v1.21.0-rc.4 - Story Chapter 2 Redux

[Akari_Enderwolf]

On 8/11/2025 at 3:41 PM, Krougal said:

Speaking of pit kilns, I am still seeing 20 hours for peat, although it has been intermittent. Also Dana updated extra info for RC last night, so I doubt it is his mod.

Do you have any of her other mods, or just that one? Idk if it could be Dana Tweaks, never fully looked into everything that the mod changes.

[Krougal]

13 minutes ago, Akari_Enderwolf said:

Do you have any of her other mods, or just that one? Idk if it could be Dana Tweaks, never fully looked into everything that the mod changes.

Yeah, quite a few, Dana being quite the prolific modder.

Tweaks got updated yesterday so I started using it again, it wasn't really working for 1.21.

Dana actually did start updating some of the mods early.

If it isn't vanilla information (because some things are, some aren't) I'm pretty sure it is Extra Info, I know it was only partially working before update as well, because charcoal pits were giving me nothing until then.

Hopefully stable will be soon though, so we'll see. I've got so many other mods it could always be a conflict too.

[clseibold]

16 hours ago, alberto_ascani said:

Maybe you're right, I didn't take notes, but perhaps I should, but I'd also like to know exactly what data is needed. Also, mind you, I'm not talking about "subjective" taste, or rather, it always is. I'm talking about how biomes are generated and how they integrate with each other, how waterways develop, and how mountain tops are generated. Furthermore, I've also done dozens of tests by changing the world generation parameters, but obviously this can cause problems, and I'm aware of that. Generally speaking, I've noticed that worlds with the default settings are much less pleasant and natural than those tested in previous versions.
I repeat, I am available to collect data or information but I would need to know what needs to be collected!

I do wonder how one would collect data on procedural world generation. That seems like a hard thing to do. I guess the best thing I could think of is to just give a list of different patterns/attributes of generation that you dislike, and then provide a bunch of seeds and locations where they can be found. Then the devs can see the patterns directly, make changes, and test those seeds to see if the changes took effect and helped solve the problem.

The other option might be for someone to create some mod that looks for different specific patterns (like floating islands for example) in the world, sees how common they are, and then generates a list of them, their locations, and from which seed they are from.

Something like a floating islands detector would have to look for contiguous land that has air blocks beneath it, for example. Not sure off the top of my head how resource intensive this would be though.

Another world analyzer tool could be something that maps out the slopes of the land. This one is probably pretty simple in theory, but I'm not sure how one would create the visual for it. Maybe something like a top-view height map that highlights particular slopes that are bad, or something.

I guess the problem with the gravel is just the general shape and size? So maybe analyze for gravel areas and gather info on how big they are (height, width, radius, or something of that sort).

As for shorelines and the dirt/grass line patterns, maybe something similar to the slopes, but on the x and y axis rather than the z axis (assuming z is the axis towards the sky).

Idk, these are just ideas.

Edited August 14, 2025 by clseibold

[LadyWYT]

7 hours ago, Akari_Enderwolf said:

Chapter 1 is probably the shortest one to my knowledge, longest part of it is just finding a Treasure Hunter since CH1 is just one dungeon.

This isn't the case anymore, since other traders will now sell you a map to the nearest treasure hunter, should you wish to skip the search.

[yanlong]

I want to know when the official version will be released？
Will this game be released on Steam?

[Maelstrom]

20 minutes ago, yanlong said:

I want to know when the official version will be released？
Will this game be released on Steam?

Frequently Asked Questions - Vintage Story

[Akari_Enderwolf]

3 hours ago, LadyWYT said:

This isn't the case anymore, since other traders will now sell you a map to the nearest treasure hunter, should you wish to skip the search.

my point was more just that Chapter 1 is really short. Since it'll probably take you longer to find a trader, and then get to the treasure hunter, than to get to and clear the dungeon.
I have occasionally had worlds where Traders were few and far between, so it took me a while to find one.

