabbing a snapshot over time, sweeping it for frequency content, then futzing with it in accordance to some algorithm they have variable control over and reinserting it into the signal path. Simple!

Not at all intuitive. I can see why it gets him hard though. He gets to play with Math and SPEAKERS!

“The Fast Fourier Transform (FFT) is a powerful mathematical tool used in audio analysis to break down complex audio signals into their individual frequency components. This process reveals the presence and magnitude of different sinusoidal waves (or tones) within the audio. By analyzing the frequency spectrum, engineers can gain insights into the characteristics of sounds, identify specific tones, and even analyze the quality of audio signal”

Edited April 19 by JGale

[Cboss]

If you can't afford it, it doesn't matter.. f****** Finance it!

[Kerry Marchman]

6 hours ago, JGale said:

Ah! Shazam! I was approaching it like a Neanderthal.

They're doing the signal manipulation in the Time Domain. Grabbing a snapshot over time, sweeping it for frequency content, then futzing with it in accordance to some algorithm they have variable control over and reinserting it into the signal path. Simple!

Not at all intuitive. I can see why it gets him hard though. He gets to play with Math and SPEAKERS!

“The Fast Fourier Transform (FFT) is a powerful mathematical tool used in audio analysis to break down complex audio signals into their individual frequency components. This process reveals the presence and magnitude of different sinusoidal waves (or tones) within the audio. By analyzing the frequency spectrum, engineers can gain insights into the characteristics of sounds, identify specific tones, and even analyze the quality of audio signal”

They lost me at DSP.. I am a Neanderthal... I'm still loving my DCM Time Windows and my JBL L-40's.. Ha!

[Jim85IROC]

I need to dig a little deeper into what he's doing.  If I'm understanding what he's doing properly, he's taking samples in the frequency domain and running a fourier transform to convert that into impulse responses, then sending each of those to the most "appropriate" speaker.  Then I assume he's using a whole bunch more DSP to manipulate the phases of the individual IRs so that they sum properly at some distance from the speaker.

 

It's an interesting approach for sure, but ultimately I'm not sure I see the value.  I'm paraphrasing here, but his biggest justification for this is that a driver can't produce the sum of multiple sine waves accurately, but that's easily proven false.  distortion measurements of some well designed speakers clearly show that THD and individual order distortions can be (and are) very low across the audible range on many well designed speakers, typically well below 1% aside from bass frequencies.  The biggest audible differences in speakers tends to be how well the individual driver signals sum across the horizontal and vertical spectrum (i.e. the polar response).  It seems that right now this new design is falling pretty short in that regard, but what I don't understand is if this has the potential to improve on the overall polar response compared to a conventional speaker.  My gut says no for the simple fact that the physical spacing between the drivers is going to create a summing nightmare that even DSP will only be able to "fix" at one point in space, which means that all of the reflected content whose energy plays heavily into the polar response is going to have major phase issues across the frequency spectrum, resulting in all kinds of problems.

 

It'll be real interesting to see where this goes over time.  While I remain highly skeptical, I'm still very intrigued by it.

[ArnieZ]

I may not have listened carefully enough but I thought he said tha each speaker only produces one hand of the sine wave. He intimated that no speaker could recover fast enough to produce both parts of the wave accurately. I wonder what wife would allow these things in her home…. Mine would not and she’s pretty easy about that stuff

arniez

[BoogieMKIIA]

Some technical clarity is in order. The complex sounds are not made up of a bunch of sine waves, but can be represented mathematically as a bunch of sine waves - using the Fourier transform. DSP (digital signal processing) uses the Fourier Transform to analyze and process all kinds of signals, from audio to high frequency. Think spectrum analyzer.

It would be interesting to hear in person. Hard to say if improvement compared to standard speakers would be due to the technique or more simply to greater efficiency of the multiple amps and speakers. Bi-amping on steroids with multiple DSP crossovers. I’m good with new ideas, why not? I don’t think there is anything revolutionary with his complex setup.

I bought my first real stereo as a sophomore in college. A Luxman R1040 receiver I still have and use. I also looked at a Harmon Kardon receiver, they claimed they could pass a square wave with no degradation.  A Fourier transform of a square wave has a lot of harmonics, implying they could cleanly reproduce even higher than 20KHz and very complex signals. The good ole Lafayette fuzz I once had fried my Princeton speaker due to all the added harmonics and power.

Would like to hear comments from pure Class A tube amp stereophiles.