The real problem with the game right now is not what you might think

Thaulos · 2019-07-30 04:50:46

The current situation

The Rift right now is an absolute shitshow: dead bodies all around; settlements unable to sustain the player base and going through cycles of starvation over and over again.

Most of us want the Rift gone for good, me included, but one has to admit the idea has some merit. Well implemented it could result in some interesting lives and I actually look forward to living in a resource deprived world where we struggle to survive. A meta that would require us to recycle and be as sustainable as we possibly can, become part of the ecosystem instead of just consuming it. We are very far from such a rich game experience.

Jason current (misguided) focus

In an attempt to determine how long civilization inside should last Jason has been counting iron and oil inside the rift area. His logic seems to be X amount of Iron/Oil = Y amount of days. If that magic ratio is figured out, and a sufficient amount provided to the players, then gameplay-wise everything should be stable during the target time-frame. The situation will become increasingly difficult by the end but that's part of the greater game loop and to be expected. Everything solved, right?

Analysis of the situation

While resource such as iron or oil being depleted is indeed a major issue, they are actually having very little impact on the lives of average players. Chances are if you played inside the rift in the past week or so you never actually died due to lack of Iron or Oil. Anyone who played consistently will quickly witness the emergence of a very strong and obvious pattern.

This pattern isn't necessarily new and has some quite useful scapegoats. Ever blamed your village death on berry munchers? Why are the berries always languishing? why no one ever takes care of them? Why are so many people just hanging around doing nothing? It's very easy to blame random players.

While it is true that there will always be a percentage of players who will contribute nothing or very little, be it for being new to the game or just lazy roleplayers, they are not something new in the game. The game was fine before the Rift. Why would berry munchers suddenly ruin every village and cause multiple starvation cycles to occur? They aren't the cause at all. You can think of them as an environment hazard. They will always be there, but game mechanics are generous enough that it's easy enough to carry them.

So, what actually changed when the Rift appeared? Three things changed!
- Limited resources;
- Higher concentration of player;
- Less viable settlement locations

The actual problem

Last I played today our little settlement was constantly getting babies. We had no food and had so many baby bones laying around it looked like a meme. We were constantly being inundated with babies. If babies were being distributed evenly through the rift you can imagine this happening in every village, over and over again, hours upon hours.

Villages have a bootstrap cost (clay, adobe, milkweed for hatchet, bow drill) and an initial ongoing maintenance cost (natural soil, water from ponds, water from early wells, rabbits, wild food, milkweed for bow, arrows, rope, etc).

There are only so many people an early town can sustain. Villages can eventually grow to sustain many more people but it will take time and infrastructure. A town that sustains its population long enough will get better forges, better wells, better farms and better food production and all these will cause it's population capacity to increase. It can also use up some of it's resources to bootstrap new villages around it. Ever increasing the overall population capacity.

Spamming Eves inside the rift means that in theory all possible settlement locations will eventually get claimed in short order. Two hours of Eve Window should be more than enough for this. Different families or different branches of said families will develop those settlements, and then as resources dwindle trade or war might occur. People will adapt to lack of resources. Life will be more difficult in some areas, maybe even becoming impossible. Civilizations will implode.

However what is happening right now it's entirely different. We have too many people and too few settlement locations. Every single village will get overwhelmed. Most lives will be spent trying to get that extra berry and living that extra ingame-year. We will be forced born in some poor village to a hungry mom along side 10 other babies all begging for food. This would happen even if there were no griefers.

I reckon most players are fine with resource depletion or for long overreaching arcs. What players are not fine with is being thrust into a half baked hunger games ripoff without proper tools to succeed.

The now famous Hammer family was able to survive relatively well inside the Rift. They actually got to experience the game as it was envisioned by Jason (as far as I understand his vision). This however was only possible because most players were actually playing outside the Rift. In the 5 lives or so I played during the last Arc I was never born inside the Rift. The low amount of Hammers was probably ideal for their Rift settlements to sustain. The outside families essentially worked as a pressure safety release valve.

The solution

Depending on how long the Arc is to last (I'll assume more than a couple days), as we are right now, we should already have more than enough Oil but a shortage of Iron. This shortage could be easily solved by making it a bit more common and not that big of a deal.

