Rubber belt alternatives - debatable?
From my experiments so far, using nylon and silk thread versus rubber belts (vpi new grey + translucent original belts), it appears that rubber has a number of good sides, that are only with difficulty improved by the thread solutions. Two nice sides with rubber is fairly robust performance regardless of platter - pulley distance (within limits, not very demanding of setup), and a fair degree of "body" and overall tonality to the music. Even if thread solutions are a step up, if set up right. Silk thread sounds a tiny bit better, I think, compared to nylon, but may be a bit more troublesome also. When things are right they sound better than belts, but they have to be tensioned "just right", and with the VPI Samas arrangement this can be a bother to adjust manually from time to time (more often than with the rubber belt). Possibly the ideal arrangement would have a light spring arrangement pushing the Samas away from the platter to "auto-tension" the thread.
As others on A-gone and elsewhere have noted, this concerns the right balance between stiffness and elasticity, transfer and isolation. We want motion going one way, but no audio noise, and nothing travelling back either.
The more the platter is able to overcome motor and pulley and string weaknesses (noise, pitch variance, statics) by its own momentum, the better. The ideal self-moving platter hasn't been invented. VPI currently offers a very expensive "hrx super platter" upgrade, this is probably a great benefit, but I have gone through three VPI platter upgrades already, and guess my current Black knight platter should do the job fair enough for awhile, there is more benefit to be had by motor tuning and drive adjustment. A direct drive for HW-19 is not available, but a good thread solution is said to be a step up in the same direction.