This is something we've talked about for a long time, but thanks to kaywhyn's videos, it's finally time to get started.

For those who don't know, Cheapo (by Peter "Essman" Spada) was one of the main engines for custom levels in the forum's early days (around 2005 when I joined). Although its place in Lemmings history was mainly as a platform for usermade content, it came with its own set of levels, which kaywhyn is playing through on video, and I'm using those as a basis for remaking the levels in NeoLemmix.

The zip currently contains Levels 1-10 of Fun, which isn't much, but it's a start 

It's time to finally get this started. After consulting with WillLem, kaywhyn and mobius about our plans for the new version of Revenge of the Lemmings (and a possible sequel pack), I've been put in charge of level selection and ordering. However, I don't want to do this entirely on my own; I want this version to be, as far as possible, something that everyone's happy with, so that it can be the final version and we won't feel the need for another re-selection ten years down the line.

Firstly, I'll start with an outline of the current situation.

* There are two versions of RotL I'm working from: version 2 and version 4 (as well as the RetroLemmini version, which is based on v4).
* v2 has 210 (7 x 30) levels; v4 has 240 (6 x 40). There are 113 levels in common to the two versions, thus a total of 337 levels between them (plus four that are only in RL).
* There is also a compilation of "RotL Outtakes", 232 levels that were considered for v4 but didn't make it in. Because these levels were already examined and rejected, there's less reason to consider them, but they are available if we need to make up the numbers, or if any of them are especially good.
* My goal is to split the pack into two packs of 4 x 30 = 120 levels each. Since we have more then 240 levels to choose from, not all will be selected. (The rest can be released as a new "Outtakes" pack so they don't just disappear.)
* The main reason why the level selection changed so much from v2 to v4 is that mobius wanted to remove all levels that are also found in MazuLems, Clammings and Insane Steve's World.

We will need to make a decision on this last point, so I've opened a poll. Should we go along with the removal of these duplicated levels, or restore them to the level pool? I'll do my best to sum up the pros and cons of each choice:

* In favour of keeping them: RotL is meant to be a snapshot of the best of the levels produced by the community in the pre-NeoLemmix era. Removing levels by three specific authors (many of which are excellent levels) goes against that goal.
* Other authors may have similar compilation packs made for them in the future, which would make it even more arbitrary that just these three get excluded from RotL. (weirdybeardy already has such a pack; one for Ben Bryant is in development.)
* The v4 levels that replace these may include levels that are not as good, and bring the overall quality of the pack down. (I am in the middle of playing through v4 so that I'll be better able to judge on this question.)
* The duplication argument doesn't entirely apply to RetroLemmini, since Clammings and Insane Steve's World are not available for RL (though MazuLems is).

* In favour of removing them: MazuLems, Clammings and Insane Steve's World are all available as separate packs for NeoLemmix, so NL players would have to play the same levels twice to get full completion.
* Removing them has allowed v4 to introduce a bunch of new levels, some of which are very high quality.
* Keeping them removed would make my task as pack compiler shorter and easier 

Oh, and just to make things clear, while I will take the outcome of the poll into account, it is not binding.

I decided to compile some trivia from the Zendo, play by forum game. Since the game moves quickly at times, I decided to do this in a separate topic so it's easy to find and update my posts as the game continues.

First, here is a list of all starter koans played so far. (Yes, round 18 had two of each colour.) Of course, there's nothing wrong with playing the same starter koans as a previous round if the rule is different! It's interesting that the only exact repeat pair we've had, WHITE/BLACK, was repeated by the same player (ccexplore).

