Don't create disjoint unions?

[WillLem]

Why make intentionally bad levels?

I didn't mean that bad levels should be made intentionally, I meant that it shouldn't be considered a bad thing if the levels vary in quality, which they almost certainly will for most creators. Sometimes, the levels that may be considered "bad" by some players can actually be interesting for some other reason; maybe they raise a particular design issue that can be a talking point, or maybe they simply make the other levels look better!

Time attack levels I would add to variety, but on the other side: Why do you specifically need to have timers or disjointed levels?

Let's say a pack could contain one or more of the following different level types:

1. Disjoint union
2. Timed level
3. Multitasker
4. Superlemming
5. Limited skillset

Some or all of these elements can be combined, but it is also possible to make a level such that it only fulfils one (and not any of the other) type criteria.

My issue is that packs are slowly but surely edging towards nearly always only fulfilling criteria #5, and barely ever any of the others. Even when one of the other criteria is included, it's usually still a "limited skillset" level.

So - I'm not so much advocating timers and disjoint unions as I am calling out any level type which looks like it's becoming too dominant. If disjoint unions started appearing all over the place, I'd probably say there should probably be the odd level which isn't a disjoint union. Also, if everyone started making any-way-you-want levels, I'd probably say the odd limited-skillset level wouldn't go amiss. It's all a matter a balance.

LemRunner for example only has locked exit levels and I wouldn't call out that pack for a lack of variety there.

That's different, since the pack clearly has an underlying concept (i.e. all locked exit levels). Not only that, but this particular concept doesn't preclude the use of several different level types, i.e. x-of-everythings, limited skillset, any-way-you-want, timed levels, disjoint unions (which, as tan x dx has pointed out, would be particularly interesting for locked exit levels), etc.

My issue here is not with repetitive theming (a lot of my own level packs are based on a single theme or concept), but with designers always building the same type of level, i.e. limited skillset & only one possible solution.

The mindset shifted to puzzle centric from execution centric as execution held solutions and ideas back

Execution-based stuff isn't necessarily the only alternative to limited skillset puzzles, and is perhaps more to do with UI than level design.

Catering to both on the same engine will result in a mess. You need to do a thing right and commit to it and not do two things unfocussed. This problem can be very serious in software developing!

Agreed, NeoLemmix (and Lix) is what it is because it caters to a particular playing style, and it does the job excellently. However, it is still possible to make very interesting levels which don't necessarily require the use of a particular skillset or a single solution. Variety and innovation are always possible no matter what the medium.

[Forestidia86]

I don't think modern engines exclude levels with alternate/a variety of solutions. It's rather a matter what is popular with the community at a certain time. x-of-all and allocating skills of a scarce skillset as Proxima indicated are still very well possible.

I actually don't see as well how execution based levels guarantee for multiple solutions or why they are not compatible with limited skillset levels?
Execution v. puzzle oriented levels is in my view a different question than having only one intended solution v. allowing for alternative ones.

[WillLem]

I don't think modern engines exclude levels with alternate/a variety of solutions. It's rather a matter what is popular with the community at a certain time

True, and it's a good point; supply and demand will mostly guide level creation. Still, there's nothing wrong with including the odd level that's something different.

I actually don't see as well how execution based levels guarantee for multiple solutions or why they are not compatible with limited skillset levels?

They are; it's possible to combine a number of different factors to create a level, as discussed above. Also, my pack SUPERLEMMINAS for SuperLemmini has puzzle levels which must be played at Superlemming speed (sure, the player can press pause, but then they haven't really beaten the level in my view ).

Execution v. puzzle oriented levels is in my view a different question than having only one intended solution v. allowing for alternative ones.

Yes, it is. I think people brought execution difficulty into this discussion because I'm known for having a proclivity towards that style of gameplay (however, I do also very much enjoy playing in NeoLemmix with the frameskip hotkeys and skill shadows at the ready, for sure!)

Regarding "intended-solution vs. any-solution" levels, the argument tends to be that the former are more interesting and engaging, and so they are more prevalent in the community. However, I'm always experimenting with ways to make open-ended levels more conceptually interesting, and I always appreciate it when I find a good one in someone else's pack. So, the search continues!

[Armani]

Can anyone give me some examples of good disjoint union levels?
It's quite hard to say if I like or dislike DUs when I haven't play them at all.(I think I played one on nepsterlem but I severely backrouted it so that doesn't count I think..)

[Proxima]

Can anyone give me some examples of good disjoint union levels?

From the original games: Creature Discomforts. Somehow, both Simon and I overlooked this one in the original discussion. It's a very clear example of what I was talking about, how the main puzzle of the level is how to divide the skills and allowed losses between the two groups of lemmings -- especially if you go for 100%.

Nepster's post gives some examples from usermade levels.