Re: how about 2 ogres
Magac said:
so how many things do you want to change in the game? how about changing not mechanics but the style you play, or are mechanics all important. (ooh contraversial)
Not contraversial [sic]; just impolite

. Although I certainly can use some tactics lessons for my game, I don't want to have to change NPCs' tactics to account for illogical rules.
Mechanics in a game need to take into consideration 4 things, I think:
1) Elegance (simple to understand and to implement)
2) Plausibility (mirror how you'd think things would work if they really were happening
3) Panache (allow and encourage cool stuff to happen).
4) Balance (there's no one killer strategy good for all situations).
Tumble, as written, is very elegant, but suffers on plausibility and panache and balance. It's plausible to me that it's harder to tumble past the hill giant barbarian than past the kobold, which isn't reflected in the rules. And all-tumble all-the-time grows stale, hurting panache. And it becomes a surefire strategy for any character than has access to tumble.
The proposed changes are less elegant: adding in an opposed roll, or a dynamic DC, makes every tumble attempt take longer. However, it increases plausibility by making it harder to tumble past the hill giant barbarian than past the kobold. And it sometimes increases panache: it's cool for the high-level monk to be able to trip the high-level rogue who tries to tumble away. And it increases balance: tumbling is no longer the default tactic for any character that can take tumble as a class skill.
So yes, it's possible to readjust tactics to match mechanics. But that's putting the cart before the horse: the mechanics should serve the scene, not the other way around.
Daniel