Switch Theme:

Share on facebook Share on Twitter Submit to Reddit  [RSS] 

Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/11 19:27:56


Post by: davou


So I was playing a game, and came down in a drop pod right next to a line of tanks. I scattered toward the edge, and landed partially off, but still clipped the tank that I had scattered through by a good margin; we rolled for it and I took a mishap, but I was wondering what the general consencus would be.

Do I get to activate the inertial guidance system rule, or does having landed partially off the table go into immediate effect first?


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/11 19:39:01


Post by: rigeld2


Off the table is off the table IMO.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/11 19:46:43


Post by: Spetulhu


I'm inclined to agree - off the table triggers a mishap. Having the guidance system save you from it would be an advantage and in general you should avoid that when a rule could go both ways.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/11 19:57:48


Post by: B0B MaRlEy


Would the position you'd be once the scatter is reduced (off the tanks) still touching the board's edge? If so you'd mishap.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/11 20:10:17


Post by: Tye_Informer


If the scatter puts you on the tank (or another model), then you reduce scatter until you are no longer on the model. If the scatter puts you off the table, and not touching another vehicle, then inertial guidance can't reduce the scatter since it only activates if you would land on the model, even if there is a model on the line from your intended landing spot to where you ended up.

For example:
* = Intended spot
- = line from intended spot to scatter point.
X= Friendly model
O = scatter point (in this case, let's make it a 12" scatter, each character is 1")
| = end of table.

*------X----O|

This would put the center of the drop pod at the edge of the table, so a good 2-3" is hanging off the end of the table. In this case, the drop pod is not on impassable terrain or another model so inertial guidance doesn't trigger.

How about if the friendly model was 2" away from the edge, so the drop pod is now hanging off the edge but on top of a model. Because the scatter is on top of a model, intertial guidance decreases the scatter to the point where the model is no longer under the drop pod, even though that pulls the drop pod over the top of the model and away from the table edge.

*----------XO|
becomes
*--------O--X-|


As you can see though, it would take a pretty large scatter to get in this situation, assuming you picked a legal landing point in the first place.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/11 20:20:14


Post by: davou


yeah, The scatter was about 9 inches. At around 4 I hit the russ, then at 8 I was partially off the table, but still sitting well on top of the russ, if I reduced to get 'off' the tank, I would have been completely on the table.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/11 20:46:34


Post by: DeathReaper


 angelofvengeance wrote:
Drop Pod mishap. End of.

Not in the situation the OP described, in light of the new evidence...
 davou wrote:
yeah, The scatter was about 9 inches. At around 4 I hit the russ, then at 8 I was partially off the table, but still sitting well on top of the russ, if I reduced to get 'off' the tank, I would have been completely on the table.

Then the Pod would not have mishapped.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/11 20:50:20


Post by: disdamn


I would agree that it's a mishap


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/11 20:51:54


Post by: DeathReaper


disdamn wrote:
I would agree that it's a mishap

Except it is not a mishap once he reduced the scatter.

re-read this post of his:

 davou wrote:
yeah, The scatter was about 9 inches. At around 4 I hit the russ, then at 8 I was partially off the table, but still sitting well on top of the russ, if I reduced to get 'off' the tank, I would have been completely on the table.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/11 20:55:36


Post by: rigeld2


He doesn't have permission to reduce mishap for ending off the table.

He's mishapping for two reasons. Drop Pods allow him to reduce for one, but not the other.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/11 21:05:08


Post by: pizzaguardian


So the drop pod will be in a superposition of both not being on the table for scattering outside and on the table for scattering on a unit ?

Do explain further please


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/11 21:12:24


Post by: rigeld2


 pizzaguardian wrote:
So the drop pod will be in a superposition of both not being on the table for scattering outside and on the table for scattering on a unit ?

Do explain further please

It's not a difficult concept.
When a Land Raider arrives from Reserves the embarked unit can get out. They still cannot charge, however, because there are two restrictions in place - disembarking from a transport (lifted because of Assault Vehicle), arriving from Reserves.
When a Drop Pod arrives and is over a unit and off the table it has mishapped. There are two reasons - final placement being over a unit (lifted because of IGS) and final placement being off the board.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/11 21:14:59


Post by: Eihnlazer


the final placement isnt off the board, cause the scatter gets reduced.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/11 21:17:16


Post by: rigeld2


Eihnlazer wrote:
the final placement isnt off the board, cause the scatter gets reduced.

Read the IGS rule again.

If you scatter off the board, you suffer a mishap. The OP scattered off the board. He just so happened to also be on top of a unit, but he still mishaps.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/11 21:21:14


Post by: Eihnlazer


Does it say:

If the final position of your scatter lies off the board you mishap
or
If the scatter roll places the pod off the board you mishap.


Because those would end in different rulings.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/11 21:27:11


Post by: Happyjew


If the Drop Pod scatters on top of impassable terrain or another model, reduce the scatter. Note that if a Drop Pod scatters off the edge of the board it suffers a Mishap.

I paraphrased the SM codex.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/11 21:35:28


Post by: Tye_Informer


rigeld2 wrote:
Eihnlazer wrote:
the final placement isnt off the board, cause the scatter gets reduced.

Read the IGS rule again.

If you scatter off the board, you suffer a mishap. The OP scattered off the board. He just so happened to also be on top of a unit, but he still mishaps.


As much as I hate being in a situation where I have to defend Space Marines rules...... In order for the rule to be interpreted in the way you are describing, then the IGS rule would never come into play. You have the mishap rule being triggered by the scatter dice, but before the final position is determined. That would mean if you scatter into a unit, you mishap, roll on the mishap table, and reduce your scatter. I don't know how that could possible be resolved for any roll other than your opponent getting to place you, and then you could reduce the scatter.

The mishap is triggered by the final position, not the scatter die roll. You roll the scatter dice, measure the location, determine if you are on impassable terrain or another model, apply relevant rules to determine the final location. Then you can resolve the mishap, if it's still relevant.

I would love it if the IGS rule was nerfed in some way (and ATSKNF, but that's a different story) but it's not going to be. Space Marines are the Emperors gift to the world and GW is their prophet.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/11 21:40:25


Post by: Fragile


Tye_Informer wrote:
The mishap is triggered by the final position, not the scatter die roll. You roll the scatter dice, measure the location, determine if you are on impassable terrain or another model, apply relevant rules to determine the final location. Then you can resolve the mishap, if it's still relevant..


Measure and apply the relevant rules. Ok, Pod scattered off the table, therefore you apply the Mishap rule. Pod is on top of an enemy, therefore you apply the IGS rule. You now have two rules that you have to apply. See the contradiction?


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/11 22:57:55


Post by: pizzaguardian


rigeld2 wrote:
 pizzaguardian wrote:
So the drop pod will be in a superposition of both not being on the table for scattering outside and on the table for scattering on a unit ?

Do explain further please

It's not a difficult concept.
When a Land Raider arrives from Reserves the embarked unit can get out. They still cannot charge, however, because there are two restrictions in place - disembarking from a transport (lifted because of Assault Vehicle), arriving from Reserves.
When a Drop Pod arrives and is over a unit and off the table it has mishapped. There are two reasons - final placement being over a unit (lifted because of IGS) and final placement being off the board.


