[NeoLemmix] NepsterLems

[Akseli]

Another pack I'm revisiting now in new formats, I beta tested NepsterLems in January 2016 and played it to some extent after the initial release if I'm deriving my replay time stamps correctly. Now after loads of backroute fixes I've solved 79/111, replays attached. This is the first time I played Moon levels 7-15, they looked a bit daunting at first but turned out to be a lot nicer in the end. I didn't attempt Black Hole levels, because I have still 12 levels unsolved before that final rating, but I still thought of dropping my solutions at this point so they're at least saved here in the forums before my hard drive dies or something. :b

Levels I suspect backrouting: Planet 16: Dying Dream, Neutron Star 10: Broken Pillars, Neutron Star 16: On the Edge of Death

Not quite sure about even some early Planet levels, I'm interested to hear about my solutions when the time comes.

[Strato Incendus]

Okay, I revisited just the final level (Black Hole 21) - pretty much exactly one year after my first attempt - after having been told by Nepster that he doesn't consider it a "good level" (anymore? ) and about his philosophy having changed.

I actually think it's pretty genious - there's always even more skills which are locked in the place where they have to be used than one would think . As far as X-of-everything levels are concerned, this is as good as they're going to get!

Spoiler

Also, once one has a grasp of the core philosophy "avoid using builders whenever you can use terrain instead", it's a lot less intimidating.

The only thing that makes it "not a good level" is the RIDICULOUS pixel precision required all over it!

Spoiler

One pixel that determines whether a blocker can be freed or not, whether a fall is lethal or not, thereby requiring an extra skill to reduce the height, whether lemmings can gain enough height by stepping over an obstacle without being turned around...

I repeatedly had to rewind over half the level for such execution-based mistakes, despite knowing exactly what I had to do conception-wise (given that rewinding is supposed to be used only for conceptual mistakes).

This is especially annoying since it forces you to do very fiddly stuff over and over again, like basher-digger staircases, or miners going through multiple obstacles connected by builders.

I've attached to replays, both centered around the core issue of how to deal with the pillars at the exit. As far as I can tell, I did exactly what IchoTolot did at this place...

Spoiler

...but the final staircase still is not high enough to reach the exit.

In one solution, I even had a miner to spare, because it didn't end up mattering whether I left a piece of the pillar intact (like IchoTolot), or whether I built from the steel ground - neither of the two approaches managed to gain sufficient height.

I know from my own levels that pixel precision can be unintended, but in the last level of a pack, I doubt it . Since this is the type of pixel precision that can prevent a solution from being successful even when it's conceptually correct, I have to object to IchoTolot's praise of this level as a prime example of "fair" difficulty.

[IchoTolot]

Since this is the type of pixel precision that can prevent a solution from being successful even when it's conceptually correct, I have to object to IchoTolot's praise of this level as a prime example of "fair" difficulty.

I think you don't understand what I personally define as "fair".

Pixel precision is fair, even if you have to rewind all of the level due to an unprecise mistake. Is it good? Certainly not, we want to prevent these things.
Is a timer unfair that cuts tons of solutions of by 1 second? No, but it's still bad.

I call levels unfair that actively hide stuff from the players eye. Hidden traps, hidden pick-up-skills etc.

Overall my praise to the level partly comes from a developed automated optimisation playstyle. Over the years I got used to automatically find and go for an ~optimal route, purely in terms of reaching the goal. This isn't always the intended one of course and it's only optimal in my current view of the level.
That way I don't mind having to do a bunch of very precise actions, as a bunch of them I do intuitively.
Does it anger me when I have to rewind a lot? Yes and I will criticise it.
Does it anger me when there is an excessive ammount of very precise actions? At some threshhold yes and I will point that out.

Also I learned to use the things at my disposal (skill shadows, save-states) to nullify the precision and rewinding to an ammount. Well placed save-states at critical points in time and making good use of replay insert can help quite a bit.

It's also the general layout of the level that made me forgive the faults and I can highly advertise those for your huge levels:
- Mostly contained crowds (only 1 active crowd would be optimal and if there are more active ones let them be close)
- No timer
- No extensive need for multitasking
- You only have to watch 1 area at the same time. --> less scrolling
These traits fit for this level, therefore making me forgive the precision.

In the end I just loved puzzeling my way through this level while doing stuff as optimised as possible.
+ The fact that it's the last boss level of a very difficult pack makes me somewhat more forgiving, but still there is a threshhold for that.

