Splat ruler: Newbies think full height is for walker, incl/excl ambiguity

[Proxima]

Split off People getting confused by NL's auto-replay

A similar problem is the splat height ruler with a zillion bells and whistles. Every new player is confused. Every new player thinks the splat height for walkers is the full length of the ruler. The obvious solution is to offer multiple splat rulers and have a plain one as default, but we can't do that because... no idea.

[Simon]

I remember that the inside markings (C in the image) are for walkers (edit: wrong, see my next reply; B is for walkers), and I don't use the outside markings. I assume both outside markings (A, B) are for climbers (edit: again, wrong, only A is).

I don't know if the red bars (C, D) at the ends are inclusive or exclusive. I use the ruler for rough measurement, and if a single pixel counts, I experiment. (= I let one splat and rewind.)

-- Simon

[mobius]

I remember that the inside markings (C in the image) are for walkers, and I don't use the outside markings. I assume both outside markings (A, B) are for climbers.

I don't know if the red bars (C, D) at the ends are inclusive or exclusive. I use the ruler for rough measurement, and if a single pixel counts, I experiment. (= I let one splat and rewind.)

-- Simon

same I use it for rough measurement, if its close I experiment anyway. Can't be bothered with those lines. IMO they should be gone. All we should need is one simple stick/line. When the feature that's supposed to make the game less complex becomes more complex than the thing you were trying to do in the first place = bad design. (imo)

[Proxima]

I remember that the inside markings (C in the image) are for walkers, and I don't use the outside markings. I assume both outside markings (A, B) are for climbers.

A is for climbers. One of B, C is for the fall from the hatch; I can never remember which but assuming you are correct about C, it must be B.

[Simon]

From the hatch! How would you even measure pixel-perfectly from a hatch? I know that the hatch show red-yellow nubs during clear-physics mode. But then the ruler supports that outside clear-physics mode.

I've experimented. Assuming we should measure from the clear-physics-mode nub: C is from the hatch. Hatch fall distance is shorter than regular walker fall distance.

If you stick C on the yellow center of the yellow-red nub and D inside the ground, the lemmings will splat. I conclude that both C and D measure death. To measure with the ruler, you must stick D into the ground (not onto the ground), and then C will tell you the first deadly pixel.

Since C is for the hatch, B must be for regular walkers.

Again, if you stick B inside the highest solid pixel of the high ground (where the walker's pin is before his fall: the pin is under the foot inside the ground) and D inside the highest solid pixel of the low ground, you have the smallest deadly fall. Again, both B and D measure death. The maximal survivable distance is: One of them is inside the ground (high or low) and then the other ground level is inside the white, not touching B nor D.

I grant:

• It's hard to visualize inclusiveness/exclusiveness. E.g., how to show that you're supposed to stick the red bar inside the ground.
• It's consistent: Both B-D and C-D measure death. Usage of B-D and usage of C-D is near-identical.

-- Simon

[∫tan x dx]

From my personal opinion, the only useful part of the ruler is the regular splat marker. I have never used the climber fall marker, nor the window fall marker.
Similar to others in this thread, I cannot even remember what those other markers do, and have to be reminded.
The fact that these other markers are present, and the USEFUL marker is in the middle of them, makes judging fall distances very irritating.
I would very much prefer if the other markers were removed.

Consider also that NeoLemmix has rewind capabilities. Suppose a lemming falls from a window and splats. This usually happens immediately at the beginning of the level, and so the player thinks "aha - need to fix that". Usually this is part of a level's puzzle anyway. In this case, the window splat marker is useless, since the player can simply rewind a few frames and assign a floater, or whatever.
There are less common cases where a level's solution involves digging directly beneath a window. In this situation, the window fall marker may be useful. However, since these kinds of level are very uncommon, this is not helpful in the majority of gameplay.

The climber fall marker seems even more useless. Suppose a climber falls from a wall. If he survives - so what? The fall marker is useless! If he splats, then the player needs to find a way to prevent that.
Most levels do not rely on the few extra pixels granted by the climber fall. It's usually the case that a "significant" safety net is required - like the full height of a builder or stacker. In this case, those few extra pixels rarely come into play, and thus the climber fall marker is utterly useless!

All in all, the specific differences in survivable fall distance do not, in my opinion, warrant the extra fall markers. The general fall distance marker is the most prevalent by a huge margin, and use cases for the other two are simply too infrequent. They just get in the way.



