The problem I have with the clipping is that it is harder to measure correctly. There will be players that'll try to use that rule and clip more than 50% of their silhouette because it's not possible for me from the other side of the table (and even in TTS) to prove that they tried to gain a movement advantage. With the "old" way of walking around corners, you will produce less errors while measuring. Imho, the intend for that squeeze through rule was to give models with big bases more access on the board, not to clip everything everywhere.
I am pretty convinced that the half base rule is only meant for narrow areas and not for cutting corners. Narrow areas come up from time to time while the corner situation would come up in every game. My main question if anyone would do this, how are you going to measure it properly? Also let's say you cut the corner of two connected thin walls , does your silhouette clip through the other side so I could draw some lof to it and declare an aro? That's a can of worms in my eyes. I would appreciate a rule clarification so that this is 100 % clear that it only applies to narrow areas. It's gamey and it feels unintended.
Doesn’t pivoting technically violate the rule about always measuring from the same part of the base? I say that because models have 360° LoF while moving, and may declare any facing at movement end—so for practical purposes you can visualize the model as becoming a silhouette cylinder at the start and reverting at the end with any facing. Measuring from any point other than the center of the base and pivoting around that point would result in either too much or too little distance traveled. To get uniform distance measurements from a math standpoint when comparing the center of the base to any other point, you have to “lock” the orientation of the silhouette and avoid pivoting it.
No. You can still measure from the same part of the base. Facing is irrelevant for this. Just imagine the base in the abstract. Like such, the fact that the facing may have changed from B to C doesn't change the point on the base you measure from.
If a true pivot where the point touching the building corner is essentially stationary troubles you, you could assume a counter-rotation to balance the movements of different parts. But at that point I think any sort of non-straight-line movement is going to give you the same problem. I don’t think the infinity rules are intended to ban movement on a curved path.
If you use the point in contact with the wall and want to change orientation to be in same situation in the other wall... it is perfectly RAW. The text says nothing about pivoting, just measuring, does it? What is more, you can get one extra base in straight movements, if you don't mind doing a 180: model aims to the left, use front point as anchor, moves all X to the left, but ends aiming to the right, the anchor point moved X (the center of base moved X+base diametre, obviously... but I picked a point :P ). RAW again. Combinations can get fun... someone with time probably can find all kind of maximized results "from Pa for orientation Oa to position Pb with orientation Ob, just claim this is your reference point". CB tried to fix "center point is hard", opened a different cancontainer of worms. Image URL has expired, or so says Farcebook.
That's an interesting answer. Follow up question to that is "how much crap on the ground merits a narrow alley; when do I have to go around and what can I squeeze between?" A box? A rock? A barbed wire that's 1,5" tall?
Something you can't vault over, I'd assume. A box or a rock is smaller than S2, so you don't squeeze through, you ignore it.
You don’t actually ignore it when vaulting. You can move over it without needing a Climb or Jump and don’t pay for the vertical movement. It does affect count as moving up and over for the purpose of LoF. You may want to squeeze through obstacles instead of vaulting them to keep your trooper’s head from popping up over a wall during the vault.
I read Nuada's "ignore" as "ignore it for the purposes of determining whether squeezing is required" but this is probably worth discussing briefly. I was thinking about it, and vaulting isn't really optional—if your model declares a move through the space occupied by something that's partially or fully the width of the base, but shorter than your Silhouette height, you vault it and move up briefly—you can't opt to vault and not move upwards—and the only way to avoid vaulting (and its vertical displacement) is to move around the obstacle. We know you aren't allowed to clip through corners when squeezing, but it's possible to shorten a corner if the corner is something low enough to vault across, like a railing. So the implication is basically: If you can vault, you vault. If you can't vault, you go around. If you have access to a narrow passageway and you can get all the way through it without stopping, you may use that option rather than go around. A wide passageway that doesn't require squeezing means you can't squeeze past things that could be solved by vaulting over them. You can squeeze and vault at the same time if needed (navigating a debris-strewn narrow passage comes to mind) but squeezing doesn't allow you to bypass vaulting obstacles, because the fact that you can vault them means they're disqualified from consideration of things that make a passage impassably narrower than your full base width, which is the effective requirement for squeezing. If you go back and re-read my comment, I talked about visualizing the model as becoming a cylindrical silhouette for the duration of the move. I probably could have been clearer, but I was explicitly talking about imagining the model (and base) in the abstract as the correct way to do it since having effectively 360 Visor while moving means model facing is irrelevant, you just pick a spot on the base and slide the model around without pivoting it for the purposes of measuring distance traveled, then rotate to final facing on arrival at the endpoint of movement. The little nubbin in the front center of the CB LoF bases is convenient for that, but it only works if you're moving in a straight line or locking the compass facing of the model and then ignoring facing for the duration of movement. *sigh* I probably should have included a picture. No, it's not about only moving in straight lines, it's about avoiding this: You can create extra distance travelled by picking a point on the base circumference and then rotating the base around that point and claiming that the model is somehow not using any movement to do that. For example, if you pick the front center of the base to move, move forward in a straight line 4", then without moving that point on the base you pivot the model to face back the way it came as if the reference point on the base were a hinge, you actually displaced the model ~5" instead of 4". Right? Mob's subsequent comment describes a form of that geometric shenanigan in action. I mean, sort of? That requires getting creative with your interpretation of the rule in conjunction with the example diagrams about measuring movement, because you end up violating the "model's MOV is the maximum distance travelled" rule in the process of misusing the "same point on the base" rule. A more persnickety way to write the latter would be "All movement distances are measured from an imaginary reference point in the center of the model's base. For convenience, you may displace that point in any direction so that you can measure from the edge of the model's base. The distance and direction between the original reference point and the displaced reference point must remain the same throughout movement, such that both reference points move the same distance in total." That is not an audience-friendly way to write a rule! I do think that some additional diagrams that show slightly more complicated examples would be helpful, the ones we have now are kind of over-minimalist.
I see; it was not clear to me that that's what you were meaning. Even if you don't factor any base rotation / pivoting in, you can still do better than moving fully clear of the corner and then making your next move at a sharp 90 degree angle to the previous line of movement. This is especially visible with large-based models. Basically, if your base ever stops touching the wall, you're paying more movement than you need to.
Visual reference: http://mammon.ghostlords.com/~jonathan/infinity/corner_move.png Red model moves fully clear. Blue model glides/pivots in contact with the wall at all times. Just from a visual comparison, it's clear that the red model moved further. Physically spinning the model is not required; that was just how I was trying to describe the concept.
