Lemmings Forums
NeoLemmix => Bugs & Suggestions => New Objects => Topic started by: namida on December 10, 2021, 10:15:17 PM
-
When a Climber or Slider has their permanent skill removed while climbing or sliding, they obviously need to fall off the wall. The question is - should they fall facing towards the wall, or away from it?
My thought is that in both cases, they should fall facing away from the wall. My reasoning is - the only time they would face the same direction at the end of climbing / sliding, is if they reach a gap in the wall (counting the top / bottom as a gap, if it's open space rather than terrain jutting out). If they encounter terrain in their way, which is the only other outcome besides "end of the wall" that doesn't inherently contain a direction change anyway, they would turn around. (The other ways they could detach from the wall are if they're pushed off by a blocker, force field or splitter - all of which inherently carry a direction change - or by wall-jumping, which implies their direction would become "away from the wall" regardless of any other potential factor that might apply.)
From Discord, Proxima mentioned that the lemming facing towards the wall would feel like he's still sliding down it, so should face away. I personally feel this is a weaker argument, though not by any means an outright invalid one. A couple more users also favored away from the wall, though gave no reason, simply agreed with an existing reason from someone else, or put it down to a gut feeling or similar.
Also from Discord, IchoTolot has a different train of thought here - that the climber continues in the same direction when reaching his goal (empty space at the top of the wall), while the slider turns around when reaching his goal (a floor at the bottom of the wall). In all other cases that don't inherently carry a direction change anyway, the climber would turn around while the slider would not. In other words, except where an external factor forces a direction, the two skills are essentially opposites. This would also fit with how they're more or less the inverse of each other in the first place. As such, IchoTolot's view is that the climber should face away, but the slider should face towards, the wall.
One thing I noted is that I cannot think of any situation that relies on one behavior or the other. The other behavior would always be possible to simulate via force fields, if nothing else.
-
I can think of 2 reasons why this is important:
1) It becomes a way to turn both a Climber and a Slider around. Whilst they do of course lose the skill thereafter, the turning around is achieved without any interaction with anything else.
2) If they fall facing towards, they have at least a 1-frame opportunity to interact with the wall upon reaching the ground (assuming it's not a deadly fall), whilst facing away means that they would have to turn around again to interact with it. Sure, the significance of this depends entirely on the design of the level, but it's clearly an important consideration either way.
There are probably other reasons as well. I favour facing away ever so slightly, but I don't feel too strongly about it at this stage.
-
I'd argue they should face towards the wall in both cases.
My reasoning for this is simple: the animation for both skills suggests that the lemming is facing the wall when interacting with it. The deassigner is a deassigner, not a turner. If the wall is still there when they reach the bottom, they'll turn around anyway, and if it's not, then they'll keep going the direction that they were visually facing when the skill was removed.
-
Two things:
1. I have added a poll.
2. For at least the first public exp release, both will face away from the wall. (The face-toward behavior can be simulated by adding a force field, if desired.) This does not mean it cannot be changed later.
-
I'd expect this to work the same way as with the Exhaustion gimmick back in the day when lemmings stopped performing their permanent skills. Of course, there were no Sliders back then, but I think Climbers turned around when falling, as if they had just hit their head.
I couldn't tell in kaywhyn's Lemmicks replays, because in the completed solutions, you usually don't see any Climbers quitting in the middle, since you have to find ways to prevent exactly that. kaywhyn, do you remember off the top of your head? ;) Otherwise I'll have to check again in the old editor.
I think the reason for the turnaround is the fact that Climbers (and now also Sliders) have their trigger one pixel inside the wall, i.e. at a point where a normal Walker or Faller would already have turned around.
-
I couldn't tell in kaywhyn's Lemmicks replays, because in the completed solutions, you usually don't see any Climbers quitting in the middle, since you have to find ways to prevent exactly that. kaywhyn, do you remember off the top of your head? ;) Otherwise I'll have to check again in the old editor.