Not the same thing actually, since IGS actually requires to you take an action "reducing scatter", while you dont have to assault when you disembark from an assault vehicle which the reserves rule would interfere and create a similar situation here. If assaulting was mandatory then yes.

Not the same thing at all.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/11 23:35:50


Post by: Jimsolo


I play Drop Pods, and I'm tempted to agree with rigeld that it mishaps.

However, I'm less than certain. If it came up in a game, I think I would point out the two relevant rules and let my opponent decide how it should play out, in the interests of good sportsmanship.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/12 00:11:45


Post by: rigeld2


Tye_Informer wrote:
The mishap is triggered by the final position, not the scatter die roll. You roll the scatter dice, measure the location, determine if you are on impassable terrain or another model, apply relevant rules to determine the final location. Then you can resolve the mishap, if it's still relevant.

Perhaps you should read IGS. The rule actually says that if the pod scatters off the board it mishaps. Using the same phrasing as if you scatter over a unit.

So IGS says you mishap.

I would love it if the IGS rule was nerfed in some way (and ATSKNF, but that's a different story) but it's not going to be. Space Marines are the Emperors gift to the world and GW is their prophet.

Adds nothing to the conversation. I'm discussing without bias.


Automatically Appended Next Post:
 pizzaguardian wrote:
rigeld2 wrote:
 pizzaguardian wrote:
So the drop pod will be in a superposition of both not being on the table for scattering outside and on the table for scattering on a unit ?

Do explain further please

It's not a difficult concept.
When a Land Raider arrives from Reserves the embarked unit can get out. They still cannot charge, however, because there are two restrictions in place - disembarking from a transport (lifted because of Assault Vehicle), arriving from Reserves.
When a Drop Pod arrives and is over a unit and off the table it has mishapped. There are two reasons - final placement being over a unit (lifted because of IGS) and final placement being off the board.


Not the same thing actually, since IGS actually requires to you take an action "reducing scatter", while you dont have to assault when you disembark from an assault vehicle which the reserves rule would interfere and create a similar situation here. If assaulting was mandatory then yes.

Not the same thing at all.

... They are the same thing. You have 2 restrictions in both cases and a rule allowing you to ignore a restriction in both cases. Voluntary or not is irrelevant.


Automatically Appended Next Post:
Eihnlazer wrote:
Does it say:

If the final position of your scatter lies off the board you mishap
or
If the scatter roll places the pod off the board you mishap.


Because those would end in different rulings.

If the pod scatters onto impassable terrain, reduce scatter. If the pod scatters off the board, Mishap.

Same wording in both places so either a) IGS does nothing or b) it mishaps if you're off the board regardless of what you're over on the board.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/12 01:49:41


Post by: DeathReaper


rigeld2 wrote:
He doesn't have permission to reduce mishap for ending off the table.

He's mishapping for two reasons. Drop Pods allow him to reduce for one, but not the other.

He fulfills the condition for reducing the scatter, so you reduce the scatter as it landed on an enemy model.

In the situation the OP laid out there is no mishap because you have to reduce the scatter because of the vehicle.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/12 01:51:21


Post by: Fragile


Only if you ignore the rule that says that if you scatter off the board, you mishap. That is also part of the IGS rule.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/12 01:54:33


Post by: Happyjew


 DeathReaper wrote:
rigeld2 wrote:
He doesn't have permission to reduce mishap for ending off the table.

He's mishapping for two reasons. Drop Pods allow him to reduce for one, but not the other.

He fulfills the condition for reducing the scatter, so you reduce the scatter as it landed on an enemy model.

In the situation the OP laid out there is no mishap because you have to reduce the scatter because of the vehicle.


Except, per the rules of IGS since it scattered off the table it does mishap. So it both mishaps, and reduces the scatter.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/12 01:54:36


Post by: DeathReaper


Fragile wrote:
Only if you ignore the rule that says that if you scatter off the board, you mishap. That is also part of the IGS rule.

That's not a rule...

At least not one I am seeing in the Space Wolves codex.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/12 01:56:58


Post by: Happyjew


Ah. So we have codex conflict. That explains it.

6th edition has a note in IGS that says you mishap if you scatter off the table. Wolves being an older codex, may nit have the same wording.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/12 01:58:50


Post by: rigeld2


 DeathReaper wrote:
Fragile wrote:
Only if you ignore the rule that says that if you scatter off the board, you mishap. That is also part of the IGS rule.

That's not a rule...

At least not one I am seeing in the Space Wolves codex.

It is part of the Space Marine codex.
Yay for consistent inconsistency with drop pod rules.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/12 02:05:11


Post by: Rapture


This is a fun one because it is novel, but I don't think it is that complicated.

It hinges on one key question. When determining mishap, do we ask:
1) Where did the deepstriking unit land?
or
2) Where did the deepstriking unit scatter to?

I don't have my book, so I can't tell you. But, if the answer is 1, then the drop pod did not mishap as, regardless of where it scattered to, the IGS rule pulled it back and made it land on the table. If the answer is 2, then there was a mishap as mishap is determined not from the actual landing position but only by the scatter position.



Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/12 02:22:05


Post by: Fragile


Well to answer those, the IGS rule says that if you "scatter" off the board, you mishap.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/12 02:23:08


Post by: Happyjew


The 6th edition IGS rule says if you scatter off the table you mishap. The 5th(?) edition codex is silent on the matter.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/12 04:27:16


Post by: Imperator_Class


You reduce scatter after you scatter, obviously. Which means they scatter off the board, which incurs a mishap, then reduces it. IGS makes you take the mishap.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/12 08:47:46


Post by: Mythra


2 things are happening at once. The players who turn decides which happens first. if it were my turn I would choose to have the distance reduced and not mishap.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/12 08:51:54


Post by: Mywik


 Mythra wrote:
2 things are happening at once. The players who turn decides which happens first. if it were my turn I would choose to have the distance reduced and not mishap.


This only applies when both players have something to do at the same time. This is not the case here.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/12 10:22:46


Post by: DOOMONYOU


 DeathReaper wrote:
Fragile wrote:
Only if you ignore the rule that says that if you scatter off the board, you mishap. That is also part of the IGS rule.

That's not a rule...

At least not one I am seeing in the Space Wolves codex.


Because space wolves are AWESOME!

I would side that it reduces the scatter from being over the tank, The rule doesn't say the Initial scatter just the scatter, so to me that means where it eventually lands after IGS. But probably a roll off is best till the sucker is FAQ'ed


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/12 13:07:32


Post by: Tigurius


Ok,

So the drop pod is both on another unit, where it cannot land it's also scattered partially off the table. Both have happened simultaneously.

It's both scatter reduced and mis-happed.

Schrodinger's Drop pod.



Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/12 14:23:38


Post by: OIIIIIIO


I do not use Drop pods but if this were to come up when playing against someone who does, I would suggest that if it is over a tank and can not be deployed (set on the table) then you reduce the scatter. If it could be deployed (actually set on table) and it breaks the plane of the edge of the table it would be a mishap.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/12 14:30:39


Post by: Gideon999


What we have here guys is an order of operations issue. I can't see any rules anywhere that state which order you check the conditions in.

