mahkra said:

Yeah, but then you're piece-mealing the rules. You're trying to use "normal LoS rules" and then switch to use the FAQ ruing and ignore blocked figures. You should either use RAW LoS or FAQ LoS, not intermix them.

By FAQ, the green zombies are visible. By RAW, none of the zombies are visible.

-shnar

As I pointed out earlier, the GLoAQ ruling prescribes a result, not an algorithm, which makes it prone to inconsistencies.

It is true that if you decide that enemy #2 is not within line-of-sight, this will eventually lead you to a contradiction with one of the rules (because either you decide that enemy #1 doesn't block LOS, which makes your assignment for #2 invalid, or you decide that enemy #1 doesn't block LOS, which is invalid when you check his LOS and get stuck on rubble).

But order of evaluation (or rather, starting assumptions) can matter, because in order to know whether monster A is within LOS you might need to know whether monster B is within LOS, which might require that you know whether yet another monster is within LOS, and there's no inherent reason that these dependencies can't be circular (they aren't for "normal" LOS with one-square monsters, but there are LOS-modifying abilities in the game, there may be more in the future, and if you don't treat large monsters as several separate small monsters you can get circular dependencies really easily).

For example, suppose heroes A and B are facing down monsters M and N across rubble X, and hero A has Spiritwalker:

X A
X
M
N
X
X B

You might decide that LOS is calculated entirely independently for "hero A attacking from his own position" and "hero A attacking from hero B's position using Spiritwalker." But by way of example, let's say you don't make that decision.

Then tracing from A, M is behind rubble, and N is behind M, so whether or not you can see N depends on whether or not you can see M. But tracing from B, N is behind rubble, and M is behind N, so whether you can see M depends on whether or not you can see N.

You can arbitrarily decide that you can see M but not N, OR that you can see N but not M, and there's no rule that says you're wrong in either case. If you can see M, it blocks your line of sight to N, so N isn't in LOS, so it doesn't block your LOS to M...and vice versa if you start the other way.

You can make up a rule for this, certainly. But it's just an example of the kind of problem that the rule promotes because it doesn't give us a well-defined procedure for enforcing itself.

shnar said:

I'm mixing the rulings intentionally. The FAQ ruling does not change how LoS works in general; it changes how LoS works for the specific instance of determining LoS for an attack . I'm not sure if tracing LoS to #2 should be considered part of determining LoS for an attack when you're attacking #7.

mahkra said:

That...is a valid and very frightening point.

And that's what I'm saying you shouldn't do, mix the different rulings like that. The FAQ ruling isn't for an attack, it's just for LOS, which happens to be use for Attacks, but could be used for anything else.

Coming from Space Hulk, which had the same ruling, I used to be a big fan of this. It made perfect sense to me, that if the creature was behind something and you couldn't see it, how could it block LOS to the creature behind it? Even in long "chains" it still made sense that if you couldn't see the center of the target square, then the creature was cowering over to one side and simply not visible (which is why you wouldn't include that creature's space when looking past it).

Space Hulk though only had 1x1 creatures, but once you apply this rule to 1x2 or 2x2 or 3x2 creatures, you get weird situations like described above. If we redid the Demon LOS to this:

I can see how the marine cannot see either Red Zombie, and how neither of those Red Zombies are blocking LOS to the Green Zombie. There's effectively a "gap" between the two Reds as they are hugging the north and south parts of their squares, allowing the marine to see the green fatty. But if the creatures are *not* separated like this, and you have this situation:

How is the marine seeing *through* the pinky? So large creatures makes what was an easy-to-understand rule (at least for myself) no longer so easy to understand...

-shnar

Looking at my diagram, I think I put the green zombie too close to the red ones, as I think he's outside LOS anyways. If you move him one space East, then the diagram works

-shnar

Antistone said:

I still don't understand this interpretation. To determine LOS for #7, you have to look at #6. To determine LOS at #6, you have to look at #5, which eventually leads you to #2. Why would you arbitrarily stop somewhere and ignore everything else that's in LOS?

-shnar

shnar said:

Read the ruling again:

" For purposes of determining LoS for an attack , ignore figures that are not in LoS themselves."

It's not a generic rule for determining any LOS--it's only when you're determining LOS for an attack , and not, say, Runemaster Thorn's teleport move. And you ignore figures that are "not in LOS".