However, the major issue is actually the lack of locations suitable for towns. While just making the map bigger might be easy and tempting, I don't actually think it would be a good thing. I believe that right now we already have a good balance of space inside the Rift to prosper, meet other families, trade or to wage war. Making the Rift much larger might turn the Rift more into an oddity than a driving force. I would even venture to say that if all major issues were solved one could even shrink the Rift a bit more.

So, how do we get more suitable settlement locations while keeping the same size? We can go at it in three ways:

1) We need a better and improved map generation algorithm. More concentration of green biome/swamp borders. (with guaranteed minimal availability of trees, soil, milkweed, wild food + ponds, clay, tule);
2) Remake green biomes and swamps. Possibly move clay and ponds to green biomes. This could turn every green biome into a potential settlement area (pending local available resources);
3) Change technology necessary to bootstrap town. We saw an attempt to incentivize this with the introduction of springs. Allowing water sources in multiple biomes could potentially allow more types of settlements. But as was stand right now no settlements can be created without soil, branches, milkweed, ponds and clay. Adding a spring on a mountain or savanna biome while keeping the resource requirements is not enough. Alternative ways of sustaining towns based on different resources could allow for a higher concentration of towns and therefore higher concentration of players.

While I'm not holding my breath on 3), I consider 1) or 2) to be of the utmost urgent necessity. I would go as far as ask Jason to put the Rift on hold while he work on any of these points.

The Rift can work and has potential but I don't believe the game is in a state that would allow us to properly experience it.

Final thoughts for Jason

Dear Jason,

While you might ridicule the idea of the game dying after it has been strong for so long, it is a real fear in the back of my mind. The game depends on casual players to sustain server population. "Veterans" will not play on an empty server. If the game isn't fun at all for casuals they will stop coming. Please don't confuse people stopping by to watch the aftermath of an accident with the accident itself being popular. People will eventually move on after they had their fill of misery.

Please remember that people play video games to relax and as a form of escapism, it is something we do on our free time. People don't play games because of "loyalty to the game". The game needs to be fun, at least at times, to keep people playing.

I completely understand your need to make the game even better. The "want" to reach that level of greatness of player experience that would be able retain tens of thousands of players. And I would definitely like to see it. Who wouldn't? The game as it is right now might need to change, possibly even drastically, and that's OK. But if the steps towards greatness aren't fun to play, players will begrudge and quit.

Could you possibly have Rift weeks and non Rift weeks? Or try out Rift from time to time and collect data on the experiment. And while you analyze and act on that data could you leave players to enjoy the already tried and tested rift-less gameplay we all enjoy? Rinse and repeat. The Rift could even become something players would look forward to. With time to plan, discuss ideas and provide feedback, but all while still being able to enjoy the game as a whole.

Last edited by Thaulos (2019-07-30 05:28:31)

jasonrohrer · 2019-07-30 06:47:44

This is a great post.

In doing a VOG survey, you are right that settlement locations are few and far between. And yes, that means you will eventually get down to just a few families getting hammered with all incoming babies (which quickly triggers the failure condition).

The other thing I noticed as VOG today is just how sparse the iron is (or even the gray biomes). With 6 biomes to spread around, there's just not that much of any one biome.

Again, this highlights problems that we've always had, but papered over with an infinite map.

Wandering around searching for a good spot isn't that interesting, after all. Wandering around searching for iron isn't either.

I hesitate to put everything in Green. You are supposed to travel to other biomes for stuff. It's impractical to do this for water, though... but maybe for clay, travel is fine? You really need a fixed amount to bootstrap. There could be a clay deposit at each spring head, and you have to dig it up first to get to the spring. That doesn't solve the pond issue, of course.

"redoing" the map gen algorithm to do something specific is pretty much impossible. It's not an algorithm, really, just an interpretation of a 2d random field as a map. So I can't tweak it to place this or that thing in this or that way. The thing is where it is, according to the math. I can adjust the probabilities for individual objects (and obviously control what biomes they are in). But that's where my control ends.

We could move away from a random map alg to something else. Like stripes or rings or something regular.