[Strato Incendus]

It's also the general layout of the level that made me forgive the faults and I can highly advertise those for your huge levels:
- Mostly contained crowds (only 1 active crowd would be optimal and if there are more active ones let them be close)
- No timer
- No extensive need for multitasking
- You only have to watch 1 area at the same time. --> less scrolling
These traits fit for this level, therefore making me forgive the precision.

Thanks for the advice! All those things happen to be the case for most of my levels from Lemmings World Tour already, though. Of course, the players are always free to bring multitasking on themselves , but especially in "The Grand Puzzle", which took a lot of inspirations from "Final Frustration", the crowds are usually easily containable as well. Contrary to late Mayhem levels, it's not that there aren't enough blockers to contain all the different crowds - it's the question of how to free all those blockers in the end again, because everyone has to be saved (or, in case of "The Grand Puzzle", you're only allowed 1 death).

That's also the reason why attempting a level like "Final Frustration" is more fun to me than timing-based levels with deliberately oddly-shaped terrain where crowd control is next to impossible.

Still, the rewinding is mainly a pain in combination with the overall duration of the level. It's not the duration per se, nor is it the fact of having to rewind due to pixel precision, but the combination of both makes things exponentially worse . With every restart, you're sitting there for quite a while holding down the space bar until you arrive back at the time stamp your currently working on. Or you're asking yourself "What's faster at this point - rewinding in 1-second-steps, or starting from the beginning in 10-second-steps?"

I already saved several replays manually (since the two I attached were ultimately not successful, i.e. not automatically saved). Replay editing helps, but at one point when I edited out a specific skill and left the rest as it was, a builder still ended up being misplaced.

[Nepster]

I actually think it's pretty genious - there's always even more skills which are locked in the place where they have to be used than one would think . As far as X-of-everything levels are concerned, this is as good as they're going to get!

I am very happy you liked the level. But that doesn't alter the fact, that my own taste changed...

The only thing that makes it "not a good level" is the RIDICULOUS pixel precision required all over it!

Sorry, but most of the precision is self-inflicted.
1) There is actually not a single place, where lemmings have to step up more than 4 pixels, as long as you don't care about the talisman.
2) There is actually no fall, that requires something close to the maximal fall height.
3) While basher-digger-staircases do help, my own solution uses just two steps like this.
4) Given that the miner-shadows now take blockers into account, placing them to free blockers is fairly easy.
Yes, of course you can do all of that, but removing the possibility of doing this (or at least trying to do) is simply beyond my designing skills in such a 20-of-everything level.

I've attached to replays, both centered around the core issue of how to deal with the pillars at the exit. As far as I can tell, I did exactly what IchoTolot did at this place...

Spoiler

...but the final staircase still is not high enough to reach the exit.

Spoiler

I hoped the height of the central tower would make it clear, that 4 builders just don't suffice and you actually need five there. But apparently not...

I know from my own levels that pixel precision can be unintended, but in the last level of a pack, I doubt it .

Pixel precision is never put there on purpose. It might be, that I simply couldn't avoid it or that I failed miserably to discourage pixel-precise attempts. But I certainly never left it in there, just to make the level harder or to annoy my players - and for this it doesn't make any difference, whether the level is the very first or very last level in the pack.

[Strato Incendus]

Thanks for the reply, Nepster!

Spoiler

Sorry, but most of the precision is self-inflicted.

So it's like with my solution to "Jump in the Ring"? I think I never even really got to see the intended solution for "Final Frustration"; IchoTolot has uploaded two different approaches to it, although the second one still has a lot in common with the first one. And I have yet to see anyone else solve it!

1) There is actually not a single place, where lemmings have to step up more than 4 pixels, as long as you don't care about the talisman.

Might be a place for me to use the spare miner then, which I had available in one of the two replays.

2) There is actually no fall, that requires something close to the maximal fall height.

Well, it turned out lucky that I didn't need an additional miner at that central square thing to lower the fall height of the digger shaft - other times I did...

4) Given that the miner-shadows now take blockers into account, placing them to free blockers is fairly easy.

I was specifically talking about that T-shaped ledge where I placed a blocker to turn around the climbers. One pixel earlier, and the climber already turns around while falling, thereby making it impossible to free the blocker. One pixel later, and the miner leaves a pixel of terrain there that maintains the ledges fatal fall height rather than removing it when freeing the blocker. That would necessitate an additional miner or digger, which I don't have at that point.

I hoped the height of the central tower would make it clear, that 4 builders just don't suffice and you actually need five there. But apparently not...

Well, thanks, at least that's a frustration-sparing hint that allows me to stop messing around with the execution of this and look for how to save a builder elsewhere.

