[1.1.59] Crash with biters spawned in water

[iännis]

Hello,

Some biters have nested directly in water (not on an island). If I try to kill them the game crash. Also since I have discovered them, even if I leave them alone, the game often crash...

Could you remove them from my game ?

Thanks

[iännis]

Capture d’écran 2022-06-08 à 20.17.21.png (687.98 KiB) Viewed 4285 times

[Loewchen]

Post log and save please, see 3638.

[iännis]

[1.1.59]

[iännis]

On the save, try to walk downward, you will find the water nest and after a few time it will crash the game...

[lyvgbfh]

Code: Select all

 11.029 Factorio initialised
  11.032 Steam Storage Quota: 3708/4768
  11.033 Loading map C:/Users/Me/AppData/Roaming/Factorio/saves/R6.zip: 40658702 bytes.
  11.072 Loading level.dat: 96349459 bytes.
  11.077 Info Scenario.cpp:199: Map version 1.1.59-0
  14.268 Loading script.dat: 425 bytes.
  14.273 Checksum for script C:/Users/Me/AppData/Roaming/Factorio/temp/currently-playing/control.lua: 3661357476
  29.379 Error CrashHandler.cpp:494: Exception Code: c0000005, Address: 0x00007ff6aa0219fd
ModuleBase: 0x00007ff6a9cd0000, ImageSize: 01db9000, RelativeAddress: 003519fd
  29.380 Error CrashHandler.cpp:500: Access Violation: Write at address 0000000000000004
  29.381 Error CrashHandler.cpp:514: Exception Context:
rax=0000000000000000, rbx=0000000000000025, rcx=0000020016b45890,
rdx=0000000000000002, rsi=0000000000000029, rdi=0000000000000001,
rip=00007ff6aa0219fd, rsp=000000ed0f97e470, rbp=000000ed0f97e4c0,
 r8=00000000000000b6,  r9=0000000000000000, r10=0000000000000000,
r11=0000000000000000, r12=0000000000000003, r13=0000020025369370,
r14=0000000000000003, r15=0000000000000003
  29.382 Crashed in C:\Games\Factorio\bin\x64\factorio.exe (0x00007ff6a9cd0000 - 0x00007ff6aba89000)

[Rseding91]

I went to load your save file however i'm getting different crashes happening when loading it. How did you get this save into this state? The biter/worm generation rules do not allow it to generate on water like you've got so something has gone terribly wrong to get into that state.

Do you have some way to get a good save file into this broken state?

[iännis]

Hi,

On this save, the water biters are still in the dark but if you walk downward you will find them !

iännis

[iännis]

Hi,

This save is far more stable, it will crash only after a long time (like one hour or two) if you don't go near the water biters.
Water biters are downward in the dark.

[Rseding91]

Both of those saves have the same corruption when loading. So what ever happened to them happened before that point.

[boskid]

I was looking at those save files (R2c, R6) but all of them have signs of a corruption. In 1.1.61 there will be an extra consistency check that will throw. From the new consistency check i got some details about the corruption: There are 12 military targets on a chunk [38, 36] but they are registered on chunk [38, 32] which is 4 chunks up, and another 6 which should also be 4 chunks above of where they are.