I could also add biome probabilities, which the alg currently doesn't have. Then snow could be rare. Right now, they all have equal probabilities. But I don't know that this will solve this specific problem too well.

I'm also somewhat interested in what some of the other recent changes (Eve window, too-few-fertile-mothers apocalypse) would feel like in a rift-free world. I imagine you'd keep going for days or weeks, though, which would be boring.

I hear you on not just making the rift bigger as a quick fix....

But that is a tempting quick fix!

I don't necessarily need people super close to prevent them from spreading indefinitely far away.

Dodge · 2019-07-30 07:53:37

"Wandering around searching for a good spot isn't that interesting, after all. Wandering around searching for iron isn't either."

You can get killed by wolves looking for iron, boars looking for clay, and looking for a spot is dangerous too.

So i wouldn't say it's not interesting.

I guess the looking for a good spot part is the one not that interesting.

You could spawn as an Eve directly in a perfect place, Green biome on a spring and adding some ponds and clay or even make it so swamp always is next to green biome.

It would make sense to have a biome order, why would a desert be next to a thundra?

If you change the biome generation probability and make a biome order where the swamp is always next to the green, it solves the issue of lack of locations suitable for towns.

Green and swamp would be more frequent than mountains and prairies then there could be jungle, desert and thundra or something like that.

With this you wouldn't even need to add ressources to biomes, since the swamp would always be next to green and that's the two required biome to start, then prairie and mountains and then the other biomes.

How frequent the biomes spawn would be proportionnal to how vital they are, at least in the beginning.

So all the green spots would be good spots, or at least good enough for a village.

But then Eve's would spawn directly in the perfect place, which would make every Eve successful, not very challenging.

So maybe Eve's would spawn randomly on the map and not necessarly on a spring in the green biome.

Each biome would have a default probability to spawn and another probability to spawn next to another biome.

So for example next to a desert the chance of a thundra spawning would be 0%.

Twisted · 2019-07-30 07:55:51

Great post, Thaulos. It's like you're reading my mind and putting my thoughts into words.

Random idea that just popped into my head: What if it rained during the Eve window, automatically watering any planted seeds every ten seconds? This, along with maybe spreading out soil deposits across the map, would make every well site a viable starting location. It would also make the Eve window visually distinct and it would feel like a big deal when you spawn into a world of rain. After the Eve window is over the rain is gone until the end of the arc, and if your family worked hard enough you should have a well and a farm up and running, ready to supply your family while you drain the surrounding environment of resources.

Morti · 2019-07-30 08:38:47

jasonrohrer wrote:

Wandering around searching for a good spot isn't that interesting, after all. Wandering around searching for iron isn't either.

Please stop doing this.

Stop assuming what people find interesting.

Stop speaking about subjective experiences objectively.

You are not Sam Harris, you do not have a graduates degree in Philosophy or a PhD in Neuroscience, and this is not the Moral Landscape.

Stop assuming you know what people find interesting or not.

You are one game designer, and we are a diverse variety of players with a range of interests.

Many of us DO enjoy scouting the landscape for potentially nice homes based on the biomes and resources.

And many DO enjoy the Easter Egg Hunt that is scouring the landscape for surface iron and hunting down iron veins, because we know how important it is from experience. Experience you don't have.

You. Are. Wrong.

Stop making changes based on these false assumptions.

Here is an exercise for you, repeat after me.
"I don't find that interesting."
"Some people might not find that interesting."
"I don't know what other people find interesting."
"I'm curious, does anyone find x interesting? Because I'm thinking about making a change to the game and I wouldn't want to sound like an enormous jackass, trying to justify that change by making blanket assertions about what you may or may not enjoy about it."

I'm starting to think that you are not the person that came up with the idea for this game.
Something has changed, your vision has changed, your ideals have changed, maybe your family life has changed, significantly, enough that you're losing sight of what you once set eyes on...

I'm guessing it's us. All the people. All the opinions.

Look, I don't mind any changes you make to this game. At least, not as much as so many other people do.
I wish you were still the person I thought you were when you sold me this game.
I don't need a Time Machine for that, I can remember.
We can all, remember.

Can you?

Do you?