[Akari_Enderwolf]

48 minutes ago, yanlong said:

I want to know when the official version will be released？
Will this game be released on Steam?

not for a long time, we still have 6 more story chapters to get through, as far as I know. As far as the steam thing, see the FAQ

Edited August 14, 2025 by Akari_Enderwolf

[LadyWYT]

14 minutes ago, Akari_Enderwolf said:

my point was more just that Chapter 1 is really short. Since it'll probably take you longer to find a trader, and then get to the treasure hunter, than to get to and clear the dungeon.
I have occasionally had worlds where Traders were few and far between, so it took me a while to find one.

Honestly you could say the same for Chapter 2 as well. How long or short a chapter ends up being depends heavily on a player's skill at the game and puzzle-solving, as well as how the player approaches the story content in general. If you focus purely on completing the objectives and don't care about loot or poking around looking at the lore and location details, the chapters are going to end up pretty short as a result.

[yanlong]

7 hours ago, Akari_Enderwolf said:

not for a long time, we still have 6 more story chapters to get through, as far as I know. As far as the steam thing, see the FAQ

I have read it. Does the production team have the latest news about Steam?

[yanlong]

8 hours ago, Maelstrom said:

Frequently Asked Questions - Vintage Story

undersand.

[Maelstrom]

That question gets asked a lot.  I don't understand why people are so enamored with steam.   It's extra annoinace to me.  Buy game from publisher on steam platform (which is only place to buy said coveted game) and have to download steam app to my computer in addition to the game.  Every so often the steam app fires off when I don't want it, I just want to play may game gosh darn it!

Maybe I'm just a gumpy old man, but all the new and "improved" way of doing things doesn't seem all that improved to me.  It was much easier to just buy a game through distributor of choice and enjoy game free from other people's fingers invading my pie.

[Thorfinn]

Another complication with compiling data on worldgen is that if you don't understand how worldgen works, you can't really add much actionable information. Ex: Is this a situation of how one region transitions into another, or is it a region's Perlin generation? If the former, which two regions are not playing together nicely, if the latter, which region is creating the problem? If you play with map, you can probably figure out if it's in inter- or intra-region issue -- the boundaries between regions are fairly obvious once you know what you are looking at. It takes a lot longer to figure out which specific regions are involved.

Some are fairly easy -- if it's swampy, or a plateau or islands, you can search the landforms.json and narrow it down to a few possibilities. That would likely be useful information. If you have narrowed it down to the fact you don't like the spires of "needledflatlands", and would prefer them shorter, that would be useful, but ideally resolved by adding a "lessneedledflatlands" that has Perlin noise damped, so that people who like the spires don't lose them. If the problem is that the transition from "Realistic mountains" to "largelake" is too abrupt, that is likely actionable. I doubt anyone would mind that, other than me -- without moveable source blocks, I like building my farm into a lake, and have my mill at the top of a nearby mountain, ideally a plateau that has a lot of buildable space. Smoothing that out would just make me choose between running further between base-lets, or building a mill on the lowlands. 

But the cool part is that once you narrow down what bothers you, you can just tweak the parameters in landforms.json to your liking, maybe even upload a mod so that others who share your preferences can skip that process. If you don't like the spires at all, just set the weight to 0 or something sufficiently small and Bob's your uncle.

Edited August 15, 2025 by Thorfinn

[Krougal]

3 hours ago, Maelstrom said:

That question gets asked a lot.  I don't understand why people are so enamored with steam.   It's extra annoinace to me.  Buy game from publisher on steam platform (which is only place to buy said coveted game) and have to download steam app to my computer in addition to the game.  Every so often the steam app fires off when I don't want it, I just want to play may game gosh darn it!

Maybe I'm just a gumpy old man, but all the new and "improved" way of doing things doesn't seem all that improved to me.  It was much easier to just buy a game through distributor of choice and enjoy game free from other people's fingers invading my pie.

Lol. As another grumpy old man, it took me the longest time to adopt Steam, but honestly, I love it. I used to go nuts patching all my games, it was worse than working as an SA. The workshop and guides sections are really useful. The convenient shopping is another thing, my wife loves the words "I bought it on sale!" I think people probably feel safer downloading from them as well, I mean when was the last time you heard about a virus spread from a Steam download?

As a consumer, there is a lot to like about it. As a developer, not so much, as Tyron listed.

I think he is making the right decision by delaying it. Gotta give it to him, he has balls to have gone without a publisher in the first place, glad to see it has paid off for him. Maybe his success will inspire other small businesses.

16 minutes ago, Thorfinn said:

worldgen stuff

All true. It doesn't help that it isn't well understood or documented. I still think something like Chunkbase would be useful, I know the ore placement and some other things are not tied to the seed, but the general map layout should be, and that is good enough to at least get an idea. Granted it would probably be a lot of work to have all the same options the game has, and without them it's fairly worthless.

I know you will work with what you get and that seems to be part of the fun for you, but it can be a real turnoff repeatedly generating maps you are not happy with. I wasn't joking when I said I got so frustrated with it I stopped playing for a good long while.

I'm sure a large part of how pleased I am with this update is tied to the fact that I got a pretty awesome map. To quote that other guy, I am happy as a two petered puppy.

 

[Thorfinn]

1 hour ago, Krougal said:

It doesn't help that it isn't well understood or documented.

Truth. I don't talk much about it except in threads like this, addressing mapgen "deficiencies", because Perlin noise, though absurdly simple, is largely misunderstood. There is a lot of bad information out there about it, but sprinkled in are some very good discussions with explanations of what the parameters mean. Someone asked me about a good one, and I'd have to go back and look, but I provided links to several YouTubes.

I think you are right about the advantages of Steam, but IMO, it still doesn't compensate sufficiently for the disadvantages, one of the foremost being their rules for Mod Workshop. (Admittedly, I'm relying on comments from several developers on GOG who explained that they needed to code things so that Steam mods would not run on non-Steam installs, essentially making servers using mods choose between being open to people from all vendors or being Steam only, since only Steam users would have access to Workshop mods. If that one thing were to change, maybe it's worth it. At least on one machine.)

Edited August 15, 2025 by Thorfinn

[Krougal]

5 minutes ago, Thorfinn said:

disadvantages, one of the foremost being their rules for Mod Workshop. (Admittedly, I'm relying on comments from several developers on GOG who explained that they needed to code things so that Steam mods would not run on non-Steam installs, essentially making servers using mods choose between being open to people from all vendors or being Steam only, since only Steam users would have access to Workshop mods. If that one thing were to change, maybe it's worth it. At least on one machine.)

Yeah, that is an issue. Which is why a lot of people still host their mods on Nexus or Git. It does suck that if you got it from another platform you can't download content, and some other things are restricted. A lot of it will work, a somewhat scummy way to get it is to buy it on Steam, download what you want, move the files to the non-steam installation and then refund the game.

Also as far as launchers, I wouldn't even consider anything on Chi-com Epic. I hate them especially with their forcing exclusives, although it has made me wait to get some games that sorely needed a year of patching anyway (I'm looking at you Hitman3!)

[Thorfinn]

26 minutes ago, Krougal said:

A lot of it will work, a somewhat scummy way to get it is to buy it on Steam, download what you want, move the files to the non-steam installation and then refund the game.

Oh, that works? I wouldn't feel at all scummy about that, since Steam is strong-arming the developers into it. But it probably means I'd have to buy/return it every time some mod updates, yea? I suspect they would catch on before too long.

[EDIT]

Humble has a launcher, too, sort of. You need to use it to download. GOG does not. Strictly speaking, itch does not either, though, like GOG, they have something that makes it a bit easier. Those three (and probably more) are essentially locked out of modded servers, unless players scam Steam.

Edited August 15, 2025 by Thorfinn

[Krougal]

6 minutes ago, Thorfinn said:

Oh, that works? I wouldn't feel at all scummy about that, since Steam is strong-arming the developers into it. But it probably means I'd have to buy/return it every time some mod updates, yea? I suspect they would catch on before too long.

