• Content Count

  • Joined

  • Last visited

Community Reputation

29 Excellent

About Simonova

  • Rank
    Junior Member
  1. Combining the cooked products of Pacu and Sanishells is a mistake when there's a clear and delightful opportunity to introduce crab patties. "Cooked Sea Meat" isn't exactly appetizing sounding, being just barely better than "Hot Fish".
  2. Yeah, sleet wheat doesn't thrill me either; I've found myself favoring food sources that aren't water bound, as I've discovered that water budget starts being very constraining. Grubfruit and sweetle farms are amazing, though a good chunk of that is due to sulfur having extremely limited uses beyond that, while water enters into nearly every other process somehow.
  3. I've never heard of cracking ethanol; I googled it a bit to see if anyone actually has tried forming natural gas from ethanol (which makes sense to some degree, because burning ethanol as fuel tends to form aldehydes, which aren't super great to be breathing), but found nothing; the only information I could find on ethanol cracking was using it as a way to coke steel (which is interesting in it's own right). It would be interesting to have an expanded role for ethanol beyond poor man's petroleum and nosh sprouts (which I find to be dubious as a food source just from the overall water requirements for tofu); at very least, some way to use it in a rocket engine. Then I read a bit more about 60 seconds after I wrote this, and saw a paper on how ethanol is being used to make regular hydrocarbon fuel cracking more efficient by inhibiting the formation of pure carbon deposits; maybe a machine that takes in petroleum and ethanol gas to output natural gas?
  4. While the infinitely more qualified folks at Klei are hard at work on the next DLC, I have an idea that I think would be enjoyable to play, which I've called the apocalypse loop. The set up is like this: you have a chain of asteroids, but you can only move to the next asteroid in the chain, never backwards (this isn't Spaced Out). After a set period of cycles, the last asteroid in the chain undergoes the "apocalypse", which I see envision as the asteroid orbiting into the Tear as a lazy "what it could be". So, the objective is to have colonized and moved to the next asteroid in the chain before the current one undergoes it's apocalypse. Meanwhile, the destroyed asteroid is renewed; it's biomes wiped cleaned, maybe some rare/DLC-exclusive resources seeded into it, so that it can be re-colonized once the apocalypse has ended and it's re-entered the loop. Later tier technologies might revolve around materials or structures that allow a duplicant to stay on the asteroid and pass through the apocalypse unscathed, picking up new attributes/research(/critter morphs?) along the way. That's my rough idea; nothing too complex, and hopefully based around enough of the existing stuff they built Spaced Out to be a modest dev effort while shaking up the stagnation that tends to set in during the extreme late game, but I'd like to know what people think.
  5. Love this, much needed buff; the place I always need the most sweeping is the bottom of high-traffic ladders, which just so happened to be where Sweepy struggled the most to sweep. Now just give me the option to bury my heroic Rovers in tasteful memorials instead of simply deconstructing them, and I will be pleased with the state of all my mechanical friends.
  6. While I was writing all that, I did have the thought, "Is it possible I am noticing these things more at high game speed because, probability-wise, running the game at high speed gives me more things to notice?". Dupes building through walls and floors seems to be an exclusively high-speed issue though; I've never seen that at normal speeds except if the action began before the intersecting tile was built.
  7. Behavior-wise, I've noticed that the game has a tendency to misbehave when running at maximum speed; debris falling through freshly-created tiles, dupes building through tiles as though they aren't there (not just freshly-created tiles though, which tosses out my theory of all these being the result of desynchronization), critters becoming "stuck" on a tile (most visible with pacu, when ~30 pacu will inexplicably decide to just hang out on a single tile indefinitely). I've also noticed that the game itself is much more likely to crash running at max speed, compared to default speed (I've never had it crash with a unity exception at regular speed; that seems to be something that happens exclusively when the simulation is running quickly). My guess is that a lot of stuff is very sensitive to order-of-operation, and as speed increases (causing "business logic" to become a larger factor in execution time, since all the same business logic must occur within a shortened real-time span when running at higher speeds), which might also explain why ONI is constrained down to just two functional threads of execution; the more threads you have, the more opportunities you have for "weird combinations" of timings.
  8. I literally just realized that my all missing aluminum was behind the asteroid back wall minutes before this update dropped; thank you!
  9. When will I be able to bury rovers with dead batteries in the tasteful memorial, and put my mechanical friends to rest?
  10. I hope they do a worldgen pass on moonlet crusts before release to make ease-of-colonization more consistent, even if the more advanced worlds still "lock out" rovers, or keep them limited to top-layer biomes. Right now, the best strategy seems to be to build a steam/rad rocket with two trailblazer modules of the same type, so you can build a rocket platform out of the materials and immediately leave (to come back later with an apartment rocket and get real work done); that seems too cheesy to be "the best".
  11. Hmm, this means that the rover can build a mini-pod; can the rover activate the mini-pod? Might not even need to build a rocket platform...
  12. Locally declared value types (like bool) don't come into play for the garbage collector, specifically because they are both locally declared and value types; they are stored on the stack, which gets it's memory management for free (the memory is allocated and deallocated at the method boundary, so there is no need for a garbage collector; that's why you don't need to manually free value types in C, which doesn't have a garbage collector at all). Declaring a bool, assigning the conditional to it, then evaluating the bool does make an extra trip to memory that doesn't need to happen, but it doesn't put any additional strain on the garbage collector; to do that, they would need to create their own class that functioned like a bool, then new it up, such as: BooleanClass flag = new BooleanClass(this.Items.Count == 0); if (flag.ToBoolean()) (I have seen professional Java developers write overly classy code like that) The easy rule of thumb is, if you have to use the "new" keyword to create the instance of the object from scratch, then it's being allocated on the heap, and it is going to be a candidate for the garbage collector. If it's a basic construct of the language, it's going to be stored on the stack, UNLESS it's explicitly declared as a ref, or reference (which is how all classes work; the object stored on the stack is a reference to a location in heap space where they object can actually be found).
  13. A primitive datatype declared on the stack is not going to interact with the garbage collector, because it's a primitive datatype; full stop. I knew a C programmer that liked to stuff non-primitive structures into stack space, specifically because it meant he didn't have to worry about any garbage collection; it would all be deallocated when the frame was popped off the stack.
  14. Balancing geysers on the classic terra cluster is greatly appreciated; the geyser density on the teleporter asteroid made building there a bit tricky (though the cool slush geyser six tiles from the copper volcano turned out to be pretty convenient for early game refined metals).