"Lemmings Redux": A "remake" of official games for NeoLemmix.

[Nepster]

I will probably post updated lists often, and draw attention to things that we probably need to start thinking about; it doesn't mean older subjects are now decided. It means (especially the former) I don't want to fall too far behind on the suggestions, and end up forgetting about some of them altogether.

You might want to compile a list of all comments sorted by level. Every time you want to update the list according to opinions posted, you have to check all previous remarks anyway to see that there aren't comments contradicting the recently stated opinion.

The main reason behind the polls is that there are often people who do have an opinion on the subject, but for whatever reason don't wish to make a post about it - perhaps they don't want to say "Do this, it's just what I feel would be best and can't explain why"; perhaps they'd rather give their opinion anonymously; and so on. 

Yes, this is a good reason to have polls at some point. But my argument still stands, that we shouldn't vote without having a discussion first.

...perhaps in the meantime we should start on some backroute fixes for levels we can be fairly certain will get in.

Why backroute fixing levels at all? Aren't most of the L1 levels meant to have lots of alternative solutions? And if we really want to backroute fix levels, we run in a few problems: For most levels, I cannot see one specific solution that I would be reasonable sure to be the "intended" one. Secondly, with the huge amount of skills most levels give the player, backroute fixing will turn out to be very hard. And thridly, even if we succeed in backroute fixing, this will likely result in huge changes to level.
All in all, I thought the intention was to essentially use the original levels, and not to create challenge versions of them that force the player to use a specific solution?

One level we can perhaps start with - this scored 3rd place in the voteoffs, so I think we can be fairly sure it's getting in - is Mayhem 13.

Usually your eagerness to get things done helps a lot to reach decisions at some point. However this time I feel you are too eager. Forum threads are a very linear structure, which makes them a bad place to have multiple discussions about different subjects (which levels to include, how many, how to backroute, ...) in them. So I wonder whether it wouldn't be better to purely focus on the level selection right now, even if that means that we only discuss and don't decide anything for the next one or two weeks.

Regarding Mayhem 13: Even with your suggested fixes, the attached solution works.

[Proxima]

I agree we shouldn't decide about backroute fixes yet, but I think (for some levels) we should do them. As I see it, the point of this project is to collect the best levels, and also improve them to make the playing experience more organic according to modern standards. This includes (as already mentioned) changes such as removing unnecessary time limits. On the last page, it was debated whether Wicked 18 is worthy of inclusion due to the hidden trap; we could consider revealing some of the more obnoxious hidden traps. I think Present 19 (Acrophobia) is worth including because the absence of direct drop does the de-backrouting work for us; if that were still in, it would not be worth considering.

We don't know about the developers' intentions, and we don't know whether any particular levels were meant to have multiple solutions or not. In some cases it's easy to guess: surely the developers would not have made two versions of Acrophobia / Rules to fall if they'd known about direct drop. In other cases, developer intention doesn't matter so much, the main reason to de-backroute is simply to improve the level as a puzzle: Sunsoft 19 (Let's go camping) is a clear example. It is really ruined by the backroute, which is so easy that no-one bothers to solve it any other way.

[namida]

Backroute fixing was more or less a consideration from the beginning for this project. If all we were going to do is reorder things and take out time limits, it would almost have made more sense to stick with the original suggestion of just making some slight tweaks to the NXPs of the official games. Obviously, we don't know the original intentions in most cases, and would have to rely on educated guesses and/or consensus. However, there's no reason we can't accept multiple solutions on levels - the cases we need to look at the most would indeed be those like "Let's Go Camping".

[bsmith]

I have been thinking recently that the best thing to do right now is to try to sort all the levels according to difficulty (ignoring major backroutes) - no decisions about what to include or exclude yet; no backroute fixes either.  That way once we reach the stage to cut levels then we will have a focus on where to cut.  I'm going to try to do that this weekend.

[Minim]

Not a bad idea, but maybe we should simplify that and categorise which levels are "tutorial", "easy" and "normal" levels before we can sort them out into difficulty ratings. At the moment, we have 8 tutorial levels (Including Fun 11). "Easy" levels are those levels with many-of-each skill, and a lenient percentage requirement. We have about 25 of them (Though I might consider "Rendezvous At the Mountain" an Easy level as well despite the layout). The others are considered "normal" levels. Here's the list of "Easy" levels.

