Surrender Flare is kicking my arse. I totally know the principle of a sub 30 cycles solution. But actually laying out the damn thing is a pain.
The 28 on Surrender Flare is the only working solution I ever made for it. It was in the phase of the game where I’d decided that there was no point in making a suboptimal initial solution, and then working on it. I know the solution to every puzzle is going to be the same kind of perfectly pipelined 4-cycle period machine, so might as well just start with it. Getting that thing working was so traumatic that I couldn’t bring myself to touch it again, even though the 1st percentile score is 26.
Purified Gold in Chapter 4 is a fun one. My initial score was 209 cycles, but improved it to 162. Of course, nowhere near jsnell’s 112, but still. It’s the first time I found a use for “ctrl-drag to copy-and-paste parts on the board”
The Glyph of Purification is a weird one. It seems to take a cycle to operate. You can’t grab the output until the cycle after the one where you release the inputs.
Edit: 113! Only one worse than jsnell, which is both frustrating and satisfying.
Another edit: 112!
What the hell is wrong with GOG?
I am venturing into the late game content, and I notice some disparate numbers, especially because I guess a bunch of us are just throwing “first shots” at those massive machineries. But I was wondering if we could maybe posts some of our first shots here, as it may be quite funny to look at. I have made a habit of trying to keep my first shots from the 3rd or 4th puzzles onward, but I am not sure if anybody else does the same?
In an attempt to kick it off, here is my latest monstruosity, on “Stain Remover”:
[closes bunker door]
I save all versions of my puzzles. Here’s my first and only attempt on Stain Remover, it’s hillbillyarious:
It’s like a drunk spider trying to walk.
I always keep the first solution around as well. This one looks like it should be trivial to micro-optimize, since that middle arm is doing too much work. Just get two arms to both do half of the program. But looking at the later solutions in the series, it’s actually a fair bit of work since the layout is so fragile.
It never occured to me to create pairs of blue+green rather than blue+blue and green+green as the base of this solution. But looking at your solutions, the former seems like the obvious right way to do this.
I thought everyone was freaking out because Steam doesn’t curate and thus PC gaming is d0med? This is curation!
My only issue with Steam is DRM. Not every Steam game has it, but when I can avoid the risk, I do. I’ll always choose a DRM-free platform over Steam if available. Humble Store it is! (If I ever buy this game.)
I finished the main campaign. I don’t get the ending of the story, so if anyone wants to explain it to me, go ahead!
Also, there’s only one person from QT3 on my Steam friends list who has also finished (jsnell, natch). If other people have finished please add me as a friend on Steam (I’m elsquonk there) to compare scores!
I’m still playing! I got to the “production puzzles” and I can’t figure them out at all, so I’m going back through the chapters and optimizing my solutions for the different categories.
Just got 30 cycles on Alcohol Separation (chapter 3)! Very pleased with myself! The histo makes it look like sub-25 is possible, but I sure don’t see it!
I got to 29 on that now, and it sure is one ugly machine. I think I see the principle for how to do it in ~20, but not sure I can actually execute it.
On the other hand, the top percentile score (which you can see by turning on an option after beating the campaign) is 25. Those are usually pretty solid, I’ve only beaten the top percentile on one level, and then just by one cycle. So odds are that I’m totally off-base here.
Curses! I mean, cool!
Oh, man, you mean I’m gonna hafta do those production puzzles??!?
No, just the main 5 chapter campaign. It should be in the options menu, “show top percentile”.
I should clearly come back to all these early puzzles every now and then. Alcohol Separation is just fascinating. I thought the theoretical limit would have been 17 or 18, but it appears hard to arrange for reasons kind of unique to this puzzle. It just has to be done with a 9 cycle period, which I don’t remember using in any other serious solution. Seeing all those wasted cycles makes it hard to cheer for the 22 that I actually could get.
Ok, I found that, but it just reminds me of one thing that bugs me, which is you can only see 5 friends in the leaderboard for each category. I have 21 Steam friends who play this now, I want to see them all!
Nice article from Tom Francis, exploring why the optimization goals can be so compelling: