Upon doing nothing but reloading, first thing that pops up are four crash reports. Iāve had none to this point.
- release-943 (x64)[M]
stonehearth/components/job/job_component.lua:111: attempt to index local ājsonā (a nil value)
stack traceback: [C]: ? stonehearth/components/job/job_component.lua:111: in function <stonehearth/components/job/job_component.lua:61>
- release-943 (x64)[M]
c++ exception: lua runtime error
stack traceback:
- release-943 (x64)[M]
std::logic_error: āinvalid reference in native get_idā
stack traceback: [C]: ? [C]: ? [C]: in function ālistenā radiant/modules/events.lua:53: in function ālistenā stonehearth/components/social/social_component.lua:73: in function ā_create_listenersā stonehearth/components/social/social_component.lua:41: in function <stonehearth/components/social/social_component.lua:34>
- release-943 (x64)[M]
get_id
stack traceback: [C]: ? [C]: in function ālistenā radiant/modules/events.lua:53: in function ālistenā stonehearth/components/social/social_component.lua:73: in function ā_create_listenersā stonehearth/components/social/social_component.lua:41: in function <stonehearth/components/social/social_component.lua:34>
Can you upload the save file?
Yes I can. Going back to my computer now, give me 2 minutes.
1545354873820.rar (2.7 MB)
The first crash is a known issue that happens occasionally 
. For the list of errors, it should be unrelated to the crash, itās possible some mods are conflicting with each other during load. Itās hard to say cause there are a good number of mods in your save. If you have a save that still works from before this, Iād recommend using that 
When you said, āHappens occasionallyā, how often did you mean? Iāve never once seen this, and now Iāve had this happen three times in less than 24 hours. 
Itās tough to say, because we never ended up being able to reproduce it consistently and find what the root cause of this was . I usually donāt see it in the majority of my playthroughs. Does it happen more for you with a specific save?
Well thatās what I was wondering, because they were all the same save. So I recently started a new one with a different kingdom and biome, to see how it goes. Thank you for replying!
It has happened in the past to some users, but we were unable to reproduce it in the savefiles that the users shared (that allegedly were crashing every now and then).
If you find a pattern for reproducing the crash let us know (like, does it happen the moment that you placed an item? what was going on in the game? etc).
Your savefile does use many mods and itās multiplayer so I wonder if thereās some kind of conflict.
From what I recall the last two since they were so recent, there was nothing similar happening. Unless it was something so menial like mining, or harvesting plants, that it wasnāt something I was specifically doing myself and not autonomously.
I donāt even remember what was going on when the very first one happened last night. Iāll try to pay attention to remember this time.
Based on the message, I might suspect a floating number precision issue which would not show up on non-release builds. Maybe Intel vs AMD FPU differences? Chances are removing or loosening that assert might fix the crash.
Would a floating point precision issue result in a NaN?
Indirectly, I suppose. E.g. 0/(1-f()) might be NaN if f() return 1, but not if itās 1+1E-20.
So I changed the assert to an early exit, which will skip updating the rendering for mobs on frames that have NaN transforms, but shouldnāt crash anymore. If it was a precision issue on the client side, it should sort itself out quickly and shouldnāt have any noticeable adverse effects in the game (i.e. might skip updating an entity for a frame or two).
TLDR: This should be fixed on the latest build on Steam. Thanks for the report @MichaelD7 & for the help @max99x
Yes Iāve been playing on the save that crashed so often with the test build and have had no issues!