One Hour One Life Forums

I'm an experienced game designer.

But this is one of the most complicated games anyone has ever attempted to design.

I get the feeling that balance will ultimately be impossible, or at least that is a kind of latent fear of mine, as more and more objects are added.

There was always a fundamental problem that I've been struggling with for years:  how to handle resource distribution and regeneration in a world where you have no idea how many people will be trying to use those resources in a given area?  If it's just right for 50 people, will it be too fat for 1 person and too lean for 100?

A few years ago, this game actually had an infinite berry bush as a wild food source.  At one point, something like that seemed necessary.  Won't people starve otherwise?  A game like Don't Starve has a single player operating in the world, so the berry bushes can be finite and a good distribution can be picked by the designers.  But when you have 1, 10, or even 100 people playing, and you're not sure how many, how many berry bushes should there be?

The game was horrible with infinite berry bushes, and got better when I made them finite.  I balanced the wild food in the game for one person, and just kinda punted the issue.

I'm not sure, in retrospect, why I was even worried about it, given that all babies come through mothers, so populations are naturally limited.  I think back in the day, close-radius Eve placement was also a confounding factor.

But I still wasn't sure if this problem was solved until after launch.  I couldn't test what would happen with 100s of players until I had 100s of players.

My original design for the tech tree and how stuff would be added was a kind of hand-wavy content-based approach.  People will craft stuff just to craft it, the tech tree will be explored because it's there, etc.  Just keep adding stuff, people love content, who cares?

I always had some misgivings about this approach.  Pointless content is... pointless... and it's not the way I've ever designed a game in the past.

And players quickly echoed this sentiment after the game launched.  No one made pies---why bother, if carrots are infinite?

I haven't seen a dyed hat in the game in a long time...

So then my misgivings came home to roost, and the looming WHY question came sharply into the foreground.  There will be a bunch of pointless content in this game (dyed hats, painted walls, etc.).  But some of the other content, the spine content, need a solid why associated with it.

I think this game has a chance to actually be a good game, as opposed to just the heap of content that I promised.

My original design for the content of the game broke everything humans ever make down into three categories:

1.  Food
2.  Shelter
3.  Entertainment

Can you eat it?  Can it keep you warm?

If not, then you're just making it to make it.  Hula hoops and monoliths fireworks and red shirts.

Items in 3 don't need a reason.

But if an item's primary purpose is 1 or 2, it needs a good reason, or else you're not going to make it.  I don't want deep wells falling into category 3.

My dear sister, you and I represent our tribe's only hope!

Blue Lagoon, here we come.

Actually, I don't think incest was my main concern when I avoided simulation of mating.  I just didn't want this to be a game about consent.  Male characters chasing female characters around, cartoon boners poking out from under loincloths, etc.

In the editor, an attacking shot wolf decays to a shot wolf in 1 second.  That content update was pushed out many weeks ago.

This has been fixed.  The fix will go live in the update later this week.

Screen shot?

Okay, fixed this in source.  The fix will go live later this week.

Just don't walk north 1000 pixels from where you were born and click there....  there's a hidden text field up there that is the underlying mechanism for chat, and if you click in exactly the right spot, it comes into focus.

You folks who were having it happen over and over were in exactly the right spot...

Yeah, I am trying to work out core principles here that everything will abide by as I add more stuff.

What are the rules by which I add higher tech?

For example, an obvious next step is some kind of pump well, maybe a hand pump, and then maybe a wind pump, eventually a motor pump.

How do these things work, relative to the existing wells and ponds?

I need some kind of design rule that ties everything together.  Worst case, I add all these things, and after a few days of role-playing pump-builder, players go back to just using ponds again.

"Everything runs out" is one design rule that works here.  If the ponds simply don't refill, players must make wells or die.  And if the wells don't refill, players must make pumps or die.  The pump wells go dry, players must make wind-pumps or die.  Same with motor pumps.  Then the petrol runs dry, and players must make solar motors or die.

"Everything runs out" works even if some things really don't run out, but simply are produced too slowly.  Wild berry bushes are a great example, since they're currently tuned so that you cannot survive in one place on them for long.  Farm or snare or die.  So ponds can use the same technique.  If you depend on them alone, you won't produce water fast enough.

So, that's great for natural resources.  Some run out entirely (iron ore), and others are produced too slowly to sustain you.

But how do man-made resources work?

The problem is that there's no limit on how many of these can be built in one area.  If wells produce X water per hour, then 10 wells produce 10X water per hour.  It would be hard to make the pump well 10X faster.  Even 2X faster is a lot.  So you could build 1 pump well or 2 regular wells, and if regular wells are easier to build, which they have to be...

If regular wells last forever, then you will probably have enough of them around from previous generations that you never need to build pump wells.

