This is actually two bugs. The first bug is that my game 'Munich Central' didn't disappear from the Browse Game Section after I stopped the headless server (Ctrl-C). EDIT: Here's the end of the log from the headless server: http://pastebin.com/7SzFyXYN The second bug appeared when I tried to connect...

it wouldn't break determinism if he was playing without RSO before. He could disable the part of the mod that is responsible for resource generation and only use the biter base generation part. But for the biter base spawning one could write a simple mod. I think one file with ~30 lines would be eno...

If you only care for enemy bases just let the game run fit a while and let the bitters expand.


If you want to use RSO then you could change the code so it will regenerate bitters and only generate bitters in newly generated chunks.

It all depends on what you want.

it wouldn't be that difficult. assuming regionsize > chunksize one would only need to check which region the 4 corners belong to. But you are right, it wouldn't give any meaningful benefit. For my rewrite i'm rethinking the whole region approach. There should be a way to create some kind of "gr...

but region size determines the number of chunks in a region.

what i wanted to say is that one could specify that one region is 128 tiles instead of 4 chunks for example. if it's changed to tiles instead of chunks one could use any number for chunksize (like 100 for example)

i'm not 100% sure but 0,0 should be the point where the 4 innermost regions touch, so you should aim for multiplies of 2 of the regionsize for the mapheight. in general i think region size isn't necessaryly linked to chunksize. in theory any value should be possible. I'll experiment with that in my ...

the limit is there because dark did some chunk checking and i think he only checked in a 32 radius if one can spawn there.

there is a cap on maximum amount per tile. if a tile would exceed 4k i think it would be reduced to a random value between 3 and 4k.

I tried to just ignore the can_place_entity step and it turned out quite well. I think i'll leave it that way for now.

Of course a proper solution would look different.

I don't know if this is a bug or intended behavior but surface.can_place_entity seems to always return false if the chunk/position isn't generated yet, even if the object wouldn't collide with anything. Simple test: /c game.player.print(serpent.block(game.player.surface.can_place_entity({name='stone...

Thanks for the fix, that will help quite a lot for debugging.

But i'd guess the working on the string would be a one time thing, so it would lag once only.

I found the problem and it is a strange one.
At one point in my resource generation algorithm i'm checking surface.can_place_entity

It turns out, this returns false for chunks that are not generated yet.

Thanks for the idea. I tried it, it still doesn't work.

My approach was to set a variable for each chunk once the on_chunk_generated event was executed and only spawn if this variable is set. i'm still missing chunks. And yes i checked if they are even calculated and they are. they just disappear.

Most of the output comes from dumping some mod related tables. But i'd say the problem is that the game tries to render a huge string(several thousand characters) and it takes some time to do it. Since scrolling isn't possible cutting long strings to just display the last part could solve the issue....

Today i experienced the same issue with 0.12.2 but it seemed like it took less time to "catch up".

Hey, as i already noted in the irc if you open the console and the console history is quite full, especially with long lines (serpent.dump() of tables with several hundred entries) the fps will drop significantly. I created this report just for reference since i think it's already fixed? for the nex...

Hey, so for my mod debugging i'm basically charting the map all the time with \c game.player.force.chart(game.player.surface, {{-1024,-1024},{1024,1024}}) After the update to 0.12.3 i noticed that when the chart is running and i'm focusing another window and then focus back to factorio, my inputs ar...

So while rewriting RSO i stumpled accross another problem that i wasn't able to solve completely yet. The idea is to spawn huge resource patches that are calculated on_chunk_generated. But these patches are bigger than the chunk that creates them so they will extend into other chunks. My first idea ...

As i already told you, i think creating your own kind of pollution would be the best. You define it as per chunk. you update it once every 60 ticks or whatever. your biters are gathered as unit_groups. each group checks the level of your pollution in the current chunk and adjacent chunks, if the lev...

One big step that came with 0.12.0 was he ability to read the map_gen_settings and it was implemented in RSO. Currently it's all about balancing that impact.