Lix 0.10.33 released.

Download for Windows 64-bit -- recommended
Download for Windows 32-bit -- fallback for ancient machines
Download for Linux 64-bit
Source code
Changelog
Issue tracker

How to update

Quick and dirty: Extract over your old installation. Level moves will then result in duplicate levels that remain in your level tree.

Clean method: Extract to new directory, then copy these directories from old Lix into new Lix:
user/
replays/
...and any levels that you've built yourself or added manually.

When we mouse-hover over the digger assignment, the bashing lix gets highlit because that lix has been that digger.

• In the tweaker, when you hover over an assignment, the line will be printed in white. The so-assigned lix on the map will also be highlit in white, regardless of where that lix is or what that lix is currently doing.
  
• Added geoo's 2 levels from Level Design Contest 30, Cycle Decomposition and Misaligned, to single/misc/contests/.
  
• Fix #499: Now, the dark background of on-map tooltips spans the full width of the text.
  
• Attribute the laser animation to Zanzindorf in doc/thanks.txt. I've forgotten that back when we added Zanzindorf's lasers. Sorry -- it's now finally in!
  
• Code: Remove the custom reimplementation of al_convert_mask_to_alpha() and revert to Allegro 5's. The custom implementation was an attempt to fix #431 (Magic Pink isn't converted on macOS), but it didn't help. In the meantime, a change to Allegro 5's image loaders have fixed #431.
  

-- Simon

Welcome to the forums! This is a pretty nice level pack, although I do have some things to point out.

Level 68 has a missing object that the style manager couldn't download. I loaded the level in the editor and the problematic object in question was an exit. For the purpose of my playthrough, I replaced it with the default crystal exit.
Level 87 has a ton of missing terrain pieces that the style manager couldn't download that I wasn't able to replace, so I wasn't able to complete this one.

I was able to complete every other level, except for level 49 which seemed excessively difficult. The replay that you provided for this one doesn't work on my version of NeoLemmix, which leads me to believe that there may be some shenanigans with different terrain pieces.
Are all your styles up to date?

While overall this pack was quite easy, there are quite a few headscratchers that I was stuck on for a while!

My favourite levels are:
13, 31, 52, 69, 70, 76, 83, 84, 86, 90, 92, 94 and 97.

Also, level 35 has the funniest replay.

My replays are attached.

QuoteLooking at the zip itself and opening the levels in Notepad and then later on in NeoLemmix itself, you have all levels titled 01-Easy, 02-Easy, etc. all the way to the final level, which is called 99-Easy. These aren't very catchy or interesting level titles. Surely you can think of better ones?

I personally didn't really mind the lack of titles. Design-wise, the levels have a minimalistic feel that I quite enjoyed, and I feel the lack of titles helped complement that. That's just my opinion, though, I'm not really sure what the consensus of the lemmings forum is on levels without titles.

Third stream will be tomorrow, Wednesday, June 10th, starting 17:00 UTC. This will be 19:00 CEST, 18:00 BST, and around noon in America.

-- Simon

Quote from: Simon on June 09, 2026, 10:15:53 AMWhen two levels in the same pack share an ID, it confuses the mass verifier, okay.

What is with levels across different packs? Is it okay to duplicate one of the IDs outside the pack, as long as the pack, among itself, has all-different IDs?

Assuming it's okay to duplicate IDs outside the pack: How do authors with N packs verify solvability? Verify each pack individually, and point for each pack to a different tree of replays?

-- Simon

Across different packs, duplicate IDs do not really matter.

Replays are being checked in the mass replay checker out of one folder for one pack, so as long as a pack itself has no duplicate IDs everything is fine.

That way we cannot really run out of IDs.
The extremely rare case it could matter is if you create a compilation pack with levels from different packs, but I have yet to see an ID-collision there.

For replay verification of N packs: I go to each of the N packs in the level selection tree and click on the Mass Replay Check button for each one and select the respective folder. So I verify each pack individually as you stated.

In addition to what kaywhyn said, when you play two levels with the same ID one after another, the engine thinks you are still playing the same level and will start in replay mode. Sure, this isn't a huge deal as you can click to cancel the replay, but it's confusing for new players and best avoided.

Ah, yeah.

They weren't gitignore'd properly in the repo, but weren't generally being copied manually when updates were occuring. They also weren't being included in the backups. I fixed the backup script to include them while doing this update, including that the backup I ran just before the update was with the new version of the script - so this time, I am able to restore them (and have now done so). And going forward, they should no longer be a problem (but can be restored from the backup if they are).

When two levels in the same pack share an ID, it confuses the mass verifier, okay.

What is with levels across different packs? Is it okay to duplicate one of the IDs outside the pack, as long as the pack, among itself, has all-different IDs?

Assuming it's okay to duplicate IDs outside the pack: How do authors with N packs verify solvability? Verify each pack individually, and point for each pack to a different tree of replays?

-- Simon

Thanks for the explanation, kaywhyn That probably explains why it didn't seem like such an issue to me, because I've never felt the need to use the MRC. If (when?) I get to creating levels, I guess I will do. Maybe players as well as creators use it to check replay folders, but I never have, so far.

Levels having the same ID is a problem when it comes to verifying your replays with the NL mass replay check because it confuses it. Basically, when a replay is generated for a level, it has the exact same level ID as the level. Thus, if you have two levels with the same level ID, you will have two different replays with the same level ID, even if the level title is different. Nxrp files are text files, and hence you can verify the level IDs by opening the replays with a text editor, such as Notepad.

The underlying idea with the MRC is when you run it on your replays, it checks the levels by finding the one with the matching level ID as the replay. When it finds a match, it checks the replay to see if it solves the level and then it can spit out three different outcomes for the replay: Pass, Fail, or Undetermined. However, in this case, the mass replay check finds two levels with matching ID and hence the MRC will instead check the first replay with the corresponding level ID on the repeat level with the same level ID instead of checking the newer replay with the different level title despite having the same level ID as the replay for the original level.

The result => It can get very messy reading the text file that's generated after the MRC is done running when there's multiple levels with matching IDs.

That's why to prevent confusion with the MRC, it is advisable to always generate a different ID by using the random ID button in the editor, even for repeat levels. In this way, the MRC will run more smoothly when checking your replay collection for a level pack.

Split off Heart66 easy pack

Genuine question, as it's something I've never been sure about, but why is several levels sharing the same ID such a problem? If level 5 and level 11 share the same ID, the auto-generated NXRP files are differentiated by name and each will still play when selected for their corresponding levels, no? If it is such a problem, is there any easier way of fixing it than by manually checking for dupes and opening the problematic NXLV files in NLEditor to re-generate the ID with the Random ID button?