h.q.droid wrote: Tue Oct 07, 2025 3:56 am
The short shelf lives of puzzles also forced them to have more depth, which I perceive as a good thing.
I suspect I've gotten far enough into the devs' heads that I understand enough about how they imagine and choose puzzles, that the new stuff I encountered in 2.0 just wasn't new enough to be a challenge any more.
But for the rest there, did you notice me mentioning that some of the puzzles had *not* been spoiled even after ten-ish years?
I mean, you're posting plausible and attractive-sounding speculation, but the long-running questions about those problems, sprinkled with the kind of complaints I mentioned, are on record and I think form a pretty solid argument that their shelf life did not match your description in any way, just that one simple fact.
I distinctly remember being in all but so many words called a liar (not in these forums, elsewhere) because I said train routing with the pre-"train limit" mechanics was solvable with combinators, no mods or train limits required, and this was after the train limits had been added (but while turning a chosen destination off was still possible and forced a repath somewhere else): the entire lifetime of trains up to then, near ten years.
Nobody ever spoiled it. The solutions people posted to the big main-sequence problems were not the best solutions, to my way because they didn't want to spoil them. I suppose it's too late to go back now, the basics of the stunt with the trains was (for pickup, delivery after was the same stunt) to code pickup types into a signal, two bits each, I never used more than, what, five? Anyway:
The low bit of the pair is (when a station sends it) "have" that pickup order, can satisfy it on arrival, and (when carried with a train) "want" that pickup order, want to go to a station that can satisfy it. Red wire along the track carries "want" signals, green wire carries "have" signals. "Have" signals are carried uninterrupted between switch/station entries and the switch/station exits that feed them; at each signal, visible "want" signals are propagated onward when the signal is reserved and cleared when its exit block goes green; switch entry signals don't wait, they immediately propagate the wants, each exit from that junction adds the inbound "have" and "want" signals and turns red/green for no/any carry bits indicating has/wants something available behind that switch exit; switch junction entry "or"s the "have" entries from all its exits and forwards that inwards. That's pretty much it, merges are analogous, the trip back for delivery is almost exactly similar just with "want" and "have" (purely notionally) swapped. With anything short of a megabase mining net you can just name your ore stations "ore" and have them send the "have" signal for whatever ore they've got a trainload of, when some production station wants more ore a train with an "ore pickup" and "ore delivery" and "ore dispatch" schedule gets dispatched with the "wants" signal for that particular ore and at every junction it chooses an exit that lets it route to a station that has what it wants, those having already been set up before it ever arrived. If you're not producing an excess this needs some further sophistication, you need to be sure there's a retry loop, if the pickup route lenghts get too widely spread the really distant ones can get ignored but there's obvious fixes for that, la la, and it can be easier to have separate pickup trains and schedules for each ore type, but the idea works and extending it to a megabase was comparatively straightforward.
I never once saw that solution posted, *every* solution ever saw used a mod, they all scripted schedule updates or just didn't work well. This is less true of the others I mentioned, but filtering a belt was also much simpler and while the splitters only showed their quirks under heavy load I've *never* understood the obsession with "belt compression", except for rail they're *by far* the cheapest thing you build, keep your assemblers running flat out, if you do it with three splitters and six 60% belts or two splitters and four 90% belts, it's the same number of assemblers and inserters either way, if they're 100% utilized in both cases the resource cost amortizes to no difference at all in seconds, the ups cost of, what a handful of splitters in an entire megabase? is a rounding error on nothing and the simplicity benefit of just "enough to handle this and maybe even one more" belt layouts is huge.
It's somewhat the same with Space Engineers and things like early prospecting or survival-mode subgrid projectoring or setting up stationkeeping, they're not hard like the global train routing was but more in "how… how did I not see that?" "duh!" territory but the genius of them is they're (going on ten years of evidence here) they only seem obvious in hindsight. I remember the moment I was thinking "it *can't* be this tedious" about hand drill mining and realized, oh, wait, maybe I can' actually get that far without a bottle.", it was the moment I fell in love with the game, just for asking me to have a little faith that a problem that wasn't explicitly called out as A Problem was in fact solvable anyway, for rewarding that faith. In both SE and Stationeers the very best problems have layer upon layer of progressively-more-clever solutions. Kinda like using assemblers used to be in Factorio, and train routing, and sushi/trash management on belts. Except those *weren't* spoiled. They were simply taken out of the game. You can't even mod some of them back in, it's not just pointless, it's not possible Which did shut the whiners up, but that consequence also factored in to why I described those specific changes, in this thread, to the question I was answering.
Whether or not I like the changes is not debatable. Whether or not the changes simplified the puzzles basically out of existence is not debatable. Whether or not Wube have a perfect right to make exactly the game they want is *also* not debatable, and if the game they want produces lots of happy players and makes them able to do more games they want to make, absolutely yay for them and their players. But the question I answered was very specific and phrased as a rude "put up or shut up" demand. I'd recommend not mouthing platitudes about brusque replies to posturing demands while staying silent about said posturing demands, it leads people to suspect you're aware you're putting your thumb on the scale.