Thanks for the information. Evidently, carbide is a lot better, with respect to concentricity of the ID, than steel or coated steel. I may try one.
As to one piece dies, I have tried a Hornady, just the least expensive, non bushing 6PPC, and the runout you get with this die is terrific, something on the order of a third of a thousandth or less at the end of the neck. There are however other issues. the reduction in diameter at the shoulder is a little over .003, and the radius at the point of the shoullder is, in my view excessive, as compared with my chamber, which I believe to be typical. Neither of these would be a problem for any other use, but because we use a small number of cases, reloaded many times, work hardening comes into play in a way that is somewhat unique as compared to other applications.
My latest daydream is to have a custom steel bushing made of steel that is hard enough to work without heat treatment. I would have it turned drilled bored, polished, and parted off in one setup. Also, I would modify the interior shape so that the unsized neck is captured within a very short cylyndrical section that would precede the lead in angle to the sizing portion. This may require a taller bushing, and backing off the top cap of my Harrell's die. One thing that you may find interesting, pull the bushing out of your FL die, size a case, with the normal amount of bump, and then check the concentricity at the end of the neck.
Also, I may get crazy and waste some money on a collet die, and see what I get with a two step process. BTW, one can change the amount of sizing on a collet die by changing mandrel diameters. Yes, I know, It sounds goofy.