No, it's not heresy at all, but it's not as if this hasn't been considered before and subjected to a fair amount of analysis and research by both Alembic and independent researchers. Ron W spent months trying to find the ideal cabling for the pickup-to-preamp connection, and went through hundreds, if not thousands, of possible solutions. That original cabling issue is largely responsible for the current design of the entire pickup, preamp and filter architecture.
The key issue here is the difference between improve and change. One can certainly _change_ the sound of an Alembic. But I'd argue that the _sound_ is a lot of what makes Alembic unique. I can build a bass guitar that can sound pretty much any way I want it to. But to get it to sound like an Alembic, I pretty much have to clone the entire electro-mechanical device, in situ.
Forinstance, I find stepped attenuators limit the flexibility of a pre-filter gain stage. You'd also alter the inter-stage impedence matching between the Alembic preamp stages. This, in turn, would affect the gain and damping factors of the filters a bit. Probably not critical, but definitely enough to alter the sound in a negative fashion (by changing the phase characteristics through the various filter poles). This isn't an improvement, it's merely a change.
Cardas tone wire won't have that much of an effect. The source and load impedences are different, for one thing. The pickup's inherent frequency response is another factor. The Cardas wire helps tremendously by improving the articulation above about 8KHz. Essentially, it has a capacitive reactance characteristic that improves its inertial slew rate. While this also affects the rise time response of low frequency signals, it also contributes to a very slight overshoot. Not audible in and of itself, but it becomes so when it's added into the phase response of the first preamp stage. It shines best when driving a 12AU7 directly; the tube gain curve dampens that overshoot and effectively masks it. The same signal driving a FET-front end IC, otoh, will exhibit a hair of ringing at certain harmonics (determined by the input impedence of the IC, the capacitive loading of the wire and the capoacitive reactance of the pickup's source load impedence).
The Alembic pickup doesn't have a lot of information up above 8k, so there's not a lot of help to be provided by the Cardas wire. Certainly not as much as a decent phono cartridge, anyway. Not so much because the pickup can't produce it, but because there's not a lot of fundamental signal above 6k coming off of the strings. Yes, you'd hear is _very_ slight glassiness in the overtones, but only if the _entire rig_ were set up to reproduce the top 1.5 octaves accurately. On stage? You'd never hear it. DI'd into an SSL board? Probably not. Maybe through a tweaked out Neve console, but only if the entire reference monitor signal path was totally time-aligned and phase corrected.
Bypassing the volume pot entirely won't be an improvement; merely another change. You'll change the both the preamp loading and the filter's input CR factor. Depending on which Alembic you try it on, you might even confuse the Q circuit.
Now, rebuild the instrument's entire electronic package with 1/4% tolerance components, and drive it straight into the above-mentioned reference studio setup, and you'll hear a difference. I wouldn't call it an improvement; that kind of tweak merely makes the tone more consistant from instrument to instrument. Which is irrelevant, as the differing wood mixes and changes in acoustic properties from plank to plank, coupled with comparitively inexact construction techniques (they don't measure the amount of glue used in any given joint in grams, for instance) mean that each instrument will be measurably different in overall voicing from any other. That's one reason Ron electronically sets up each instrument.
By all means experiment, though. Just keep in mind that difference between improve and change. You may well (almost certainly will) discover a sound you like better. It may even be relatively close to the standard Alembic sound. But it will be _different_, and will contain tonal elements not initially present in the core audio generator that is an Alembic instrument.
Now, that's not necessarily a bad thing. And I guarantee that if you DO experiment along those lines, especially using an Alembic as your test bed, your understanding of ALL acoustic phenomenon will increase. You will make discoveries and you will increase your own toolkit of tricks and techniques. You WILL find sonic changes that are improvements, and that you can apply in other real-world situations.
But I'm betting it won't sound quite like an Alembic anymore.
:-)
