Content not appearing is the worst, 'cause all too often the devs know it’s supposed to be there but don’t know it’s not showing up, while the players know it’s not showing up but don’t know it’s supposed to be there. And if they do get wind of it, they don’t know if it’s a bug, or deliberately dummied-out code. As may or may not be the case here; I have no idea.
I’ve always thought that the files loaded by the games was the ones without “_import”, since were the one i’ve usually use to mod things.
Currently i’ve seen and killed Blueprophets just once, at Crying Heights.
Looking at the game file there should be 4 spawner of them. one “bugged” that will spawn just Jillyfish.
For tree of Ages and Eater of Names, the should be spawned ONLY in one location each, with a 50% probability (splitted with the other spawner i think).
However i’d like some words form the devs, since the spwaner routine seems a bit erratic.
Interesting. I just tested, and you’re absolutely right. _import must be what comes down when you get a new story available, and it merges them.
It’s strange, then, that the non _import files have so much repetition/redundance in them. But yeah, in that case, it should be able to spawn.
p.s. speed boost x5 on full power is extremely funny
Alright. I think I may have picked this one apart.
There are two major versions of each tileset in the tiles.json file. For the Chelonate (at least), there are 3 minor variations on each of the two, giving 6 possible combinations. I can’t verify this, but I’m guessing when the map is generated, it randomly picks one of the two major sets, and then one of the subsets. Most differences are types and amounts of spawn, and location of minor terrain like buoys and outlying islands.
Anyway, for the Chelonate, only one of the major tilesets mentions the eater. The other has a crab instead. So, if I understand this right, if you have a crab to the west of the chelonate (see second image), you’re not getting the eater, and need to suicide your captain, and hope it spawns the right tileset next time. Of course, I could be completely wrong, and it never spawns the other tileset. Who knows? But, with some minor file editing, here’s the eater:
Eater: http://i.imgur.com/xHFrRKp.jpg
Location: http://i.imgur.com/FKDTrSP.png
Die, Eater, die: http://i.imgur.com/rIunSqN.jpg
Looting Options: http://i.imgur.com/q1xTohi.jpg
Figurehead: http://i.imgur.com/1uyM5RE.jpg
(if you pick the other option, it drops a cache of curiosities with a value of 30-50).
Well, that seems odd. Is that the case for the Blue Prophets as well?
What’d you edit to make the eater spawn? I want to kill it too!
After downloading today’s content bump, I had another poke around; it seems like all of the spawn points should be spawning what they’re supposed to be spawning now, except one of the Blue Prophets points is still spawning Jillyfish, I guess? The _import files say that the spawnpoints are all bust, which is weird because they’re usually in lock-step, but they’re not what the game uses anyway so it’s no big
Oh, so it may be working now? Time to go hunting with my new Frigate.
Would that be these guys?
Major Version 1:
"Name": "CarnelianBlueProphets",
"SpawnName": "Jillyfish",
Major Version 2
"Name": "CarnelianBlueProphets",
"SpawnName": "BlueProphets",
CryingHeights has the same deal, with blue prophets vs tyrant moths.
Given that we’re all seeing no blue prophets and no eater of names (and the prophets/eater are always in the second set of tiles) it’s either a (very) rare chance to use the second tileset, or there’s something wrong.
tiles.json in C:\Users<user>\AppData\LocalLow\Failbetter Games\Sunless Sea\geography
I replaced the references to chelonatecrab with chelonateeaterofnames. You’ll probably have to put it through a json formatter, though, because by default the whole file is on one line.
I would have expected the Prophets to be around the Crying Heights only, considering that’s the text for that region. Huh.
It’s the same with Crying Height and Carnelian Coast (and, maybe, with other Tiles for Behemustache, TreeOfAges etc).
Each one have two main version.
One with two Tyrant moths and another with two blue prophets, in the case of Crying Heights.
One with Behemoustache and Blue Prophet and another with two Jillyfish, in the case of Carnelian Coast.
That’s for "static" tiles.
For dynamic tiles it’s a bit more complicated.