Code: Select all

  25.966 Error Chunk.cpp:536: On chunk [38, 32] (surface nauvis) there is a MilitaryTarget (for entity name=big-biter, type=unit, position={1240.0000000000, 1181.0000000000}, setup=true, owned-by-ghost=false, to-be-deconstructed=false, to-be-upgraded=false, simulation=false) which is not registered on this chunk but on chunk [38, 36] (surface nauvis)
  25.967 Error Chunk.cpp:536: On chunk [38, 32] (surface nauvis) there is a MilitaryTarget (for entity name=big-biter, type=unit, position={1243.0000000000, 1181.0000000000}, setup=true, owned-by-ghost=false, to-be-deconstructed=false, to-be-upgraded=false, simulation=false) which is not registered on this chunk but on chunk [38, 36] (surface nauvis)
  25.968 Error Chunk.cpp:536: On chunk [38, 32] (surface nauvis) there is a MilitaryTarget (for entity name=big-biter, type=unit, position={1221.0000000000, 1178.0000000000}, setup=true, owned-by-ghost=false, to-be-deconstructed=false, to-be-upgraded=false, simulation=false) which is not registered on this chunk but on chunk [38, 36] (surface nauvis)
  25.968 Error Chunk.cpp:536: On chunk [38, 32] (surface nauvis) there is a MilitaryTarget (for entity name=big-biter, type=unit, position={1220.0000000000, 1168.0000000000}, setup=true, owned-by-ghost=false, to-be-deconstructed=false, to-be-upgraded=false, simulation=false) which is not registered on this chunk but on chunk [38, 36] (surface nauvis)
  25.969 Error Chunk.cpp:536: On chunk [38, 32] (surface nauvis) there is a MilitaryTarget (for entity name=big-biter, type=unit, position={1224.0000000000, 1178.0000000000}, setup=true, owned-by-ghost=false, to-be-deconstructed=false, to-be-upgraded=false, simulation=false) which is not registered on this chunk but on chunk [38, 36] (surface nauvis)
  25.969 Error Chunk.cpp:536: On chunk [38, 32] (surface nauvis) there is a MilitaryTarget (for entity name=medium-biter, type=unit, position={1218.0000000000, 1178.0000000000}, setup=true, owned-by-ghost=false, to-be-deconstructed=false, to-be-upgraded=false, simulation=false) which is not registered on this chunk but on chunk [38, 36] (surface nauvis)
  25.969 Error Chunk.cpp:536: On chunk [38, 32] (surface nauvis) there is a MilitaryTarget (for entity name=big-biter, type=unit, position={1237.0000000000, 1181.0000000000}, setup=true, owned-by-ghost=false, to-be-deconstructed=false, to-be-upgraded=false, simulation=false) which is not registered on this chunk but on chunk [38, 36] (surface nauvis)
  25.970 Error Chunk.cpp:536: On chunk [38, 32] (surface nauvis) there is a MilitaryTarget (for entity name=big-worm-turret, type=turret, position={1232.0351562500, 1175.4414062500}, setup=true, owned-by-ghost=false, to-be-deconstructed=false, to-be-upgraded=false, simulation=false) which is not registered on this chunk but on chunk [38, 36] (surface nauvis)
  25.970 Error Chunk.cpp:536: On chunk [38, 32] (surface nauvis) there is a MilitaryTarget (for entity name=biter-spawner, type=unit-spawner, position={1221.0000000000, 1181.0000000000}, setup=true, owned-by-ghost=false, to-be-deconstructed=false, to-be-upgraded=false, simulation=false) which is not registered on this chunk but on chunk [38, 36] (surface nauvis)
  25.971 Error Chunk.cpp:536: On chunk [38, 32] (surface nauvis) there is a MilitaryTarget (for entity name=big-worm-turret, type=turret, position={1229.0000000000, 1183.0546875000}, setup=true, owned-by-ghost=false, to-be-deconstructed=false, to-be-upgraded=false, simulation=false) which is not registered on this chunk but on chunk [38, 36] (surface nauvis)
  25.971 Error Chunk.cpp:536: On chunk [38, 32] (surface nauvis) there is a MilitaryTarget (for entity name=small-worm-turret, type=turret, position={1227.9765625000, 1171.1601562500}, setup=true, owned-by-ghost=false, to-be-deconstructed=false, to-be-upgraded=false, simulation=false) which is not registered on this chunk but on chunk [38, 36] (surface nauvis)
  25.971 Error Chunk.cpp:536: On chunk [38, 32] (surface nauvis) there is a MilitaryTarget (for entity name=medium-worm-turret, type=turret, position={1225.7460937500, 1178.6992187500}, setup=true, owned-by-ghost=false, to-be-deconstructed=false, to-be-upgraded=false, simulation=false) which is not registered on this chunk but on chunk [38, 36] (surface nauvis)
  25.972 Error Chunk.cpp:536: On chunk [38, 33] (surface nauvis) there is a MilitaryTarget (for entity name=big-biter, type=unit, position={1243.0000000000, 1184.0000000000}, setup=true, owned-by-ghost=false, to-be-deconstructed=false, to-be-upgraded=false, simulation=false) which is not registered on this chunk but on chunk [38, 37] (surface nauvis)
  25.973 Error Chunk.cpp:536: On chunk [38, 33] (surface nauvis) there is a MilitaryTarget (for entity name=big-biter, type=unit, position={1240.0000000000, 1187.0000000000}, setup=true, owned-by-ghost=false, to-be-deconstructed=false, to-be-upgraded=false, simulation=false) which is not registered on this chunk but on chunk [38, 37] (surface nauvis)
  25.973 Error Chunk.cpp:536: On chunk [38, 33] (surface nauvis) there is a MilitaryTarget (for entity name=big-biter, type=unit, position={1221.0000000000, 1184.0000000000}, setup=true, owned-by-ghost=false, to-be-deconstructed=false, to-be-upgraded=false, simulation=false) which is not registered on this chunk but on chunk [38, 37] (surface nauvis)
  25.973 Error Chunk.cpp:536: On chunk [38, 33] (surface nauvis) there is a MilitaryTarget (for entity name=big-biter, type=unit, position={1237.0000000000, 1187.0000000000}, setup=true, owned-by-ghost=false, to-be-deconstructed=false, to-be-upgraded=false, simulation=false) which is not registered on this chunk but on chunk [38, 37] (surface nauvis)
  25.974 Error Chunk.cpp:536: On chunk [38, 33] (surface nauvis) there is a MilitaryTarget (for entity name=medium-biter, type=unit, position={1218.0000000000, 1184.0000000000}, setup=true, owned-by-ghost=false, to-be-deconstructed=false, to-be-upgraded=false, simulation=false) which is not registered on this chunk but on chunk [38, 37] (surface nauvis)
  25.974 Error Chunk.cpp:536: On chunk [38, 33] (surface nauvis) there is a MilitaryTarget (for entity name=biter-spawner, type=unit-spawner, position={1240.0000000000, 1184.0000000000}, setup=true, owned-by-ghost=false, to-be-deconstructed=false, to-be-upgraded=false, simulation=false) which is not registered on this chunk but on chunk [38, 37] (surface nauvis)

