In your example, it is already possible to choose corners of C and T that pass through actual blocked squares (try lower-right on both C and T), so this example isn't really applicable. Choosing corners isn't a contest between the attacker and the defender, it's a state check. If any two corners exist such that a line between them is blocked, the defender has cover. If all such lines are blocked, the defender has total cover.
I had never considered this. I read "borders" as the four lines of the square. Your point about the wording for flanking makes me question this definition. With no higher authority to give a ruling, and without a general consensus of the board, I would say no, the corner does not provide cover. If it DID provide cover, I would give the defender no more than +2 to AC, instead of the normal +4.
|
I don't see where the problem is.
"If any line from this corner to any corner of the target's square passes through a square or border that blocks line of effect or provides cover, or through a square occupied by a creature, the target has cover."
Imagine a perfectly 45 degree line being drawn from the corner of your choice to the most extreme corner of your opponent's square. This line will connect with one corner of every square on its path, but it will never cross a border or go through that square.
*thinks to himself that this explanation doesn't help, and wishes he could just draw a picture to help illustrate his point*
Yes, just a corner provides cover. Though the GM is free to adjust just how much of an AC bonus the defender gets. In your above example, I would grant +4.
Both examples provide cover. I see no reason that you shouldn't count corners. The corner is the point where two lines meet, so by definition if it passes through the corner it is passing through the line.
|
The inclusion of corner re: flanking is for flanking, not for cover.
Blahpers, ranged cover is determined from the attacker's best corner. There is cover if the line is blocked to any of target's corners. You're "any two corners" idea is false with respect to ranged cover.
In OP's diagram, there is no cover on a ranged attack from the wall; the corner doesn't matter.
Whelp, seems we have some disagreement.
cor·ner [kawr-ner]
noun
1.
the place at which two converging lines or surfaces meet.
I think it is safe to say if the lines meet at a specific point, and you draw line of effect through that point, you are drawing line of effect through a border of the square.
|
|
Lest there be any confusion, my post replied to the first example only.
I don't understand what physical situation the second one is supposed to represent. Either there is line of sight based upon the architecture or there isn't. If there is line of sight via a gap, then it would be improved cover. If the corners touch, there is no line of sight: total cover.
In your example, it is already possible to choose corners of C and T that pass through actual blocked squares (try lower-right on both C and T), so this example isn't really applicable. Choosing corners isn't a contest between the attacker and the defender, it's a state check. If any two corners exist such that a line between them is blocked, the defender has cover. If all such lines are blocked, the defender has total cover.To determine whether your target has cover from your ranged attack, choose a corner of your square. If any line from this corner to any corner of the target's square passes through a square or border that blocks line of effect or provides cover, or through a square occupied by a creature, the target has cover (+4 to AC).
Based on the rules for ranged cover, it is a contest between the attacker and defender. The attacker chooses a single corner of his square and draws the line to all corners of the target square and then sees if any of these lines are blocked.
Hmm, you're right, and that borks things. The attacker shouldn't have to choose a corner to determine something like this; cover ought to be something that is or isn't. Plus it creates a corner case (haw!) like this thread. I'm pretty certain I can come up with better wording that means the same thing as above (no disrespect intended to the designers and editors!), but it wouldn't get rid of the corner case.
I'd have to rule on a case-by-case basis. Celanian's logic re: total cover makes sense if you're really looking at giant 5x5 cubes of stone inexplicably meeting at a corner, and I'd rule the same way. If those obstacles were tree trunks (not Minecraft tree trunks, but good old fashioned round tree trunks), I would rule the other way, as you aren't really clipping the corner.
| 1 person marked this as a favorite. |
It makes sense to me that you choose the corner you wish to shoot from. That is how you can shoot around corners.
Archer1 is shooting around the corner. Archer2 is in open space in a corridor. Archer 1 has cover while Archer2 does not.
As for the two corner issue...if the corners meet then it is total cover. If the corners do not then it is probably improved cover. Select your corner, draw lines from your corners to the target's corners and if there is a clear space to shoot through (even if it is only a fraction of a grid square) then it is a possible shot. From that point it just depends on what the amount of cover is.
Remember, buildings and the like do not have to match up with the grid. There could be a gap. Otherwise, shooting through windows and archer slits would not be possible.