There are again two main versions, each array has different (usually three) "orientation".
I Really don’t know if it’s working as intended since it’s REALLY a poor implementation and randomization, it would be really nice if every setted spawn point has a % of spawning some monsters.
edited by Frenzgyn on 2/18/2015
Do the _import.json files only have one version of each? Because I have a feeling the import system is duplicating content somehow.
WELL WELL Gregg, i was checking this just while you were posting, and i think we have just found something here.
YES, defiitively the import system it’s faulty and duplicates entry. It’s so for all the geography files, but i guess it’s the same far all the files.
I need some testing from you gentleman, i’m using the test build of steam, however, deleting Tiles.json and the _import version, seem to FIX the issue (no duplication and correct spawn points, and i have to say, Behemoustache and BlueProphet have really nasty moves :S, they quite killed my merchant cruise).
I need someone check this also on the stable build.
This could be also part of the issue with officer/journal entries duplication and framerate issue.
I’d like some words from a dev, before pointing it out in a bug report.
I’ve noticed duplicate entries both in the encyclopaedia files (where all the weapons are stored) and in the tiles, and possibly a few others, but the entities files are surprisingly free from duplicates. Those are so large because the import system fills each entry with useless null/default values.
I’ve never heard of the officer/journal entry duplication issue. I’ll try to mess around with the files you mention when I get home.
I’ve been tinkering some more, and it seems to me that the _import files are downloaded when there are new stories, put through some kind of processing to expand them, and turned into the used .json files (there’s generally a lot more data in the non _import files, but nothing different, exactly, just more redundant data).
As for the random generation, I’ve run through about 30 captains, killing some via terror, via fuel, supplies, hull and retirement. Haven’t tried crew death. I’ve tried a handful of brand new captains, and one from someone who’d basically completed everything. None of them spawned the Eater - just the crab. At this point, I have to assume that it’s either a small random chance of generating the eater tileset, or you need to do something specific with a previous captain.
It seems (although it’s impossible to confirm) that the blue prophet tileset is linked to the eater, since I haven’t seen that either. But if both have a small random chance of independently appearing, I’d be seeing more or less the same results. So at this point, I have no idea what to do to spawn it, but can confirm if you have a crab in the Chelonate, then you’re not going to see the Eater.
Not at all Sporks,
Tiles_import.json just try to "merge" itself with "Tiles.json" during a story update, but "fail" in some degree, resulting in just copying itself at the top of the old Tiles.json.
Proably when the games is searching the info for the tile, it pick the first value, and that’s why you have always the crab.
If you delete Tiles.json and Tiles_import.json and start the game again, it recreate the former, as the one from release (without the story updates modification, if there are any) and in this case, there are no duplication and the spawners are the "correct" ones, so you will find Bluprophets and all the other beasties.
I’d like you will do some tests to confirm mines.
edited by Frenzgyn on 2/18/2015
Right. I did this a few days ago. If you delete the tiles.json and tiles_import.json, it rebuilds the tiles.json from whatever defaults exist.
What we don’t know (and can’t know) is how it’s selecting items in the whatever.json. Is it random? Is it dependent on something a previous captain has done? Or is it, as you say, actually just broken? And if it’s broken (and just adding _import to the top of each file) then why are acoupe of _import files bigger than their non-_import counterparts?
I’ll email the support team, and see what they say, because at this point, we’re all just making blind guesses based on incomplete data.
Don’t look at the raw size of the files, because it depends also on how the text is formatted.
*.json is treated like a single line textfiles, while the majority of *_import.json are in some degree formatted with spaces and new line.
If you use Jstool to format the files and then compare them with a program like Winmerge, you’ll found out IF there are difference.
If you don’t format first, Winmerge treat the two files as something completely different, also if it’s not the case, like TileRules.json and TileRules_import.json, who are the same, the former just with duplicated entries.
I’ve already dropped a mail at the support team, linking also this discussion.
edited by Frenzgyn on 2/18/2015
edited by Frenzgyn on 2/18/2015