Yeah, although for some games it can be such a hassle that it really isn't worth the effort.

[Thorfinn]

2 minutes ago, Krougal said:

Yeah, although for some games it can be such a hassle that it really isn't worth the effort.

Right. I just don't feel really good about rewarding them for mulcting users and developers.

[clseibold]

8 hours ago, Thorfinn said:

Truth. I don't talk much about it except in threads like this, addressing mapgen "deficiencies", because Perlin noise, though absurdly simple, is largely misunderstood. There is a lot of bad information out there about it, but sprinkled in are some very good discussions with explanations of what the parameters mean. Someone asked me about a good one, and I'd have to go back and look, but I provided links to several YouTubes.

I think you are right about the advantages of Steam, but IMO, it still doesn't compensate sufficiently for the disadvantages, one of the foremost being their rules for Mod Workshop. (Admittedly, I'm relying on comments from several developers on GOG who explained that they needed to code things so that Steam mods would not run on non-Steam installs, essentially making servers using mods choose between being open to people from all vendors or being Steam only, since only Steam users would have access to Workshop mods. If that one thing were to change, maybe it's worth it. At least on one machine.)

Imo, this is why I feel like it would be good to write tools to analyze the *result* of a map gen, so that you can see what your perlin noise stuff is actually generating in the end. The goal should be to have a general idea of generation in mind, and then make your perlin noise and other generation methods match that, not the other way around. It's harder to tell how things are going to generate with just the perlin noise functions, but it's much easier to look at what those functions are generating in the end result, and tools that analyze the end result map help to make that much easier.

I don't have that much experience with procedural generation, but I was working on a project recently that did quite a bit of it. The biggest thing, imo, was knowing and reminding myself of the ordering of overlapping generations. For example, knowing that I generate lakes before rainfall (yearly amount for the region's climate) is extremely important, especially when the rainfall is generated based on the lakes, lol. Or that I generate general land with mountains and plains, then plateaus, then lakes, then rivers. Then I'd know better which layer will need to be tweaked (e.g., tweaking lakes would actually change river placements). It becomes easier when you're very careful in how you layer things, putting things you want to tweak directly on the bottom layers (like the rivers) and things you'd rather be calculated from other things rather than tweaked directly (like climate) further up.

This type of information, though, is not very accessible to users. Which means, imo, it's much better to just give a list of world gen features that you don't like, where they occur, and on what seed (and gen settings), and the devs can figure out whether some specific layer of generation needs to be tweaked or not, and how it needs to be tweaked.

Edited August 16, 2025 by clseibold

[Thorfinn]

3 hours ago, clseibold said:

Which means, imo, it's much better to just give a list of world gen features that you don't like, where they occur, and on what seed (and gen settings), and the devs can figure out whether some specific layer of generation needs to be tweaked or not, and how it needs to be tweaked.

I think it's safe to say they modeled each of those landforms, and decided that was what they wanted the game to be. If I were in charge of developing the landforms, I'd put only one landform into landform.json so it repeats that over the entire world. That way I can see the variations in those Perlin parameters give, without caring about seed or settings (other than uplift, obviously), simply by flying around. Look at the map to see if there's anything that stands out but otherwise, in a few minutes, you have a pretty good idea what that does.

That, BTW, is what I would do if I were interested in modding the worldgen, for those interested. 

[EDIT]

Entries like this are why I think they spent some time on these parameters

			"terrainOctaves": [0, 0.81, 1, 1, 1, 0.6, 0.3, 0.15, 0.1],
			"terrainOctaveThresholds": [0, 0, 0, 0, 0, 0, 0.5, 0, 0, 0],
			"terrainYKeyPositions":    [0.397, 0.415, 0.438, 0.459, 0.548, 0.602, 0.684, 0.711, 0.752],
			"terrainYKeyThresholds":   [1.000, 0.770, 0.517, 0.287, 0.197, 0.117, 0.047, 0.047, 0.000]

There's no obvious patterns that I can see at a quick glance. There is a difference of 90, then 80, then 70 un the middle of the last line, but the pattern ends there. They look like they have been tuned to get a specific result.

