Spawn interval fixed per level

[Simon]

Hi,

tl;dr:

Spawn interval (SI), a.k.a. release rate, should be fixed per-level. Can't change SI during play. Find counter-arguments: Suggest levels that can't be adapted with runners, jumpers, initial/required, or other features in Lix.

Why remove variable spawn interval (SI)?

Variable SI doesn't fit into the game design.

You control lems/lix by clicking on them, assigning skills. Right now, there are two extra ways to affect solutions: The nuke, and variable SI. In practice, the nuke assigns skills for you. Variable SI is an oddball, it affects lix-independent level variables.

The level author sets external variables, the player assigns skills.

Lix already allows the player to adjust spacing between lix, by walker/jumper/runner assignments. Each works differently than variable SI, but supports the game core much better. Supporting the core while solving a problem, that's the correct direction in game design.

Spawn-interval-changing solutions are prone to trial-and-error. Several players (Clam, NaOH, myself) consider SI-changing solutions less fun.

Replacements for variable SI

• Runners: I considered these superfluous before, but their use for clustering and separation has redeemed my opinion. Runners jump further, this enables level designers to enforce certain lix to become runner.
  
• Walkers or jumpers: These aren't permanent, they need different backrouting considerations than runners.
  
• Initial/required: Stream enough, but not too many after the route is built. With a fixed SI, number of initial lix becomes an important screw to adjust. Lemmings 2 had too many lems, 60 lems per level, so L2 was forced to use a fast SI. Levels became a bit bland. In Lix, you can lower the number of lems and have better fast-forward.
  
• Fast fixed SI, then let several die: Possible; but keep mass-murdering close to the beginning of a level. Players like to estimate how many lems may still die, to plan the solution.
  

What levels can't be adapted?

We have about 500 levels. My claim is that most of them can be adapted.

I believe the strongest counter-arguments (= arguments for keeping variable SI) will emerge from discussion of some existing levels, and how they use variable SI.

Proxima's earlier list of levels where the spawn interval is part of the intended solution:

• Thomas the Climber
  
• Changing of the Guards
  
• Buridan's Lix
  
• The Hotel in Hell
  
• The Continuum Hypothesis.
  

I would like to investigate these with Proxima, and with people who have solved them.

Everybody, suggest levels that can't be adapted well with other Lix mechanics, and would lose lots of their spirit.

When effective?

I don't want to change the C++ version. If I pursue fixed-only SI further, it would become a feature of the D port. So, nothing would change in the upcoming months.

Previous discussions

Discoverable UI (scrolling, dir select), where I presented the idea to the forum for the first time. Fixing SI was a proposed solution to something else, there.

Proxima suggested in here the 5 levels from the list above.

IRC: Clam, Simon, 2015-12-12

<Clam> I'd happily do without spawn interval being changeable in-game
<Clam> use FFF and don't have ridiculous amounts of rodents in the level
[...]
<SimonN> the variable SI is a fucking bug
<SimonN> heavy heavy bug, and there is a simple argument for it, about why it doesn't belong in this game
<SimonN> you control lemmings by skill assignments

IRC: NaOH, Proxima, Simon, 2015-12-20

<Proxima> Speaking as someone who's done a lot of challenges, it's a tool I use all the time
<Proxima> though of course, the point of challenges is always to do as well as you can with the tools available, removing a tool changes the game but doesn't make it impossible to challenge yourself
<NaOH> I have seen that it is very useful when attempting pixel-precise tricks
<NaOH> But I feel guilty when I use it somehow.
<SimonMath> this is dangerous :)
<SimonMath> basher staircases were removed on a similar hunch, with great effect eventually, but that wasn't all expected
<NaOH> I feel much much more guilty with basher staircases :3

