Given some recent (and not so recent) requests for various L2 graphics, I thought it'd be good to have all of them in one place, and hopefully easy to find.
All of this is more than a decade old, but I'm not sure I've ever posted an archive of all the L2 graphics that I managed to extract, so here we go.

Tribe specific graphics:

• l2gfx: 16x8 pixel pieces that level maps are composed of (including palette in top row)
• l2tiles: Arrangements of the above into sensible terrain pieces
• l2obj: Object graphics and animation
• l2ss: Special tribe specific animations

Other graphics:

• vlemms: all lemmings graphics (using cavelem palette -- in the game, these are recolored according to tribe palette)
• intern: additional universal animations
• cutscenes: Graphics from various cutscenes

If there's interest, I can dig around some more and try to find some relevant code pieces that we used for extraction.

Welcome to the 3rd installment of this contest!

This is a bit of an experiment, but I created the same level in both Lix and NeoLemmix. I understand that there are physics differences and that some solutions that might work in one engine might not work in the other, but I hope most solutions will work the same and that many minor differences can be worked around, making it a fair game. At least I don't think there are any systemic biases giving one engine a clear advantage.
That said, if people are unhappy with this, I will restrict the contest to a single engine (based on popular opinion).

EDIT: I added a second version of the NeoLemmix level. The only difference is two little steel bumps on the top steel platforms. This is to account for the difference that you cannot assign miners/diggers on steel in NeoLemmix, while in Lix you can. You can submit solutions to either of the two NeoLemmix levels.

Goal: Try to solve the level using as few skills as possible. Tiebreaker: Higher number of lemmings/lix saved.
(Note that this is different from the previous 2 contests, the main criterion and the tiebreaker have been swapped!)

Submission deadline for solutions: Sunday Feb 17th, 6:00 am UTC
You can submit solutions for one or both engines. Submit to me via PM. When you submit, submit a replay, as well write down the number of skills used and number of lemmings/lix saved.

At the end, I will post a complete ranking list (determining the next contest host) as well as engine-specific ranking lists in case some solutions from one engine end up not working in the other.

PS: I made the NeoLemmix version in a text editor, because the NLEditor in Wine doesn't display any objects or tiles, and placing invisible tiles is kinda useless. That's why it's lacking any decoration.

Submissions so far:
Minim
SQRon
Simon/Forestidia
Crane
Ramon

Results in this post further down the thread.

Issue: (1) on 45 degree slopes, and as far as I'm aware also 60 degree slopes, the behaviour when a tumbler/jumper encounters terrain is inconsistent.
(2) the tumbler code is a messy blob.

Simon is addressing (2) with a solution that cannot address (1). Thus I'd propose a slightly different solution, which is rather modular, and the modules are easy to comprehend in what they are doing. I think the proposed rewrite mixes moving and collision checks a lot more than my proposal, which should make the code clearer.


Subroutine (1):
Determine first point of impact via going along the trajectory 1 pixel vertically or horizontally at a time.

Subroutine (2) [takes coordinate as argument], or probably better even 3 separate subroutines.
Let's call the coordinate that we got as argument (0,0).
Is there space below at (0,-1) or (0,-2), if yes, we've hit a ceiling.
Is there space above at (0,+1) or (0,+2)? if yes, we've hit floor.
Is there space one pixel behind at (-1,0)? If yes, we've hit a wall.
Return those three booleans.
Remark: Notice that as (0,0) is the first point of impact by only going vertically and horizontally, at least one of the three will be true.

Subroutine (3):
Call (1)
Call (2)
If we've hit floor and ceiling, use y-speed to determine whether to call landing or bumping subroutine (this only happens if you have a 1-px layer of terrain).
Else If we've hit floor, call landing subroutine.
Else If we've hit a ceiling, call bumping subroutine.
Else we've hit a wall (and nothing else), call reflection subroutine.

Note: The reason I propose checking 2 pixels in the ceiling and floor check in (2) is to handle 60 degree slopes consistently. Inconsistent behaviour will now only happen at slopes going at least 3 pixels up/down per 1 pixel horizontally, which I don't believe exist as tiles.

In the past, a variety of gameplay mode ideas have come up for multiplayer, and Simon recently brought up the issue again now that the coding for multiplayer in D Lix commences.

In this post I'll try to list and discuss most feature ideas that have popped up in the past. The main considerations are how much complexity they'd add to the game (and the UI), and how much gameplay value they add.


Survival mode. There there are no exits, and it times the death of your last lix. The objective is to outlive all opponents. I think this always boils down to the same gameplay at the end, a 1 vs 1 fight/dexterity fest, possibly with some strategies such as bunkering. Clones had it and it was called Quantum Quarrel, and I think it didn't see much use (and wasn't particularly fun).
I don't think it warrants a special game mode as it will probably not spark many new level ideas. And it's possible to kinda make it work anyway as in "Last Man Standing".

NaOH's Ark-like mode. You play an asymmetric level twice, with the player positions swapped. Whoever saves more cumulatively wins.
I think this can be done by consensus and doesn't warrant a special game mode, especially if it's possible to pre-assign hatches to colors in the lobby.

Race mode. The game times your first save, or maybe your n-th save. The objective is to be fastest. There are two questions here.
(1) How interesting is "first to save n wins" where n > 1. My hunch is that it'd only particularly be used in levels with little interaction, where the goal is to be the fasted to solve the level (basically each player plays a singleplayer level at the same time).
(2) Are we also interested in 2nd, 3rd, ... place and their times?
If the answer is yes to both questions, a game mode where in the final standing players are listed with their time might be warranted. If the answer is no, and only first to the finish is of interest level-design wise, then just having a nuke that finishes the game instantly if you've saved at least one lix and the overtime is set to 0:00 seems simpler and would just do the job.

Neutrals. A post above mentions the capture-the-clone mode from clones. There is one or a few designated clones that nobody controls, and you win if you get one of them. I'd propose the following generalization which I believe has a lot of gameplay potential: Neutral lixes. They could be pre-placed in the level, or come out of neutral hatches. Nobody can control them, they just wander around. One caveat is that all neutral hatches should release lixes simultaneously rather than sequentially, to be able to achieve perfect symmetry in 2P levels. These could even be an interesting addition to singleplayer. Either way, they could be a game-mode flag "only neutral lixes count", or this counting could be implicit whenever neutral lixes are present. I don't consider the latter simplifying gameplay though, as the player would still have to check which mode is in effect; a flag instead would also allow to have everyone count which I believe (but I'm not as convinced) is also interesting. Note that with an instant nuke (see below) this also implements the capture the clone mode.

Nuke modes. I think a lot of options have been floating around here. Cuber nuke, burner nuke, stunner nuke... Note that the idea is always that nuke triggers overtime if at least 1 lix has been saved by the player. Let me discuss 4 of the nuke ideas here:

• Regular nuke: It's a bit disruptive, and you can badly mess up other players' routes. While this could be a strategic element, a lot of the effect is quite random. Maybe we just like it because we're used it, and it's gimmicky because "fun! flinging!".
• Cuber nuke: Just as gimmicky but not as cool. Don't see much use in it.
• Burner nuke: Player can set a timer for the game to end, but without disrupting the game. Other players can still finish their routes and nuking with the hope of messing up the other players' paths doesn't work anymore.
• Concede: Trigger overtime, but your lix keep walking around, you just can't control them anymore. You'll think twice here if you nuke, and also you can't bail-out nuke when your bunch is walking towards someone else's exit. You can even nuke before your lix have entered your exit if you think they are safely walking into it. Could be implemented by turning the lix into neutrals. But at the same time they shouldn't be neutrals for counting purposes in the "only neutrals count" mode.

I think at most two different nuke modes should be implemented, and the last one is my clear favorite here. It could be visualized with white flags on the nuke button instead of the explosion. (Though that might be associated with losing the game, so maybe checkered flags for finishing the game?) I think I almost prefer the burner nuke over the regular one, but it's closer to the concede option while the regular nuke is the polar opposite, and has fun flinging. So I'd propose concede in addition to the nuke that we already have. Concede with overtime of 0:00 could also immediately end the game, thus indirectly implementing the first to the finish mode.
In particular, I feel that 0:00 in the editor being infinite over time is an artefact, and the editor should offer infinity as a time option (now that the cool infinity symbol is in).