Spoiler

Covox Mayhem 02 "JackLemmings"
Genesis 11 "Where do you see Lemmings?"
Present 05 "Darkness of the royal family"
Sunsoft 13 "Doomsday"
Sunsoft 18 "Sunsoft Special"
H93 Flurry 09 "A Block from Home"
Tame 01 "Down and Out Lemmings"
Tame 04 "Downwardly Mobile Lemmings"
Tame 08 "Dangerzone"
Tame 09 "And Now This..."
Tame 10 "New Lemmings On The Block"
Tame 11 "With Compliments"
Tame 12 "Citizen Lemming"
Tame 13 "Thunder Lemmings Are Go"
Tame 14 "Get A Little Extra Help"
Tame 20 "Custom Built For Lemmings"
Fun 22 "A Beast of a level"
Mayhem 22 "A BeastII Of A Level"
Taxing 02 "Watch out, there's traps about"
Taxing 15 "What an AWESOME level!"
Tricky 11 "Lemmings in the attic"
Tricky 14 "Menacing!"
Tricky 21 "All the 6's..."
Tricky 28 "Lost something?"
Tricky 29 "Rainbow Island"

[namida]

If we're going to use special graphics levels, I think we should keep the whole "one per rank" thing. I believe other people have said the same. Therefore, they should only be considered relative to each other, not relative to any other levels.

[Proxima]

Also, levels with several of each skill are not all easy and we definitely don't want to put all such levels before others. Watch out, there's traps about is much harder than Catch more floaters.

[ccexplore]

I feel like rather than seeing an intended solution for Mayhem 13, I want to see some of the solutions that are supposed backroutes (especially Minim as he seem to have rather specific criteria) and see how truly easier they are.  I'm not opposed to eliminating certain classes of solutions such as removing the blockers, but I guess with the Fun repeat-counterpart being so much easier, I had never even really considered there are backroutes at all in the Mayhem-repeat version.

[namida]

This would be the first one to come to mind; although it could very well be argued that this solution is clever enough to be worth leaving as well. Also, see Nepster's replay above.

[ccexplore]

Incidentally, the possibility that we might include just a subset of Tame levels just feel very strange to me.  I think I'd rather we cull more levels rather than ending up with a situation where we may be forced to use any Tame levels as padding.

[namida]

Unless we cut down to a lower amount of levels (or at least lower per rank, with a larger number of ranks) - which is, of course, also an option - the problem that arises from doing so would be that even in the first rank, things might get too hard too quickly. There's a lot of great levels among the harder ones, not so much so among the easy ones. If also keeping repeats to a minimum (currently, IIRC, we have only 5 repeat pairs), this quickly results in a situation where easy levels are lacking.

[Proxima]

Well, my opinion is that we should cut down to a lower number (I'm going to go through the list and make suggestions, as soon as I have time to do this). I'd like to use some repeats to pad out the easy set -- while not all of the original Fun set are worth including, there are some that are really good fun levels -- lots of solutions, potential for challenges, and the hard version has particular requirements that ensure it presents something distinctly new. At the moment I'll just say that Fun 27 is a clear example of the kind of thing I mean.

Also, I'm still in favour of using a small number of the PSP levels, as there are some excellent easy levels in that set.

I think, if those changes are made, we will need fewer Tame levels, but we should still include some. They do the job of bringing the gap between the skill tutorial levels and the first true puzzles much better than the original early Fun levels. In particular, Tame 1 and 4 stand out as good levels to include early in the revised Fun; Tame 5 is worth special consideration as it introduces steel. Tame 14 is a good inclusion as it's a bit more demanding than other Tame levels, and introduces the rock tileset.

[Minim]

I think cutting it down to a lower number is a good idea, rather than increasing it to make sure it reaches 200, because there are some levels, like "Welcome to the party pal", that are remarkably tedious and builder heavy. These sorts of levels - unless if they have an innovative feature, like "We All Fall Down" for example - should be removed without question.

[ccexplore]

Hmm, I guess I can at least accept the idea of using Tame as a way to introduce ONML tilesets.  I guess I'm so used to original Lemmings that I don't feel like there needs to be anything in between the tutorial levels and the Fun levels following right after them, but I get others may feel differently and I'm not oppose to the idea.  Anyway, maybe I'm not the best person when it comes to opinions related to easy levels.

And you're right namida, if we also want to minimize repeats, we probably have no choice but to replace some of the easy but repeated levels with Tame levels, even though IMO in doing so we may actually be replacing a level rejected merely due to repeats with a "worse" one from Tame.

[Proxima]

I guess I'm so used to original Lemmings that I don't feel like there needs to be anything in between the tutorial levels and the Fun levels following right after them

That's not quite what I'm saying. Between the tutorial levels and the first true puzzles (levels like "Catch more floaters", "Just 17", "Easy when you know how") there should be some easy N-of-everything levels with no or few hazards that allow the player to mess around and get a better grasp of what the skills do and how they combine. Using a few Tame levels is a great way to do this.

The original Fun 8-16 do the same job, but overall they don't do it particularly well. Fun 8 and 9 are completely trivial, worse than anything in Tame. 10 is annoying because the most obvious solution is to make everyone a floater. 11 and 12 are good levels. 13 is We All Fall Down. 14, 15 and 16 are quite good, but unnecessarily long and builder-heavy.