For example, we have said that the pod is both on an enemy model AND off the table, so the question should be which condition is checked first, wouldn't it?

If you check for off the table first, then it mishaps. If you check for impassable/other models first, then you reduce the scatter.

I play pods a lot too and I have to admit I have never had this come up so I am curious to see if there is an official ruling myself....


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/12 14:37:31


Post by: OIIIIIIO


Only thing that I can see for order of operations is that in the book, it has it written to reduce scatter in the first sentence, the second sentence covers the mishap.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/12 14:40:26


Post by: Haighus


 Gideon999 wrote:
What we have here guys is an order of operations issue. I can't see any rules anywhere that state which order you check the conditions in.

For example, we have said that the pod is both on an enemy model AND off the table, so the question should be which condition is checked first, wouldn't it?

If you check for off the table first, then it mishaps. If you check for impassable/other models first, then you reduce the scatter.

I play pods a lot too and I have to admit I have never had this come up so I am curious to see if there is an official ruling myself....

If this is a case, then a D6 should be rolled to decide which takes precedence each time and it therefore mishaps on a 4+, is scatter reduced otherwise.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/12 15:19:57


Post by: nosferatu1001


Neither takes precedence; you have 2 rules one of which leads to a mishap, the other to no mishap

You, either way, have a mishap


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/12 15:40:09


Post by: Pyriel-


Final position (scatter was lowered to no longer end up of the table) is not of table.
No way in hell I would call that a mishap no matter on which side of the game I was on!


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/12 15:45:15


Post by: DogOfWar


As far as RAW is concerned, I was in the mishap camp at first, but now I'm leaning the other way.

RAW - The way I see it, a deep strike mishap can logically only occur after the final placement of the model (or models) has been established. This must happen after the scatter has been fully determined or any rule that reduced or modified the scatter would be pointless (we don't determine if a blast template hits until we reduce the scatter by the BS of the firer, for example).

In the OP's example, the DP's initial placement was both on impassable terrain and off the table. Since the IGS rule specifically addresses the issue of impassable terrain, this rule comes into effect and the DP's scatter distance is reduced until it can be legally placed (or not, as the case may be, depending on your view of the initial DS placement rules). Once the final placement it determined, the check to see if the DP is still either in impassable terrain or off the table is made and a determination is made on whether a mishap occurred or not.

I don't believe this 'violates' a rule because both the IGS and the mishap rules have been satisfied (the former due to the initial scatter of the pod and the latter due to the final determination of the scatter).

HIWPI - The DP should probably mishap despite the wording of IGS. Anyone who manages to place a DP such that it scatters off the table—especially with pre-measuring and the IGS rule to protect them—really deserves to mishap... but that's another discussion.

DoW


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/12 15:52:35


Post by: helgrenze


 davou wrote:
yeah, The scatter was about 9 inches. At around 4 I hit the russ, then at 8 I was partially off the table, but still sitting well on top of the russ, if I reduced to get 'off' the tank, I would have been completely on the table.


It's a mishap.
Look at the distances.... there is a scatter of "about 9 inches", at 8 inches, it is "partially off the table". There is one more inch to go on the scatter before determining the final position. Since the Russ is hit at 4 inches, the table edge at between 6 and 8, the final inch would move the pod off the table by more than half its footprint..... thus mishap.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/12 16:00:04


Post by: DarthOvious


In the end it all boils down what occurs first. The mishap for being off the table or reducing the scatter because you're over another unit. Also the wording of the IGS rule will be important as well where it mentions a mishap for being off the table.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/12 16:31:14


Post by: liturgies of blood


 Mywik wrote:
 Mythra wrote:
2 things are happening at once. The players who turn decides which happens first. if it were my turn I would choose to have the distance reduced and not mishap.


This only applies when both players have something to do at the same time. This is not the case here.

Nope sorry this is wrong. When numerous things go off in your turn at the same time you decide the order of opperations. That way I can choose in what order I bring in reserves, cast maledictions and blessings and roll for outflank. There is a clear part in the rules on it and even an FAQ to boot.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/12 18:28:40


Post by: davou


 Pyriel- wrote:
Final position (scatter was lowered to no longer end up of the table) is not of table.
No way in hell I would call that a mishap no matter on which side of the game I was on!


Thats the thing; final position is not 'scatter'.

its another one of those words that need a definition.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/12 18:32:00


Post by: White Ninja


Inertial Guidance System: Should a Drop Pod scatter on top
of impassable terrain or another model (friend or foe) then
reduce the scatter distance by the minimum required in
order to avoid the obstacle. Note that if a Drop Pod scatters
off the edge of the board, it suffers a Deep Strike Mishap as
per the Warhammer 40,000 rule book.

Directly from the new marine book. Following this rule because you do have to follow it in its entirety and not just the parts you like you will need to first check if it is on top of an enemy model and then before you do anything the rest of the rule would kick in and tell you you misshape. The mishap must still happen. This is not a case of two rules interacting and trying to figure out which goes first this is a case of one rule with two clauses triggering and since the second has triggered it must be resolved.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/12 18:37:52


Post by: raiden


Guys, if it cannot be placed (IE, its partly on top of the tank) then it does NOT LAND THERE. you reduce the scatter, then see if it is off the board (most likely not) its really not that hard. you determine mishaps after IGS takes effect, otherwise their is no real point to IGS. if you scatter off the board, or on board edge w/out landing on another model, THEN you mishap as IGS does NOT take effect.


Automatically Appended Next Post:
 davou wrote:
 Pyriel- wrote:
Final position (scatter was lowered to no longer end up of the table) is not of table.
No way in hell I would call that a mishap no matter on which side of the game I was on!


Thats the thing; final position is not 'scatter'.

its another one of those words that need a definition.


as I said, you determine "scatter" results after IGS takes effect.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/12 18:39:57


Post by: nosferatu1001


Raided - retread, note that scattering off the board triggers a mishap, regardless of the presence of other models

It isn't that hard when you read ALL the rule.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/12 18:41:51


Post by: juraigamer


If you go off the board, unless it reduces scatter when you do that, you mishap. You can't have 1/2 of a rule work.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/12 18:42:18


Post by: raiden


but its not the actual scatter, because IGS takes effect before mishaps do (via REDUCING) the scatter.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/12 18:43:36


Post by: rigeld2


 raiden wrote:
but its not the actual scatter, because IGS takes effect before mishaps do (via REDUCING) the scatter.

So you're just ignoring half of the quoted rule then?
Cool story.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/12 18:48:31


Post by: raiden


rigeld2 wrote:
 raiden wrote:
but its not the actual scatter, because IGS takes effect before mishaps do (via REDUCING) the scatter.

So you're just ignoring half of the quoted rule then?
Cool story.


no, if the pod scatters off the table it takes a mishap, but because it landed half on a model (the tank in the OP situation) IGS takes effect reducing the pods scatter so it does not scatter off the table edge. because it CANNOT land on another model therefore would not even be ALLOWED to land their to begin with.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/12 18:51:45


Post by: rigeld2


 raiden wrote:
rigeld2 wrote:
 raiden wrote:
but its not the actual scatter, because IGS takes effect before mishaps do (via REDUCING) the scatter.