Edited August 16, 2025 by Thorfinn

[clseibold]

10 hours ago, Thorfinn said:

I think it's safe to say they modeled each of those landforms, and decided that was what they wanted the game to be.

I don't dispute that. But developer vision doesn't always match users expectations, which is why issue trackers and user feedback is so important.

The seed is absolutely useful for looking at worldgen. Firstly, things are semi-randomly generated, so seeds help you get a consistent generation that you can look at to see how something changes. Looking at a different seed every time you change world gen settings isn't nearly as helpful (because you'd be changing multiple things now - generally you want to change one thing and see how it compares). Secondly, it allows you to pick out specific coordinates that you can send to other people, and then they can just look at that specific coordinate at that specific seed to see exactly what you were seeing. That seems extremely important for user feedback and for tweaking generation in a team.

As for looking around vs. using tools, well, just looking around (going by eye) becomes harder when your land is more spread out. If the devs did make the land more traversable, they could have done so by spreading the land out so things aren't as steep, rather than just changing amplitude. That way you aren't changing the variation in height, just the movement between heights.

But all of this is just speculation, I guess. It's not like I can see the code, lol.

There's also so many different ways you can use perlin noise. In my game, I generate land first and assign biomes later, but I guess one could assign biomes and then generate land based on them. I suppose Minecraft does the latter, and simulation games tend to do the former? I don't know which way VS does it.

Edited August 16, 2025 by clseibold

[Thorfinn]

3 hours ago, clseibold said:

If the devs did make the land more traversable, they could have done so by spreading the land out so things aren't as steep, rather than just changing amplitude.

That's inherent in the Perlin function. All it does is define a curvy line between two points. The parameters determine how curvy the line is, and, consequently, the "traversability" of the terrain generated. A good way to think of it is a radio wave. Imagine the first octave is the base frequency, the carrier wave, and the second octave is the signal you are piggybacking on top of that, to add finer details. Now recurse that; add another octave to the signal, in essence making the signal itself a carrier wave for this third octave. The more octaves you use in your Perlin function, the finer the degree of detail the resulting wave exhibits.

3 hours ago, clseibold said:

But developer vision doesn't always match users expectations, which is why issue trackers and user feedback is so important.

It seems odd to have to repeat it; obviously, they did not intend to have the "realistic" terrain some users expect, but rather something that conveys the level of catastrophe the world endured, the un-worldliness of the setting. This wasn't just some minor zombie apocalypse that affected only the people; this was truly earth-shattering, shaking the very land itself in bizarre ways. There's no other reason to have landforms like "realisticflatlands" and a plethora of other flatlands, by implication, non-realistic. If they wanted something more earthlike, they would have done so. Perlin noise parameters that give earth-like results are pretty easy to find on-line. Have been for at least the last 25 years. At one time I used one to generate D&D maps.

You are right that many do not want the experience Tyron envisions, thus the plethora of more earthlike mapgen mods. For Homo Sapiens mode, it might make sense to have a different landform.json. Homo Sapiens mode did not have the same catastrophe.

Going to have to call @Rudometkin in here so he can explain the intent is an uncompromising survival game.  

[EDIT]

Quote

It's not like I can see the code, lol.

Unlike pretty much everything else out there, you can. Tyron has gone that extra mile and then some to be mod-friendly.

Edited August 16, 2025 by Thorfinn

[Krougal]

1 hour ago, Thorfinn said:

Unlike pretty much everything else out there, you can. Tyron has gone that extra mile and then some to be mod-friendly.

Not to mention it is all there on github.

 

[Thorfinn]

Just now, Krougal said:

Not to mention it is all there on github.

That's what I meant, though I see I revised it too much to be able to get that from what I said.