Analogue interfaces and cables, a few musing for thoughts
..
Once upon a time there were only valves / tubes. The poor little transistors hadnt been invented by Bell Labs yet. As valves are good voltage amplifying devices a standard evolved where by voltage driven system was used to connect pieces of hifi equipment together. Unfortunately this meant that unless the interfaces was designed well it was subject to the vaguaries of the cables parameters like capacitance and inductance (C in particular). An output therefore has a very low output impedance and an input has a very high input impedance. When properly engineered these can work REALLY well in either balanced or unbalanced form.
Now along came some boffins at Bell labs and invented the transistor. These took over the world but for the reason of compatibility the same interface was sensibly kept so we could all mix our old valve equipment with newer transistor equipment (or the other way round for some of us). Unfortunately for us transistors are REALLY good at switching current. [It should be noted that valves and transistors can do both but what they are best at is important].
Now some really bright sparks had another look at the old analogue interface to see if they could make it better. They designed a pre out with a high output impedance (an ideal current source) as transistors are good at this. They then design a power amp with a low input impedance. This may sound slightly odd (bare with me on this). This then appears to short circuit the connection. This is the oppose of what we currently do with minimal current flowing. They have in effect designed an interface based on current and not voltage. It is incompatible with current kit but does it work and what are the benefits of changing?
Changes of voltage are minimal in this new method as the input impedance is so low, therefore no negative feedback is needed (good thing). [Remember this is what valves makers often toute as a forte of their technology]. As voltage differences are in effect zero, cables ceases to have any effect. There is no voltage across the input and therefore C has NO EFFECT. Any residual voltage that is developed owing to cable inductance is tiny and also has NO noticeable effect on the now current source. Therefore the cables, no matter how silly in C figures, have NO effect on the sound of cables. I.E. Completely independence from the various factors that make cables different (C, L and R). Granted you will still want well screen cables etc but this solution has significant technical benefits re sound quality. So is this myth?
Krell call this CAST. Tag would have used it in their F1 equipment if they were still going
strong. I suspect many others have had a play with this system as well in the past.
Unfortunately cables and interfaces are all to often intimately mixed and the interface standard often comes much before the desire / knowledge to make the best connection method. Equipment and knowledge both being noticeably better in the future. Hopefully when we all go properly digital with well designed interfaces (HDMI / IEEE1394) hopefully we can lead behind these limited methods behind us as an anachronism from the past. Cables will cease to become an issue for all of us and costs will come down.
It is all about how it is engineered, we can make really good voltage interfaces in analogue now but perhaps we can make even better current one but the future is digital. For now I am happy with 'my' bargain cables and good old voltage based interfaces