In summary, personally I'd propose the following game-mode additions:

• Concede in addition to regular nuke
• Neutral lixes, and "only neutrals count" mode

I think this is a rather lean addition without too much extra complexity, but a lot of gameplay potential.

Level designers often like to cut terrain pieces into bits and paste them together into new shapes. Currently this needs to be done using black (eraser) terrain and the no-overwrite flag which is confusing and cumbersome, especially for more complex arrangements.

I propose to do away with the no-overwrite flag and support terrain composition instead. This would work with the following grouping functionality:

The designer selects an arrangement of terrain pieces (which may include black pieces), and can group them together into a single new terrain piece. Thus a new terrain piece is defined that behaves like a regular terrain piece. Parts of the composite piece that are covered by black terrain also part of the composite piece will become transparent. The original terrain pieces involved in the grouping are removed, but the game remembers which pieces a composite piece is comprised of, and also creates a new memory bitmap for this composite piece. Optionally, there could also be an ungroup functionality, which decomposes a composite piece into its parts.

Composite pieces can be defined recursively using existing composite pieces. The level stores how each defined piece is composed, and creates an internal memory bitmap of each such piece as it reads its definition from the level file. If a composite piece is not used in the level, the editor should ensure that the definition is also removed from the level file. Once the player loads a different level, internally the bitmaps of composite pieces can be deleted.


Backward compatibility.

Existing levels exploiting no-overwrite can be converted automatically.
Currently, the drawing algorithm draws the pieces in the order in which they are in the level file from back to front, black terrain erases, no-overwrite tiles don't overwrite already drawn tiles and are only drawn over void.

We achieve the same result as follows:
Input: An ordered list of terrain pieces, optionally with black terrain or no-overwrite flag.
Output: Definitions of composite terrain pieces, and an ordered list of (composite) terrain pieces (free from black or no-overwrite pieces) to be drawn from back to front.
We read the input list in order. At any given time, the current output O is an ordered list of (composite) terrain pieces. Three cases, depending on the next piece T in the input list:

• T is normal terrain: append T to O.
• T is no-overwrite terrain: prepend T to O. (Note: This works because O contains no black/no-overwrite pieces.)
• T is black terrain: Initially, define S as the set of tiles in O that T overlaps. While there exists a tile T' in O\S such that there are tiles F and B in S with F overlapping T' and T' overlapping B, add T' to S. While there exists a sequence of tiles T1 -> T2 -> ... -> Tn in O\S such that there are tiles F and B in S with F -> T1 and Tn -> B, add T1, ..., Tn to S. Here -> denotes the overlapping relation. Group the tiles in S into a new tile N, and remove S and T from O. Let F be the tile with the lowest index overlapping N, and let B the tile with the highest index that is overlapped by N. Add N to O somewhere between B and F. (The following is Nepster's idea, with a slight fix:) Define a set S+ of all terrain pieces of O\S that are reachable via -> from S (i.e. via a sequence for forward edges). Then reindex them starting from max(S), while keeping their order. Put N in the list right before S+, and delete T and S from O.
  

Implementation notes:

• If we want to allow composing steel and normal tiles, you'll have to remember the mask of each composite tile with bits for solidity and steel.
• If a level does not contain no-overwrite pieces, no conversion is necessary.
• Multiple consecutive black pieces can be processed in step, resulting in only one grouping. Black pieces that are not overlapped by anything else can just be added to the end of the list without grouping.
• While processing terrain pieces, you could maintain a directed graph modeling the overlap relation within O. As a new piece is added to O, the graph needs to be updated. When a grouped tile is added (and thus some pieces deleted from O), I think the edges pointing to the new tile can be inferred from the edges belonging to the deleted vertices.

Why are we thinking about a server move?
We haven't managed to contact the server admin, Adam, in a very long time. He seems to be paying the server bill, but there's uncertainty for how long that would last, and the forum could go down at any point in time, and data might be lost. Without server access we also can't update the forum software, which right now is outdated, buggy and probably prone to exploits for spambots like the recent invasion. On a new server with someone active being server admin we have complete control over everything.