- Gauss
| 1 person marked this as a favorite. |
Lines are 1 dimensional even though we have to draw them in 2 dimensions. A one dimensional line cannot provide cover with respect to a parallel attack, such as the open 5' corridor. The attacker can choose one side of the line or the other to attack down with no impediment.
To provide cover, the attack must pass through a border. The point at which two lines meet is the vertex, in our case a 90 degree corner. A third line, our attack, passes through the vertex outside of the 90 degrees that makes the corner. Math states that the attack does NOT pass through the border.
The attacker chooses one corner to represent the attacker setting up the clearest shot possible. It is check against all of the defender's corners to represent the defender using cover to his best advantage.
This represents that an attacker that hides behind a tree and shoots around the tree at a target in an open field has cover from that target. If the cover was measured from all of the attacker's corners, the target in the field would have cover from the attacker as well.
Example 2 could be as simple as a wizard putting a wall of stone or force up against an existing wall or maybe with an inch or so of spacing. That would be 2 separate and distinct walls.Sigh. I guess we can't let you use the word wall either. Think line of effect blocking object. If a line of effect blocking object is made, part of wood, part of stone, part of iron, and part of force you cannot shoot the gap in the atoms between one material and the next. If it blocks line of effect it blocks line of effect.
If there is an "inch or so" of space there is an endpoint. Where 'object that blocks line effect' meets 'space not blocking line of effect' there you hand an endpoint. You effectively have an area to make a ranged attack through. And you would draw this but not having the small gap in the line. When you use W to represent a line of effect blocking object I assume its continuous. If it has holes in it, then you should state such.
Let's take your example. Suppose a wizard puts up a 5' wall in the space between the walls. This is a separate and distinct wall from the other 2 and may have an inch or so of space between them.If it connects to existing line of effect blocking objects its not separate and distinct. If you leave gaps between the new line of effect blocking object and the old line of effect blocking object then you have line of effect moving between the gaps. If you do not leave gaps then it become part of the same object line of effect blocking object regardless of what material it is made off.
I think you're quibbling semantics. So let's modify your example to show how silly you can get in not counting corners as blocks.
000|000
000|000
0000000
A00|00B
000|000In this particular case, I think it's pretty obvious that A and B have total cover relative to each other. Yet if we take the RAW, they would only have regular cover relative to each other since...
In your example, the attack would have to pass through a solid barrier as the attack is contacting the vertex at 90 degrees or less. As such it must pass through the barrier to continue. If it cannot, total cover is achieved.
I hope this helps. Years of mechanical drafting, paper style, helps when looking at these.
In example 2 the vertical line passes through the vertex and does not stop. To go from C to T you have to pass through the barrier. Total cover is clearly achieved.
As to the most recent example, again, a line has only one dimension. This is an important concept. The attack must come from within the square and end within another square. The line between squares has no thickness.
If the attack and the target are on the same row, the attack MUST be 90 degrees or less. This is not a game rule, this is the math of the game. It is only discussed here to show you how the rule and math coincide. In a similar fashion, there is nothing in the rules that state 1 + 1 = 2. That can only be achieved by following the rules of math.
Any attack that touches the vertex at more than 90 degrees cannot enter a square past that vertex that is in the same row as the original square.
Following the above mathematical rules, an attack that starts in square A must hit the barrier you depicted before it can end in square B. Likewise, an attack that starts in the square directly above A would NOT pass through the barrier before it reached any corner in the square directly above B.
I hope this helps.
It appears that you are taking to many data points.
In your example 1, no information from squares 1 2 3 4 5 or 6 is relevant. All of those squares are completely outside of the lines that are used to measure anything. The terrain in those squares affect nothing.
Square 7 to square 9 has a solid barrier in 8 that gives total over relevant to each other.
In your example 2, no information from squares 1 2 3 7 8 or 9 is relevant. All of those squares are completely outside of the lines that are used to measure anything. The terrain in those squares affect nothing.
Square 4 to square 6 has no barrier in 5 that gives any cover relevant to each other.
Your example 3 does not have numbers, but imagine them being numbered the same way. A = 1 and B = 16.
The best corner for the attacker is the top right corner of 1.