IRC: geoo, Ramon, Simon, 2016-01-06

<SimonN> ARRRRRR, who wants to guess what feature I want to kill from Lix
<SimonN> With exciting new reasons, too
<SimonN> Cull, cull, cull. Summon ccx! Rise like a phoenix from deallocated memory and let us argue!
[...]
<SimonN> what affects physics, but is not assigning skills?
<Ramond__> something the player does?
<SimonN> yes
<Ramond__> wow no idea. there must be some hidden stuff I didn't know about
<SimonN> you can assign skills to make stuff happen differently, we agree on this?
<Ramond__> yes
<Ramond__> ...
<Ramond__> oh no...
<Ramond__> NO
<Ramond__> YOU DON'T MEAN
<Ramond__> NOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOoooo

-- Simon

[Proxima]

I'll reply to the general arguments at greater length when I have time. For now, a quick look at the specified levels.

Detailed discussion follows in spoiler tags, as it reveals several intended solutions. A brief summary:

* Most levels will be okay with fixed SI.
* The main exception is levels with a time limit, where fixing the SI would prevent enough lix spawning.
* Compression Method 2, The Hotel in Hell and The Continuum Hypothesis have other problems, which may be fixable.
* As a separate issue, Buridan's Lix should have initial SI raised to 60 due to new digger behaviour.

Spoiler

Compression Method 2: The entire puzzle here is to use the SI to merge lix into a single location. Unsalvageable without variable SI.

Thomas the Climber: Solution requires sending up two climbers very close together. Easiest way to achieve this is setting SI to 4. However, without changing SI from its initial 50, you can wait until at least 19 lix are out, when there will be two close enough together. (Could set initial SI to 4 but this might make the trick too obvious?)

Changing of the Guards: Originally intended solution is to set SI to 4 initially so that last lix from left hatch can start digging a.s.a.p. allowing floater from right hatch + others from left hatch to be saved. However, digging with second-last lix from left hatch allows this solution to work without changing SI.

Buridan's Lix: Actually has no problems, except that initial SI must be raised to 60 due to new digger behaviour.

The Hotel in Hell: The problem here is that the initial SI must be kept, to allow the gap to be bridged without losing lix (there is no way to hold the crowd), but the SI must then be changed to 4 to allow all lix to spawn in time. (This actually should affect several other levels, such as Toccata.) Geoo has a crazy elegant solution involving tight SI control to merge most of the crowd.

The Continuum Hypothesis: Tight SI control is necessary to merge most lix (SI 49) then send up climbers for the I am A.T. trick appropriately spaced to gain the maximum possible height (SI 36). The only way to deal with this with fixed SI would be to give more climbers and diggers, or else to lower the height of the wall. Not the best level ever due to highly demanding execution, so I'm okay with it being toned down.

[Proxima]

The Hotel in Hell works with a fixed spawn interval with the following changes:

Spawn interval 52 (unchanged)
Skillset: 5 walkers, 1 floater, 1 imploder (no blockers), 5 builders, 5 bashers
3 minutes
Save 27/30

[Nepster]

A summary at the beginning:
I believe that most, but not all, levels can be adapted. However SI-changes are a feature to help players with executing their solutions and removing this feature will only move the difficulty/annoyance to other game elements or even newly create such difficulties.

Replacements for variable SI

• Runners: The most useful replacement and should be used more often. However consider We're in this one together (Hopeless 25): I certainly do NOT want find the correct place to assign runners in order to get them bunched up for the squisher trap!
• Walkers and Jumpers: These are very powerful skills, because they allow to stop bashers/miners/diggers/... and allow Lix to get to places, they normally cannot reach. Both uses for walkers and jumpers are MUCH more powerful than creating/closing gaps between Lix and so adding these skills will almost surely open backroutes in more complex levels.
• Initial/Required: This doesn't seem to help for levels that actually require SI-changes. Giving overly great numbers of Lix is to be discouraged as bad game-design anyway and doesn't have any impact on whether to allow SI-changes or not.
• Fast fixed SI: Not sure if there is even one level, where this can be used and does not get even more annoying than SI-changes. As reducing the save requirement often opens backroutes, the number of lemmings to kill due to missing SI-changes has to be carefully computed (another place where backroutes may turn up!) and players will never know, whether to save a Lix more on the rest of the level or at the SI-change point.
  Moreover at least I have a mental block when it comes to killing lots of Lix. E.g. When first playing OrigLemmings, Cascade was one of the most difficult levels until end of Taxing, because I always tried to stop the horde (which I didn't manage at that time). So I find changing the SI much easier than finding a solution requiring mass deaths at the beginning of a level.