This is what has happened so far:

• With Mr. K's help, we now have access to the admin control panel, but nobody of us has server access
• Temporarily, to fight the spam bots, user registration was turned off for a day, which I changed to new users requiring approval by an administrator. It seems like there haven't been new members trying to register within the past two days, so I'm setting in back to normal now and see what happens.
• We can't get hold of an up-to-date version of the database (that'd contain all user information, posts, everything we'd need to just move the forum over without issues)
• Adam (and maybe Liam, whoever he is) is the only one able to get us the full database and the domain www.lemmingsforums.com, we have tried to contact him, but without success so far

This is what we have access to right now:

• Threads and posts from the Lemmingswelt forum (2004-2007 or so)
• Complete database from the isamedia forum (2007-2009 or so)
• Database of this forum from January 09th, 2014. All newer data is missing
• namida can get the complete set of theads/posts/attachments of this forum at any point in time (I believe)

My suggestion is to allow some more time to contact Adam, discuss and not rush things, while still keeping posting in this forum and leaving namida's forum as the emergency option should the forum suddenly go down. (I don't think it's very likely that the forum will go down within the next month.) We should prepare everything for a swift move, and once preparations are finished set a date (and announce it a week in advance) for when the move shall take place. That day namida would grab all the data off the forum and import it over on the new forum, and then this forum would be closed (or at least become read-only to ensure new content will be posted on the new server).

The reason I suggest to keep posting here for now is that moving to the new forum would encourage reposting topics which would have to be merged later, which becomes messier the more time passes and thus puts us under time pressure.

If we can somehow get hold of Adam, this would make things a lot easier as we could just import the whole database and get a full copy of the forum, and in addition the domain www.lemmingsforums.com. So I encourage everyone to help us in our effort to contact Adam!
Should this not work, I'd suggest purging the few posts from the new forum and importing the members from the January 2014 database (this means we'd also have most PMs, and almost all posts would automatically already be linked to the users), while taking all the posts that namida will grab from here over. Furthermore, we could add the posts from the isamedia and lemmingswelt forums and try to link them to the user accounts (something that I feel very strongly about). If you forgot your password from January 2014, or if you registered later than January 2014 you could identify yourself and get access to your account/get it linked to the posts that you made. Maybe we could also get the PMs from the isamedia forum working, I personally find having an archive of my PMs important, but I don't know what other people think about that.

The plan I outlined is just my opinion.

Feel free to discuss!

This is an incomplete alpha version of the Lix Community pack. We're getting close! Thanks everyone for contributing so far, and keep it up so we will finish soon!

The last 4 ratings (Cunning, Daunting, Vicious, Hopeless) are ordered. Please comment on any level that you think is in a place it doesn't belong. I adapted my ratings to take Nepster's comments into account, and then determined a positon based on my and Proxima's ratings. I also went for diversity, i.e. don't put similar levels too close together. The last few levels of a rating might be harder than the first few of the next rating.
I gave NaOH the task to come up with an ordering for the first 40 levels, possibly with adjustments to some levels to smoothen the difficulty curve.
Once that is done (maybe after some discussion), I'll compile a version where all levels are sorted into their ratings (probably still with excess levels).
Please also comment on any other issue there might be. There might be inconsistencies in capitalization of the level titles or author names, please point them out and suggest a fix.

A few levels need the new terrain updates: http://www.lemmingsforums.com/index.php?topic=1002.msg20541#new
Latest version of the level bundle (.zip): https://github.com/geoo89/lixlfpack/zipball/gh-pages
Level list (html): http://geoo89.github.com/lixlfpack/
Level list (xls spreadsheet): https://github.com/geoo89/lixlfpack/blob/gh-pages/lixlfpack.xls?raw=true
Date of last update for each level: https://github.com/geoo89/lixlfpack
If you want all the levels viewable in lix in alphabetical order, just copy all the levels from the rating folders into the main folder.

This is a bit rushed, but I won't be home for the next three weeks and won't have much time to work on this. Please don't feel discourage from posting in the thread by my likely low activity for the next three weeks in this thread.

Right now there are 17 excess levels, meaning we have more than 40 per rating.
Below I compiled a list of levels that I consider in potential danger of being scrapped.
You're encouraged to compile your own list of levels you wouldn't mind seeing gone.

