Recently, we had the following situation:

OHOOOO
OOOXFR

H = Ranged Hero, X = Rubble, O = Empty, F = Front half of a Hellhound, R = Rear half of the same Hellhound

Does H have LoS to R and can attack, i.e., does F block the LoS to R?

It is clear that if F was also Rubble, LoS from H to R would be blocked .

If F and R would be two one-space monsters, then LoS from H to R would not be blocked according to the ruling from the GLoAQ ("for LoS purposes, ignore figures that are not in LoS themselves").

Finally, we have the example from the basic rulebook on pg. 10: Sir Valadir has LoS to the nearer two spaces occupied by the ogre, but not to the farther two spaces, obviously because the nearer spaces of the same figure block LoS.

Based on this ruling, I would say that in my example, LoS from H to R is blocked by F.

Opinions? Are there any LoS rules I might have missed that would say something different?

This is just my personal opinion, but I calculate LoS on a space by space basis, even for large monsters. You don't need LoS to the entire figure, just one space it occupies, in order to attack it. Using the GLOAQ ruling, I would say F does not block LoS to R since F itself is not in LoS. Therefore the hero may attack the two-space monster, but he must target the R space to do so - range will be calculated to the R space, not the F space.

Edit: This may or may not be supported by the fact that attacks target spaces and not figures. Thus a single two-space monster or two one-space monsters should be treated the same for targetting purposes.

So wait, you are trying to say that the Hell Hound is blocking LOS to itself? I honestly don't see how that actually works.

The "northern spaces" of the ogre in the basic rulebook (pg.10) does block LoS to its "southern" spaces!

Parathion said:

That I can understand, but he can see the two other spaces near him. If you moved Sir Val back a space and put a rubble token right in front of him, he'd still have LOS to the other space. I don't know, something about that doesn't seem right that the Hell Hound would be blocking LOS to itself in that case. I fully understand why the Ogre blocks LOS to its "back" half in that example, but I'm not sure it works here for a Hell Hound.

I don't think that attacks or LOS dealing with spaces rather than figures helps us here. The only possible grounds for ruling that the rear square of the hellhound is within LOS is the GLoAQ ruling that you should ignore figures not in LOS when determining LOS. Since the hellhound isn't in LOS, it doesn't block LOS, therefore the hellhound is in LOS, therefore it blocks LOS, therefore it isn't in LOS...

In addition to being a questionable idea purely on gameplay grounds, the ruling in question is algorithmically unsound: it says you should modify your LOS calculations based on the result of your LOS calculations, which is circular. Even if you address the hellhound, it's a trap just waiting to spring on any future ability that affects LOS. This is just one of the reasons that I think it was a bad ruling and I completely ignore it in my games.

If you want to keep the ruling, you'll have to house rule this particular case. I've little doubt that anyone who supports the ruling in the first place will support an extension to allow the hero to target the hellhound here, thoughthat seems like exactly the sort of thing it's designed to allow.

Here is what we do, if it helps:

Take a piece of yarn or string. Hold one end down in the middle of the hero's square. Now hold the other end down in the middle of either one of the hell hound's squares. If either one of the resulting lines is unblocked by the rubble, then you have LoS to the hell hound and can target it.

We got this reasoning from the basic rulebook on p.9 about Determining LoS: "In order to attack a space, the attacker must have line of sight to it. In other words, the attacking figure must be able to trace an uninterrupted straight line from the center of its space to the center of the space it is targeting." In this instance, it is perfectly reasonable to target the "rear" half rather than the "front" half.

This is all, of course, only my opinion.

I'm not sure you followed the problem, Joorsh. I'm pretty sure everyone agrees that you can target either space of the hellhound and only need an uninterrupted line to the space you're targeting. However, in the example given, the line to the front half is blocked by the rubble, and the line to the back half is blocked by the front half (figures block LOS, even to other squares of themselves--see the diagram on page 10).

yes we have come to this problem before. and then we reached an agreement:

assuming that the helldog has two hearts in the very center of each space it occupies(only attacking its heart will cause damage), aiming at the front heart will be blocked by the rubble and aiming at the rear heart will be blocked by the front body of this dog(imagining that the dog uses its claw to deflect arrows). my point is that LOS is decided by whether you can see the center of the target space.

so the conclusion is "the hero do not have the LOS of the hellhound under this condition". this is our house rule, hope it might help.