With the exhaustion gimmick, climbers turn around when they lose the skill. This was my exact thought that when climbers go through the deassigner that they should fall facing away from the wall as well due to that’s how it was with the “exhaustion” gimmick. Basically, climbers lose the skill at the point just before the climb becomes a fatal fall so that the faller survives but just barely.
-
While it's true that according to the physics, a climber/slider is inside the wall, and thus a walker would have turned around, ultimately that's an implementation detail that shouldn't matter most of the time. It doesn't really matter for climbers in typical situations, but for sliders it matters a lot more, and I really do think it makes the most sense to make the decision based on the information that's visually shown to the player: why would this one specific situation cause a lemming to turn around when it interacts with the object, whereas in other situations it doesn't? From the perspective of how the lemmings are treated internally, it makes sense for them to turn, but this requires players to understand a more complex rule (deassigners don't turn, unless climbing or sliding), whereas adding some additional rules to the internal physics to prevent the turn allows us to give the player a simpler rule: deassigners don't turn, ever.
-
Thanks, kaywhyn! I've voted accordingly now, i.e. that both Climber and Slider should turn around and face away from the wall if the skill gets removed.
-
Thinking again about it, I agree with Dullstar now.
A skill remover should only remove the skills from the the lem and not do anything additionally.
As a result in both cases the lemming should still face towards the wall.
Both the climber and the slider are facing the wall and even though they (usually) turn after the action they have indeed not turned yet and the skill assigner should not do that for them. That would be up to the wall after the fall if it reaches the ground.
-
I assume that "skill remover" means "ability remover" and not "activity canceller". With this given...
I wouldn't cancel the activity in the first place. :P
You will produce this wonderful discrepancy where climbers cancel, shimmiers keep doing what they do merely because of nonpermanence of a much-earlier skill assignment, and sliders again cancel.
All four options feel awkward: Ditch interesting gadget idea over this, make gadget but not have it cancel, make activity-cancelling gadget and turn lemming, make activity-cancelling gadget and face the wall.
If the choice is purely between cancel and turn, or cancel and not turn, hmmm... I feel like it's more natural to make the gadget non-turning, leading to falling lemmings that face the wall, always. But still unnatural.
What are the cases where we turn? How natural is the turning to the design?
Let's go back to the roots: Why is there turning at all in the climb activity? Climbers that hit ceiling turn. Ceil-hitters must turn because the skill's design doesn't work at all otherwise: Non-turning ceil-hitters would re-climb the wall over and over. Thus the turning is already tacked-on. necessarily part of at least one transition.
The exact same reason applies to exhaustion. The exhausted climber must turn, otherwise the gimmick design doesn't work and he would re-climb eternally.
Jumpers assigned to climbing lemmings make them turn and then jump from the wall; this design is unnatural in the first place, why can you even assign jumpers to climbers. That's a grotesque break from the assignment rules.
An argument for turning cancelling that doesn't depend on transitions: Nowhere else in the game do you produce falling facing the wall. You always fall facing away from the wall.
Another argument for turning: Since the pin of NL is in the wall during climbing, which is terrible leftover from L1 but, well, you're stuck with it, you would move the lemming backwards to have it fall. Nowhere in the game do lemmings move backwards. They always turn first and then move forwards.
That leaves us to look at the cases where the lemming already doesn't turn. Hoisting obviously, but we aren't near the top, so that shouldn't weigh too much.
Assume we assign exploder to the climber or shimmier. Which direction does an ohnoer face? I suppose they face the wall? Of course it doesn't matter much, but it's one of the few things that come close to what we want to decide.
Very close is also the vanishing of the wall while we're climbing/sliding. The climber will think that he's at the top and hoist, but that's leftover cruft from Lemmings 1 when they didn't consider much that you can do all kinds of weird things to the terrain. What does the slider do when another effect suddenly removes the wall?
Yeah, vanishing wall seems like the closest scenario, and I think I would expect the gadget to behave like that. Everybody face the wall after cancelling. There is no need in the design to turn, unlike for the ceil-hitting. Let's see if I still think this after sleeping over it.
