Quote from Øyvind Kvålsvoll on 08/06/2026, 07:12

What is left to improve, if anything, on sound quality and experience?

While sound differences in amplifiers and DACs can be argued to be too small to be possible to hear, most people and certainly most audiophiles agree loudspeakers do indeed sound different. But is that because some speakers have faults, while others are better. Is it even possible to make the perfect speaker.

I realize that it is difficult to make this perfect speaker, simply because we have different preferences for how we want our sound to be like. Some properties are universal, but the difference between say a directional horn speaker to a omnidirectional, will present a very different sound, and there is no established rule that says one is better than the other. They are just different. Which means, desired properties from one concept may not be possible to achieve from the other. It will always be a compromise, one has to choose.

Sound character: Soft & smooth or sharp & detailed: 

This is a choice, but can you have both from the same speaker?

That is not so easy. A typical hifi speaker, if it is otherwise well-made, a box with woofers, midrange and dome tweeter, this can sound very smooth and delicate, but lacks any resemblance of transient impact and dynamics, it will also be limited in 3D rendering, and very sensitive to room acoustics. A larger horn system can provide what is missing, but at the cost of softness, it will sound sharp and detailed and open, but not so smooth and soft, and tonality will also be affected by the radiation pattern, which will tend to give it a brighter tonality compared to the dome tweeter speaker.

Some speaker designs actually does give more of all desired properties. Some dipole panels, some well designed horn systems, even a trad hifi box, larger, stretched to the limits, can provide more of everything.

So if some speakers have more of everything, maybe it is possible to achieve a sharp and defined and dynamic sound with impact, while keeping all of the softness and smoothness and tonality of a "soft" speaker.

Rendering: Look into the recording, or have the band in your room

Another aspect where what you have is more of a choice, rather than simply best.

Back in the days, I had two systems set up in the same room; one was the magnestatic dipole planars, then I had trad boxes with domes and box. The planars, with dipole line-source radiation, gave large presentation with the instruments in the room, the trad speakers presented a more precise view into the recording.

What if we can have both.

The instruments in the room, with the reverb and acoustic environment from the recording present, in a way that places the recording in the room, with this very real and physical presence of real instruments.

Solved: Transient impact & dynamics, loud enough, full frequency range coverage

The F205/F205-H with a decent bass-system is one example of a solution that has all this.

This is where all typical hifi box speakers fall short; they simply ca not play loud enough, and when comparing to something that actually works, it is apparent they lack transient impact and realism, making the whole presentation boring to listen to.

Less sensitive to room acoustics and consistent performance across different rooms

This is difficult. The F205 was designed to work well in less than optimal rooms, but reality of physics sets in, if the room is not suited for sound reproduction you can only do so much. Yes, it works better, but still a huge difference and performance boost in a properly treated space.

More presence and realism

Can there be even more presence, more realism, like you can reach out and touch the instruments.

A speaker like the F205 has quite good presence, but requires acoustic treatment to bring out this last level of performance.

More instrument separation

Already very good on the F205, but can it be even better.

Separation is how clearly defined each instruments is rendered, and how easy it is to pick one instrument and follow this, when there are many instruments in the recording.

Works across a larger seating area

This is very difficult, due to the nature of stereo/2-channel reproduction and how our hearing works. And multichannel does not solve it.

Bass

Yes, it is not too difficult to achieve full frequency range; at least down to 16Hz with full capacity.

To get really good bass quality, with the desired tactile feel, and be able to replicate that across different rooms, that turned out to be difficult. No, you will not get it right by simply applying eq to flatten the frequency response and then adjust delay to get timing right. DSP processing with eq and delay on both bass and main speakers is a requirement to be able to get even close, but it does not solve consistent and predictable bass across different rooms, it needs to be tuned correctly.

 

Quote from Øyvind Kvålsvoll on 10/06/2026, 09:05

What are the properties of the future loudspeaker:

Much of my work with loudspeakers has been controlled listening tests for the purpose of finding correlation between what we hear and technical properties of sound. This gives information on what is important, and what properties affect different aspects how sound perception.

Only when properties of the sound are known and defined, can the technical properties of the loudspeaker system be specified.

Now that we have, from the previous post, some desired properties we want to improve, we can establish requirements for the sound that the loudspeaker must reproduce.

We need to go beyond frequency response.

Radiation pattern: Both wide and narrow

Is that even possible.. maybe it actually is.

We need to radiate just the right amount of sound into the room, in all directions, so it can be reflected and create some smoothing and some decay. Not too much, because that will destroy realism and 3D rendering.

We need a stronger direct sound with just wide enough pattern to allow for some movement of the listener and ideally a wider listening area. Create a wavefront with higher intensity, not like a small point source, but a sound field where the velocity potential is in phase with the pressure, more like a planar wave. This preserves dynamics and impact. And the pattern must be narrow enough to eliminate very early reflections.

The size of the source must be larger than point source, but not very tall like a true line source. This defines the size of the soundstage and instruments, we want some size and volume, but not overly large. This perceived size relates to the size of the source, so that a small point source will project small instruments while a very large planar line source will create too large instruments. We want something in between.

The radiation pattern is what defines the sound character, it is the most important property of the speaker, and it can not be fixed in DSP afterward, it is an inherent property of the speaker.

...