Richard is dead on with his post, in my opinion. Our ears are critical for good reason, so let's go back and take a look at the thread as its originator intended. It is about the three different drive categories, assuming the string isn't in one all its own.
Direct drive:
How well it works depends upon how its speed is controlled and how it is coupled to the platter. Of course, there are many other variables, but these are most critical for this drive method. I believe a great direct drive is the most difficult to achieve because of audible artifacts introduced by most implementations, but some have accomplished the task beautifully. The Mitchell Cotter turntable comes to mind instantly, but others, like Brinkman have done an admirable job. Still, my ears hear a lack of dynamics and a "jitter" with 90% of the direct drives out there. One thing I have noticed about them is that the plinth isn't quite as important as it is in other designs. That's probably due to how the spindle is coupled with the rest of the system. It is a system that can be stellar, but it is difficult to do.
Belt drive:
As we all know, these run from horrid to fantastic. It seems that most are copies of another one, in that they typically use essentially the same motors and same belt treatment, more or less. Others, however, employ ingenious workarounds to avoid any pitfalls introduced by belt creep. The SG Spiral Groove comes to mind with how the motor is oriented to the platter. There is very little spare belt hanging out there in air with that turntable. That differentiates it from the pack, and the results show it. That's only one example, but I gave it to illustrate that there are designers who build belt drives who think outside the box in extremely practical and creative ways. There are others, but most go down the path of the status quo. Anyway, I believe we will see a lot more done with belt drives because there are a lot of ideas that haven't been tried, yet. There are also ideas that have been tried, but forgotten for some reason. Possibly the finest belt drive I have ever heard is the Fairchild Studio 750. This old beast went against convention by using an incredibly high torque motor. Maybe someone will repeat that with a modern incarnation. Even if they don't, great belt drives exist in fairly large numbers.
Idler drive:
I realize that idlers are foreign in many circles, and that biases exist which are based upon vintage units that display what appear to be inherent weaknesses. Most of those weaknesses, however, are due to resonating linkage, clunky top plates, noisy idler wheels, etc. One thing to bear in mind is that the idler wheel tracks with greater inherent accuracy than a belt. The trick is to make it track quietly, which can be done. If one eliminates the noise of the vintage units by totally rethinking how the technology should be applied, an idler can be a wonderful turntable.
So...
Any of the three can be at the top of the food chain. It all depends upon how the design is approached and how practical ideas are implemented that improve on what was done before. We will never settle the discussion of which is best, as long as people fail to realize that drive methods are nothing more than points of departure, and that the devil is in the details, regardless of the method chosen.
I submit to you all that turntables exist, or can be made, of all three methods that are high performers, and that those turntables can satisfy anyone's personal preferences when the problems facing each drive discipline are addressed properly.
.