BladeWoods · 2019-07-30 14:13:14

jasonrohrer wrote:

This is a great post.
In doing a VOG survey, you are right that settlement locations are few and far between. And yes, that means you will eventually get down to just a few families getting hammered with all incoming babies (which quickly triggers the failure condition).
The other thing I noticed as VOG today is just how sparse the iron is (or even the gray biomes). With 6 biomes to spread around, there's just not that much of any one biome.
Again, this highlights problems that we've always had, but papered over with an infinite map.
Wandering around searching for a good spot isn't that interesting, after all. Wandering around searching for iron isn't either.
I hesitate to put everything in Green. You are supposed to travel to other biomes for stuff. It's impractical to do this for water, though... but maybe for clay, travel is fine? You really need a fixed amount to bootstrap. There could be a clay deposit at each spring head, and you have to dig it up first to get to the spring. That doesn't solve the pond issue, of course.
"redoing" the map gen algorithm to do something specific is pretty much impossible. It's not an algorithm, really, just an interpretation of a 2d random field as a map. So I can't tweak it to place this or that thing in this or that way. The thing is where it is, according to the math. I can adjust the probabilities for individual objects (and obviously control what biomes they are in). But that's where my control ends.
We could move away from a random map alg to something else. Like stripes or rings or something regular.
I could also add biome probabilities, which the alg currently doesn't have. Then snow could be rare. Right now, they all have equal probabilities. But I don't know that this will solve this specific problem too well.

I'm also somewhat interested in what some of the other recent changes (Eve window, too-few-fertile-mothers apocalypse) would feel like in a rift-free world. I imagine you'd keep going for days or weeks, though, which would be boring.

I hear you on not just making the rift bigger as a quick fix....
But that is a tempting quick fix!
I don't necessarily need people super close to prevent them from spreading indefinitely far away.

*cough* There's 7 biomes not 6.

I find searching for a good spot to start a civilization to be interesting. It's not a simple thing to do or a simple decision to make where to setup a camp that you want to be in a viable spot to develop into an advanced town. It's the only part of the game where you have a nomadic lifestyle too. Afterwards most people settle down and don't travel too much.

I think biome probabilites would be a great improvement. There's a lot of 'wasted' space in the snow, desert, and jungle biomes. It's another issue that wasn't an issue until we got a finite play area.

miskas · 2019-07-30 14:52:37

One long stripe of Green biome across the rift with random little squares of biomes bordering it, will make it easier to spot the right spot for a town. It will also improve communications between towns, migration and maybe trade.

Last edited by miskas (2019-07-30 14:53:05)

jasonrohrer · 2019-07-30 15:55:17

Whoops 7 biomes, yeah! That's a lot, and that's part of the problem. It will only get worse as more are added.

A lot of good ideas here. Rain is interesting, though I hesitate to add a whole new system before trying to tweak the existing stuff more.

You don't need Jungle or Desert or Arctic until late stage.

You need green, yellow, swamp, and grey to bootstrap. Those were the original 4 biomes when the game shipped. But they've been "watered down" by the other ones. Currently, they are close to half as common as they once were.

You need green and swamp immediately for sure, and maybe yellow (for carrot seeds and rabbit fur to get a forge going). Then after that, you need gray.

So, maybe biome probabilities would help.

Or biome classes. So we roll 1-5, and place green, yellow, swamp, grey on 1-4. If we roll 5, we pick ONE of Jung, Desert, Arctic.

Unforunately, a proc-genned map isn't as simple as that. You don't just roll numbers and place patches. You have an oracle that you can ask a question for a single tile: what biome is at (x,y)?

Someone suggested biomes being "near" each other always. I'm not sure exactly what that means (if swamp is always next to green, and vice versa, then there'd be nothing but swamp and green).

There could be an altitude ordering, though. Altitude as a virtual concept, of course (it's all flat), but the random 2d field could specify an altitude (it kinda does already), and then that could pick the biome based on some ordering.

Something like:

desert, jungle, swamp, green, prairie, mountain, snow.

So these things would occur in the map in topo rings instead of patches. Every topo ring of green would have swamp around it and prairie in the center of it. But it would still be proc-genned and not tightly planned. There wouldn't necessarily be jungle and desert between each green "hill". Some of the "valleys" would be deeper than others. Some "peaks" would not be high enough to have snow.