To my way of thinking, attacks target spaces, not figures. To make an attack one must have LoS to the target (which is a space), therefore LoS must be determined to a space, not to a figure. In the OP's example, the hero has LoS to the R space precisely because he does not have LoS to the F space. He may attack space R, which happens to contain part of the hellhound, therefore the hellhound will take damage assuming the attack is successful.

I can understand how this situation may be confusing to some, but to me it seems clear. All the confusing arguments seem to center on the fact that F and R both contain part of the same figure, but this is irrelevant becuase you don't attack figures, you attack spaces.

Steve-O, if it was irrelevant what is in the spaces F and R, then you would allow LoS to R in the case that F was a space containing rubble ??

Parathion said:

It's not irrelevant what's in the spaces, it is irrelevant that both spaces contain part of the same figure. As I said above, you need LoS to the space. If F contained rubble then the hero would not have LoS to R.

Steve-O said:

But the ruling about being able to see through figures that are not in your LOS works with figures, not spaces. At least as written.

Without that ruling, it is very simple. But you don't have LOS.

There's a handy line of sight sheet at http://www.boardgamegeek.com/filepage/35882 which shows all traceable line of sight within 5 spaces. It's not official, but it's rather exhaustive, considering that tracing los is just that - tracing an uninterrupted line from the center of the hero's space to the center of the target space, and there's only one way to do that. According to that sheet, your hero doesn't have los to either the front or back half of the hound because the rubble token interrupts any lines starting from the center of the hero's space. So in this particular example, it's not the hellhound blocking los to its rear, it's the rubble. If there was no rubble, then yes, the front half of the hound would block los to its rear half because figures do block line of sight as some have pointed out already. You would have to target the front half instead.

_Loki_ said:

True. I was focused on the description of the situation and didn't look at the diagram carefully. I assume Parathion was actually interested in what would happen if you didn't have that empty column between the hero and the rubble, like so:

HOOO
OXFR

In that case, tracing LOS from H to R requires that you go through F but not through X.

Oh, sorry - my example in the OP is wrong indeed.

I mean exactly the situation Antistone depicted.

I hope all the answers so far are still valid, though.

Antistone said:

Figures do block LoS (except spawning), but they block LoS to spaces, not to other figures. This is the only way I can consolidate all the rules for determining LoS, because otherwise you would be drawing LoS to something other than what you are targetting. If you can draw LoS to a figure but can't draw LoS to the space its in, how do you know whether or not the attack is valid? For that matter, how do we define what blocks LoS to a figure and what blocks LoS to a space? I choose to believe there is only one type of LoS. That LoS is drawn to a space because spaces are what you need to target. This is a simple rule that answers this OP's question and many others I see on these boards quite simply.

Again, this is just my opinion. You don't have to agree with it. I prefer to keep things simple when I play, so I look for a simple way to break down the rulings that have been made into a uniform system. If you prefer to split hairs on how LoS reacts to figures versus other obstacles or spaces themselves, that's your business.

On that note, I think I'm done here. I've explained my position as clearly as I can, so if you still can't understand it you'll just have to find a different answer. Good day gentlemen.

Um...sorry, but your clarification helps me not at all.

Also, there are clearly at least 2 or 3 types of LOS/LOE in Descent (possibly more), even though they try to pretend there's only one. There's LOS for purposes of attacks (which is normally the most restrictive, but some abilities like Precision and Crack Shot only affect LOS for purposes of attacks), there's LOS for purposes of spawning (which isn't blocked by figures), there's LOS for purposes of Blast AoE (which also isn't blocked by figures, and works backwards by RAW), there's LOS for sweep attacks (which has yet a third, independent rule that says it ignores figures), and there's the rules for hitting people with Bolt/Breath attacks (which isn't called LOS, but it's yet another rule for determining whether you can trace a valid path from source to target). There's also "basic" LOS, for special rules and abilities that specifically requrie LOS but don't specify any further details, which works like LOS for attacks except it doesn't get the specific "for attacks" abilities like Precision.

But I don't see how the different kinds of LOS are relevant to this discussion, either, since I don't think there's any dispute over which type is being used.