I didn't abandon this thread, but I had to stop and think about this for while. Mistwell presented a very compelling argument that caused me to reconsider my viewpoint. My complements to Mistwell.
So, let me respond to only a couple of points first (there are way too many to cover everything), and then I will respond to Mistwell's argument at the end.
Artoomis said:
Remember that the actual language is "...First, you must move to the closest space from which you can attack the opponent. " This means that with a reach weapon you do not really have more choices than with a regular weapon - or more chance to veer further off-course.
The 'closest space' term is very poorly less defined term without 'directly towards', however (as I pointed out in post 24). It is one of two things: (a) closest to the designated opponent, or (b) closest to the starting position of the attacker. As I noted earlier, if you just read the SRD (e.g. you only use OGL), it is clearly (a). This in fact does mean that you can charge to any square around the target that you wish, because they are all equally distant to where you can attack the opponent. (Assuming no obstacles are in the way, including your target.) You have to open the PH to see the picture (not external text) to come to the intended interpretation that it means (b) due to the additional text of "shortest path." The alternative to using that picture in the PH is to incorporate 'directly towards' with 'closest space'.
Hypersmurf said:
The square closest to A's current position that threatens X is marked in red, if you're talking piece-of-string distance.
Why are you using the square closest to A's (the attacker) position and not just any square closest to X's (the target) position? See my response above.
Hypersmurf said:
The ending space isn't 'the square X is in', though. The ending space is the space he attacks from.
I agree with you, though obviously I disagree about the legality of the ending space itself. However, I just wanted to point out (IIRC -- and wouldn't it be embarrassing if I didn't) that the PH picture actually shows that this rule applies to the target's space, not the ending space. It is IMO a contradiction in the rules, but text clearly trumps pictures/graphs/tables, so you are in the right.
Mistwell said:
So which is it. Do you mean center point to center point, or do you mean any point in the PC square measured directly towards any other point in the NPC square.
I say neither, and I'm not just saying this to whine (thanks for that Endovior) or to be obstinate. I've given it a lot of thought and I admit I have to really dissect the words "directly towards" in game terms that are not spelled out anywhere. In doing so, I have to concede my stance and offer support for your position. Let me elaborate.
That said, I say that "directly towards" means any point (not just vertex) in the PC (A) square measured directly towards any other point in the NPC (X) square such that the line (not line segment) must pass through the NPC square. If it doesn't pass through, then it can't be considered directly towards, but directly to the side. In other words, given your example, drawing a line from the bottom of A to the top of X is invalid as the line does not pass through X. I don't consider that the line overlapping the edge counts because, like I said, that would be to the side and not towards.
The one caveat to this is that when you draw the line, then that point of A that you used should end up in the same relative position when you reach the target. Thus, if you use the top left vertex of A and draw a line to the top right vertex of X (which passes through X), then your movement should reasonably approximate that the top left vertex of A will be attacking the top right vertex of X. In other words, you can't move A to the top right corner of X because that would have A's bottom left vertex (or so) attack. And that isn't the line you used.
So, you instead can have A's bottom left vertex attack a point just below X's top right vertex. For sake of argument let's call it 1/8th of the way from the top of X. This will illustrate the example that most accurately rebuts my previous stance. It should be obvious that in order to have A attack X in such a situation, A's ending position must be the top right corner X, and NOT the space immediately right of X. Although the line always will have a non-zero slope (in this example), the best approximation of the line is moving A straight west.
Note further that this also offers support in the following case:
O
OOOOO
X
OOOOA
O
OOOOO
A could charge X to any of the three squares. However, as Thanee points out the top and bottom squares are not shortest distance (one diagonal could be called 7.5ft), but they are equal path lengths in the D&D grid movement system. I would not disallow them as valid ending positions, though, because I would want to make it consistent for ride-by attacks. I'd rather not houserule it just for non-ride-by attacks.
PS I reserve the right to change my position again at the drop of a hat.