So you're just ignoring half of the quoted rule then?
Cool story.


no, if the pod scatters off the table it takes a mishap, but because it landed half on a model (the tank in the OP situation) IGS takes effect reducing the pods scatter so it does not scatter off the table edge. because it CANNOT land on another model therefore would not even be ALLOWED to land their to begin with.

If it scatters on top of a model, reduce.
If it scatters off the table, mishap.

The same wording is used for both. You're attempting to follow one and not the other. Please explain why.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/12 19:02:26


Post by: raiden


rigeld2 wrote:
 raiden wrote:
rigeld2 wrote:
 raiden wrote:
but its not the actual scatter, because IGS takes effect before mishaps do (via REDUCING) the scatter.

So you're just ignoring half of the quoted rule then?
Cool story.


no, if the pod scatters off the table it takes a mishap, but because it landed half on a model (the tank in the OP situation) IGS takes effect reducing the pods scatter so it does not scatter off the table edge. because it CANNOT land on another model therefore would not even be ALLOWED to land their to begin with.

If it scatters on top of a model, reduce.
If it scatters off the table, mishap.

The same wording is used for both. You're attempting to follow one and not the other. Please explain why.


the reason it doesn't mishap, as I have said several times now.. is because the IGS takes effect for it being on top of a model and therefore reduces the scatter, before determining mishaps. if, the pod goes off the edge w/out landing on a model. it mishaps as no IGS takes place. If pod reduces its scatter, and lands off the edge, it mishaps. the OP situation doesn't trigger the mishap because 1) it is on top of a model, which is not allowed, 2) IGS takes place reducing the scatter because it is on top of the model after reducing scatter if it is still over the board edge, it mishaps.

are you trying to say that mishaps happen before scatter is reduced?


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/12 19:04:44


Post by: rigeld2


 raiden wrote:
are you trying to say that mishaps happen before scatter is reduced?

I'm trying to say that the IGS rule says, and I'll quote it again so you hopefully see it this time:
Note that if a Drop Pod scatters off the edge of the board, it suffers a Deep Strike Mishap as per the Warhammer 40, 000 rule book.




Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/12 19:06:32


Post by: nosferatu1001


The mishap for scattering off the table does, yes, because it states so. Scattering off the table, at all (it does not state before or after igs, so it is for both), means you kishap


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/12 19:08:10


Post by: raiden


rigeld2 wrote:
 raiden wrote:
are you trying to say that mishaps happen before scatter is reduced?

I'm trying to say that the IGS rule says, and I'll quote it again so you hopefully see it this time:
Note that if a Drop Pod scatters off the edge of the board, it suffers a Deep Strike Mishap as per the Warhammer 40, 000 rule book.




okay, so if it ends up off the board or halfway off it mishaps... its clarifying that you DO mishap. order of operations, its on top of a model therefore IGS takes effect before determining ANY mishaps. it reduces the scatter (note this is NOT A SEPERATE SCATTER) If it still winds up off the table edge it mishaps, if not, nothing happens, as the pod was never in the earlier position to begin with.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/12 19:13:07


Post by: rigeld2


 raiden wrote:
rigeld2 wrote:
 raiden wrote:
are you trying to say that mishaps happen before scatter is reduced?

I'm trying to say that the IGS rule says, and I'll quote it again so you hopefully see it this time:
Note that if a Drop Pod scatters off the edge of the board, it suffers a Deep Strike Mishap as per the Warhammer 40, 000 rule book.

okay, so if it ends up off the board or halfway off it mishaps... its clarifying that you DO mishap. order of operations, its on top of a model therefore IGS takes effect before determining ANY mishaps. it reduces the scatter (note this is NOT A SEPERATE SCATTER) If it still winds up off the table edge it mishaps, if not, nothing happens, as the pod was never in the earlier position to begin with.

... No.
Why do I say no?
Should a Drop Pod scatter on top of impassable terrain or another model

Note how it's literally the exact same wording (Should a Drop Pod scatter, do X. If a Drop Pod scatters, do Y.).
See how using actual rules in a discussion is important? The trigger for both is the same - if you scatter somewhere, do something. One allows you to decrease the scatter. The other says you mishap. So reduce all you want, but if you don't mishap you're not following the actual rules.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/12 19:14:07


Post by: nosferatu1001


 raiden wrote:
rigeld2 wrote:
 raiden wrote:
are you trying to say that mishaps happen before scatter is reduced?

I'm trying to say that the IGS rule says, and I'll quote it again so you hopefully see it this time:
Note that if a Drop Pod scatters off the edge of the board, it suffers a Deep Strike Mishap as per the Warhammer 40, 000 rule book.




okay, so if it ends up off the board or halfway off it mishaps... its clarifying that you DO mishap. order of operations, its on top of a model therefore IGS takes effect before determining ANY mishaps. it reduces the scatter (note this is NOT A SEPERATE SCATTER) If it still winds up off the table edge it mishaps, if not, nothing happens, as the pod was never in the earlier position to begin with.

Except the actual, written rule does not state that. Try again


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/12 19:18:45


Post by: raiden


I will disagree, but I am done with the argument, I don't use pods really so IDC what you guys determine. If I was playing vs someone with pods and the situation in the OP happened, it reduces its scatter before determining mishaps as it cannot land their anyway.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/12 20:06:20


Post by: davou


 raiden wrote:


as I said, you determine "scatter" results after IGS takes effect.


Part of IGS taking effect, is suffering a mishap for being off the table. If IGS happens before you do anything else, there's still a provision in IGS for you to mishap


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/12 23:02:32


Post by: nosferatu1001


 raiden wrote:
I will disagree, but I am done with the argument, I don't use pods really so IDC what you guys determine. If I was playing vs someone with pods and the situation in the OP happened, it reduces its scatter before determining mishaps as it cannot land their anyway.

You are entitled to your opinion, but you have failed to back it up with any actual rules.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 00:29:14


Post by: pizzaguardian


I understand that nos's and rigeld's points that the dp suffers a misshap when it scatters off the board and why not misshaping can be interpreted as not following the whole rule.

But i think the reverse of the situation applies as well. When we follow the second sentence and make the pod misshap we are not following the first sentence and not reducing scatter distance.

As i somewhat suggested earlier, this creates somewhat a simultaneous situation where if we reduce the scatter distance we will not obey the second sentence. And if we cause a misshap for scattering off the table we fail to complete the first sentence of the drop pod rule where we were supose to reduce scatter.

I dont see a rule where one part triumphs the other, no matter how i read it. They have the same certainity.

If i had to push a decision my decision would be not misshaping per 2 rules from the brb.

Page 36. first bulletpoint of "arriving ...

to determine the model's final position : this is surely not done yet because IGS still requires us to reduce scatter.

Page 9 . "Exceptions"

This rule we all know only has any matter in this subject if we accepted that ; the pod both misshaps and not misshaps at the same time. And then the player whose turn it is decides.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 00:48:36


Post by: DJGietzen


My take on this.