Example levels (with a few additions):

Spoiler

• Hotel in Hell: I am really interested in seeing alternative versions of these levels, where not changing the SI is required and skill assignment needs not to be overly precise. I fear that any change of the level will be adapted to precisely one of the solutions, making other approaches either much harder or even impossible.
  EDIT: Just saw Proxima's latest post. Will try that.
• Over the Hump (Daunting 24): I don't see how to fix the level. Reducing the number of Lix will not work, because it opens other solutions. Adding more skills will only open other backroutes.
• Conundrum (Vicious 2): Adding walkers or jumpers will open backroutes. When adding a runner one certainly has to make the level shorter which removes lost of other possible approaches. On the other hand, changing the SI to 4 is trivially done once the solution is figured out.
• A matter of Perspective (Vicious 28): Likely possible to fix with terrain changes, but to figure this out will be a very non-trivial task for the level designer.
• Heed the Traffic Light: (Hopeless 16): This is already possible without changing the SI, but easier with. To achieve even this state, I needed more than one hour on the original Lemmix level just to adapt the first working version to remove the dependency on SI-changes.
• Tinker Tailor Soldier Lix (Hopeless 24): Perhaps this level can be fixed with reducing the number of Lix, but this might open up old backroutes.
• We're in this one together (Hopeless 25): Apart from the runner problem mentioned above, the number of Lix need to be drastically reduced. If I remember correctly, one can save a lot more Lix than required. As we don't want to change that, after solving the level the player has either to nuke the remaining Lix or wait endlessly.

It should be noted that most example levels only require setting the SI to 4 at some point and possibly back to the original SI again. This is relatively easily done, so any fix to allow constant SI could easily make the level much more difficult.

[Proxima]

As Nepster surmises, my suggested "fix" for The Hotel in Hell centres around making a close variant of the original intended solution work. I don't mind that there are other ways to solve the level (so long as there is no trivial solution) but I'm not that concerned about whether all the various solutions that have been found remain possible.

In case any backroutes turn up with the new version, here is another alternative (actually, in some ways I prefer this one):

Spawn interval 52
Skillset: 2 walkers, 2 floaters, 2 imploders, 2 blockers, 5 builders, 5 bashers
No time limit
Save 25/30

[Simon]

Thanks for the exhaustive studies of individual levels, and general ideas. I know that Proxima would like to give another general argument these days, for keeping variable-SI.

The levels' designers can tell best how close in spirit an adaption is. I like Proxima's Hotel in Hell with unlimited time, any intended route seems to take long anyway.

I agree with Nepster on how mass-suicide feels shaky and uncomfortable, let's try to avoid it. Another idea to adapt: Place several hatches, stacking all but one. The stacked hatches will spawn a crowd that follows the few workers spawned from the lone hatch.

Over the Hump, I'll take a look at it. I know how its basic idea is straightforward, I have solved this in the past. Interesting that such a seemingly-streamlined level poses adaption difficulties.

Breaking a single level is OK, violating general principles is not.

-- Simon

[Nepster]

Do I really need to post a rant about how bad stacking hatches is? If players see two hatches, then they expect that both are used alternatingly, nothing else.

