Update: Live from PAX West
August 30, 2019
I write to you from rainy Seattle.
If you're at PAX West, please stop by and say hello. I'm in the PAX Rising area, and I've got:
--500 Limited Edition OHOL purchase cards with unique QR codes (on sale for the first time at 25% off for PAX attendees only)
--Mofobert stickers
--Sandara posters
--Diamond Trust of London
The update went out on Wednesday evening, before I left, and it includes several improvements, which I will simply list here:
--Eve window remains open for 8 hours.
--Arc ends if we ever get down below two families.
--The bug causing the apocalypse whiteout to stall has been fixed, so families can easily survive the end of the arc now.
--Fixed a imprecision accumulation in server walk speed code that could potentially allow WASD mods to walk slightly faster.
--You can spend YUM to do hungry work, and you must always have a 5-hunger buffer when doing hungry work (for safety, so you don't starve immediately after).
--After the Eve window closes, babies are distributed to families round-robin, to prevent one family from getting baby starved.
--A new family population log has been added, so I can make nice graphs of families over the arc.
--Pine trees are no longer hungry work (softwood, naturally)
--You can eat carnitas straight.
--You can eat onions and tomatoes straight.
--You can deconstruct the track cart kit.
| Update: What We've Learned
August 24, 2019
On the heels of my return from vacation, there will be no gif this week, but there's still an update.
During my vacation, as many of you might recall, the arc wore on an on and on for something like 8 days, and none of the end conditions were triggered. Meanwhile, the space inside the rift became an over-griefed hell-scape. I had to kill that arc manually from vacation, and disabled the rift barrier temporarily until I could return home and tweak things.
So, this week, we take what we've learned from the arc runs so far and tweak a bunch of stuff that needs tweaking. Keep in mind that this game has been live for 18 months at this point, and we just learned about new problems that never mattered too much before. Bears have been in the game for 18 months, and it never mattered that their caves respawned new bears every 24 hours. I didn't even recall that it worked that way. But inside the rift for 8 days, this suddenly mattered. There were hundreds of roaming bears by the end.
The point here is that an infinite map kinda "solves all problems" magically. If something is broken, you can always walk away from it. If something runs out, you can always find more of it further out. I can't even really answer a question like, "Is there too much oil in the game?" How much oil is in the game? An infinite amount. What if I cut the oil spawning rate in half? Then there's still an infinite amount. We have to talk about resource allocation in terms of what's within a reasonable walking distance from a given town. But towns can move and spread over time, so any distance becomes reasonable, given enough time.
I can find answers to questions about a finite map, however. I can cut the oil spawn rate in half, and there will be half as much oil.
The biggest thing we're testing this coming week is a new arc failure condition: when we get down to less than 5 surviving families, the apocalypse happens and a new arc begin. In conjunction with this failure condition, new Eves only spawn for the first 2 hours of the arc. After two hours, the families are fixed, and the arc will last as long as at least 5 of those families last.
The great thing about this failure condition is that it puts the arc end squarely in the hands of the players. Do you want the arc to keep going a bit longer? Then help that struggling family stay afloat.
The other interesting thing is that the arc end does not kill anyone. Every living player survives, which means family lines can outlive the arc. In fact, with these changes, there's nothing stopping a family from living forever.
We're also testing a new curse system. The old, global-tally Donkey Town has been removed and replaced with a personal cursing system that lets you determine, unilaterally, who you no longer want to play the game near. Once you curse someone (you get one token per hour), that person will not be born within 50 tiles of you for the next 48 hours. If someone is cursed by so many people that there's nowhere left for them to get born, then they do go to Donkey Town, but there's no longer a fixed-length sentence there.
Finally, a bunch of little issues that came up during the last long arc have been fixed:
--The default state between families is neutral instead of war. To use war swords, an elder must declare WAR in the target family's language. Elders cannot hold swords themselves, though, blocking solo crusading.
--Cross-family curses can now work without saying it in their language.
--The shortcuts CURSE YOU (closest player) and CURSE MY BABY have been added.
--Babies now keep their family name even if adopted.
--A bunch of things that were blocking and unremoveable (Newcommen towers and looms) have been made deconstructable.
