Kallor
Level: Master Delver
Rank Points: 277
Registered: 04-02-2007
IP: Logged

As always, not sure if bug or feature (intended or not).

Pushing a spearman into tarstuff breaks the tarstuff if the hit counts as a stab, just as anyone would expect.

However, if the spear in already on top of the tarstuff, as a result of turning for example, pushing the body of the spearman forward doesn't break the tarstuff. Similar move by the spearman himself counts as a stab and breaks the tarstuff (as it does with a sword for that matter).

I find this a bit odd, is this behavior known?

____________________________
Kallor

Chaco
Level: Smitemaster
Rank Points: 3659
Registered: 10-06-2005
IP: Logged

Fundamentally, the idea here is that having some entity A pushing some entity B with a weapon into an obstacle, doesn't count as that entity B bumping into that obstacle; it just counts as entity B waiting instead.

Another example: if you have a clone with a dagger get their dagger on top of a bomb (or orb), having that clone bump into the bomb detonates/activates it, but having something else push the clone into the bomb will not detonate/activate it.

It's probably not too late to change this behavior, but the implications of changing waits into bumps could be far-reaching.

____________________________
Quick links to my stuff (in case you forgot where it was):

skell
Level: Legendary Smitemaster

Rank Points: 3764
Registered: 12-28-2004
IP: Logged

Note to self: check how fixing this would affect demos.

____________________________
My website

Dragon Fogel
Level: Smitemaster
Rank Points: 2467
Registered: 06-21-2014
IP: Logged

This seems like a very risky thing to change, and it's not obviously wrong that a push that doesn't move the enemy should be "wait" and not "bump". It could be explained in-universe as a push in place not generating enough momentum to make the target's weapon do anything.

navithmastero
Level: Smitemaster

Rank Points: 621
Registered: 01-03-2012
IP: Logged

Besides the fact that changing the way pushing works has the potential to break hundreds of rooms, it doesn't really make sense in terms of game logic to do so.

We know that pushing something is not the same as that thing moving. For example, if we push a roach it will not change direction. This is despite the fact that in every other situation when a roach moves it faces the direction it is trying to move in. This makes perfect logical sense and is well established.

Due to the above, we know that pushing something into a wall is not the same as that thing trying to move into a wall (and thus bumping it). Using a similar example, if the player is swordless and gets pushed into a wall, it does not make sense for the player to reorient and face the way he has been pushed (which is what happens when a player bumps the wall). This makes sense because the player is stunned at this time and so could not possibly have rotated. Currently, being pushed forces something to wait and not move (something which is pushed only moves because something is pushing it). If we start considering something pushed into a wall as also bumping that wall then what we are saying is that something which is being pushed can also move at the same time as being pushed (since a bump move is the same as a regular move except something is in the way). This clearly cannot, and should not be the case and would have so many ramifications to published rooms, as well as interactions with other monsters. For example, if we say that pushed characters can also move on their turn, then monsters would reorient when they are pushed. This could break any demo which involves a pushing weapon and any monster who's orientation is important (aumtlich, stalwart, guard, evil eye etc.). What is even worse is that this could also extend to swordless players and doubles which introduces a whole heap of issues.

It would be very very dangerous and ill advised to change this behaviour. There also seems to be no reason to do so as it's not a bug - there are no inconsistencies involving the current behaviour, it makes perfect sense, is discoverable, is used in many, many rooms and the ramifications of changing the way things work would be massive and game breaking.

____________________________
Member of the Snake Appreciation Society

One of your local HAs.

My stuff:

---

[Last edited by navithmastero at 10-13-2020 05:24 PM]

skell
Level: Legendary Smitemaster

Rank Points: 3764
Registered: 12-28-2004
IP: Logged

I generally disagree with the interpretations above. If we make an analogy to real-world physics it really doesn't make sense that pushing something that's close to a wall to have less momentum than if pushing it further from the wall, yet pushing normally causes stabs.
As for pushing-is-actually-waiting, while it seems to be internally consistent it's also very technical (though if you went technical you'd see that pushes into obstacles are not waits - they're non-moves). I am also confused by this passage:

If we start considering something pushed into a wall as also bumping that wall then what we are saying is that something which is being pushed can also move at the same time as being pushed (since a bump move is the same as a regular move except something is in the way).

Pushing something is already making it move, and bump is as much a move as, well, moving or being pushed. A much simpler explanation for why pushing does not reorient is because by analogy to real world, when you push something it doesn't move on its own, which is why there is no need for it to change orientation.

It all sounds like an attempt retroactively adding logic to something that's inherently illogical . We could make pushes-into-obstacles be treated as bumps without changing anything and I find it really difficult to believe anyone would find it confusing or counter-intuitive (beyond being confused by already learning that the rule is opposite).