If you have any steps to reproduce this issue i would like to know but until then i am moving this to "1/0 magic".

[boskid]

When looking into this more i think i know what happened. There is quite huge stripe of entities that are all shifted down by 4 chunks:

Entities as in the save file:

before

screenshot_2_corrupted.png (8.61 MiB) Viewed 3868 times

Entites when chunks were regenerated:

after

screenshot_2_regenerated.png (6.4 MiB) Viewed 3868 times

This aligns with the entities saving order on a surface (where surface is saved going with stripes from left to right and each stripe is saved from top to bottom) and the MapPosition saving optimization which remembers previously saved position inside of a save file and if next position is nearby it only saves a position delta. In that case there is only 1 bit of data that would have to be corrupted (first position or one of the deltas) inside of a save file at some point to cause this type of corruption where all entities saved next are loaded at wrong position up to a point where MapPosition is too large to be saved with a delta and a full position is given that would fix loading of future positions.

[iännis]

What a glitch !!!
It's interesting and amazing to see what a single bit error can change !
Thanks

[iännis]

Hello

Since factorio 61 I can't play on my saves anymore...

viewtopic.php?f=7&t=102632

Is it possible to correct that glitch ?

Should I go back to 59 ? How ?

Thanks

iännis

[Rseding91]

Hello

Since factorio 61 I can't play on my saves anymore...

...

Is it possible to correct that glitch ?

...

Your computer is broken. You need to fix your hardware. There is nothing we can do to fix your hardware.