DS: What's the technical advantage of this connection?
HD: I wrote a paper 20 years ago about the advantage of having a fully matched connection. I chose 50 ohms because it was the termination that was already there. So I didn't need to to develop a special cable because there were already some 75 and 50 ohms. 75 ohms were used for digital. I didn't want to cause confusion, so I chose 50 ohms for myself. 50 ohms cables were also used for telecommunication networks and the internet. If they were in coaxial they were also used in the measurement labs and for antennas for broadcast. Everything was 50 ohms then. I choose this because so I had a wider choice of cables in this industry. If you would choose 75Ω, you would not have the same choice. So I wrote my paper and I showed that with end-to-end impedance matching, you cancel all the echoes in the cable.

A lot of cable manufacturers say that their cables have a rise time. If you have a rise time in the cable, it is only because of the echoes, because the echoes go back and forth. If you enlarge the image on an oscilloscope, you can see that it is not a slope, but steps. I have shown this in my paper. But if you match the impedance, you don't have any echos and the square wave is a square, no matter how long the cable is. The signal has a rise time of of five nanoseconds at the other end. The signal is delayed in time because the propagation time in the cable is about five nanosecond per meter or so. That's it. When I published the article in Stereophile in 2001, it was funny: All the cable manufacturers told me that I was a crook and I was telling lies. But it was just physics. Then they said that the echoes only appeared at very high frequencies. It's okay. But, you know the human hearing system is quite complex and you can hear things you can't measure, for sure. If you listen to music with a matched or an unmatched impedance, you can hear the difference immediately. As soon as the impedance is matched, it sounds as if you lifted a veil: It's just clean, crystal clear. So I decided to make all my connections that way.

It's different with loudspeakers because unfortunately speakers don't have a constant impedance. So it's difficult to build a suitable cable or you have to do it with a network at the back of the speaker to control the impedance. But that's a bit complicated because you would have to make one for each speaker model individually. Anyway, I will introduce a speaker cable soon that's based on different physical approach. I don't like making things that I can't explain. Snake oil is not my cup of tea, you know? I'm not criticising people who make speaker cables: There are many very good speaker cables out there, but we don't have an explanation why one cable sounds different from another, because we can't measure it. That's the challange to find the explanation why. You know, I'm close to the answer.