Yothiel

  • Content Count

    27
  • Joined

  • Last visited

Community Reputation

54 Excellent

About Yothiel

  • Rank
    Junior Member
...

Recent Profile Visitors

596 profile views
  1. Great find! I've totally overlooked the other asteroids and was only searching for oddities in the first one! I don't think it's a generation issue, but rather a liquid replication bug caused by a very specific configuration. The kind of stuff @mathmanican just loves to understand and implement in some ingenious contraptions. However I have been out of touch with the game for too long and have no idea if this instance is old news or not. Looking back at previous autosaves (16 being the oldest I still have), it seems the interesting stuff has been occuring since cycle 18. Before this, some weird stuff did occur (namely, most of the crude in the pocket getting concentrated in a single tile, but the total mass was pretty much the same as at game creation). Then at cycle 17, this natural cooker got hot enough for making some sour gas and moving things again. Watching from this cycle, I can observe very different results, such as: the pocket stabilizing back normally the pocket being somehow supercooled (a different bug, or just a different symptom of the same cause?), see attached screenshot a more persistent ballet between petroleum, crude oil and vacuum, possibly leading to duplication at some point Different simulation speeds, excessive pausing/resume and/or leaving the area unexplored may or may not play a role in which result you can get. It's actually quite random at this point. In my game, I seem to have stayed in the third case for a long time. I've noted the total mass of the pocket over the various autosaves: cycle 18: 28.7 t cycle 19: 32.6 t, duplication at some time started cycle 20: 52.7 t cycle 21: 134.4 t cycle 22: 346.4 t cycle 23: 566.1 t cycle 24: 944.7 t cycle 25: 3349.4 t I've attached a couple saves in case people want to have a look / play with this. This occurs on asteroid Jeanina, near the bottom right corner of the map. Most of the time in those saves, when observing directly the area, it can evolve both into mass duplication, or supercooling which brings some calm (at least until the area heats up again). Running at x1 sim speed seems to keep the duplication from going haywire, but at cycle 25 it won't be enough to stop it altogether. Note that I've only used vanilla speeds during my playthrough, and mostly speed 2. It seems a specific tile has a role of catalyst in the duplication bug, as I've seen duplication starting after it breaks (either naturally from liquid pressure or being debug-deleted / dug out). I've tried repainting this tile with neutronium at a later time (cycle 24-25ish), but it's not enough to stop the duplication. You can see it in the attached screenshot, it's the tile already damaged, just bottom right of a sour gas tile. I think the staircase-shaped pocket with a couple tiles of vertical clearance may play a role for causing the very specific striped pattern we can see in your screenshot, as well as the massive duplication. Most of the mass is in a column of crude oil stuck into two columns of petroleum. Curiously, I've tried observing the area starting from a fresh game with that seed, and it seems the magma pocket is solidifying after turning a couple tons of crude to petroleum =( Breathless Pit Cycle 18.sav Breathless Pit Cycle 19.sav Breathless Pit Cycle 22.sav Breathless Pit Cycle 25.sav
  2. At some point, performance starts to degrade heavily (micro-freezes, increases in CPU & memory loads). Saves done during the bug also have bloated sizes. Eventually, the game process will likely crash by lack of memory. Exiting the game & reloading produces the same results. I've tried validating the files via Steam, but no file issue was detected nor did it impact the bug. No workaround found. Player logs (also attached to this bug under the logs section) are flooded with messages such as "[20:31:08.779] [1] [WARNING] Too many water particles to render. Wanted 163395 but truncating to limit" I've attached 3 saves that were done during the generation of these logs: "prebug" is a couple seconds before the first "Too many water particles" message "postbug" is a couple of second after these first messages, when micro-freezes also started to occur "postbug-2" is taken a dozen seconds after. Note the bloating savefile size (almost x5, and it can keep increasing way more in my tests) Breathless Pit-prebug.sav Breathless Pit-postbug.sav Breathless Pit-postbug2.sav Player.log
  3. After some googling, it may actually be a duplicate of the following issue already detected in previous versions:
  4. A meteor is able to destroy bunker doors a fraction of second after its construction is done. Molten steel at 2000+°C will be generated (and typically lost to space). A minable cell of steel can also appear. Issue detected during normal gameplay, but difficult to normally observe. Using instant build will greatly help (see steps to reproduce) Reproduced in QoL 1 preview branch. Also occurs with bunker tiles in similar condition: the tile building itself may remain, but the underlaying steel will be destroyed, and a cell of minable steel will be created nearby.
  5. A small pool of petroleum is loosing heat somewhat quickly. I couldn't find any source of cold (the only pipe through the pool is warmer). This keeps happening after reloading. This seems to be caused by the pump interaction with the liquid (it seems to be either the middle cell of bottom row or the left cell of top row). Deconstructing the pump stops the cooling and rebuilding it causes the cooling to resume. Issue detected in the live branch, and also reproduced in QoL 1 preview branch Subterranean Friends.sav
  6. It looks like the strange behavior for cell/building heat transfers hasn't changed since OIL/AT versions (including the part where building isolation is not part of the formula). Until it's adressed, the only difference for insulated piping is the mass of the building, there is no actual insulation (unlike insulated tiles). So as R9MX4 said, you'll want to use abyssalite piping (doesn't have to be insulated) for the best results.
  7. I've also encountered this bug (actually wheen seeing it exploited for the following build). It seems that the part responsible for handling some of the vertical liquid interactions has a couple issues. Naively, the behavior I'd expect would be to exchange top & bottom cell contents if the elements are different and the top has higher density, or to homogenize their temperatures if the elements are the same. What it actually does: if the element are different or if the top cell is warmer / same temperature, it will exchange cell contents if the top cell has higher density if elements are the same and top cell is colder, top temperature is set, using your notations, at the value (T1* M1 + T2 * M2) / (2 * M1) and bottom temperature at (T1 * M1 + T2 * M2 / (2 * M2). Both temperatures are clamped to [min(T1,T2),max(T1,T2)] for safety. With T1 < T2 and M2 greatly higher than M1, it will indeed swap T1 & T2 So it would seems there are at least two issues highlighted above: handling the case were T1 >= T2 differently using a wrong formula for equlibrium temperature: fraction denominators would need to be M1 + M2, not 2*M1 or 2*M2 (and indeed, the current formula would only have the same results when M1 = M2)