Completed Armani's R1 level - let me just say that was absolutely beautiful... definitely my favourite level so far.  You were right about the last few skills being the puzzling ones - I worked out where everything else went, but with the remaining skills...

...it begged the question:

How do you release the crowd?!

R2 and R3 still giving me trouble unfortunately, despite your clues.

I've decided to release a V2 for my R3 not because of backroutes (no one has backrouted it yet) but to fix it up a little more to make it look better. In particular, I've replaced some steel pieces with its corresponding green piece and vice versa. Yea, I think I'm at that point where even small stuff like that in my levels bugs me

R1 remains V3!

Changelog History of R1:

V2 - Removed some terrain to fix a backroute I discovered. That's what I get for going overboard with the decorating!

V3 - Shifted steel blocks on the right side to exactly line up with the thin terrain pieces above to fix a backroute thanks to Crane. Another dumb mistake on my part because I thought they were already perfectly aligned!

---

R2 remains V2!

Changelog History of R2:

V2 - Added a button to fix a backroute thanks to Crane.

---

V2 of R3 uploaded!

Changelog History of R3:

V2 - Fixed up the level some more to make it look better.

OK, so after a failed attempt to embed the resources (which would only solve the default style anyway), I tried exploding lems to replace the missing Sleeper. This is fun, we get the fireworks, but it takes longer for the animation to play out, requires extra code for having the explosion not leave a crater, and fills the screen with particles which may obscure important level detail that the player may need to see.

So, let's try the Vaporizer (Burner) animation instead, effectively turning the exit into a fire trap when time runs out. This is shorter even than the Sleeper animation, less code, and less impact on the visuals (which, again, may be important).

That handles the missing sprite. We also need to handle missing animation data from scheme.nxmi. After a bit of unsuccessful tinkering with various bypassing methods, I've opted to simply add the missing data to the scheme.nxmi file at runtime. This is relatively low impact, and in fact helps any sprite set designers because they won't then have to manually add the data themselves should they wish for their style to support the Sleeper sprite.

This seems to be the best solution overall. It's highly unlikely that any other sprites will ever be added to CE, so the special handling in this case is acceptable. And, it only applies to users who opt to run CE from their original NeoLemmix directory. If doing things the other way around, the Sleeper sprite will be there anyway.

Implemented in commits 2651404 and 7c9de20.

I also went for pillar. Crystal is the easy pick for sure, but definitely overused and so 2022. You're Armani or not ?

Quote from: jkapp76 on November 16, 2025, 02:56:39 AMCould you add some code so if the sleeper sprite is not present something else happens? (explode comes to mind)

This is actually a great idea: have the lems explode rather than exit.

I'll have a go at embedding the Sleeper sprite (and default\lemmings\scheme.nxmi) into the .exe first, and fallback to default sprites if a custom set doesn't include the Sleeper (I think this happens anyway) just so that we have this as a potential option. Even so, explode might be the best way to handle this: it removes the lemming, updates the "available" count correctly, and it's more fun/novel than some awkward fallback behaviour.

Are you using CE or NL 12.14? Please note that no further updates will be made to NL 12.14 - However, since this involves user data loss, namida may wish to make an exception in this case (although this can't be guaranteed).

In the event that this issue isn't resolved in NL 12.14, I'd recommend giving CE (Community Edition) a try. There is now a Show/Hide button in the Level Select menu which shows/hides the Advanced Options tab (so, no need to change the settings). This will at least make it a bit quicker to get to the Mass Replay Check (now called "Replay Manager").

Also, CE's Replay Manager allows you to append the result of each check to the replay's filename, making it much quicker to find replays that haven't passed validation for whatever reason. This should at least help to streamline the process a bit.

Regarding the checkmarks being lost: next time it happens (in either version, NL or CE), please check the following immediately:

1) Is the settings\userdata.nxsv file present?
2) Is the settings\userdata.nxsv file populated with your level progress data?
3) At the top of the settings\settings.ini file, what are the values for "LastVersion=" and "UserName="?

This should help to narrow down exactly what the issue is. Best guess is that the userdata.nxsv file isn't being read from in certain circumstances, but if the data itself is being wiped that's a pretty big problem that we could do with fixing.

Added a button for "Save As Custom Skillset" which allows the user to save the currently-applied skillset as a custom preset, as per the method described above. When saving, entering the name of an existing custom skillset will overwrite the existing one.



Implemented in commit 5bc7fea

The "Set All Skills To Zero" button has now been changed to "Set All Non-Zero Skills To" + [N], where [N] is a number control that's set to "0" by default. This essentially does exactly the same thing, but now the user has the option to specify a non-zero amount to be applied to non-zero skills:



So, for example, if you wanted to have 4 of each Floater, Blocker, Basher, Miner:

1) Set Floater, Blocker, Basher and Miner to "1" by clicking the ^Up arrow on the [N] number control for each of the chosen skills.
2) Set the [N] control (shown above) to "4"
3) Click the button. This would set all of the chosen skills to 4.

To achieve "Set All", you'd non-zero all skills first, then set the amount you want, then click the button.

Implemented in commit f0bd7a0

The first benefit is proof of solvability.

The second benefit is that you document the intended solution.

You'll forget a few of your own intended solutions within 5-10 years. Or you pass maintainership of your pack to another community member. In both cases, your documented solution helps judge backroutes.

-- Simon