IGS makes the drop pod mishap if it scatters off the board edge, but also requires us to reduce the scatter if we are over a friend/enemy model. This reduction is part of the scatter process (I feel you have to infer this because you are reducing the scatter and not moving the drop pod after it scattered). You only mishap after the final position is determined to be completely/partially off the table.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 00:58:50


Post by: Fragile


The "Final" position is not taken into account. You place the model and then roll the scatter dice. If you scatter over an enemy model / IT, then you reduce to avoid. If you scatter off the table you mishap. Did the scatter take the Pod off the table ? Yes. Then you mishap.

The rule does not say if the final position.... it says "if you scatter off the table" which you did.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 00:58:58


Post by: rigeld2


 DJGietzen wrote:
My take on this.

IGS makes the drop pod mishap if it scatters off the board edge, but also requires us to reduce the scatter if we are over a friend/enemy model. This reduction is part of the scatter process (I feel you have to infer this because you are reducing the scatter and not moving the drop pod after it scattered). You only mishap after the final position is determined to be completely/partially off the table.

But that's not what IGS says.

You would trigger a mishap and trigger a scatter reduction. Reducing the scatter is fine as long as you also mishap.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 01:14:52


Post by: DJGietzen


rigeld2 wrote:

But that's not what IGS says.


What IGS says is "...if a Drop Pod scatters off the edge of the board..." and I don't think it can until after anything that modifies that scatter has been applied.

I also would like to point out that IGS says "...it suffers a Deep Strike Mishap as per the Warhammer 40,000 rulebook." and the BRB says "...would land partially or fully off the table...". If the model's scatter is reduced in this situation it would not land partially or fully off the table and per the BRB it would not mishap.

A model cannot land anywhere until after its final position is determined . Its final position can't be determined until after you reduce the scatter. If that final position is partially or fully off the table it will mishap.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 01:26:15


Post by: rigeld2


You left part of your quote out conveniently.
The Mishap is triggered if you scatter off the table, meaning you don't need to reference the BRB trigger.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 01:30:23


Post by: Nilok


 DJGietzen wrote:
rigeld2 wrote:

But that's not what IGS says.


What IGS says is "...if a Drop Pod scatters off the edge of the board..." and I don't think it can until after anything that modifies that scatter has been applied.

I also would like to point out that IGS says "...it suffers a Deep Strike Mishap as per the Warhammer 40,000 rulebook." and the BRB says "...would land partially or fully off the table...". If the model's scatter is reduced in this situation it would not land partially or fully off the table and per the BRB it would not mishap.

A model cannot land anywhere until after its final position is determined . Its final position can't be determined until after you reduce the scatter. If that final position is partially or fully off the table it will mishap.


If you are required to mishap if you scatter off the table, why don't you still mishap after your scatter is reduced?


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 01:42:24


Post by: DJGietzen


rigeld2 wrote:
You left part of your quote out conveniently.
The Mishap is triggered if you scatter off the table, meaning you don't need to reference the BRB trigger.


I see what your saying but if you reduce the scatter for some other reason did you still scatter off the board edge?



Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 01:45:16


Post by: rigeld2


Yes. The same scatter that triggered the reduction triggers the mishap.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 01:54:46


Post by: raiden


so, your whole idea of thought is that, because it is stating if you do scatter off the table, the IGS does not take place and you take a mishap, that even if it is in a spot that a drop pod could never be. IE on a unit. it was still there, no matter what? thats like saying ok, I mishap, then IGS takes place. I land here, now let me roll on the mishap table. I see your reasoning, but I don't understand it...


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 02:17:36


Post by: rigeld2


 raiden wrote:
so, your whole idea of thought is that, because it is stating if you do scatter off the table, the IGS does not take place and you take a mishap, that even if it is in a spot that a drop pod could never be. IE on a unit. it was still there, no matter what? thats like saying ok, I mishap, then IGS takes place. I land here, now let me roll on the mishap table. I see your reasoning, but I don't understand it...

IGS does take place. Since the mishap is part of the IGS rule and all...


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 02:28:04


Post by: DJGietzen


rigeld2 wrote:
Yes. The same scatter that triggered the reduction triggers the mishap.
In my opinion the reduction alters the scatter and any results from the un-altered scatter should be dis-guarded becouse it is incomplete. The model does not scatter then move. It either scatters X inches or it scatters X minus Y inches, not both.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 02:52:37


Post by: raiden


 DJGietzen wrote:
rigeld2 wrote:
Yes. The same scatter that triggered the reduction triggers the mishap.
In my opinion the reduction alters the scatter and any results from the un-altered scatter should be dis-guarded becouse it is incomplete. The model does not scatter then move. It either scatters X inches or it scatters X minus Y inches, not both.


this is my reasoning as well.


Automatically Appended Next Post:
rigeld2 wrote:
 raiden wrote:
so, your whole idea of thought is that, because it is stating if you do scatter off the table, the IGS does not take place and you take a mishap, that even if it is in a spot that a drop pod could never be. IE on a unit. it was still there, no matter what? thats like saying ok, I mishap, then IGS takes place. I land here, now let me roll on the mishap table. I see your reasoning, but I don't understand it...

IGS does take place. Since the mishap is part of the IGS rule and all...


but you are saying it takes place at the same time or after the mishap. If this was true, what is the point of the IGS?


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 03:30:57


Post by: Nilok


 raiden wrote:
rigeld2 wrote:
 raiden wrote:
so, your whole idea of thought is that, because it is stating if you do scatter off the table, the IGS does not take place and you take a mishap, that even if it is in a spot that a drop pod could never be. IE on a unit. it was still there, no matter what? thats like saying ok, I mishap, then IGS takes place. I land here, now let me roll on the mishap table. I see your reasoning, but I don't understand it...

IGS does take place. Since the mishap is part of the IGS rule and all...


but you are saying it takes place at the same time or after the mishap. If this was true, what is the point of the IGS?


Beside avoiding any models and impassable terrain during deep striking?
Honestly, unless you are really unlucky, or have very poor placement, you should almost never scatter off the table. I think only Tyranid Spore Pods correct for scatter like this.
This is simply stating a weakness of IGS as it still mishaps when it scatter off a table, just like how blast do nothing if they scatter off a table, even it they would be covering a model.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 03:34:36


Post by: raiden


 Nilok wrote:
 raiden wrote:
rigeld2 wrote:
 raiden wrote:
so, your whole idea of thought is that, because it is stating if you do scatter off the table, the IGS does not take place and you take a mishap, that even if it is in a spot that a drop pod could never be. IE on a unit. it was still there, no matter what? thats like saying ok, I mishap, then IGS takes place. I land here, now let me roll on the mishap table. I see your reasoning, but I don't understand it...

IGS does take place. Since the mishap is part of the IGS rule and all...


but you are saying it takes place at the same time or after the mishap. If this was true, what is the point of the IGS?