[Proxima]

To keep everything together, here's Nepster's list from the older topic:

After a closer look, the following levels require RR-changes:
S19, C19, C30, D10, D19, V2, V7, V28, V29, H4, H12, H25, H31, H34
(and perhaps some more Hopeless levels that I haven't solved yet)

There are quite a few more, where it helps (sometimes depending on your solution):
S21, Q11, Q19, Q31, Q37, C5, C14, C17, C23, C27, C37, D4, D6, D13, D25, V9, V10, V17, V21, V25, H2, H16

(Some levels may have changed position since this list was made?) Assuming level numbers are the same as in my copy of 2014-07-02, here are some comments on my own levels that are on this list:

S21 (Lix Potion Number Nine) -- While it's intended that you can solve the level by setting SI to 4 and not assigning any skills, perhaps this is a bit cheap, and requiring the player to find ways around the traps is more interesting.
Q31 (Prelude) -- Should not need any SI changes.
C23 (Come on down to my place) -- Same.
C27 (Compression Method 2) -- In spite of what I said above, Nepster is correct to list this as "where it helps" rather than "required". The given initial SI of 20 does allow merging with a particular placement of the initial cubes.
V17 (Lixster Quadrille) -- Should not need any SI changes. [EDIT: This has a fixed spawn interval, so Nepster *must* have meant some other level.]
V21 (Thomas the Climber) and V25 (Buridan's Lix) -- Already discussed above. Both are fine with fixed SI.
V29 (Brute Fours) -- I can see where changing the SI helps, but I can also see a way to avoid it.

[Simon]

Do I really need to post a rant about how bad stacking hatches is?

Agree. The solution is to put the hatches close to each other instead. Lix's hatches are smaller than typical L1 hatches, this works well for small numbers, even without overlap. Anyway, listing ideas to fix levels won't get us ahead much -- I have to consider if and how much executional problems fixed-SI would impose on the player.

Link: Solution to Over the Hump that doesn't touch the spawn interval. This wasn't done with too much precision, but maybe I got lucky, simply. In that case, we can always go from 29/30 to about 27/28, or speed up the SI by 1.

-- Simon

[Nepster]

And here are backroutes for both new versions of Hotel to Hell.
Test1 is Proxima's second suggestion. Note that the SI-change was only done to speed things up - it is not required.
Test2 is Proxima's first suggestion.

EDIT: I had another solution in mind for Over the Hump, likely from an earlier version (it used a blocker). So, yes, the current level is fine.

[Proxima]

Nice work    For the first (no blockers) version, how about setting bashers to 3? (No version ever actually used more than 3 in the intended solution, the 5 bashers were just there for symmetry.)

[Nepster]

How about this solution: 28 Lix saved, 1 walker and 2 bashers spare and one pixel-precise basher assignment.

[Proxima]

Excellent. Like your 39/40 in Brute Fours, I regard that as a minor variation of the intended solution and not a backroute.

[Nepster]

Just found a completely different 29 Lix solution.

[ccexplore]

I feel like there hasn't been sufficient argument presented as to why this couldn't simply be left to the level designer's discretion as it is today.  Most levels probably don't need variable SI and the level designer may well already restrict the range of SIs anyway for backroute-avoidance purposes.  Time limits are best avoided generally speaking, and the number of levels where it serves a good purpose are perhaps few, but we have managed not to cull that out yet.  Variable SI feels very much like in that vein to me.

I'll concede that even the historic evidence suggests the game designers at DMA do not value variable SI as a level solving tool, starting with the fact that they removed it as soon as they got to Lemmings 2, and IIRC it didn't make a reappearance until Lemmings Revolution (not sure if Lemmings 3D has it), though in that game you actually get to control the SI for individual entrances (because lemmings don't alternate across entrances, they come out in parallel).  It seems very likely that it was originally conceived merely as a poor version of fast forward, though as more levels are developed for the L1 game mechanics, even the original games' level designers have caught on to ways of using it more as part of the level's solution.

Yes, if the level requires fiddly, precise changes of SI it wouldn't be too fun.  Then again, the same is true when a level requires a lot of precision moves with skill assignments.  Bridge stretching for example.  We handle these issues by recommending best practices for level designs.

I just feel like it's a little strange to cull away a gameplay feature that, while perhaps not that crucial overall, would seem to eliminate otherwise fine levels in the original game like "Just a Minute (Part II)" and "Flow Control".