     White         Black
1   LEMMINGS        CHEAPO
2   INSANE          STEVE
3   SIMON           GEOO
4   WHITE           BLACK
5   STOOGE          GATES
6   NEWPORT         CARDIFF
7   BROWN           BLACK
8   BUDDHA          NATURE
9   WHITE           BLACK
10  CLAM            SPAM
11  YELLOW          BLUE
12  QUILLSHWAMMY    LIGHTYAGAMI
13  LEMMING         PROXIMA
14  ELEUSIS         ZENDO
15  NO              YES
16  EUCLID          EULER
17  ALABAMA         MISSISSIPPI
18  FIRE, MIGHTY    FIRES, MIGHT
19  EVEN            ODD
20  RAT             RAGE
21  BUBBLE          WRAP
22  ENOUGHTHINGS    OTHERSTUFF
23  GABRIEL         CHARLES
24  EDITOR          DOSBOX
25  BANANA          APPLE
26  PINEAPPLE       LOOPING
27  THOUSAND        MILLION
28  MARY            POPPINS
29  BANANA          APPLEPIE
30  LEMMING         LIME
31  POTATO          CARROT
32  JOY             SWIM
33  END             START
34  WRONG           INCORRECT
35  NORTH           SOUTH
36  BUCKLE          UP
37  TAXING          MAYHEM
38  NOT             KNOT
39  CITADEL         CASTLE
40  TELEPHONE       CELLPHONE
41  CASHMERE        SILK
42  TRIFURCATION    ASYMMETRIC
43  LIGHT           BRIGHT
44  FERRET          WEASEL

Back in 2019, there was an overhaul to contest voting, because the old system often led to a huge number of voting rounds, which is a lot of work for the organiser and puts a heavy burden on voters if they want to commit to taking part in the whole thing, which can lead to excluding those who don't.

The current voting system, while definitely better, is still flawed. We now have only one vote for each rule; that's great. But if the numbers fall out unluckily, we can end up with a situation like that in Contest #29, where a mixed group of six levels needs five or even six voting rounds to establish a winner.

So what's the solution? Well, first we need to decide what the goals of the voting system are. At the moment, they seem to be:
* To establish an overall winner -- obviously the most important goal.
* To establish a winner for each rule, since this carries a certain amount of prestige in itself, especially for rules like "Make a Tame level" where entries have no chance of winning overall.
* To establish overall 2nd and 3rd place -- is this still felt to be important? There are no prizes for these places, but do those who regularly take part in contests care about them?
* To separate out the top three and then rank them, so that the overall winner is not decided until the last voting round.

If all of these are still felt to be necessary, my suggestion would be:

* As in the current system, each rule has one vote only, no tiebreakers.
* The first round of the mixed group aims to establish a top 3. If this isn't possible because of ties, a tiebreaker is held.

For example, with the votes from contest #29 (6 for A, 2 for B, 3 each for C, D, E, F), A would be into the final, B would be out, and a tiebreaker would be held to find the top two of C, D, E, F.

Ideally, this system limits the mixed group to three rounds plus one possible tiebreaker. Unfortunately, because of the requirement that an exact number of entries qualify for the top three, it's still possible for the tiebreaker to need a tiebreaker (e.g. C, D, E, F get 7, 4, 4, 3 votes respectively). I would suggest that in that case, we could just be okay with not having a third place (or having a tied third place), so that the mixed group cannot have more than four votes.

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.

As a reminder, we decided on a meet-up in Leeds on 5 August, which is next Saturday. I'm starting a new topic so that we can keep track of who is definitely coming, so that on the day we know when we've got everyone.

Unfortunately it looks like it might be raining on the day, but I'm sure it will still be great to see everyone and we can have a good time.

The plan is to meet at 11:30 on the main concourse of the railway station, then when everyone has arrived, go for lunch together. There is a Wetherspoon's at the station, so we could eat there unless it's too crowded -- quite possible on a Saturday -- in which case there are plenty of places nearby.

After lunch we could walk around town and chat if the weather is okay; if not, maybe we could establish a base at the Waterstones bookshop? I would be happy to take any other suggestions if anyone has ideas.

Confirmed:
Proxima
WillLem
Ste Woz Ere
∫tan x dx
Flopsy
Kingshadow3
Pooty
Crane (confirmed on discord)

Here is a spreadsheet of almost all content posted on the NeoLemmix Levels board as of today (5 July 2023). The only packs I've excluded are any that were fully incorporated into later packs, and migrations of the original levels. (If I've missed any packs I should have included, let me know!)

In many cases, I've trusted authors as to their claims of how many ranks and levels were in a pack, so let me know if any of these are wrong. The "date posted" column has the date in the first post of the forum topic; for NL conversions of older packs, this may not match the actual date of publication. I haven't gone to the trouble of looking through the long Lemmings Plus topic to get a publication date for each pack.