Beside avoiding any models and impassable terrain during deep striking?
Honestly, unless you are really unlucky, or have very poor placement, you should almost never scatter off the table. I think only Tyranid Spore Pods correct for scatter like this.
This is simply stating a weakness of IGS as it still mishaps when it scatter off a table, just like how blast do nothing if they scatter off a table, even it they would be covering a model.


you can't compare a blast template with a drop pod. They have to completely different goals, the Drop pod does not want to land on the squad, the blast does. the drop pod, if it would land on a squad, reduces its scatter causing it to not land where ever it would have been. I have never had this come up tbh. I run 1 pod regularly and never put in withing 12 inches of a table edge. That being said if it winds up on a table edge after IGS or IGS not being activated its a mishap.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 03:57:26


Post by: rigeld2


 raiden wrote:
rigeld2 wrote:
 raiden wrote:
so, your whole idea of thought is that, because it is stating if you do scatter off the table, the IGS does not take place and you take a mishap, that even if it is in a spot that a drop pod could never be. IE on a unit. it was still there, no matter what? thats like saying ok, I mishap, then IGS takes place. I land here, now let me roll on the mishap table. I see your reasoning, but I don't understand it...

IGS does take place. Since the mishap is part of the IGS rule and all...

but you are saying it takes place at the same time or after the mishap. If this was true, what is the point of the IGS?

It avoids units and impassable terrain. Just like the rule says.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 04:34:52


Post by: raiden


rigeld2 wrote:
 raiden wrote:
rigeld2 wrote:
 raiden wrote:
so, your whole idea of thought is that, because it is stating if you do scatter off the table, the IGS does not take place and you take a mishap, that even if it is in a spot that a drop pod could never be. IE on a unit. it was still there, no matter what? thats like saying ok, I mishap, then IGS takes place. I land here, now let me roll on the mishap table. I see your reasoning, but I don't understand it...

IGS does take place. Since the mishap is part of the IGS rule and all...

but you are saying it takes place at the same time or after the mishap. If this was true, what is the point of the IGS?

It avoids units and impassable terrain. Just like the rule says.


yes. and in avoiding a unit, reduces the scatter so it would not ever have landed near a table edge, therefore doesn't mishap


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 04:57:24


Post by: rigeld2


 raiden wrote:
yes. and in avoiding a unit, reduces the scatter so it would not ever have landed near a table edge, therefore doesn't mishap

You act like the rule doesn't specifically address scattering off the table. It's not a get out of jail free card.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 04:59:57


Post by: Nilok


 raiden wrote:
rigeld2 wrote:
 raiden wrote:
rigeld2 wrote:
 raiden wrote:
so, your whole idea of thought is that, because it is stating if you do scatter off the table, the IGS does not take place and you take a mishap, that even if it is in a spot that a drop pod could never be. IE on a unit. it was still there, no matter what? thats like saying ok, I mishap, then IGS takes place. I land here, now let me roll on the mishap table. I see your reasoning, but I don't understand it...

IGS does take place. Since the mishap is part of the IGS rule and all...

but you are saying it takes place at the same time or after the mishap. If this was true, what is the point of the IGS?

It avoids units and impassable terrain. Just like the rule says.


yes. and in avoiding a unit, reduces the scatter so it would not ever have landed near a table edge, therefore doesn't mishap


The problem is that IGS says when you scatter off the table, you mishap. It dose not give you permission to not mishap that if you somehow scatter back onto the board. I can see both going off, you scatter back onto the board, but you still triggered IGS's mishap.

Reading it, nothing in the first part denies the second part and vice verse. It tells you that when you scatter past the table, you mishap. It dose not say when you finish IGS, just when you scatter.

If viewed like programming:
Did you scatter over a model/impassable terrain?
reduce scatter

Did you scatter out of the table?
misshap


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 07:29:40


Post by: Dat Guy


Ok like my self I think people just like to hear themselves talk or in this case like to type and read their own writing again counts for me as well.

It is really simple if the drop pod was to scatter and its end scatter distance were to land on a model or impassable terrain the distance scattered gets reduced, if any part of that bottom of the drop pod is off the table edge it's a mis hap, if it ends fully on the table it's fine, the models final placement is what determines where it landed.

It's like deep striking terminators, the centered model can scatter and land near the board edge, if you can't place every model in that squad with at least the minimum touching the original models base then it is a mis hap.

Why do you people make this game so complicated and drawn out?


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 07:55:21


Post by: davou


Dat Guy wrote: the models final placement is what determines where it landed.



You have a page number for that claim? Because IGS dissagreees with you


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 08:25:08


Post by: marv335


Regardless of the fact that you can reduce the scatter, you still scattered off the table.
If you're following one part of the IGS rules you have to follow all of them.
The rule says if you scatter off the table, you mishap.
Even if you reduce the scatter to avoid the tank, you still scattered off the table, and the mishap is still triggered.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 08:59:22


Post by: nosferatu1001


 DJGietzen wrote:
rigeld2 wrote:
Yes. The same scatter that triggered the reduction triggers the mishap.
In my opinion the reduction alters the scatter and any results from the un-altered scatter should be dis-guarded becouse it is incomplete. The model does not scatter then move. It either scatters X inches or it scatters X minus Y inches, not both.

Except you are not follwoing the rule as written, but what you infer. It does not say "if you scatter (after reducing scatter...)..." or anything even remotely like that. It just states: if you scatter

The rule says "if it scatters". DId it scatter off the table? Yes. Did it also scatter over a model? Yes.

You trigger a mishap from the first, and you reduce scatter for the second. You still, however, mishap.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 09:34:39


Post by: Dat Guy


The distance on the dice refers to the movement it would scatter to, but because of igs it prevents it from going off the table if impassable terrain or models reduce it enough. Its simple. The key word in the rule is "if"it doesn't say "when" it does it says "IF" their I even capitalized it for you! Your welcome in advance.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 10:28:19


Post by: Crimson


If a blast weapon fired by a BS 5 models rolls enough on scatter dice for the centre hole to land two inches off the table before reducing the scatter by the BS, what happens?


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 11:10:40


Post by: Dat Guy


You minus the distance before you center the blast.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 11:15:18


Post by: Crimson


Dat Guy wrote:
You minus the distance before you center the blast.

Yes, and this drop pod thing seems to be a similar case.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 11:19:14


Post by: nosferatu1001


Dat Guy wrote:
The distance on the dice refers to the movement it would scatter to, but because of igs it prevents it from going off the table if impassable terrain or models reduce it enough. Its simple. The key word in the rule is "if"it doesn't say "when" it does it says "IF" their I even capitalized it for you! Your welcome in advance.

So, no rules to back up you assertion?

IGS activates on the same trigger as the off the table mishap. By activating one you MUST activate the other. Failing to do so is breaking a rule


It says SCATTER, not "Scatter after reduction for landing on a model or impassable terrain", it just says scatter.

If you disagree, page and para wher e you get to ignore the same initial condition "just because"


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 12:50:33


Post by: rigeld2


Dat Guy wrote:
The distance on the dice refers to the movement it would scatter to, but because of igs it prevents it from going off the table if impassable terrain or models reduce it enough. Its simple. The key word in the rule is "if"it doesn't say "when" it does it says "IF" their I even capitalized it for you! Your welcome in advance.

Do you agree that IGS has two parts to the rule?
Do you agree they have literally the same trigger?
Why are you applying one and not the other?


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 13:02:08


Post by: Happyjew


I would just like to point out that if the off-table clause doesn't kick in until after you determine final location, then neither would the scatter reduction. This would mean that IGS does nothing. I'm OK with that.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 13:03:11


Post by: rigeld2


 Happyjew wrote:
I would just like to point out that if the off-table clause doesn't kick in until after you determine final location, then neither would the scatter reduction. This would mean that IGS does nothing. I'm OK with that.