Current code takes the 2d-field altitude and maps it differently, to avoid rings, and achieve truly variable and independent patches instead. I can't remember what it's doing exactly, but something clever. It was like 3 years ago that I wrote it. Topo rings is very easy by comparison. It's sort of the default way for such an alg to work.

Anyway, you could imagine "one village per hill" if this was the case. Almost every hill would be viable.

HOWEVER, we still have a huge problem in actually testing this stuff, because people are escaping every time.

I guess they didn't escape during run1, (301 years), so we have at least that data point. I.e., it sucked!

Dodge · 2019-07-30 16:42:45

First a green tile is placed, if it's the first one then the next is 100% green until there is at least 50 green tiles placed then it's 40% green 30% swamp 10% prairie 10% mountain and the rest desert,jungle or thundra

When a swamp tile is placed if it's the first then the next is 100% swamp until there is atleast 50 swamp tiles then it's it's 40% swamp 20% prairie 20% mountain and the rest desert,jungle,thundra or another green

But when a thundra tile is placed then it's 100% thundra for 20 tiles then 40% thundra but 0% desert and 0% jungle and the rest at choice

The 50 tiles and 20 tiles are just an example i dont know how big it looks but basically it makes sure the biome is at least a certain size and not just 1 tile big

It would still be random but green would have a very high chance to have swamp next to it.

And there would be no desert or jungle next to thundra.

Not sure if it work in terms of coding with the engine though.

The percent could even get lower as the tiles are placed to not have huge green or swamp biomes or others

So if 200 green tiles are placed then the next would only have a 10% to be green for example.

Peremptive · 2019-07-30 17:34:46

I don't understand how someone makes game changes to force a behaviour on players. There is no perfect outcome. I remember a few months ago we had a huge town going. For days it kept growing, with two bell towers, great farms and bakeries. Roads leading out of it at all directions, people building houses and adding buildings for the fun of it (it was a short time before the temp update). I wish we could see again something like that. It felt like Rome or something. I remember walking 2-3k and dodging so many dangers to get to it, then spending a few lives expanding it and fighting griefers.

I don't understand why there should be a boundary limit. If I want to walk to the east for 50 minutes why should I be blocked?

The_Anabaptist · 2019-07-30 17:38:48

In regards to crowding in the Eve window:

I had a recent rift eve life where I found a good spot, got the starter tools, setup a kiln and started firing clay. Had another Eve wander in, who liked what they saw. If you said something to them, they would repeat it. If you fetched kindling, they would immediately make the biggest fire they could with it. They ate whatever was within screen view. I don't know if they thought they were helping, but they certainly were not. I know that we had a couple additional Eve neighbors in the greater area, from what my kids reported to me. None of them directly visited the kiln area, and certainly none of them wholesale moved in like this.

Now, do I drop everything, use valuable time and resources to make a bow and arrow and murder the interloper? Every time they burned my spare kindling, so help me, I wanted to. This is part of the reason I recommended adding rocks as a weapon. A well placed rock or two might have given them the hint that they weren't welcome anymore without outright killing them.

As a result, I do think the current map dimensions are too small. And that Eves need to be separated more, so that what occurred here is the rare exception and not the rule.

The_Anabaptist

jasonrohrer · 2019-07-30 18:09:11

Dodge, I don't think you understand. It's an infinite map. Nothing is "placed" anywhere. There is no database of natural map tiles (how could there be?)

There is simply a question: what is at this particular (x,y) location.

That question has to be answerable for each (x,y) in isolation. It can't depend on neighboring tiles, etc. (or you have infinite regress.... if answering (x,y) requires knowing (x-1, y), how do you compute (x-1, y) without computing (x-2, y))

There are many ways to do this... many ways to answer that question in isolation.

One way is a grid. So you take ((x/100) mod 7) and use that to pick the biome. Then you have vertical stripes 100 tiles wide. You can even "jiggle" x a bit with some noise, so like ((noisifiy(x)/100) mod 7). Then you have vertical stripes with bumpy edges.