Now, it could be that the author intended the first part of the sentence to set up the context of the remark, and that all references to LOS were supposed to be constrained by that context. That wouldn't even be much of a stretch. But strictly speaking, that's not what it says. What it says is that tracing "LOS for an attack " ignores figures that are "not in [generic] LOS", which means that it even ignores figures that are within LOS for purposes of an attack if they are not within LOS for other purposes .

This is all, of course, more confusing than it has to be, because the Descent writers like to pretend that "line-of-sight" is a well-defined monolithic concept, when they actually have like 3-5 different ways of determining line-of-sight depending on the purpose for which it is being used and none of them are clearly defined anywhere.

Admittedly, we've already established that doing exactly what it says is totally unworkable, since there are actually no rules for determining whether a figure is in LOS in the first place, and it leads trivially to paradoxes with large monsters. So we should probably be discussing "what house rule produces the best results?" instead of "what precisely did that one designer mean when he made that one informal, vaguely-worded, snap ruling that one time?"

Antistone said:

Isn't that what Descent is all about?

I see now where the mix up is coming from. Silly worded ruling.

-shnar

Ironically, Mahkra's reading of the rule is actually cleaner, because at least it's not recursive , so it's possible to apply it without getting tangled up in paradoxes.

It also produces less intuitive results, which is a major problem for a new, complicated ruling whose sole purpose is to make the results more intuitive, but it is cleaner .

The way I see it, there are two 'reasonable' house rule options:

1 - Strictly follow the RaW from the instruction book, ignoring the FAQ ruling.

2 - Make a judgement call on a case-by-case basis; i.e. allow an attack if it seems appropriate.

The FAQ ruling is just too confusing, even without considering attacks going 'through' large monsters. For instance, take the Space Hulk example above. While there is a logical reason to allow the attack on the green baddie, what happens when you remove the two walls? The walls did not affect LoS to the green baddie at all, but removing them makes it impossible to target him because the red baddies are now in the way. But nothing moved, so why are they in the way? Why can't you still shoot 'in between' them?

Likewise, let's go back to the row of 7 zombies hiding behind a 1x1 obstruction. According to shnar's calculations, only #2 and #5 can be targeted. But what if we remove #2? This makes it so that only #3 is targetable. LoS from the hero to #5 actually goes through #2, but you can only attack #5 if #2 is in the way?

The 'problems' I see with the original RaW algorithm can be thematically explained by saying a hero needs a "clean shot" at the monster's body (not just limbs) and that a high-angle shot against a large monster would just glance off without causing appreciable damage.

mahkra said:

Oops. This is wrong. The green zombies should have been #2 and #6, not #5. (The fact that it's taken this long for any of us to notice that error just goes to emphasize how difficult it is to determine LoS under that FAQ ruling.) When you remove #2, you can now target #3 and can no longer target #6.

After listening to all the points here, I have to change my stance. The GLoAQ ruling does make sense on a certain level to me, but ignoring it does make for cleaner, simpler rules. With everything else going on in Descent, K.I.S.S is a welcome concept for me.

mahkra said:

I have no problem with this rule for 1x1 creatures. I just picture the 2 red zombies as "hiding" just enough out of LOS you can't see them. But since they're doing that, you can then see "through" them. Remove the walls, the zombies have nothing to hide between, so they are visible and can now be targeted and the green zombie is "hiding" behind the reds.

But eh, just come to a general ruling with your group and play it that way. It is, after all, just a game, no need to get too worked up over it.

-shnar

shnar said:

This reasoning gives me an idea - what if the FAQ ruling is interpreted this way:

Ignore figures out of LoS due to an obstruction (wall, rubble, etc not other figures) when determining LoS for an attack. LoS may still never be traced between the spaces covered by a large monster.

This would simplify LoS determination, but I'm not sure yet if it would give more logical results.

With the line of zombies, #1 is hiding behind cover, so #2 is targetable. #3 through #7 are all obstructed by other zombies.

Thematically, this assumes that when you are about to attack, figures actively hide behind fixed cover, which could reveal a figure behind. Other figures are not fixed cover, though, so do not entice a monster to 'hug the edge' of its space. (This exception would only apply to "LoS for an attack", not "generic LoS", because the monster only hugs the cover when a weapon is raised for an attack.)