Yeah, evidently it's difficult for some people to realize that


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 13:15:20


Post by: raiden


 Crimson wrote:
Dat Guy wrote:
You minus the distance before you center the blast.

Yes, and this drop pod thing seems to be a similar case.


if this is true.. why is a drop pod any different? because the phrase was poorly written?


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 13:16:28


Post by: nosferatu1001


How is the phrase poorly written?

Both triggers are the same, and neither rewrites the other....so why only follow one, and not the other?


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 13:22:10


Post by: pizzaguardian


rigeld2 wrote:
 Happyjew wrote:
I would just like to point out that if the off-table clause doesn't kick in until after you determine final location, then neither would the scatter reduction. This would mean that IGS does nothing. I'm OK with that.

Yeah, evidently it's difficult for some people to realize that

Because until happyjew explained it you were saying the pod misshaps because igs rule, not that two parts of the rule cancel each other out. You keep saying "why follow one part on the other " and one can say why follow misshap part and not the reduce. I agree with happys way that the rule is bypassed in this situation and the pod mishaps, not with rigelds explanation of why follow half a rule which creates a superposition when two sides of the coin is taken into account.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 13:24:45


Post by: rigeld2


 raiden wrote:
 Crimson wrote:
Dat Guy wrote:
You minus the distance before you center the blast.

Yes, and this drop pod thing seems to be a similar case.


if this is true.. why is a drop pod any different? because the phrase was poorly written?

They aren't similar.
Blasts you subtract BS before you measure.
Pods you measure and then reduce if required.

Pods mishap if they scatter off the table. Not "if their final position is off the table".


Automatically Appended Next Post:
 pizzaguardian wrote:
rigeld2 wrote:
 Happyjew wrote:
I would just like to point out that if the off-table clause doesn't kick in until after you determine final location, then neither would the scatter reduction. This would mean that IGS does nothing. I'm OK with that.

Yeah, evidently it's difficult for some people to realize that

Because until happyjew explained it you were saying the pod misshaps because igs rule, not that two parts of the rule cancel each other out. You keep saying "why follow one part on the other " and one can say why follow misshap part and not the reduce. I agree with happys way that the rule is bypassed in this situation and the pod mishaps, not with rigelds explanation of why follow half a rule which creates a superposition when two sides of the coin is taken into account.

http://www.dakkadakka.com/dakkaforum/posts/list/562615.page#6241153
You obviously missed that post.
And I've said - from the start - that you do both. Reducing the scatter is fine, but you also mishap.
I said the triggers are the same. Advocates of not mishap ping are saying you only do one.
You can walk away saying Happy is right and I'm not but since we're saying the same thing...


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 15:27:05


Post by: nosferatu1001


 pizzaguardian wrote:
rigeld2 wrote:
 Happyjew wrote:
I would just like to point out that if the off-table clause doesn't kick in until after you determine final location, then neither would the scatter reduction. This would mean that IGS does nothing. I'm OK with that.

Yeah, evidently it's difficult for some people to realize that

Because until happyjew explained it you were saying the pod misshaps because igs rule, not that two parts of the rule cancel each other out. You keep saying "why follow one part on the other " and one can say why follow misshap part and not the reduce. I agree with happys way that the rule is bypassed in this situation and the pod mishaps, not with rigelds explanation of why follow half a rule which creates a superposition when two sides of the coin is taken into account.

The pods mishap because of the IGS rule, which has 2 parts. There is no superposition, the pod mishaps and doesnt mishap - however mishap forces a state (rolll on table, deploy, destroy, etc) which means it is irrelevant what the other result is.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 15:57:03


Post by: blaktoof


If the model ends ontop of another model or off the board and IGS cannot prevent that, or causes it. the model has mishapped as per the rules for IGS


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 20:18:24


Post by: pizzaguardian




I agree with the end result not with the way you reach it.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 20:34:01


Post by: rigeld2


 pizzaguardian wrote:
I agree with the end result not with the way you reach it.

So... how do you reach it then?


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 21:03:06


Post by: Dat Guy


Ok, you have to follow the deep striking rule, when you deep strike you place the model on the table for the spot you want them you roll 2d6 and a scatter die. So with igs if the distance would scatter on top of impassable terrain or another model friend or foe then reduce the scatter dice to avoid the obstacle.

For igs PG 99 out of C:SM and PG 36 out of Warhammer 40k 6th edition basic rule book. Particularly read deep strike mishaps it notes a mis hap occurs if the model or models cannot be placed.

Enough said.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 21:05:47


Post by: rigeld2


Dat Guy wrote:
Ok, you have to follow the deep striking rule, when you deep strike you place the model on the table for the spot you want them you roll 2d6 and a scatter die. So with igs if the distance would scatter on top of impassable terrain or another model friend or foe then reduce the scatter dice to avoid the obstacle. With IGS if the pod scatters off the board it mishaps.

For igs PG 99 out of C:SM and PG 36 out of Warhammer 40k 6th edition basic rule book. Particularly read deep strike mishaps it notes a mis hap occurs if the model or models cannot be placed.

Enough said.

You're misquoting the IGS rule and not applying it when you should. I added the sentence you forgot in bold.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 21:29:35


Post by: Dat Guy


How does the pod scatter off the board? Its the scatter distance in inches that is off the board not the drop pod because if impassable terrain or a model were to be where the scattering drop pod is placed the distance gets reduced. Obviously if their is no impassable terrain or models in its final scatter then of course its a mis hap you have to apply all the rules together and read the words, then sentences, and paragraphs as a whole not a separate entity.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 21:31:02


Post by: rigeld2


Dat Guy wrote:
How does the pod scatter off the board? Its the scatter distance in inches that is off the board not the drop pod because if impassable terrain or a model were to be where the scattering drop pod is placed the distance gets reduced. Obviously if their is no impassable terrain or models in its final scatter then of course its a mis hap you have to apply all the rules together and read the words, then sentences, and paragraphs as a whole not a separate entity.

Read the trigger to reduce scatter.
Now read the mishap trigger.

Notice that they are identical. Now please support your assertion that one happens and not the other.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 22:09:27


Post by: Dat Guy


The mis hap trigger says if you can't place the model or models fully on the table.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 22:17:56


Post by: rigeld2


Dat Guy wrote:
The mis hap trigger says if you can't place the model or models fully on the table.

So you're ignoring half of the IGS rules then?
That's a fine house rule, but the actual rules disagree.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 22:36:41


Post by: Dat Guy


Please state your assertion that both can happen because that sounds stupid because if both happen it can't because a mis hap never ends with the unit on the table. And if both happen like you say that's a false statement.

Your making up a rule saying if you land on the table but your scatter distance is off and your final placement is on the table you still mis hap....tsk tsk


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 22:49:07


Post by: Nilok


Dat Guy wrote:
Please state your assertion that both can happen because that sounds stupid because if both happen it can't because a mis hap never ends with the unit on the table. And if both happen like you say that's a false statement.