What I'm doing is using (x,y) as seeds for a smooth 2D random fractal field (1/f noise, or Perlin noise, or whatever you want to call it). You can call fractalNoise(x,y) and get a value out, and that value is related to neighboring values due to properties of the algorithm itself. Fractal noise is like a semi-smooth mountain range (altitude values, really). It's in fact used all the time to generate 3D terrain for games.

If you just map that altitude value to biomes, you will get rings.

I'm doing something different, which is my "clever" innovation:

I'm generating a SEPARATE mountain range for each biome, and then "crossing" them, and figuring out which one is the "tallest" in a given area.

So imagine you've got seven 3d mountain ranges of different colored plastic. Then you superimpose them, and look down from the top. You can ask, "Which color plastic is visible in each spot"?

So they are really separate "hills" of each color poking up above the other hills. If you're in a green biome, that's a spot where the green hill was the tallest hill.

Dodge · 2019-07-30 19:03:26

Couldn't these mountains have some sort of logical order and size?

The thundra,desert and jungle mountain would be smaller than the green,swamp,prairie,gray.

There is always four sides North,East,South,West

So if a thundra mountain is created the four other mountains cant be desert or jungle.

It would start with the original tile the 0,0 this one would be either always green or randomly selected.

Then 100 tiles (or something) to the east another tile is generated (a new mountain), this tile would ask what is the other mountain tiles to the west,east,north and south if there are already generated then there is a logic if none is generated then it could be any biome.

So it would be like a grid, every 100 tiles (or 200 or more) a new biome tile (mountain) is created but it would always ask what are the 4 other tiles in every direction (what is the tile 100 north from here, 100 east, 100 south and 100 west)

Depending on the biome it would have a different size so it wouldnt look like an actual grid which wouldn't look good

jasonrohrer · 2019-07-30 19:26:27

Okay, here's the existing map for the last arc:

And here's my first pass at a topographical version of this:

It feels okay to walk around in, because you know your way around. If you want to get to the jungle from green, you walk across the swamp. Swamp is always next to green, so there are "good spots" everywhere.

Obviously, the scale would need to be adjusted, and maybe per-biome scale (or have them spaced using some kind of curve, instead of linearly, with altitude). We can tweak various things to affect the "spread" and make deserts bigger and snowcaps smaller, etc.

Right now, green strips are really skinny, so scale changes would fix that.

arkajalka · 2019-07-30 19:29:28

oh man that looks so sweet compared to what we have now.

jasonrohrer · 2019-07-30 19:32:08

Note that this does look a lot more like real land!

However, I tried this three years ago and determined that it wasn't interesting enough. The whole map feels very same-ish, and your never very far away from anything. The patchy version has a much more "who knows?" feeling to it.

Maybe I could adjust this further to have large scale gradations on top, so that there can be some larger-scale features (like a big desert swath, or whatever).

Spoonwood · 2019-07-30 19:40:29

Morti wrote:

jasonrohrer wrote:
Wandering around searching for a good spot isn't that interesting, after all. Wandering around searching for iron isn't either.
Please stop doing this.
Stop assuming what people find interesting.
Stop speaking about subjective experiences objectively.
You are not Sam Harris, you do not have a graduates degree in Philosophy or a PhD in Neuroscience, and this is not the Moral Landscape.
Stop assuming you know what people find interesting or not.
You are one game designer, and we are a diverse variety of players with a range of interests.
Many of us DO enjoy scouting the landscape for potentially nice homes based on the biomes and resources.
And many DO enjoy the Easter Egg Hunt that is scouring the landscape for surface iron and hunting down iron veins, because we know how important it is from experience. Experience you don't have.
You. Are. Wrong.
Stop making changes based on these false assumptions.
Here is an exercise for you, repeat after me.
"I don't find that interesting."
"Some people might not find that interesting."
"I don't know what other people find interesting."
"I'm curious, does anyone find x interesting? Because I'm thinking about making a change to the game and I wouldn't want to sound like an enormous jackass, trying to justify that change by making blanket assertions about what you may or may not enjoy about it."
I'm starting to think that you are not the person that came up with the idea for this game.
Something has changed, your vision has changed, your ideals have changed, maybe your family life has changed, significantly, enough that you're losing sight of what you once set eyes on...
I'm guessing it's us. All the people. All the opinions.
Look, I don't mind any changes you make to this game. At least, not as much as so many other people do.
I wish you were still the person I thought you were when you sold me this game.
I don't need a Time Machine for that, I can remember.
We can all, remember.
Can you?
https://i.imgur.com/tRrWUPl.png
Do you?
https://i.imgur.com/hjsHjnh.jpg