So is IchoTolot's solution a backroute then? Because the managed to go up the shaft with four builders, plus dig-mining a bit off the pillars, and the fifth builder then only was required to get to the platform with the exit.

[kaywhyn]

Currently playing through this levelpack. Finished the first 3 ranks and currently on Sun 4. Wow, what mind-boggling levels. Making slow but steady progress. If it already gets this difficult at the beginning of Sun, I can't imagine how difficult the last two ranks will be. As a matter of fact, Planet 13 managed to stump me and I took a break from the pack for a few weeks before coming back to the pack today and managing to beat the level in an unusually fiddly way. Replay attached.

Spoiler

Planet 13 - While I still ended up using all the skills except for a few climbers, this required extreme precise timing in order to close the gap after all the lemmings are above the bridge currently in progress but before they fall past and splat. When I watched Icho's replay of this level on Youtube, it really surprised me that it's possible to build through the thin floor with just one builder. I think I kept trying to see if that was possible but the lemming kept hitting his head and so I gave up and didn't come back to this level and the levelpack for a few weeks. Tbh I'm surprised that my solution works but I solved the level in a much harder way. As a matter of fact, it really appears in the replay that the last climber isn't supposed to get past the completed builder bridge, like it glitched through it or something. In any case, when it comes to solving levels, any solution that works is good enough for me. 

Also curious with the attached image of Sun 2, the first Tomb Raider level. This also applies to Sun 3, as it's the exact same level but with one extra builder and one less basher. I was able to solve both levels before watching Icho's replays, but in both levels, the arrows on the OWWs are pointing in the wrong direction. I got so confused why I couldn't bash in the direction of the arrows, but it does work when I bash against the direction shown, so I was still able to solve Sun 2. When I watched Icho's replay of Sun 2, the arrows are pointing in the correct direction, so it turns out that the direction of the arrows are shown incorrectly when I played through the level. This is v2 in Neolemmix 12.8.2. The engine told me updates to styles were available today and to update them via the style manager, which I did do earlier today by clicking on Update All. If it's due to an update, not sure why it would be the cause. Can someone else check if this level also has the arrows pointing in the wrong direction? If so, then I'll know that I'm the only one who has the problem.

Edit: Just checked Sun 2 today and it seems the arrows are now displayed properly when playing the level. So, was it something with the level or was it just me? Come to think of it, I did restart my computer several times between the time of this post and just before today, so maybe that fixed it? Also, I just checked the marble levels from earlier in the pack and the arrows on the OWWs are correct, and I'm not close to finishing the pack, but most likely the arrows are fine in the rest of the pack.

[kaywhyn]

Unreal! I have finally beaten Sun 7 - Scrap the Builders from Nepsterlems, the level I was stuck on for a really long time! Until now, I had beaten every level up to Sun 7. See attached replay. According to the date of this replay and the date of the previous level, I was stuck on Sun 7 just a little over 5 months, which is an additional month longer than the time I was stuck on Neutrality 41 of United. Oh well, I finally solved this extremely brutal level.

My first impression from the save 30 out of 40 lemmings requirement was "can lose up to 10? That's pretty generous." However, as it turns out it's not generous at all, as there's literally no room for error on the level. Besides figuring out how to get the bottom crowd up, the hardest part for me was avoiding losing so many to the needle traps, as they just trigger way too fast. For the very first time last night, I finally figured out how to get both the top and bottom crowds to the exit, only to find to my disappointment that I was 4 short of the requirement. I fiddled around and tweaked my solution a bit and by adjusting a miner assignment going in the other direction, I was able to do 1 better, but it was still a no solve. I was wondering that I don't think I can improve my result any further without losing too many to the needle traps on either side, but I decided to call it a night. I almost wanted to give up and probably attach my best replays for namida and Icho to see and possibly ask for a hint. Just before I dozed off in bed, I was literally thinking that the needle traps are evil and thought why did Nepster have to include them when the level is already really hard anyway. If they weren't there, I already would had passed the level. Then again, it would make the level a lot easier instead of it being very difficult to the extreme, so why remove them?

Determined not to give up, I just took another look at the level and had another go at it starting an hour ago. I realized that I can indeed improve my result.

Spoiler

I had to time the first miner and basher in such a way that when I bounce the miner with a blocker, that only one lemming is nearby to release the basher with a digger. Likely the position of a blocker on the ramp leading up to the exit is necessary to tweak as well. Not only that, when bouncing the miner, it had to be in a position such that the miner makes a tiny hole that would allow me to bomb close to the wall after a lemming falls two pixels so that the bottom crowd can get up. All of this was really hard to do, as the timing is extremely difficult, and I had to constantly rewind and adjust which lemming bashes. Once, I got two lemmings to be behind the basher but then one would get killed by the needles trap. With this one, my result was improved to being 2 lemmings short, due to the 2nd miner being too far from the bunched up crowd.

