Repton 1 speedrun in 17:52

I didn't notice until I opened this topic that this is an anagram of my first ever time, 25:17

I'm finished with Repton 1 for now. The run has some small inaccuracies that lose a couple of seconds here and there, but I'm pretty sure there are no more major improvements to find, and it doesn't seem worth it to keep grinding runs for a slightly better time.

Instead, I'm moving on to Repton 2, which is a much harder game. The main new game element is the "spirit", which follows the walls to its left until it reaches a cage, and there can be up to eight of them per level. So now, routes have to be carefully planned to manipulate spirits' paths and get Repton around as fast as possible without collision, and there are many more places where a small slip can have large consequences. In addition, eight of the levels have a planetary surface that you can go above (and you have to, in most cases); and six of those drop meteors randomly on you when you are above the surface The meteors are pretty slow but can fall in clusters that are tricky to dodge, and of course the unpredictability makes it harder to plan consistent routes for handling the spirits.

Still, I'm making decent progress; I've recorded six videos of routing and practice on the individual levels (which isn't quite six levels; it's complicated, but some levels are in chunks that are accessed separately, so one of my videos contains half of one level and half of another). My plan is to finish going through all the levels like this and then start full game runs. The only previous speedrun is a 57:07 TAS by George Foot, so it will be quite an achievement if I can beat that with a real-time run

I am not aware of any video recording of the 37-minute Cheapo level, and a quick Youtube search doesn't turn up anything. This is not surprising, though. Cheapo was current around the time of my visit to Australia (I remember this because I worked on Cheapo graphics while I was there) -- and that was in the same year that Youtube started. It would be a long time before I had a computer setup capable of recording gameplay at a decent frame rate, and by then, Cheapo was long dead.

It's frustrating when levels get broken by updates, for sure, but I think it's important to remember that Lemmings (including all clone engines) is a game, and everything we do, including engine changes and level design, is primarily aimed at giving the player the best possible experience.

It's not fun to be stuck on a level, maybe for a long time and coming back to it every so often to see if this time you can finally beat it, and then discover that the solution is to take advantage of a bug in the engine that you didn't know about.

And it was absolutely a bug that oh-noers could enter a mid-air exit while normal lemmings cannot. As mentioned, SuperLemmix allows all lemmings to enter mid-air exits, so that is the intended behaviour for that game (and the same for Cheapo and early versions of NeoLemmix). When NeoLemmix moved from the "one executable per pack" model to having one central executable, it was essential to remove behaviour options and have just one behaviour. After a forum discussion, we collectively decided on "no direct drop for normal lemmings; floaters/gliders can still enter mid-air exits". It was purely accidental that oh-noers were overlooked when other states were banned from exiting; this was fixed in the next update after it was noticed. Much later, someone noticed that stoner oh-noers could still exit because they hadn't been specifically prevented; this was also immediately fixed.

Fortunately, NeoLemmix development is planned to come to a final version very soon, after which there won't be any more bug fixes, so you won't have to worry about levels being broken again. However, I would still say that if you find any behaviour in the engine that is clearly a bug, it's not a good idea to use it in levels.

In all honesty, questions that are akin to "How many pixels are there in all Mayhem levels of the DOS version of Lemmings?" probably shouldn't be asked.

After all this time, thanks to this post, this "impossible" question can finally be answered.

30 x 160 x 1648 = 7910400.

Given its comparative emphasis on real-time challenges, I would have thought that SLX needs the minimap, in a way that NL does not. If performance is an issue, just go back to the monochrome minimap as default or even as the only version offered.

Dynamite / Mine - You can set these somewhere and run away before they explode. Like the non-lethal Bomber in Lemmings 2.

This reminds me that I had a dynamiter as one of the skills in The Light of Horderland, a Lemmings-like game I drew on paper when I was a teenager. It worked as you'd expect: dropped the dynamite, which was non-interactable and would explode after a countdown. The idea, I suppose, was that it would be tricky for the same "horder" to take advantage of the bomb hole; you'd have to either get a second worker, or find a way to bring the first one back. Even though this didn't make your shortlist, I think it could still have puzzle potential in modern Lemmings.

NES Lemmings has different physics and terrain from Amiga/DOS Lemmings, so all the NES levels available for NeoLemmix/SuperLemmix are unofficial recreations by Ron_Stard, not rips of the actual level data like the Genesis levels.