This is a great comment. I know I've more-or-less tried to tell Jason the same thing (though not the part about hundreds of craftable objects... though that is true) before and he didn't care and kept on telling people what is interesting as if he were some sort of supernatural being capable of knowing what is interesting for all people. His hubris knows no bounds.

My personal recommendation is that people move on with their lives and play some other game or do something else with their time. That one of the longest time players has encourage people to use his account (instead of purchasing the game) should also make for a clue as that doing something else with your time is probably better than playing this game.

arkajalka · 2019-07-30 19:46:28

Maybe you can adjust the resourse drops in some way, so they spawn in some kinda clusters and that way would be far from each other. That map just looks so good. I can imagine you could actually learn your way trough the surroundins as in now its just big plain areas and you just follow the edges to even have some clue of your surroundings. Actually you are moving between biome lines (atleast thats how i travel) even now cause the midle parts of the biomes are unrecodnizeable and you get easily lost without homemark.

Last edited by arkajalka (2019-07-30 19:47:26)

jasonrohrer · 2019-07-30 20:14:35

Resources already spawn in clusters that are separate from the biomes.

There are "dense" areas and "sparse" areas as a separate overlay, regarless of what biome falls there. So, for example, you'll notice that a dense swamp area (lots of cypress trees) will continue when you cross into the prairie (will be dense with juniper trees).

So it works like this:

1. Decide whether some object is here at all, based on density function.
2. If so, ask what biome is here, which will give you a subset of possible objects to pick from
3. Turn that object subset into a probability distribution (based on their map likelihoods)
4. Pick an object from that prob distribution and return it.

Dodge · 2019-07-30 20:16:42

Spoonwood wrote:

My personal recommendation is that people move on with their lives and play some other game or do something else with their time.

Why you dont follow your own advice and leave?

Yet you're still here posting the same generic boring shit that adds no value to the discussion.

We get it you dont like the warsword but instead of complaining all the time why you dont ask yourself why you dont like it, what makes it bad in your opinion and think of a way to make it better.

Like for example you dont like it because it's made to kill other lineages/races instead of "enemies", so why not make the warsword kill people based on the number of words you know from the other family, if different lineages spend enough generations together they learn each others languages and are at peace.

So multiple lineages could live in the same village peacefuly without fear to be killed by the sword.

But NO it's better to whine and make personnal attacks all the time.

Amon · 2019-07-30 20:34:59

Jason! The new map gen looks pretty. And yes, I think sprinkling in some pattern breaking, overlaying biomes would do good in adding some extra diversity to keep us guessing.

Thaulos · 2019-07-30 20:51:22

Wow the map looks awesome!

I'm not great on map generation but if you are overlaying noise maps maybe you could somehow overlay 2 different maps?

Something like this:
- humidity: desert => savanna => green => swamp => jungle
- height: [inherited] => mountain => snow

If it is possible, you could potentially have a greater ability to define biome size ratios depending on where you set the thresholds for each biome.

Last edited by Thaulos (2019-07-30 20:58:34)

jasonrohrer · 2019-07-30 21:03:34

Yes, Thaulos, that's interesting.

Here's a little counter-intuitive quiz for you all, though.

Look at this map:

What is the most common color on it? What is the least common color on it?

SirCaio · 2019-07-30 21:06:42

So neat! Please go forward with this idea, the map looks great like this! (Although penguins and seals on top of mountains does feel a little odd)

jasonrohrer · 2019-07-30 21:08:09

Spoiler answer (highlight to see):

Grass = 86,307
Yellow = 81,053
Swamp = 76,696
Snow = 76,103
Mountain = 62,906
Jungle = 62,576
Desert = 55,623

One Hour One Life Forums

#1 2019-07-30 04:50:46