Your making up a rule saying if you land on the table but your scatter distance is off and your final placement is on the table you still mis hap....tsk tsk

When dose the IGS rule for reducing scatter occur? When you scatter over a model or impassable terrain.
When dose the IGS rule for mishap occur? When you scatter off the table.
Both trigger from the same scatter.

Could you please state which part of this rule happens first and why?


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 22:50:23


Post by: rigeld2


Dat Guy wrote:
Please state your assertion that both can happen because that sounds stupid because if both happen it can't because a mis hap never ends with the unit on the table. And if both happen like you say that's a false statement.

No - both absolutely happen. You reduce your scatter but you also mishap.

Your making up a rule saying if you land on the table but your scatter distance is off and your final placement is on the table you still mis hap....tsk tsk

I'm making up nothing - I've cited the rules. You are literally ignoring a rule for no reason.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 23:51:24


Post by: Dat Guy


If igs is reducing the distance the drop pod never scatters off, how the feth can you have something deep strike and land on the table and still freaking mis hap?

The answer is no.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/13 23:57:50


Post by: Happyjew


Dat Guy wrote:
If igs is reducing the distance the drop pod never scatters off, how the feth can you have something deep strike and land on the table and still freaking mis hap?

The answer is no.


Before reducing the scatter, did the DP scatter off the table?


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/14 00:11:27


Post by: raiden


 Happyjew wrote:
Dat Guy wrote:
If igs is reducing the distance the drop pod never scatters off, how the feth can you have something deep strike and land on the table and still freaking mis hap?

The answer is no.


Before reducing the scatter, did the DP scatter off the table?


no, because it never really landed to begin with.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/14 00:19:31


Post by: Powerguy


Dat Guy wrote:
If igs is reducing the distance the drop pod never scatters off, how the feth can you have something deep strike and land on the table and still freaking mis hap?

The answer is no.

The issue is that you are missing a step. IGS doesn't care if your final position is off the table, only that you scattered off the table - a subtle but important difference. So you place the model, then roll scatter - taking you off the table. This is the standard process for the Deep Strike rules and would normally be your only two steps (place then scatter). Then at the same time both parts of the IGS rules kick in meaning you have another two steps which resolve at the same time so you both a) mishap because you are off the table and b)are allowed to reduce your scatter. Note there is nothing stating that you can't mishap and end up on the table (quite the opposite in fact, this is precisely what happens if you don't have IGS and land on another model etc).


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/14 00:30:31


Post by: Happyjew


 raiden wrote:
 Happyjew wrote:
Dat Guy wrote:
If igs is reducing the distance the drop pod never scatters off, how the feth can you have something deep strike and land on the table and still freaking mis hap?

The answer is no.


Before reducing the scatter, did the DP scatter off the table?


no, because it never really landed to begin with.


In that case you do not have permission to reduce the scatter.

If the DP is not considered to have scattered off the table until you determine the final location, then the same can be said for impassable terrain/models. As such, you have no permission to reduce the scatter ever.

Good job, you've made IGS do absolutely nothing.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/14 00:34:52


Post by: raiden


No. The igs says SHOULD a models scatter place it over another model or terrain then reduce the scatter. The mishap part simplely stats if it scatters off the table (they are probably assuming no IGS activation which to be fair I would never have thought "what if" for this scenario lol) then you mishap.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/14 01:12:36


Post by: hyv3mynd


Dat Guy wrote:
If igs is reducing the distance the drop pod never scatters off, how the feth can you have something deep strike and land on the table and still freaking mis hap?

The answer is no.


You're making a logical leap but missing the point.

To reduce the scatter due to hitting an enemy tank, you must first scatter. Scattering onto the tank is what causes the reduction. Without the act of scattering, you cannot trigger the cause for reduction.

The same trigger (scattering[off the table edge]) causes a mishap. You cannot reduce the scatter because you've mishapped.

I've got 15 years of fog between me and my last logic class, but it would be something like this:
If S(e) then R(S)
If S(t) then M

S= scatter
e= enemy or impassible
t= table edge
R= reduce
M= mishap

The only variable is e (enemy unit or impassible terrain) versus t (tble edge). Roll your scatter and plus your result into the equation before arriving at a reduced scatter result or mishap.

Edit- actually they're not mutually exclusive but if you arrive at both situations you must apply both results thus ensuring the mishap as reducing the scatter won't matter as you've already mishapped.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/14 01:45:15


Post by: AndrewC


Of course this could simply be a holdover from a previous edition when the edge of the board was considered impassible terrain, and GW couldn't be bothered to proof read their rules......

Cheers

Andrew


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/14 04:47:25


Post by: rigeld2


 AndrewC wrote:
Of course this could simply be a holdover from a previous edition when the edge of the board was considered impassible terrain, and GW couldn't be bothered to proof read their rules......

Or it could be that the rule under discussion is printed in a 6th edition codex and is a marked difference to 5th edition codexes.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/14 05:05:10


Post by: Dat Guy


Yeah I can see now, both do happen but igs doesn't help it at all. So really all in the end it's just a mis hap.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/14 10:05:15


Post by: nosferatu1001


 AndrewC wrote:
Of course this could simply be a holdover from a previous edition when the edge of the board was considered impassible terrain, and GW couldn't be bothered to proof read their rules......

Cheers

Andrew

The edge of the board has not, as far as I know from at least 4th onwards, ever been impassable terrain. You just do not have permission to move OFF the edge of the board.

If it were impassable terrain in 5th (or 4th, etc) how did you move reserves on? Magic?


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/14 11:11:53


Post by: DeathReaper


nosferatu1001 wrote:
 AndrewC wrote:
Of course this could simply be a holdover from a previous edition when the edge of the board was considered impassible terrain, and GW couldn't be bothered to proof read their rules......

Cheers

Andrew

The edge of the board has not, as far as I know from at least 4th onwards, ever been impassable terrain. You just do not have permission to move OFF the edge of the board.

If it were impassable terrain in 5th (or 4th, etc) how did you move reserves on? Magic?


It is only impassible from on the board moving off, not from off of it moving onto the board. Like a one way mirror that...

Lol ok seriously I have no idea where they got that from. Maybe from the previous FaQ that talked about the edge of the board being the end of the world?


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/14 14:36:55


Post by: Nem


Why is this 5 pages. IGS reducing the scatter distance can only be used for;
1. If you have Scattered onto another model
2. If you have Scattered onto impassable terrain.

If you scatter off the board, you are neither on another model or impassable terrain and unable to use IGS rules to reduce scatter, you are off the board, you mishap.

It shouldn't need to state in its rules that you mishap when you scatter off, you have a permission. Scattering off the table isn't within that.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/14 14:39:30


Post by: rigeld2


The confusion is if you're off the board AND over another model/impassable terrain. Does reducing scatter mean you don't mishap?

The answer is no, but it's a fine question.


Drop pod scatters off the table partially, but still connects with a tank parked on the edge. @ 2013/11/14 14:43:00


Post by: Nem


rigeld2 wrote:
The confusion is if you're off the board AND over another model/impassable terrain. Does reducing scatter mean you don't mishap?

The answer is no, but it's a fine question.


Ya, scatter placement then work back taking each obstacle as it comes.