The best corner for the defender is the bottom left corner of 16.
Draw a line between the two.
Now draw a line between the top right corner of 1 to the top right corner of 16.
Shade in the area between the two and all of square 16. This is the only are that should be checked for cover. Not anything touching the lines, just things within the lines.
I think I figured out how to draw this so it should make sense. I will do so in a few hours when I get to my PC (working on iPad now).
A line that hits the vertex at greater than 90 degrees is pushed outside of anything behind it. If the line continues past the barrier, it can never get back into a square behind it. This is not a Pathfinder rule, this is geometry.
In this game, with this type of line, you can be on one side of the line, or the other. You cannot be on it. The line has no thickness. None, zilch.
The same line can indicate cover, or not cover. It depends on the side of the line you are on.
I think it will become more clear when I post graphics.
Example 2 could be as simple as a wizard putting a wall of stone or force up against an existing wall or maybe with an inch or so of spacing. That would be 2 separate and distinct walls.Sigh. I guess we can't let you use the word wall either. Think line of effect blocking object. If a line of effect blocking object is made, part of wood, part of stone, part of iron, and part of force you cannot shoot the gap in the atoms between one material and the next. If it blocks line of effect it blocks line of effect.
If there is an "inch or so" of space there is an endpoint. Where 'object that blocks line effect' meets 'space not blocking line of effect' there you hand an endpoint. You effectively have an area to make a ranged attack through. And you would draw this but not having the small gap in the line. When you use W to represent a line of effect blocking object I assume its continuous. If it has holes in it, then you should state such.
Let's take your example. Suppose a wizard puts up a 5' wall in the space between the walls. This is a separate and distinct wall from the other 2 and may have an inch or so of space between them.If it connects to existing line of effect blocking objects its not separate and distinct. If you leave gaps between the new line of effect blocking object and the old line of effect blocking object then you have line of effect moving between the gaps. If you do not leave gaps then it become part of the same object line of effect blocking object regardless of what material it is made off.
I think you're quibbling semantics. So let's modify your example to show how silly you can get in not counting corners as blocks.
000|000
000|000
0000000
A00|00B
000|000In this particular case, I think it's pretty obvious that A and B have total cover relative to each other. Yet if we take the RAW, they would only have regular cover relative to each other since...
This is not interpreting the rules as written. "To determine whether your target has concealment from your ranged attack, choose a corner of your square. If any line from this corner to any corner of the target's square passes through a square or border that provides concealment, the target has concealment."
I think when the rules say "pass through a border that provides concealment" they mean from one corner of a square to another corner. Just passing through the corner without passing through another corner does not pass through the border. This handles both the example at the top of this post and the one quoted in my reply.
If this is not clear, think of a coordinate axis with lines connecting points (0,0), (1,0), (1,1), and (0,1). The lines connecting these points count as the borders of the square. If a line goes through any two corners, it must either pass through the square itself (such as a line connecting 0,0 and 1,1) or through the border (a line through 0,0 and 0,1). Just passing through one corner does not pass through the square nor through a border of the square.
I see why you see it as an issue. You are counting the vertex as a 2 dimensional object when in fact, it only has on dimension. It is so small that it can only exist as an idea and not as a physical thing.
The vertex does not block anything. It is only the point of distinction of what does vs what does not.
You will see in about 2 hours when I draw it up.
Oh, that was a cut and paste and using iPad on a boat error.
The point was only 3 squares matter in each example. One example gives cover, the other does not.
What's so hard about lines? Seems simple enough to me. I made a diagram that I thought might help others to better understand (see link).
Blue lines represent the line. Sky blue are the squares effected by the line.
Note that the PC could easily have zapped the bottom right NPC simply by adjusting the angle of the line (call it a warning shot).
I have attached two graphics.
The first is the 16 square grid that was depicted earlier. The top right of square 10 is the corner in question. Grid
Here you can see the area that the attacker must check to see if there is cover. Three lines are drawn.
Line 1: From the top right of square 1 to the bottom left of square 16.
Line 2: From the top right of square 1 to the top right of square 16.
Line 3: From the top right of square 1 to the bottom right of square 16.
Those lines formed 4 points, all of which were connected to form a perimeter. The area was shaded yellow in the "Grid" graphic.