Either way, the more important questions are:
1. What would be the consequences of this change for published holds? I'd like something more specific than "hundreds of rooms will be affected".
2. What types of dangers would it introduce? Clearly unintended solutions might be one (aren't they always?), but it feels like the situations where this could be used for that are rather limited. I get the feeling that it's a rather obscure interaction anyway.
3. What edge cases would we have to deal if this were to change?
4. Puzzle potential?

One edge case I can think of is imagine a situation, where a pickaxe wielding monster is pushed sideways into an obstacle, but there is a bomb under the obstacle - is that a sideways stab or not?

____________________________
My website

Dragon Fogel
Level: Smitemaster
Rank Points: 2467
Registered: 06-21-2014
IP: Logged

From my perspective, I don't see a reason to change it.

The biggest problem is that this is a change to a basic action, pushing. There are a lot of rooms that involve pushing. There are a lot of rooms that involve pushing and doubles. There are almost definitely a lot of rooms that involve pushing and doubles and orbs, which would all be potentially impacted by this. And that's just one category of rooms, although I wouldn't be surprised if it were the largest.

Maybe very few of them would be seriously impacted! But it sure seems like it would be a huge amount of work to go through them all and confirm that. I would say that's the main issue I have with making this change at this point.

disoriented
Level: Smitemaster

Rank Points: 2401
Registered: 08-07-2007
IP: Logged

I also vote to not change this. I know at least one room I've made that would not work as designed if this were to be changed.

____________________________
34th Skywatcher

Best to PM me, since I might miss your message on CaravelNet chat.

---

[Last edited by disoriented at 10-15-2020 01:18 AM]

Dragon Fogel
Level: Smitemaster
Rank Points: 2467
Registered: 06-21-2014
IP: Logged

On further thought, I believe this has potential to get more confusing if implemented.

This is because you can potentially push in any direction and have it be blocked. You could push an east-facing mimic with a staff over an orb towards a wall to the west, and I sure don't think that should make the mimic activate the orb.

And extending that, I'm not comfortable with asking players to extrapolate from the direction of pushing whether the pushed entity will attack or not. Especially with a pickaxe.

(For the record: if you have a pickaxe over an orb, and bump any direction other than the one the orb is in, the orb doesn't activate.)

skell
Level: Legendary Smitemaster

Rank Points: 3764
Registered: 12-28-2004
IP: Logged

Dragon Fogel wrote:
On further thought, I believe this has potential to get more confusing if implemented.

This is because you can potentially push in any direction and have it be blocked. You could push an east-facing mimic with a staff over an orb towards a wall to the west, and I sure don't think that should make the mimic activate the orb.

And extending that, I'm not comfortable with asking players to extrapolate from the direction of pushing whether the pushed entity will attack or not. Especially with a pickaxe.

That's an interesting observation. I both feel like you're right and that if pushing into an obstacle worked exactly as bumping into it on your that it should not be more confusing. But maybe the fact it involves pushing adds another layer of complexity that is just enough to make it hard to reason about.

The biggest problem is that this is a change to a basic action, pushing. There are a lot of rooms that involve pushing. There are a lot of rooms that involve pushing and doubles. There are almost definitely a lot of rooms that involve pushing and doubles and orbs, which would all be potentially impacted by this. And that's just one category of rooms, although I wouldn't be surprised if it were the largest.

Remember that it would only affect rooms where you're pushing doubles and monsters with weapons other than staff and that likely greatly reduces the number of affected rooms.

I'll leave it up to Mike to make the final call on this one - I'd personally prefer if pushes into an obstacle with a weapon in that obstacle caused the same effect as the entity bumping into that on their own, as it seems to be much more of a consequence of other rules than the current behavior. But I guess I am worn out from all the changes and while I say it's an exception from rules, it's also one that's fairly simple to figure out and understand.

I may or may not do a check on my own with the spider to see how much demos are affected if this were to change.

____________________________
My website

navithmastero
Level: Smitemaster

Rank Points: 621
Registered: 01-03-2012
IP: Logged

Can I ask what reasons we have to think that this is even a bug? I'm struggling to understand what about the current interaction is wrong and might make people think that it should be changed.

EDIT: Also, to clear up what I was trying to say earlier I will post a video later today explaining the ramifications of changing this behaviour as it is far easier to do this visually than in words

____________________________
Member of the Snake Appreciation Society

One of your local HAs.

My stuff:

---

[Last edited by navithmastero at 10-15-2020 08:05 AM]

skell
Level: Legendary Smitemaster

Rank Points: 3764
Registered: 12-28-2004
IP: Logged

navithmastero wrote:
Can I ask what reasons we have to think that this is even a bug? I'm struggling to understand what about the current interaction is wrong and might make people think that it should be changed.