--Chopping all trees, including juniper, is now hungry work.
--Fences now orient themselves relative to other fences automatically. No more ugly fences.
--Wildcard phrases for gate ownership, like MY OFFSPRING OWN THIS (all living children) and MY FAMILY OWNS THIS (all living family).
There will be one more update next week, and then I'm off to show the game at PAX West.
| Vacation Update: A Reprieve from the Rift
August 16, 2019
I'm still out of town on my family vacation, but it was clear that life inside the rift had gone completely haywire in my absence, mostly due to newly-discovered forms of griefing that I cannot fix until I get back. We have a lot of great data an ideas to work with, though, and the path ahead is clear.
In the mean time, so I can enjoy the rest of my vacation worry-free, the Rift is disabled, and the ever-growing Eve spiral is back in action, with no forgotten-area culling.
Thanks for hanging in there during these rocky times, and I'm looking forward to getting back in the saddle next week.
| Update: Whole New World
August 3, 2019
What...a...week!
Thank you all for bearing with me as I continued to adjust and perfect what is now the most massive set of fundamental changes in the history of the game. Like I said in the last update, we got off to a very rocky start, but by the end of this week, it was almost completely smooth sailing.
Containing all player activity in a bounded area, instead of letting it spread infinitely on the map, revealed all sorts of problems, particularly in terms of resource distribution, settlement locations, and griefing. On an infinite map, there are always more resources available if you walk far enough in any direction, which is what people used to do to find settlement locations, and you can also easily hide from trouble-makers in the vastness. A bunch of important features in the game, like fences, were rendered unnecessary by the unbounded map.
So, first up, let's take another look at that now-ancient map generation algorithm. It placed biomes independently in patches, and there was no structure to that placement. That meant that the biomes that were useful together, like swamps and grass (a prime settlement location) almost never spawned next to each other. Finding a prime spot used to require a very long walk. This also meant that jungles could border the arctic areas. The independent placement resulted in a lot of map variety, but there were obviously some trade-offs.
The new algorithm uses a more naturalistic topographic layout, with biomes in altitude rings. This means that each biome always borders the same other two biomes. Swamps always border grass, for example. Now prime settlement locations are all over the place. I also added per-biome likelihood controls, so the really-necessary biomes can be more common---they have wider topographic bands. Finally, I classified three biomes as "special": arctic, desert, and jungle. These aren't needed quite as much as the others--they're only needed for advanced tech---and it's more interesting if they are far-flung on the map. They don't occur in regular topographic rings, but instead at the centers of each topographic peak.
This one change resulted in a dramatic improvement in the survival rates of settlements on the map. Suddenly, the bounded arc area became quite livable. Even better, the old long walks to find a settlement location were gone.
Living close together highlighted a bunch of new problems. Families often live in the same village for generations, yet are still logically separate due to war swords and inability to curse each other. After many generations together, they might even speak a common language. At that point, they really are one village.
Now, if you can curse someone in a language they understand, it will work, whether or not they are in your family. And elders from two families can declare PEACE to each other, as long as it's in an understood language, thus disabling the war swords. They can also declare WAR to each other again later if need be. Note that both these features also work before you learn a common language across generations if you do the work to actually type the other family's language.
Now that people live near each other, fences are everywhere. This is good. Towns are more interesting with fences. However, rogue fences can also be a problem. The idea with fences is that they homestead unclaimed land with a waiting period to ensure local consensus. But what about out in the wilderness? In an infinite map, it's all unclaimed land, but in a finite map, it might need to be used by someone in the future. A fence bisecting a large wilderness area is a real problem. The 2-hour decay period for an abandoned fence is too long in this context. So, I've given you a way to remove a fence, with the help of an elder and a brief waiting period. The idea here is that you'll only be able to remove abandoned fences, because of the waiting period. If someone cares about the fence, they will intervene and cancel your removal notice.
And regarding resources, the only non-renewable so far that has been a real problem has been iron. So I've given you a high tech way to produce more iron by burning oil. Iron never runs out now, but oil is finite, so there still is an eventual limit. Those diesel mining outposts are extremely valuable, both in terms of production and the expensive capital improvements that are installed there.
