Glen, it's taken me 10 years to reach 2,000 so I draw the line at 5,000 if I'm looking at somebody else's tree. That might be a mistake, but you've got to draw the line somewhere. If, as it's widely believed, one in a hundred births are not as expected, or rather conceptions not as expected, there's a lot of potential for error in 5,000.
Without DNA I know I would have three mistakes in my 2,000. And that makes everything prior to that irrelevant. If ever I get in casual conversation with somebody who says he's researched back to some obscure Saxon King, I remind him of that fact, and watch his jaw drop.
Zaph