For one, Kallor was confused by this. And I argue that the way it works right now is not a natural consequence of other rules - in all other cases, pushing an entity has exactly the same consequences as the entity moving on their own (barring token activation but that's quite literally a separate topic). There is no reason for pushing into an obstacle to suddenly deviate from this rule but it does, and this is why I consider this to be something that's worth discussing changing.

I suppose I also argue that it's really easy to internalize certain awkward rules and retroactively explain them as something that actually makes sense. As I've been going through the bugs thread I've seen a few times a situation where players who started with KDDL were caught off guard by seemingly obvious stuff.

The end goal being cleaning up some of the more obscure exceptional rules to make a bit more sense to those of us who are not living encyclopedias of DROD rules . The reason I want to get it all discussed and possibly corrected (or officially declared as OK) now is because (I feel like this might be the last time we'll ever get a chance to do so with the existing elements.

Finally, this was reported as a bug back in the day. No one chimed in back then to say "Actually it works as intended" (I'd say Chaco's comment is more about explaining why it works that way) which means that back then people either did not care or agreed that it's something that needs to be fixed. I am partially working from that assumption.

Ultimately though these are not a hill I am willing to die on, but I want to give those reports a fair try.

____________________________
My website

skell
Level: Legendary Smitemaster

Rank Points: 3764
Registered: 12-28-2004
IP: Logged

You know, I've got another argument to NOT change this. A thought I had while seeing about implementing this. The code infrastructure for pushing is not really prepared to make either this or that change. If I were to implement it it'd be more changes in the gritty details of interactions than I have energy to handle now, and it would risk breaking other things related to pushing.

I will not be attempting to make this change.

____________________________
My website

---

[Last edited by skell at 10-15-2020 08:30 AM]

navithmastero
Level: Smitemaster

Rank Points: 621
Registered: 01-03-2012
IP: Logged

skell wrote:

navithmastero wrote:
Can I ask what reasons we have to think that this is even a bug? I'm struggling to understand what about the current interaction is wrong and might make people think that it should be changed.

For one, Kallor was confused by this. And I argue that the way it works right now is not a natural consequence of other rules - in all other cases, pushing an entity has exactly the same consequences as the entity moving on their own (barring token activation but that's quite literally a separate topic). There is no reason for pushing into an obstacle to suddenly deviate from this rule but it does, and this is why I consider this to be something that's worth discussing changing.

I suppose I also argue that it's really easy to internalize certain awkward rules and retroactively explain them as something that actually makes sense. As I've been going through the bugs thread I've seen a few times a situation where players who started with KDDL were caught off guard by seemingly obvious stuff.

The end goal being cleaning up some of the more obscure exceptional rules to make a bit more sense to those of us who are not living encyclopedias of DROD rules . The reason I want to get it all discussed and possibly corrected (or officially declared as OK) now is because (I feel like this might be the last time we'll ever get a chance to do so with the existing elements.

Finally, this was reported as a bug back in the day. No one chimed in back then to say "Actually it works as intended" (I'd say Chaco's comment is more about explaining why it works that way) which means that back then people either did not care or agreed that it's something that needs to be fixed. I am partially working from that assumption.

Ultimately though these are not a hill I am willing to die on, but I want to give those reports a fair try.

Fair enough, perfectly reasonable!

I did not see your second post when I was recording and uploading this video just so you know.

Check out this video as an explanation as to why I feel that pushing something is not the same as something moving and some other points as to why I think this behaviour shouldn't change. If you've definitely decided that you don't want to try and change this then this can act as some sort of historical record and maybe a learning resource if anyone else comes across this thread. It's a little rambling as it's unscripted but hopefully it does the job!



tl;dr (or watch, in this case): pushing something *is not* the same as that thing moving. Therefore, pushing something into a wall is not the same as having that thing bump a wall. Therefore, this should stay as is IMO.

____________________________
Member of the Snake Appreciation Society

One of your local HAs.

My stuff:

---

[Last edited by navithmastero at 10-16-2020 09:59 AM]

mrimer
Level: Legendary Smitemaster

Rank Points: 5138
Registered: 02-04-2003
IP: Logged

Thank you for the discussion and the video, which has some good illustrative cases played out.

I don't consider the behavior described in the original post as a bug. Edit: I agree pushing something is not the same as having the pushed entity move/bump in that direction. That's something we could have designed into the game rules, but we didn't. The current behavior is working as designed.

It sounds like there's general agreement to leave this behavior as-is, and I concur.

____________________________
Gandalf? Yes... That's what they used to call me.
Gandalf the Grey. That was my name.
I am Gandalf the White.
And I come back to you now at the turn of the tide.

---

[Last edited by mrimer at 10-16-2020 12:37 AM]