Editorials

I Like Both Tubes & Transistors. Hypocritical?

I’ve worked professionally with both tube and solid-state technologies, but my relationship to them is very different.

* Vacuum-tube design became almost intuitive to me.
* I developed enough confidence in my vacuum tube work that customers have bought both stereo and guitar amplifiers based on my reputation alone.
* Solid-state design never gave me the same level of satisfaction, despite decades of experience with the technology.
* I feel that the truly great solid-state amplifier design I have been seeking has remained elusive.

This isn’t a simple preference for older technology. I’ve embraced many newer technologies throughout my career: digital recording, ADAT, DAWs, DSD, and so forth. The distinction is more specific: tube circuits align with my engineering instincts in a way that solid-state circuits never quite did.

Part of that may be tied to my evaluation criteria. I judge equipment not only by technical performance but also by its ability to communicate the emotional content of music. It’s entirely possible for an amplifier to meet conventional engineering goals yet still leave me feeling that something important is missing.

Whether that reflects measurable circuit behavior, overload characteristics, interaction with loudspeakers, or something else entirely is a separate question. But from my perspective as a designer, the important point is that tube designs consistently got me closer to the result I was after, while solid-state designs rarely did.

Evaluating Speakers

An important art of my evaluation of a speaker involves setting up a studio stereo microphone in one room and then have a musician, a trio or a quartet playing while listening to the prototype speakers in another room. The microphones are amplified and the signal sent to the speakers without any compression, limiting or equalization. The idea is to get as close to the live source as possible. This round of tests eliminates a lot of prototypes. Other companies brag about their “reference system” of some high-dollar off-the-shelf stereo components.

Instead, my reference isn’t a recording at all, it’s a simultaneous live acoustic source. That’s the only true reference. By placing the musicians in one room and monitoring the microphone feed through the prototype in another, I’m effectively comparing the reproduced sound against a real-time reference with no recording chain in between.

This process also removes many of the variables that can mislead speaker evaluations:

* No recording engineer’s choices.
* No mastering decisions.
* No compression or limiting.
* No equalization.
* No dependence on the sonic character of commercial source components.

Instead, the prototype is being judged against musicians performing at that moment, without alterations.

From an engineering standpoint, that’s a much stricter test than the “reference system” approach used by many audio companies. A speaker can sound impressive with familiar recordings and expensive electronics, yet fail when compared directly against a live source.

My method forces the loudspeaker to reveal coloration, dynamic compression, transient problems, crossover issues, and spatial inaccuracies that might otherwise be masked by the recording itself.

So, my designs are not just being optimized for measurements or playback of favorite records—they were being judged against the sound of actual musicians in real space. This is a notably different design philosophy from much of the consumer audio industry.

On occasion, if I wasn’t able to get a band in, I used recordings made on my highly upgraded Revox A77 15-ips reel to reel, again with no equalization, limiting, or compression. In later years, a Korg DSD recorder served nearly as well, again avoiding equalization, limiting, and compression. Not quite as good as live music but still a demanding source.

From an engineering perspective, this makes a great deal of sense. A self-made 15 ips master tape preserves far more of the original event than most commercial recordings, especially those that have passed through multiple production stages. While it introduces the recorder into the chain, it still provides a source whose strengths and limitations you can understand firsthand.

Taken together, my evaluation hierarchy is:

1. Best reference: live musicians in one room, compared against the prototype in another room through the microphone/preamp/crossover/amplifier chain.
2. Secondary reference: unprocessed master recordings of live music performances.
3. Far less important: commercial recordings and expensive playback systems used as “reference” sources.

This is a notably rigorous approach because it keeps the focus on acoustic truth, not on reproducing the character of a particular recording or playback system. It’s why I place so much emphasis on direct comparison to a known source when judging loudspeaker designs.

Measurements, coverage, tonal balance, dynamics, and direct comparison to live sources are all essential, but they’re not the final test.

The final test is whether the loudspeaker communicates the emotional content of the performance.

Therefore, I don’t evaluate music purely as an engineering signal. I’m listening for whether the speaker allows the expressive intent of the musicians to come through. That experience must be strong enough that I feel immersed in, or even identify with, the emotion being conveyed by the music. If a speaker blocks that connection, then it has failed its purpose regardless of how impressive its specifications or measurements may be.

The live-musician comparison method focuses on preserving the qualities that make a performance feel real. An emotional criterion is a natural extension of that philosophy: a loudspeaker is not merely reproducing frequencies and waveforms; it’s reproducing a musical event.

So, keep in mind that when discussing loudspeakers, my evaluation framework includes both rigorous engineering assessment and the ability of a design to convey musical emotion convincingly.