Since this is a spreadsheet, you may find it interesting to sort by author or by number of levels to get some interesting statistics. You could also use this spreadsheet to track your progress on packs solved (I would be interested to see kaywhyn's especially!)

Is there any other data you'd like me to add to the spreadsheet? Please bear in mind that other data (such as talismans, or yes/no on whether the pack exclusively uses the original 8 skills) might be a lot more time-consuming to collect, as I would have to install all these packs and examine them in the Level Select menu, so I make no promises about when I might get round to this.

Following on from a chat on the discord today, we are thinking of organising a meetup for those Lemmingsforums folks who live in the UK. It will most likely be in the north, since we discovered that at least three of us are located in Leeds and Liverpool, but nothing is set in stone yet. Some points to discuss before we go ahead with making definite plans:

* Who's interested?
* Anyone responding, please say where you live and whether it's possible for you to travel to a meetup location.
* What dates would be best / are there any dates we should avoid?

I know that no new suggestions are allowed now, but this is such a small tweak I felt it couldn't hurt to ask...

When viewing a pack or rank in level select, the talisman window is just short of being able to fit information on six talismans with names, and it would be nice if it fitted. (Also, the "p" of "completed" just above the talisman window is cut off in a weird way.)

Spoilers for GemLems level titles and talisman names, I guess....

I've started watching Ramond's streams and every so often I'm going to pause and note any comments I have, whether it's issues I think should be fixed or just general reflections. This will include my feedback on Simon's comments from the other topic. (Simon -- feel free to merge the topics if you think it's better that way.)

L4 Diamond Dash: This should provide a few walkers instead of imploders and blockers.
L8 Climb to Freedom: It was suggested that we should separate the two snow levels. L8 is very simple so maybe it could even move earlier?
L10 Escape the Pit: Unintended solution by digging and bombing horizontally through the terrain. I checked and this is even possible with imploders alone. I think it's fine to leave multiple solutions so the player can exercise some creativity, but the replay collection should showcase the intended solution (platforming over the pit).
L11 Beneath the Lab: There was a suggestion to move this later. Maybe switch it with L17 Solar Eclipse?
L15 Put Your Lix on Ice: Hacky unintended solution by repeated mining to break through the pillar. Again probably best to leave this alone. Pillar just before the exit should be less precise to find the place where a miner can break through and then stop herself.
L19 Let's All Go Down the Strand: The crowd can be turned around at the very top or on the second step. Any solution is fine here, but the "main" solution is to have the entire crowd go down and build back up.
L20 The Borderland: Requirement of 30/60 allows saving either group while ignoring the other, with the assumption that doing so is a bit tricky and first-time players will be nudged towards saving most of both groups. Is this okay, or should we just make it 40/60?

I recently added these tiles to the NeoLemmix version of my tileset. Do we want to have them in Lix as well? One problem is that the NL versions are scaled down so that each pair of tiles has a 32x32 and a 16x16, but for Lix we would want them to be 64x64 and 32x32.

In the level select menu, clicking on a level sometimes doesn't display its preview, but continues to show the preview of the old level.

I observed this with the pack BadLems, and it might be because some of the levels have the same ID. I spent some time clicking on levels in Lemmings Redux and they all previewed correctly.

If it's decided that this won't be fixed because having levels with the same ID is just something you should not do, then I would suggest that packs on the NeoLemmix Levels board should be fixed to avoid this issue.

I opened an unfinished level file (originally created with an older version) in the editor, finished and saved it, and edited my GemLems levels.nxmi to refer to it. The level loads fine in the editor and playtest mode, but trying to open it in NL causes a crash. I've screenshotted the error message, and attached the level file just in case that helps to pinpoint the problem.

[EDIT: Since I am no longer experiencing this problem, I've removed the level file.]

During building of a level, it is quite common to have no exit, simply because you haven't yet built the part of the level that will contain it. It is also quite common to want to playtest the level, for instance to check how many skills it takes to get from A to B, before the exit has been placed.