Square 10 is not indicated by the shaded area. There is nothing in it to check.
Now we shall zoom in to the corner in question. Zoom Each black square in the picture is a pixel in the line. Each square = 1/100 of an inch. As you can see, none of the line actually touches square 10 (the grey square in the bottom left of the picture). I did not draw the black squares. I just chose both ends of the line and my computer drew them at a dpi of 100.
I will show how it works on a straight line next.
For checking cover on the same row, we start with the same grid. Grid-2.
I added both situations to one grid.
The top situation shows a wall between the attacker and defender.
Three lines are drawn:
Line 1: From the bottom right of square 5 to the top left of square 7.
Line 2: From the bottom right of square 5 to the top right of square 7.
Line 3: From the bottom right of square 5 to the bottom right of square 7.
Those lines formed 4 points, all of which were connected to form a perimeter. The area was shaded yellow in the "Grid-2" graphic.
As you can see, it clearly passes through the wall and thus provides cover. As it must pass through the wall at all angles, it is total cover.
The bottom situation shows no wall between the attacker and defender.
Three lines are drawn:
Line 1: From the top right of square 9 to the top right of square 11.
Line 2: From the top right of square 9 to the bottom right of square 11.
Line 3: From the top right of square 9 to the bottom left of square 11.
Those lines formed 4 points, all of which were connected to form a perimeter. The area was shaded blue in the "Grid-2" graphic.
As you can see, it does not pass through a wall at all. Even though it touches the wall, it does not pass through it.
Here is the zoomed in graphic which again, each little square is 1/100 of an inch. Grid-2-zoom
I think this should show how the situations we discussed fit within the current rules.
Two corners that touch still touch. The line is not broken at the vertex. As such, your line to identify cover passes through a barrier and is stopped. If there is a tiny gap, then there is a tiny gap that the GM can adjudicate. It cannot be depicted by the battlemat.
As to walls touching walls, there is no space between them. If there is, they are not touching. If you want to get that detailed on your map where 1" = 60" and you are depicting a 1" gap, then understand that it would be drawn on the battlemat as a .017" gap, or roughly the thickness of 5 pieces of paper. My $70,000 CNC router only has a tolerance of about .005"
It is my opinion that you are looking for computer precision in a place that you are not going to find it. It takes me about 10 minutes to set up the grid to figure it out to the 1/100". Does your game need to be that precise?
At my table, any doubt goes in favor of the PCs. This is because they are the heroes and the NPCs don't get upset when they die.
I hope these drawings help.
No, that is not it at all. I do not see a vertex as a 2-dimensional object. Heck, I don't even see it as one-dimensional. A vertex is a point.
Let's zoom in a bit and expand.
------
------
AB##EF
CD##GH
------
------This is nine squares. One square of wall, two squares containing critters. Critter one has corners ABCD, critter two has corners EFGH.
Obviously, a solid wall between ABCD and EFGH provides total cover. So, pick a point, say, "A". We draw the lines (AE, AF, AG, AH). Two of them (AG, AH) unambiguously intersect the wall. Two of them (AE, AF) go along the edge of the wall. They never cross the wall. But that has to be total cover, because it is obvious that it is total cover.
Now consider:
--##--
--##--
AB--EF
CD--GH
--##--
--##--We draw the same four lines (AE, AF, AG, AH). Obviously, AG and AH do not intersect any walls, so we don't have total cover. Do we have partial cover? AE and AF are exactly the same as they were in the previous sample -- they are lines which go parallel to, but never cross, the side of the wall.
Does a line which overlaps, but does not cross, the side of a wall count as cover? If it does not, then the first example is only partial cover, even though it's obviously intended to be total cover. If it does, then the second example is partial cover, even though it's obviously intended to be no cover.
As a line has no thickness, it cannot overlap. It can only cross or not cross. It cannot do both. It is impossible.
What's so hard about lines? Seems simple enough to me. I made a diagram that I thought might help others to better understand (see link).
Blue lines represent the line. Sky blue are the squares effected by the line.
Note that the PC could easily have zapped the bottom right NPC simply by adjusting the angle of the line (call it a warning shot).
I think you are talking about lines as in spell areas and not lines as in cover?
Oh. The OP was talking about casters, so I thought...