-- Simon
-
What does the slider do when another effect suddenly removes the wall?
He becomes a faller. His position remains in line with the wall (ie: the pin does not move horizontally), and he faces towards where it was. This is also what would happen if the Slider reaches the bottom of the wall without encountering a floor (akin to either the climber reaching the top without encountering a ledge he can climb onto; or the climber reaching the top without encountering an overhang; depending on whether you follow the "like for like" or "like for inverse" logic).
-
I think the reason for the turnaround is the fact that Climbers (and now also Sliders) have their trigger one pixel inside the wall, i.e. at a point where a normal Walker or Faller would already have turned around.
Whilst this makes sense from a physics POV, I'd prefer a more generally intuitive reason to keep the turnaround behaviour.
I'd argue they should face towards the wall in both cases.
...
The deassigner is a deassigner, not a turner. If the wall is still there when they reach the bottom, they'll turn around anyway, and if it's not, then they'll keep going the direction that they were visually facing when the skill was removed.
This feels more intuitive, the more I think about it. I tentatively +1 this and have voted accordingly. Simon's discussion expands upon and pretty much sums up my own thoughts around it as well. Except for these points:
Jumpers assigned to climbing lemmings make them turn and then jump from the wall; this design is unnatural in the first place, why can you even assign jumpers to climbers. That's a grotesque break from the assignment rules.
This, I massively disagree with. Wall-jumping is a common videogame trope nowadays, and a Climber jumping away from a wall seems a perfectly reasonable expectation from both real-world and videogame physics.
Also, my own reason for initially favouring turning away was the real-world scenario of exactly that: jumping away from a wall during climbing usually involves turning around.
However, lemmings entering a skill de-assigner are not actively jumping/moving away from the wall, they are simply having their skill removed, so they should fall (and, probably not turn).
There is no need in the design to turn, unlike for the ceil-hitting
This raises a question: why do lemmings turn upon reaching a ceiling? I assume it's the physics reason (the dot being inside the wall) and pretty much nothing else. The only reason to keep it, then, is maintenance of original physics, which NeoLemmix isn't primarily concerned with.
Probably too late to change that now, but "existing physics" seems to be the only reason to keep the ceiling behaviour, so maybe it's also a good enough to apply it to the "fall after de-assigner" behaviour.
As you can see, I'm very close to the fence on this one. I've leaned to "both face towards" because it's the most intuitive, but "both turn away" would be more physics-consistent.
Maybe, don't cancel the existing skill action. Climbers and Sliders continue until they reach their destination, albeit with the permanence of the skill removed. This seems to make some sense, and would certainly solve the problem of face towards/turn away, but is less satisfying.
-
This raises a question: why do lemmings turn upon reaching a ceiling? I assume it's the physics reason (the dot being inside the wall) and pretty much nothing else. The only reason to keep it, then, is maintenance of original physics, which NeoLemmix isn't primarily concerned with.
Simon already addressed this:
Let's go back to the roots: Why is there turning at all in the climb activity? Climbers that hit ceiling turn. Ceil-hitters must turn because the skill's design doesn't work at all otherwise: Non-turning ceil-hitters would re-climb the wall over and over. Thus the turning is already tacked-on. necessarily part of at least one transition.
-
This raises a question: why do lemmings turn upon reaching a ceiling?
Simon already addressed this:
Non-turning ceil-hitters would re-climb the wall over and over
So he did, this makes a lot of sense (I'm not sure how I missed that part of Simon's post ??? but anyways...); in the case of the de-assigner, I suppose the need for this behaviour is removed along with the skill.
-
The existing exp build has given plenty of time to play around with the "away from the wall" facing behavior.
In the next exp build, I'll instead implement the "towards the wall" behavior, and see how people feel about that in practice.
Note that this does not mean either behavior is final yet, just that the next exp update will trial a different option.
-
Okay so, the build with the "towards the wall" fall behavior has been out for a while.