Even though a level, by default, starts with 40 lemmings and a save requirement of 20, when there is no exit then NL changes the save requirement to 0. This means that any playtesting of the level will be a successful completion, and will result in saving a junk replay. While these don't take up much space, generating junk files can be an annoyance and can make it harder to find replays that you actually want.

I suggest that either exitless levels don't have the save requirement changed to 0, or if there's some other reason why that has to be kept, don't autosave replays on levels with a requirement of 0.

If a level has 100 lemmings and I save 100, when I view my records, the "100/100" runs awkwardly into the clock icon. The clock could easily be moved to the right as there is lots of unused space.

I asked namida over discord for permission to make this topic, in view of the decision about "no new physics suggestions". This is actually a suggestion I made a while back that got sidelined, so namida said it was okay to bring it up.

The attached level (requires proxima_greenhill style) shows a problem I have with the jumper: on rough terrain, if you assign it just before a step of at least 2 pixels, the jumper immediately stops, effectively wasting the skill. This is uncomfortable, and my suggestion is to relax terrain checks at the very start of the jump.

namida points out that any change could affect existing content, but I think this is such a small tweak that it would be unlikely to break anything, especially as it's very undesirable for backroutes to be blocked with single pixels.

Any support for this?

Let's say I want, for whatever reason, to combine two or more packs into a mega-pack, so that the structure is:

Mega-Pack > Pack A > Ranks A1, A2, A3
            Pack B > Ranks B1, B2, B3

This is relatively simple in NL: put both packs inside a mega-pack folder, remove "BASE" from their levels.nxmi files, and make a levels.nxmi for the mega-pack.

However, there's one small issue. NL's main menu displays the name and author of the current pack, just above where it displays NL's version number. This information is, correctly, read from either Pack A's or Pack B's info.nxmi, rather than that of the mega-pack.

Now, if I use the arrow keys to select a rank, I move from A1, A2, A3 to B1, B2, B3, but when I move onto the B ranks, the name of pack A is still displayed.

Taken from WillLem's "Ideas for challenges" topic. The idea is to find the fewest number of skills per level to kill all lemmings instead of saving them. Nuke is obviously disallowed, and so is running out of time.

Lemmings Redux

--------------------------------------------------------------------------------------------------------
|Rating | 1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 |
|-------|------------------------------------------------------------------------------------------------|
|Gentle | 2  0  0  *  1  2  6  2  1  1  1  0  1  0  0  0  1  0  1  0  0  0  0  0  1  1  0  0  1  0  1  0 |
|Quirky | 4  0  1  0  1  1  1  2  7  0  0  1  0  1  0  0  1  2  2  0  0  2  1  0  1  0  0  0  1  1  0  2 |
|Zany   | 1  1  0  8  0  3  1  2  0  0  0  0  1  0  2  1  0  0  1  1  0  0  1  0  0  0  0  0  4  0  0  0 |
|Manic  | 1  0  1  0  *  1  0  1  0  0  0  0  0  0  0  0  0  1  0  1  1  0  0  1  0  2  0  1  0  0  0  0 |
|Lunatic| 0  0  0  2  0  3  1  1  1  2  0  0  0  2  1  0  0  0  0  0  1  0  0  0  0  0  0  1  0  1  3  1 |
--------------------------------------------------------------------------------------------------------

* On this level it is impossible to kill all lemmings.

I would like to add high-res graphics to proxima_tile, which is after all my "flagship" tileset in the same way that willlem_lemminas is WillLem's.

The biggest problem with doing so is the spheres:



(Left: the high-res tile as seen in NL. Right: the actual tile, pasted in with GIMP.)

As I understand it, NL deletes pixels where the low-res graphic has an empty pixel. I understand the argument for doing this -- that it makes the physics clearer -- but choosing high-res is in any case a decision to prioritise appearance over physics (look at the lower-right of the sphere; you cannot tell exactly which pixels are solid at all), and the player always has the option of turning on clear physics mode if they need to check the precise physics for any reason.

I want my high-res tiles to look hi-res rather than having the janky, pixellated edge seen in the above screenshot.