Some levels are not final. You can still update your levels, in fact some levels still need work. See below.
There's definitely no shortage of levels, so in order to add a level after this alpha has been released, please put with your levels a suggestion or a few suggestions (preferrably not in the top 12 of my list) of which level your new level should replace (not in the same position, they can end up in different ratings, it's just that one level goes and another level comes in).

You're encouraged to add hints to your levels that a player can iteratively view when stuck in a level! Most levels don't have hints yet, I only have a list of hints from Proxima so far. While not a must, please consider adding hints to your levels! Send them to me, preferred format: Take this spread sheet from https://github.com/geoo89/lixlfpack/blob/gh-pages/lixlfpack.xls?raw=true and in the 4 tiny columns at the end there's space for 4 hints (if you need more, just continue in the next columns). At the end I will write a script to add the hints into the levels.

Please refer to levels with their name, not their rating as the latter might still change.



Some special ordering ponderings:
Switch Chasm (Vicious 40) and MERDEMERDE...(Early Hopeless)? - Chasm looks completely impossible, and it extremely hard to execute, that's why I thought it might fit Vicious 40, the most vicious of all levels. MERDE... is definitely easier, maybe a good breather in Hopeless. Another alternative for Vicious 40 might be Metal City Mayhem.
Which level last: The News Struck at Ten (formerly Stroke at Retirement Age), You Only Get One Bash at It, Brickout - It seems the Brickout will be a simplified version, and YOGOBAI has my easier alternative solution, so I went with TNSAT for now.
First level of Cunning: LIX, The Razor's Edge, Merge Sort - which one?

Hints:
LIX: Add 1 minute challenge (already added in spreadsheet, please say if objecting)
Too Far to walk: 0:55 left on the clock challenge (already added in spreadsheet, please say if objecting)
You only get one bash at it: Have to add hints (Don't want to read them)
Put your lix on ice: Hints says to turn a lix in a miner tunnel (I guess), but you can just turn on steel.
Labyrinth of despair: Add hint re. 100% challenge

Levels that need a design overhaul in order to stay:
All over the place
Heavenly skies
Escape the Pit

Levels that could do with improved design:
The Giant Mushroom
Three Days of the Condor
Roundabout
The Lix Who Japed
Well OK Then
Bulldozer
Nepster's complaints:
Babylon Fading
It's a long way up
Digging the Air (layout is good, but it doesn't fill one screen)
To Destroy is to Construct
Division of Labor.

Levels that are still WIP:
wontgetfooledagain.txt
thenewsstruck.txt
brickout.txt

Levels with known backroutes (backroutes attached):
lixcannon.txt
buridan.txt
3111.txt
nomoreheroes.txt
betchacantsave.txt (can save 2, though I wonder if the level is open-ended so there are no real backroutes?)

Suggested changes:
Lixology: replace with Breackout (attached), maybe with prettier water at the bottom once available
Dances with Lixes: Enforce trickier sol? (alternative version attached). Rationale: There are two quite different solution approaches. One of them is already used in 100% Built by Lixes and Ascending and Descending, while the other one is, I believe, novel. I attached a version with buzzsaws enforcing the second approach, I wonder whether I should include this version instead, as the ideas from the first approach are already featured in two other levels. (Spoiler: Approach one is building a bridge from the left, but not completely to the steel at the right, and have another lix build a bridge from the top of the steel to the left, thus falling onto the lower bridge. Approach 2 instead builds a bridge from left to right all the way to the steel, and then uses a bomber to simultaneously bomb through the bridge and make a dent in the block at the left that makes climbers turn around.)
It Takes Time to Build...: Cut down to 3 bashers/diggers?
Snowball Battle: Remove walker and increase spawn interval?
Just Stop the Bleeding: Still needs Steve's 'hole in the head gang' treatment


Below is a list of levels that are in danger of being scrapped, in order of magnitude of the threat that is exerted by me.
Other players are welcome to compile similar lists. Right now I think it's a good idea (once we have a first rating arranged) to compile an alpha with all 257 levels sorted, and have Akseli give this version with too many levels a playthrough and wait for comments, if Akseli is ok with that.
We need to scrap 17 in total:

