The error will be pretty much the same regardless of repetitions till you do some
serious wear to the servo, and some of these specs probably do give an 'absolute
accuracy' spec.
  If you want some good first hand knowledge though, get on the SRS mailing list.
Nothing punishes a servo like a robot, so they can tell you which ones in which
price ranges hold up the best with best accuracy etc.  If you truly need precision
the JR super servos will probably win out, they have less dead band and are more
active around the dead band.  May be something better by now though, I haven't kept
up on that front lately.  RC Heli links may be better for that, a good competitive
flyer can tell not only which start out best, but even about how many hours are on a
servo by how much extra slop it has.
  What are you going to do with the motion?  In most cases there are better
positioning methods, and even if a servo is the best answer, modifying the servo
will often give much better results.  (And is pretty easy to do..)

John Walker wrote:

> manufacturers specs but have been unsuccessful so far. I can run the test
> myself but I doubt the results will hold for all the types of servos
> available. I want the information in order to determine which servo to purchase.