Speedrunning the Lemmings games has been getting quite a bit of attention on the forum and Discord lately, but I don't think we have a topic for speedrunning in general, so this feels like a good time to start one! So... who else here has experience of speedrunning other games? Anyone have any achievements or anecdotes to share?

I know we have at least one speedrunner in our community: Simon is (at the time of writing) the world record holder for Jazz Jackrabbit, and has posted about his experiences in this topic. Of course, trying to go fast is still speedrunning no matter at what level, so please don't feel you have to be a world record holder or contender to participate in this topic!

Some games have time as the sole metric for scoring or competing, so I suppose that my introduction to speedrunning was trying to surpass my personal best times in Minesweeper, which occupied a lot of my free time when I was in college. My best achievement, and my first introduction to the thrill of setting a good time, was an 18 on Intermediate -- my best score on Expert was a much less satisfying 76.

Around the late 2000s, I discovered that my all-time favourite game series, Repton, had been remade for PC, so I bought all three games, completed them within a month, and then started to compete for best scores. In Repton 3, your score has three components: in-game score (diamonds, monsters and crowns); lives bonus (1000 per life remaining after each level); and time bonus (900 minus time taken in seconds). However, if you complete a full scenario (8 levels) in one sitting, then the in-game score will always be the same, since all its components are required for completing the levels at all; lives bonus will always total 24000 for a deathless run; so the difference in two scores will always equal the difference in times. In other words, Repton 3 scoring is speedrunning, just without an explicit statement of time in minutes and seconds.

When I first started scoring Repton 3, the online high scores were dominated by Chris Paul Andrews, except for the "Prelude" scenario, where Peter Van Ek's score was top. I remember that at the time, I felt Peter's score might be completely impossible to surpass -- the Repton games are entirely grid-based, so there are much more limited options for what you can do to improve a time, and the "Prelude" scenario doesn't have the luck-based elements that some other scenarios have -- for example "Egyptian", where on Level 5 one monster must be killed by fungus, which grows randomly. Still, I persisted with working out better routes, eventually surpassing Peter's score by 36 seconds, and claiming first place on all 32 scenarios, with a total score beating CPA's by just over 10,000 points (almost 3 hours!)

The trouble with Repton 3 scoring/speedrunning was that I never really had any competition. My friend Jonathan also did runs (and beat CPA's time on "Oceans" before I did, so he briefly held a world record), and Matthew Thompson was a spirited competitor, but neither was a serious threat. It's now over ten years since I improved any of those scores, and all my records are still standing. I have half a feeling that I'd like to submit the Repton series to speedrun.com so that I can motivate myself to play them again, but without the factor of competing with others, I'm worried that it would feel like drudgery and I'd lose interest quite quickly.

Recently, I've gotten a bit more into speedrunning through Steam achievements. Both Legend of Grimrock games have an achievement for completing the first level under a set time -- LOG 1 gives you four minutes, which is very tight; LOG 2 gives you six, when four would be about the same difficulty as it was in the first game.

But by far the game I've spent most time on is Hollow Knight. Again, my interest was kicked off by Steam achievements: there is one for beating the game in 10 hours, one for 5 hours, and one for 20 hours with 100% completion. The first and third of these are trivial to achieve on a second playthrough when you know where everything is; the 5-hour achievement is the interesting one. That may sound odd, considering the world record is a mere 33 minutes. But Hollow Knight is a metroidvania, with upgrades all over its map. The WR achieves this time by skipping everything that can be skipped -- which makes the boss fights extremely hard. In order to make the 5-hour achievement doable, I had to plan out which upgrades I would get, and what order I would get them in so that every boss in the run would be within my ability with the upgrades I had available. This was a very puzzle-like challenge, which I enjoyed immensely. I intend, if I am ever able to afford a faster computer that can cope with recording HK, to make a video showing the route I constructed, so I have continued to practise on the same route, and got my time down to 2:49. I believe that with my current skill level, I could get my time down even further by skipping some of the upgrades, but again, I don't feel motivated to work on this right now when I am unable to record my progress. I have also done a 100% run in 6:54, which is far above a good leaderboard time, but serves as a starting-point for bringing my time down.

Well, that's my wall of text Anyone else want to take a turn?