Systematic Separation (Hopeless) - could go into RubiX' Epic Adventure
Low Profile (Vicious) - could go into RubiX' Epic Adventure
Tick Toxic (Daunting) - could go into RubiX' Epic Adventure
lixesinarms (Vicious) - part of MaZuLems
Bibbidi (Cunning) - Part of PimoLems
Diggin the Air - Part of PimoLems. Mobius wants it scrapped, Proxima likes it.
Confusing Fractals - Can't turn it into a decent tutorial level
Leap of the Locust - Yeah, very repetitive, and already uses the basher staircase trick. Proxima rating C. Nepster says it's not too repetitive, and actually you don't need the basher staircase trick.
Paragliding to the Pyramid - the level is tricky enough that it's well possible to stumble over significant portions of the solution of Endeavor in this version already.
The Giant Mushroom - nothing strictly bad with it, nothing outstanding though
Walk them golden stairs - Solution is kinda spammy and annoying having to go through the mesh with the given terrain removers. Once You Pop (sequel) is similar except that now you have bombers to work with instead of the other terrain removers, and the lack of other skills except for builders makes it a lot more interesting.
Wrong way - Proxima rating C. I don't think it's that similar to Minimalism (Part 1), still nothing really new here. Nepster says it's better than Minimalism (Part 1). I don't.
Three days of the condor (Cunning) - The only thing really is turning the miner with a blocker here, and it's a huge and ugly level for just showing this trick. Proxima rating A, so I don't know though.
Tower of Babel (Cunning) - Proxima rating C+. Again, pretty unique so I personally find it to be a good one to throw in the mix. The main idea is obvious, but there are a few details to work out. Execution isn't bad actually, especially considering the kind of level this is.
A fearful symmetry (Cunning) - A lot of unrelated stuff for a level that just revolves about one main trick.
Over the Hump (Daunting) - I feel there are a lot of levels with tons of climbers that are similar now
When the Levee breaks (Cunning) - kinda straightforward, mostly about not getting any details wrong.
Laser Deathroom 3 (Vicious) - main idea same as Part 1, just harder setup
Soaring (Cunning) - Proxima rating C+. It's quite unique, but not much of a puzzle and similar to Clam's 'Stall Tactics' there's the weird issue that sometimes they don't go straight up.
The Ring of Fnargl (Vicious) - Unique, but extremely technical, revolves around a precisely timed bomber
Heavenly Skies (Daunting) - This one is pretty tricky, though I personally don't like exercises in estimating how far your builders go. The route I found uses a move that's pretty hard to execute. Needs design overhaul.
Comfort Levels (Daunting) - rename for one.
Cornerstone (Daunting) - With Nepster's route I realize the level is a bit open ended, and there are probably better open-ended levels around (And even the intended route is a little tricky to execute).
Babylon Fading (Cunning) - solid but nothing outstanding about the solution
Play Ball (Daunting) - It's a good level, but the main part is realizing that you have to do synchronization compression. Then it's not so hard, but there are already a few levels that use this, so I wonder if this level is a bit redundant.
Well Ok Then (Cunning) - Proxima rating C+. I don't know something different yet simple, and I quite like it for that. Design could maybe be improved though. Also a bit of a precursor to Chasm.
Snowball Battle: Decent puzzle, not super special but decent. Makes use of the same bland design of the prequel.
A lix on the edge (Cunning) - I don't know what to think about it, it's tricky and interesting, so I guess it's a good thing. Again, I don't see much of a point in the time limit.
A completely ridiculous level -  Like Ramen Masterchef, I think it's nice to have a level or two like this, even if it's kinda weird.
Blocked by a snowball - It's a decent easy level and you have to take care about a few things without risking killing lix. Depends on what else we'll have at the beginning.
Escape the pit - perfectly good if it gets a design overhaul.

Other people would be ok with scrapping:
Proxima: Leap of the Locust, Confusing Fractals, Wrong way, Escape the Pit -- and Diggin' the Air
mobius: when the levee breaks, a lix on the edge!, play ball!, watch ye step (IS's but I remade it and he never made any comments about it), the last laugh, Division of Labor, bibblidi-bobblidi-boo (pieuw), Leap of the Locust



Mod Edit: Restored attachments.