I've had some ideas regarding alternatives to this system: Use the skill shadow system instead.
Suppose the player places their cursor over a falling lemming. What about a shadow that traces out the path of the falling lemming, starting out grey before turning red after splat fall distance is reached?

[Proxima]

The hatch splat marker is mainly useful when judging where to start building from to make the fall safe.

The climber marker is occasionally good to have, like on "The runaround" (Redux, Quirky 7), where the walker splat marker would falsely suggest that a particular fall is fatal, and since you don't have enough skills to make it safe and also complete the route, would discourage players from trying a solution that does actually work.

Agreed, of course, that these are very rare use cases.

I've had some ideas regarding alternatives to this system: Use the skill shadow system instead.
Suppose the player places their cursor over a falling lemming. What about a shadow that traces out the path of the falling lemming, starting out grey before turning red after splat fall distance is reached?

Interesting thought, but I don't like this as a replacement for the splat ruler, because it's very often useful to know whether a particular fall will be fatal or not when planning a route, whether or not you have a lemming nearby.

[∫tan x dx]

Interesting thought, but I don't like this as a replacement for the splat ruler, because it's very often useful to know whether a particular fall will be fatal or not when planning a route, whether or not you have a lemming nearby.

Why not have both? The splat ruler could be changed to remove the climber/window behaviours, and the skill shadow system would cover those cases by itself.

[Simon]

Interesting approach to move some tasks away from the overloaded ruler.

The hatch drop is 100 % local and doesn't need a mobile ruler. You can stick the splat marker for a hatch into the hatch's hover-during-clear-physics information, i.e., it appears whenever the red-yellow nub appears.

Climber ceilings are mostly local, too. From only a gut feeling, ∫tan x dx's annotated trajectory sounds like it's enough, that's cool. Still, you have to fetch a climber first. Can we add more clarity than the red color, for accessibility?

For the walker part, the movable ruler sounds like largely forced design. When you want to measure, you don't necessarily have all three of (1) the high ground, (2) the low ground, or (3) the walker nearby. Often 1-2 of these are missing. Sometimes all 3 of them are missing. The case when both grounds are missing is an argument against aggressive snap.

Downside to all this: You scatter the information across 3 unrelated features.

An argument for snap: Snap removes your need to remember whether you have to stick the red bar (D) in the ground (correct), or whether you have to put it on top of the ground (wrong), or, likewise, the top end for walkers.

But the red marker (D) is itself the source of inclusivity/exclusivity questions. Cut off the red marker, and keep only the white above. Obviously (I hope?) you must place the wite area on top of the low ground then, and you've reached the design of mobius's simple stick/line.

-- Simon

[WillLem]

The obvious solution is to offer multiple splat rulers and have a plain one as default

Consider also that NeoLemmix has rewind capabilities. Suppose a lemming falls from a window and splats. This usually happens immediately at the beginning of the level, and so the player thinks "aha - need to fix that". Usually this is part of a level's puzzle anyway. In this case, the window splat marker is useless

Why not have both? The splat ruler could be changed to remove the climber/window behaviours, and the skill shadow system would cover those cases by itself.

The hatch drop is 100 % local and doesn't need a mobile ruler. You can stick the splat marker for a hatch into the hatch's hover-during-clear-physics information

The solution to the problem lies in these four comments, IMO.

For hatches, it's nearly always obvious when it's splat distance. And if it isn't, let the game start playing and find out very quickly! For the die-hard "picture puzzlers" (i.e. those who absolutely don't want to start playing until they have all the information they want/need), we can add the info to CPM as Simon has suggested (and, we can add it as a non-CPM helper image at the very start, along with pre-assigned icons, etc).

For climber fall distance, I like Proxima's idea of a separate ruler. A shift key modifier for the regular ruler should be sufficient, or we can allow its own hotkey, of course.

Walker fall distance can use the no frills splat ruler without additional markers - we just need to decide exactly how big to make it. Personally, I like mobius's "only show fall height" ruler (with no markers) - the top side of the terrain piece at each end of the ruler must meet the ruler for the fall to be safe, simple*.

Meanwhile, Tan's idea of showing fall distance when hovering over a faller is brilliant, and very easy to achieve alongside everything else (as opposed to instead of it) - I'd suggest simply showing the safe fall distance, no need for extra markers/information. If the shadow meets terrain, we know the fall will be safe.


*I've prepared a version of the ruler which fulfils this. I'll be making this one the default for SuperLemmix, but obviously anyone is free to use it in their copy of NeoLemmix. For reference, the fall on the left image is safe, the fall on the right image is fatal:



The first attachment is low-res, the second is hi-res. Place them into gfx/helpers and gfx/helpers-hr.