Finally, I had success with getting only one lemming behind the basher such that it turns around in the basher tunnel and could release the basher interrupted midstroke assigned a blocker with a digger. The only tricky part left was making sure the 2nd miner isn't that far behind from the bunched up crowd. If it was, then I would still be a lemming short. However, I finally succeeded with getting the miner bunched up with the crowd, and thus level finally solved!

In any case, now that I finally managed to solve the level after 5 months and how I struggled tremendously with it, I'm in absolute agreement with namida and Icho that Sun 7 is way too hard for its position. In addition, so far I've been struggling a lot with all the Sun levels, so if the rank has been really difficult for me since the start, I can't imagine how much trouble I'll have in the Neutron Star and Black Hole ratings when I get to them.

Well, as I got up really early, I'm quite tired, so I'm off to bed some more before proceeding onward with the pack. At least I can sleep satisfactorily now knowing that I finally beat Sun 7. In the meantime, I'll be watching namida's and Icho's Youtube video solution of the level to compare mine with theirs.

[ericderkovits]

Yes, I just watched the replay on Scrap the builders. The replay I Have shows that 32 of 40 were saved, so 2 over the requirement. Also Yes I can see that being out of place.

[IchoTolot]

In addition, so far I've been struggling a lot with all the Sun levels, so if the rank has been really difficult for me since the start, I can't imagine how much trouble I'll have in the Neutron Star and Black Hole ratings when I get to them.

In my experience the pack gets to the hard stuff very quickly, but it doesn't get too much harder once you got over the shock. Don't get me wrong, it still gets harder and there are 100% really hard nuts in the last 2 ranks, but compared to other packs the difficulty curve is a lot flatter. Well, at least in my experience.

[kaywhyn]

I just watched the replays that are available for Sun 7. AHAHAHAHAHAHAHAHA as expected, out of all the solutions I've seen, mine is the most difficult of all, as well as the most unintuitive. So far, mine is the only one that goes through the area with the needle traps. On top of all that, aside from two of the blockers, every assignment needs extreme pixel precision and is extremely timing heavy and dependent. Seriously, why do I persist with finding such very difficult solutions for hard levels? In addition, mine meets the save requirement exactly, while all the rest managed to go over it, with the best one being 4 over by Icho. Icho's and josh's are pretty much the same, although the latter lost two lemmings to the bottomless pit because he forgot to place blockers. Ok, forget what I said about the save requirement not being very generous. After watching Icho's solution, I instantly went "oh gosh, I can't believe I never ever saw that. So much easier and cleaner." At the same time, I think regardless of the solution you use, none are particularly easy or trivial to find. Especially with namida's, I never would had come up with using a

Spoiler

small digger pit and a blocker inside it to make the crowd turn back to the right, as my thought would be that surly it won't work at all, but his solution does require some extreme precision.

In my experience the pack gets to the hard stuff very quickly, but it doesn't get too much harder once you got over the shock. Don't get me wrong, it still gets harder and there are 100% really hard nuts in the last 2 ranks, but compared to other packs the difficulty curve is a lot flatter. Well, at least in my experience.

Thanks for the input. That's reassuring Maybe it won't be so bad then, but that remains to be seen for me.

Anyway, it was cool to see other solutions to this very brutal level. Onward for me in a few hours.   

[kaywhyn]

All rightie, after 5 months the Sun rank has finally fallen. Replays attached up through the Sun rank. Up to this point in the pack, I have to say the most difficult levels so far for me were Planet 13 - 50% is not enough!, Sun 7 - Scrap the Builders, Sun 12 - Death of a Working Lemming and Sun 20 - Empty Walls, although the last two were nowhere near as hard as the brutal Sun 7. I thought both Sun 12 and Sun 20 were hard due to the fact that the solutions in each one are difficult to spot, with the latter the way more difficult one. Also, I thought that the levels after the extremely brutal Sun 7 were much easier, although I guess the level immediately after, Sun 8 - Minimalistic Platforms was somewhat hard, but since there's only 2 different skills, a blocker and 5 builders, it's tricky more than difficult given the timing.