How do people feel about the two after seeing both in practice?
-
I still think it feels a bit more natural when the skill remover simply removes the skill and does not provide an additional mechanic, so "towards the wall" felt better for me.
-
Okay - we've had time with experimentals featuring both variations, we've had probably all the discussion we're going to have.
I think it's pretty unanimous that both skills should have the same behavior; it's just a debate over which behavior that should be. So let's have one more poll with just those two options. If it's clearly in favor of one option, we'll go with that; if it's a close call, I'll make the final decision.
-
I already stated my opinion here, although it is not a hard stance.
I just simply would say a skill remover should only remove the skill and do not add any additional things like a turnaround. And we would have no point for a turn as no head is being hit here.
I can see the other way working out as well.
-
slider turns around when reaching his goal (a floor at the bottom of the wall).
Edit Simon 2023-07-09: My first paragraph in this post doesn't hold water; see namida's/my replies below. -- This turning (slider reaches the floor) isn't strictly necessary design-wise. It's mostly a shortcut because the subsequent walker would stand in front of the wall and turn anyway.
Certainly, there are corner cases where the shortcut makes a difference:
####.... <- Slider slides down this wall facing leftwards
####....
........ <-- 1-pixel-high walkway: Should slider turn here or not?
######## <-- floor that ends the slide if the wall's corner hasn't yet
Here, I don't feel that either design (walk into the 1-pixel-high walkway, or walk away from the wall) is overly compelling. As a new player, I'd have to test how the game behaves in this corner case.
Contrast this with the climber's turning at the ceiling. The climber's ceiling-turning isn't a mere design shortcut; it's necessary in the design to avoid re-climbing the wall over and over.
Voted now for non-turning, but it's a weak opinion. Now the poll is 3:3.
-- Simon
-
^ Another, perhaps more common, case to consider is if that slider is also a climber.
-
Ah, right. That makes it necessary design-wise that the slider turns when he reaches the floor.
-- Simon
-
...now of course, on further thought - while important for the general case, it's not so important specifically for this case. The skill remover would remove the Climber skill too. Of course, a new climber could be assigned, but if immediately climbing the wall is not desired, this assignment could simply be delayed until after he turns around.
-
Whilst it is important to think of different in-level possibilities/cases, it shouldn't be the only thing to base the decision on. It's important to think about it from a conceptual/aesthetic point of view as well - i.e. which simply seems most natural, or likely, based on the action itself, and the effect of the skill remover on that action.
-
If I disregard level design possibilities, I'll recommend against introducing the remover in the first place. Don't open the unnecessary can of worms. People already don't agree on the design of the remover and now you don't have to force one way or the other. Everything will continue to look natural during play.
Okay, I half-miss the point here. You can certainly be interested in the level design possibilities that the remover creates (regardless of whether it turns or not) and uninterested in the difference (in level design possibilities) between the two remover designs.
-- Simon
-
As this is one relatively minor detail in what appears to be a worthwhile and otherwise straightforward object, I don't think we need to cancel the idea over this. At worst, I flip a coin to decide.
-
Poll results are fairly close. A slight lead for preferring towards the wall.
I guess the last thing to ask about is actual practical experience. To those who have played around with both setups in experimental - which one felt more natural? And which one seemed to have more puzzle potential (or less backroute potential)?
I'll make a final call this weekend most likely.
-
I think for this behavior "level possibilities/cases" can mostly be left out as it is only a turn and we got tons of way to achieve that already.
So I would go purely by "does this behavior make sense for players" and "does it make sense in terms of engine logic".
From my logic and expectation, even after trying out the exp versions, a "skill remover" removes the skill - and therefore the ability even if it is currently active - and should not do any more extra stuff like a turnaround.
Sliders and climbers are facing the wall and therefore should still face the wall after the ability got removed. Also, for simplicity sake here a turnaround like after the climber hitting its head is not needed as he would not climb it again at the bottom.
For the lovers of "level possibilities/cases" I would still got this:
Maybe there is a hole at the bottom after the climb now and he gets to walk into a new tunnel.