[Proxima]

Thanks for those, WillLem.

In case anyone wants them, I've attached here the NL splat rulers modified to keep only the walker fall distance. To use these, put them into your helper folders and rename them to just "fall_distance".

[Simon]

All we should need is one simple stick/line.

My first reaction: I caveman, I only understand simple stick.

Second reaction: This is far deeper insight than it looks. Whenever you put 1-pixel thick lines anywhere on the ruler, you produce those inclusive/exclusive questions. Shall you stick the marking into the ground? Shall you put it on top of the ground?

Clearly wrong	Maybe correct?	Maybe correct?	Clearly wrong
	(but still wrong)	(indeed correct)

Compare this with WillLem's previous post (reply #9, which is 2 posts above this). Only one positioning looks correct there: Put the ruler's triangular base onto the ground, but not into the ground.

Thus: Let's ditch all 1-pixel thick markings. You can still have a complex ruler, e.g., by making the climber extension thinner (across its entire extra length) than the main ruler. But no 1-pixel thick markings.

Thanks, mobius! I knew that the ruler had inclusivity/exclusivity issues, but without your comment, I wouldn't have connected the issues to the 1-pixel thick lines.

Now, WillLem's lo-res ruler stumbles into the same pitfall: The ends are 1-pixel thick rods. Please either replace those with thicker ends, or make the entire ruler as thick as the ends.



I've attached one possible lo-res ruler.

-- Simon

[WillLem]

Thus: Let's ditch all 1-pixel thick markings. You can still have a complex ruler, e.g., by making the climber extension thinner (across its entire extra length) than the main ruler. But no 1-pixel thick markings.

You could also simply make the entire "Climber" portion a different colour, but have the ruler the same thickness all the way along, as per the current NL default; still ditch the 1px marking, only show the actual fall distance.

For SLX, I'll probably make 2 versions of the ruler and allow them their own hotkey, whilst also making the Climber one callable via the Shift modifier. This makes the most sense to me personally, and eliminates all possible confusion.

Meanwhile, just as a point of discussion, it could be argued that even my 63px "fall distance only" ruler is incorrect, because in the example I posted, the top of the ruler aligns with the lemming's foot position whilst they're still on solid ground (i.e. they haven't fallen yet, so it's actually fall distance + 1px). However, I think we can agree that this is reasonably meaningful: we need a place to measure from, and the topside of the topmost pixel of solid ground is as good a place to start as any.

For measuring whilst a lem is in the faller state, a skill shadow showing only the fall distance (i.e. no added pixels!) as proposed by tan_x_dx is a great idea; note that such a shadow must only show how much further a lemming can safely fall at any time - so, if a fall is fatal, the shadow shouldn't touch the ground. This will require simulation and ongoing calculations, so isn't as simple as just drawing a line.

I've attached one possible lo-res ruler.

Yes, this is better, thanks. I've added it to the next SLX update.

[Simon]

even my 63px "fall distance only" ruler is incorrect
top of the ruler aligns with the lemming's foot position whilst they're still on solid ground (i.e. they haven't fallen yet, so it's actually fall distance + 1px).

Certainly the top of the ruler aligns with the top edge of the ground. Safe fall distance is 63 pixels, deadly fall distance is 64 pixels. The 63-pixel ruler (126 in hi-res) is exactly as long as the safest fall. What should be incorrect here?

It's a distance in pixels, not a number of pixels not a set of certain air pixels. If I sit on Just Visiting and I roll a 10, and I tap the pawn on each space along the way to Free Parking, yes, I'll have touched 11 spaces during the turn. But that should be irrelevant.

Or do you want to measure the first deadly fall?

-- Simon

[WillLem]

What should be incorrect here?

It's a distance in pixels, not a number of pixels. If I sit on Just Visiting and I roll a 10, and I tap the pawn on each space along the way to Free Parking, yes, I'll have touched 11 spaces during the turn. But that should be irrelevant.

Or do you want to measure the first deadly fall?

No, you're right - we only need to show the safe fall distance. If the ruler doesn't touch the pixel at the bottom of the fall, we know it isn't safe. Anything else is potentially misleading, as we've already demonstrated.

I maybe thought my ruler was technically incorrect because I was counting the starting space - so, recognising "11 spaces including Just Visiting" rather than just thinking of it as "10 spaces from Just Visiting", as per your example. It's the same difference either way though, Free Parking is reached and that's what matters. Well demonstrated.