Lately I've been daydreaming about designing, manufacturing and selling a line of audio equipment (dare I say "HiFi"?).
I've designed and built such things since childhood with varying degrees of success. It started out of necessity as even as a child I was very sensitive to music and the way it was recorded and reproduced. I was fortunate to have been exposed to Quality recordings and stereos at a young age which for better or worse gave me high standards and expectations for the way music should sound. I also spent a fair amount of time listening to live music as many of my friends and family members played in bands and orchestras. This gave me a very natural understanding of how different recorded music is from a live performance and how the equipment used in either case is truly an instrument in and of itself. Over these early years I cobbled-together some good sounding systems out of nice-but-broken gear that was discarded or donated to thrift stores and learned enough in the process to even fabricate some equipment of my own.
Later I became a musician myself and again necessity became the mother of my inventions. I not only learned to play, but to record and perform live, all using second-hand or repurposed equipment. Some of my more notable accomplishments from this time was the use of Betamax video recorders (cheaply available once VHS won the format war) for recording high-quality audio and a modem-like device that allowed our bass player to attend practice remotely via phone lines (as he was grounded more often than not).
Later in life I became employed and could finally afford to purchase at least some level of new medium-to-high-end equipment, but I was often disappointed. The off-the-shelf audio gear of the 90's often fell short of what I experienced with the recycled or home-made gear of my youth. This could have simply been due to the fact that even with a real job the gear I could afford new wasn't as good as the broken-and-repaired gear I grew-up with, but I don't think that explains it completely. If I had to guess I'd say that the gear of the 90 and later leaned-into the inherit (and let's face it, easy) fidelity of digital music, and since we no longer had to contend with the noise and mechanical complexity of analog sources we could get away with cheaper parts "downstream".
This is just a guess.
Regardless of why, this left me longing for the sound of the older gear and I went back to collecting and repairing equipment from the past.
Fortunately a lot had changed since I was a kid. First of all, the Internet made documentation available for much of this gear. The old stuff usually included a schematic or at least a simplified diagram inside the chassis but nothing like what enthusiasts had found or produced in niche Internet forums. Second, I now had access to high-quality repair equipment like signal generators and oscilloscopes; things I couldn't have dreamt of when I was ten.
The Internet also made it possible to hunt-down and procure specific pieces of equipment in a way that was entirely based on luck before, and I was able to finally experience gear that I had always wanted to audition but could never get my hands on. This was a significant learning experience because many of the pieces I desired the most turned-out to be less profound when I finally had a chance to listen to them. Conversely, some of my favorite pieces turned-out to be some of the most modest, like the Henry Kloss-designed Tivoli Model One and Two.
In parallel with this, my knowledge of not only electronics exploded but my knowledge and experience with fabrication in general grew dramatically as well, and when I got into 3D printing around 2010 my ability to produce physical objects using the software skills I've honed my entire life resulted in an explosion of creativity, including a return to designing and building audio equipment.
For me 3D printing had a profound impact on one specific are of this work: speaker design. I had designed and build speakers over the years using traditional woodworking methods and achieved satisfactory results, but it was hard for me to really experiment with speaker design this way. My carpentry skills are limited (especially in the finishing department) and anything beyond a basic box is really beyond me. I certainly wasn't going to be bending wood into curves or other exotic shapes needed to try-out some of the more interesting areas of speaker design. It was also harder to fine-tune a design when every change meant starting over or trying to build test jigs that were adjustable. 3D printing makes it possible to create a parametric design and then run-off a range of variations that can be auditioned, and do this without endless rework. It also "makes complexity cheap", so if you want to try something weird you can do so without spending months or years learning exotic construction techniques and/or burning through a lot of expensive materials.
But can a 3D printed speaker sound good? Or even great?
That's the question I'm working on answering now. There's certainly plenty of speakers that have been designed on computers and modeled this way, and while I'm not sure if any are currently produced via 3D printing I'm sure that prototypes have been made this way. I could take the same route, work the kinks out with rapid prototyping and then switch to something that is more traditional or more compatible with mass production but I'm more interested in finding a sweet-spot where I can produce a great sounding speaker using the "manufacturing" capacity and scale I already have access to. This allows for a few neat things in addition to keeping it manageable like producing both off-the-shelf models as well as one-off custom pieces.
I've started to experiment with this and so far have produced a prototype that I am testing now as well as a few more designs that I'll be testing soon. My goal is to dedicate time to this work over the winter (getting more serious after Halloween is behind us) and minimally have at least one design that meets my standards.
From there who knows?
I have been thinking a lot about what I want to do next, and in particular how I want to get away from the AI/LLM-infested field of professional software. The idea of starting a company and manufacturing a physical product seems almost impossible to me, but the idea of writing software for a living once seemed impossible to me as well.
Jason J. Gullickson, 2026