The goal in all of this is to enable a collective challenge: How long can you all survive together before civilization collapses globally? The most recent record was 44 hours. But there's enough oil on the map to support farming for 100 people for at least ten days. I'm guessing that 44 hours is just the beginning, and you'll all be gradually getting the hang of it over time.
It took a while to come together, but this really does feel like a wholly new and improved game. There's something going on at any moment in this world now. There's a story to tell.
Welcome to IMPULSE FOLLY.
And with that I'm off on a two-week vacation with my family. There will be no updates for the next two weeks.
| Update: Grand Arc
July 27, 2019
We got off to a rocky start with this week's update. It was hard to predict exactly what would happen when going from an infinite map with infinite resources to a limited area with limited resources, but in the first few runs, all hell broke loose.
But before I get into that, I should explain the goal of this update: To give the world in the game an arc beyond the arc of individual villages. When you join the game tomorrow, it should be fundamentally different than when you played today. I want things that you do in the game to matter on a grander scale than just your individual life or even just the life of your village. Finally, I want players to be engaging in a real collective challenge over the course of several real-world days in the game. With this update, that challenge becomes surviving as long as possible, collectively, with limited resources. Before this update, the game just went on and on, endlessly and always the same, with an endless supply of natural resources and and endless supply of very similar village rise-and-fall arcs.
Now on to the details of the rocky start:
Most of this actually had to do with the failure detection code being thwarted, but how quickly we got to what should have been a failure state was also troubling. The first time, it happened in about three hours.
During the first run, people exploited a low-tech glitch to escape from the limited area, which allowed them to exploit unlimited resources outside while the remaining people faced a resource-stripped wasteland inside. The failure condition at that point had to do with baby survival, and babies were certainly surviving just fine on the outside, so on we went. However, things got pretty bad inside in about three hours, because a few people realized they could ruin all the wild berry bushes.
The next run was much better. The escape glitch was patched, and the area made 2x larger, and the berry-ruining method was blocked. Eves could only spawn in the first two hours, so there was no longer an endless flow of nomadic families. This time, they went after all the maple trees, systematically chopping down every one. Even so, people managed to bootstrap and survive for more than 16 hours.
The failure condition in the second run was an absence of fertile mothers inside the limited area (so even if people escaped by plane, they wouldn't thwart failure detection). At the end of 16 hours, we got down to just one family surviving, and eventually just two fertile mothers, each with a trail of about 20 babies chasing them. Still, there were fertile mothers around, so no failure was triggered. I eventually had to kill it manually.
Now we're on Arc Run 3. Chopping down most of the more useful trees has become hungry work, so you can't just go berserk with the chopping. The failure condition is now the ratio between helpless babies and fertile mothers. When that ratio gets too high, the world resets. Finally, the map seed is chosen randomly for each run, so you can't just memorize the best spots without actually exploring each time. Also, Donkeytown has been fixed so that it is located outside of the limited area, correctly. There was a glitch before that caused the donkeys to mingle inside.
Things are going much more smoothly in Run 3.
But in general, I've now realized how much of a design (and player) band-aide the infinite map really was. So many problems can be solved in a hand-wavy way by saying, "Just walk further out and _______", where the blank is "find more resources," or "evade griefers," or "hide from your neighbors."
Why did no one ever chop down all the maple trees or ruin all the berry bushes before? Because there were an effectively infinite number of them, of course. But an infinite quantity of any resource is simply not that interesting. Actually there are around 36,910,000,000,000,000 maple trees on the map. That's 36 quadrillion.
There have always been (and will always be) huge opportunities for griefing in this game. I can add work-arounds for some of the big ones, but at the end of the day, people are always going to be able to do stuff that you don't want them to do. That is the fundamental challenge of building a civil society from scratch with a group of strangers. That is this game, full stop.
Now, with limited resources in place, we can see these fundamental challenges very plainly. There's no way to escape from them or paper over them. The only way to succeed is to actually trust and interact and cooperate and optimize and protect and conserve and balance and lead. You can't just head into the hills to escape from this challenge.
|
|
|