It would also prevent sliders from sliding into a hole as the lemming is now falling on the outside again.
-
Putting a bit more thought into this and reading through the older replies:
I still agree that turning should in no way be part of the interaction; it just doesn't make sense why this would be a special case where deassigners climb.
Climbers and Sliders have a mechanically similar counterpart which isn't a permanent skill: the Shimmier. It wouldn't get cancelled by this object since it's not permanent, which is technically consistent, but it does make me wonder if perhaps it *should* cancel because the skills are so closely related in function. But then why stop at cancelling Shimmiers -- why not cancel *any* skill, *including* the removal of permanents? Or, on the other hand, there's Simon's earlier suggestion of having the skill remover only remove the skill but not cancel it: so a climber would *finish* climbing the wall, but it wouldn't be able to climb again unless another climber was assigned. But -- this would cause a unique situation in which the climbing/sliding/floating/gliding/whatever state is decoupled from the status of actually having the skill. On the other hand, it's at least a simple rule.
The problem with potential turning is that it's a complicated rule: Usually, the deassigner won't turn lemmings, but in certain cases (which you need to memorize), the desassigner will cause a turn. Non-turning is a simple rule, but there's one "oh, I didn't think it would work that way but I guess that makes sense" exception you have to memorize with the shimmier ("it should cancel a shimmier too, right?" seems like a reasonable misconception even if logically, it's not a permanent so it doesn't get cancelled). If you cancel shimmiers too, then that's a different actual exception that you have to memorize. Non-cancelling (the permanent skill will finish its activity but can't be used again) or all-cancelling (including non-permanents) seems like the simplest rule.
I can't remember if it was an option in the experimental or not, and I could see the UI for it being a problem, but assuming you could find a reasonable way (maybe if you hover your mouse over it?), there's also the option of allowing arbitrary combinations of cancelled skills, which I actually think would be more interesting for puzzles than a simple deeassigner that deassigns everything (whether it affects non-permanents or not): you could imagine a puzzle where you're given multi-athletes and you have to navigate them through a maze of deassigners that requires retaining at least a few of the skills to navigate, so you have to figure out which ones you can afford to pass through.
-
From my logic and expectation, even after trying out the exp versions, a "skill remover" removes the skill - and therefore the ability even if it is currently active - and should not do any more extra stuff like a turnaround
I still agree that turning should in no way be part of the interaction; it just doesn't make sense why this would be a special case where deassigners climb.
+1 for this.
("it should cancel a shimmier too, right?" seems like a reasonable misconception even if logically, it's not a permanent so it doesn't get cancelled). If you cancel shimmiers too, then that's a different actual exception that you have to memorize. Non-cancelling (the permanent skill will finish its activity but can't be used again) or all-cancelling (including non-permanents) seems like the simplest rule
+1 for this. If it's skill-interrupting, then it should interrupt all skills. If it only de-assigns permanents, then perhaps it shouldn't cancel the action currently being performed. Then, there's no reason to expect any other skill to be interrupted by a de-assigner.
Also, if the de-assigner doesn't cancel the action, then the turn-around question is answered anyway.
-
The remover already does more than merely cancel abilities and activities: It pries lemmings horizontally out of the wall. Climbers have the pin inside the wall; this matters for oh-noers who don't fall. If the cancelled climber shall fall facing the wall, you must pry it horizontally out of the wall first to not have it walk and turn inside the wall.
Icho recommends firmly that removing the climbing ability should immediately cancel the climbing activity. Reason: The lemming has forgotten how to climb, and we shouldn't care if the game code tests that knowledge mid-climb or not. I'm not as dead set on activity cancelling, but there is good substance to Icho's argument.
-- Simon
-
Alright. Based on the slight preference towards it in the votes, and the fact that the arguments for it are a bit more compelling, we're going to go with faces towards the wall (ie: the behavior seen in the current experimental).
I personally thought facing away felt more natural, but I think at this point towards is the way to go.