In particular, the 1-of-everything levels in this rank, Sun 4 - Just a Quickie!, Sun 11 - It's Hero Time Again!, and Sun 20 really stand out as being difficult, with Sun 4 probably the most difficult of the three due to the extremely strict time limit, only 30 seconds. Without it, it's a really easy level, but the time limit really ramps up the difficulty considerably. Of course, in Dos and Lemmini this isn't a problem, as the amount of minutes can only be an integer, which means it has to be given at least a 1 minute time limit in those engines.

I've noticed that a lot of the levels in the Sun rank were small in size, being only about a screen wide, but they were still hard levels. It goes to show that small levels may look easy in appearance, but they can be surprisingly hard. Indeed, don't expect to solve through the pack very quickly! The levels get hard starting with the Planet rank, and they get harder really fast. The Planet rank is the first rank after the X-of-everything levels in the Comet and Moon ranks, 5-of-everything in the former rank, 10-of-everything in the latter rank. Even the other levels I didn't list were challenges, and my progress slowed considerably starting in the Planet rank.

In any case, after I solved the Sun levels on my own, I compared my solutions to those of Icho's on Youtube. I find it quite amusing that in almost every level in the rank, I was able to save a skill(s), while Icho used all of them. For example, in Sun 9 - Silence on Earth, I have a blocker leftover, Sun 11 I have a floater leftover, Sun 13 - Cannonballs in the Air I have a bomber leftover, Sun 15 - Revival of the Lost I have a blocker leftover, Sun 16 - The Marble Temple I have two miners leftover, and Sun 18 -Five Survive, I have a climber leftover. That's 6 levels where I did better in skill usage. I swear, I wasn't even trying to be more efficient in my solutions than Icho was. I couldn't be, since I only watch solution videos after I beat the level first, so I was quite surprised I ended up saving skills on many levels compared to him in the Sun rank. On the flip side, he was much better in skill usage than me on both Sun 10 - We All Dig Down! and Sun 19 - Just a random heap of junk!. I especially love his solution to Sun 7. It's definitely the easiest and cleanest solution I've seen to the level which escaped me completely. I think Sun 20 was the only level after Sun 7 where we had the exact same solution.

Finally, Icho's solution to Sun 4 might not work (key word: might) in the current NL version, since lemmings no longer can exit when the time limit runs out. In it, the last lemming exits exactly when the timer hits 0. So, I got to say that my solution is more efficient in that I have 3 seconds left on the clock, although it's not a trivial solution at all. Also, in Sun 12, I managed to save an extra.

edit: I just watched Icho's replay. Yup, the timer hits 0 at the exact moment the last lemming finishes exiting (easiest way is for the last lemming to exit and then use the N key until you hear the ringing alarm), but the game still gave him the credit for passing the level. Wow, imagine if it was one frame too late instead.

Well, onto the Neutron Star rank, the second to last rank of the pack. Already nervous about what the rank has in store, particularly if the Sun rank was any indication. In particular, every level up to Sun 7, inclusive, were all difficult and tricky, while the levels after Sun 7 generally got a bit easier, although not by much. Here goes nothing! 

[ericderkovits]

Good job Kaywhyn passing Sun 7, even though I have all the replays for NepsterLems I still downloaded yours(I always like to compare yours to others). I looked at your solution for Sun 7, It's quite a bit different than the one I have. Also the one I had saved 32 instead of 30, but I tweeked the one I had so now it saves 4 over the required, since the replay I had 2 were lost that shouldn't have been lost.



Now since you passed Sun 7, here is the solution I have for it.

[kaywhyn]

Hey eric, I know you appreciate replays from me for the pack even though you already have a complete set. I, too, love to compare my solutions to others. Most of the time, I seem to have made the solution more difficult than was necessary. I think I know which solution you're referring to. josh's solution is the one that loses two lemmings unnecessarily. I guess he forgot to block at the ends at the top and bottom. And yes, mine is completely different in that mine is the only one that goes through the needle trap area. Not only that, mine is the most difficult solution because it is completely timing dependent. Bash too early and you won't be able to release the blocker with a digger and mine out the bottom crowd. Bash too late and same thing. Similarly, put the first miner even slightly earlier or later and you won't be able to rescue the bottom crowd. Obviously, I don't recommend anyone my solution for Sun 7 other than maybe a "wow, this person has a crazy solution for Sun 7" reaction. It is still a wonder how the much easier solutions managed to escape me on that level.

As usual, be on the lookout for when I post my Neutron Star and Black Hole replays when I finish each rank.   

[IchoTolot]

Finally, Icho's solution to Sun 4 might not work (key word: might) in the current NL version, since lemmings no longer can exit when the time limit runs out.

That makes no difference. Lemmings that enter after the timer went down did not count before. I also can confirm that my replay still works.