So, maybe infinite, with rate increases for higher tech, doesn't work.

One of my local design friends thinks this is the way tech should work.  The ponds in an area cannot sustain more than 1 person.  Build a well, and you can sustain 2 people.  Build 2 wells, and you can sustain 4 people,  Upgrade a well to a pump well, and you can sustain 8 people, etc.

The problem is that, in practice, anything more than about 8 people in a village is superfluous to the village's long-term survival.  Maybe that will change as things get more complicated, and more roles are needed.  But it doesn't seem like "sustaining a bigger and bigger population" is really anyone's core goal.

Furthermore, we can't control even how many ponds you have access to, let alone how many wells you build.  So you might luck out and find yourself in a natural area that can sustain 8 people right from the start.  Or an area with a bunch of loose stones, so you can easily build 3 regular wells and be set forever.

What about finite with increasing capacity?  While the natural things regrow over time, albeit slowly, maybe the man-made things do not?

That is the current difference between a carrot farm and a wild berry bush.  The berry bush produces food slowly forever, and you have no control over how many you have access to.  The carrot farm produces 5 carrots per row, but that's it, and you can have as many rows as you want.  After a row is gone, you have to build a new one, which slowly consumes resources over time.

So, this is the version of the well that is like a tank with 50 water in it.  Once it's empty, it's empty.  Build a new one, or upgrade it to get 100 water.  A cistern becomes useless (because a well is essentially a cistern that starts full), though a cistern is still useful for ponds.

The question with higher and higher tech shifts from "how many people can you sustain" to "how long can you last?"  This gives us a lot of design flexibility, because we can tweak these numbers a lot without breaking anything.  We can easily give some higher tech 100x the capacity of lower tech, which just means it lasts 100x as long.  (On the other hand, if we were to give something 100x the production rate, that would support 100x the population, which would effectively mean the resource is unbounded.)

We can also be much more generous with natural sources, if they run out in the same way.  If a pond produces X water per hour, our hands are tied, and we have to keep X low, so that higher water tech is worth building.  But if a pond simply produces X water total, then X can be whatever we want it to be.  No matter how high we make X, it will eventually run out, and people will need to move on to wells.

For carrots, this would mean that putting fertilizer on your carrot row would make the same row grow 10 carrots instead of 5, and using a deep plow first would make the same row produce 20.

The problem is that it kinda feels weird for some resources, and runs contrary to real life expectations.  Real wells don't have fixed capacity like this.  It does make some sense for the highest tech, because a petrol pump well would eventually run out of fuel and need refilling.

What about temporarily infinite with increasing timespan?

A well could produce infinite water for 5 minutes, and then go dry.  A pump well could produce infinite water for 10 minutes.  A wind pump for 20 minutes, etc.

During that time, there would be a frenzy of water use, harvest, and storage.  Cisterns would be crucial.  Then the well would go dry, and there would be drought.  Perhaps the well could run on a cycle, 5 minutes water, 30 minutes dry, 5 minutes water.  Perhaps the higher tech could make the dry time shorter and shorter.

For carrots, during a brief harvest window, there would be infinite carrots to pick and store.

This would also seem to eliminate the benefit of more than one of any resource-producer.  Storage would be king.  Though building wells at different times might give you infinite water all the time, as the well cycles would form phase patterns.

This "feast or famine" model is pretty out-there, but it's an interesting corner of the design space to think about.

Then, there are hybrid approaches.   Fixed capacity, with some refill time in between.  A well with 50 water.  Once drained, it takes 30 minutes to recharge.  Then you can take another 50 water.  Better wells can have greater capacity and/or shorter recharge times.

This is currently how domestic berry bushes work, along with some maintenance.

Another hybrid approach is to have everything recharge VERY slowly, pretty much at the same rate regardless of tech, but modulate starting capacity for higher and higher tech.  This again gives us flexibility with ponds, because we can give them higher starting capacity.  Essentially, this could be tweaked so that, no matter how many wells you have, you always need to build more eventually, but wells never become totally dry and useless.

Also, I'm pretty sure this was due to the server1 disk space issue, where map tiles were becoming corrupted.  The container wasn't there, but the contained items still were, or something...

Okay, I think this has been fixed now.

Sorry for the trouble!

Until I get some better client differentiation in place, I've disabled saving these logs server-side.

I'll post a new bounty thread when the better client code goes live.

Yeah, I've thought about this.

The problem is that the client currently outputs all the relevant messages to stdout and doesn't save them in RAM at all.  On Windows, this is conveniently saved in stdout.txt, so people can email it to me.  I don't think Windows would let me open this file for reading if it was still open for writing.

Farm milkweed?

Knitted clothes require no thread, right?