LovePosted by Eskil Steenberg Mon, February 01, 2010 19:37:16How do you find order in chaos? Love is and will always be a chaotic game in some ways. Sometimes I get mails like "I found a hole on 23:56:E, can you come fix it?". How do you make sure that there are no holes to fall down in when players can build what ever they want? How do you know that there really is a way for the players to destroy the AI settlements? The truth is you cant and you don't. You can minimize the problems, you can give the players alternatives like spawning home or using rare items, but it can never go away completely, its not that kind of game. Freedom has consequences.
Most procedural games try solve this by running an algorithm that makes sure the levels are solvable, but love is too big, too complex and the players and AI alike have too much freedom to in anyway be able to ever know if something is solvable or not. It would be pointless to try to make sure the world was perfect because by the time I would have an answer it would be different world. Even most algorithms in the game that create content has to use a random sampling to "guess" if a specific design is good or bad. It will fail, sometimes making it too hard and sometimes too easy for the player, but hopefully it will work most of the time.
Programming love is to never solve any problem, but to improve the chance of success. A good example is the code that the AI uses to rout power from power wells and other sources to their destination. The AI needs to place a number of relays, and each relay needs line-of-sight to the next. Since the world is so large and complex the system can in no way test every conceivable routing, but it can guess. So it guesses that a good placement would be in the direction of the destination at a reasonable distance, and then it chooses a random block in that area. If it cant get line-of-sight with that block it randomly selects another block, this time a little closer and makes the random area a bit larger. It keeps doing this until it finds a spot that it has line of sight with and then it continuous from there. You may think choosing a random block will cause it to draw crazy routs, but it turns out that it is excellent, and often attempts and succeeds with things a human designer wouldn't think of. Yes things can go wrong, but 99.9% of the times it works. Whats especially good with this algorithm is that if the world changes and the lines becomes obstructed, it can start over from that point and fix it. The algorithm is not in anyway dependent on that the world is designed in a specific way.
This type of code (while very robust) makes it very hard to program because you don't know what is really going on. Any tiny change in the world will cause the route to be different. Its hard to know why the artillery unit has stopped working, Is it really broken, or did the randomly generated power route just take longer this time? or was it upset by some other event like an AI destroying a relay by accident or a new building blocking it? Things very often have this knock on effect where you may see a problem in one area and it will be caused by something completely different, or maybe it even "should" be broken as things should be able to break in a real world.
May players have noticed this knock-on-effect with the ruin algorithm that "cleans" the world of unwanted geometry like old settlements. When the game decides to kill a settlement it tags all blocks in the settlement to be cleaned. The game also regularly cleans all areas of the game since even though they may not have settlements they can still get messed up (by bomb pods for instance). The system is also used to build new geometry, say change the geography and build settlements and other things.
The thing is that the system is limited to "Clean" no more then 2 blocks per second. The reason for this is that when ever the it changes the environment it needs to send it to all clients, and sending too many of them would cause lag, and slowdown on the client side. Also, you want the destruction of old settlements to take some time.
The "ruining" happens in a few steps, like first removing edges and roofs, then having caves fall in and then the ground get lowered, so in practice it takes 2-3 seconds to restore each block. (This is still FAR faster then it used to be, but now that the blocks sink faster i get complaints about the holes it creates instead...)
The world has 100.000 blocks, so it would take about 3 days to clean every block in the world. It is therefore very important that the game doesn't give the cleaning system more work then it can handle. If it messes the world up faster then the cleaning system can fix it, eventually the world becomes one big mess.
If you look back at my
twitter feed you will find that the ruin algorithm has been tweaked a lot, for instance it now handles the ruining of things that aren't buildings much faster by avoiding going through all stages that make the ruining of buildings look good. Another addition is that the system now cleans the world far faster if no one is logged in to the server. This has made the system far more robust. Still, if something else goes wrong like the AI starts building too many bridges, or the AI settlements start self dieing, the result will be a very messy world, not immediately but eventually (making it hard for me to catch, yes it looks great now, but what about in 2 days?). I get e-mails like "Why haven't you fixed the messyness yet?" when I really have fixed it. I'm now trying to fix the issue of IA being able to attack the players even if there is no real path to get to them by building bridges, and a bug in that code has caused too many bridges to be built, that causes the mess...
Players who have been around since the beta will know that Love is a game that has changed not just in terms of features and fixes but that one small change in the AI or the way the world is generated can completely change the way the game is played. They also know that the solution to a problem often can be something completely different from what you would at first think.
Building a Procedural game is a very unstable affair when very small changes, can change the game completely. Slowly i am getting a handle on things, but things will always be added that will break other things. The game is just far too complex to ever be able to anticipate everything and it doesn't get easier to fix when the thing that is broken usually is broken because of something one would think is unrelated. Yes its annoying, but its also whats fun about it, You don't know what it will be like in a week, you just know it will be different.
Do you know why there hasn't been much artillery lately? Well people complained that the AI built Settlements too close to player settlements, so I pushed the minimum distance, but then most artillery got out of range. Again I get bug rapports that the Artillery isn't working, when its really doing precisely what it should, its just a different world again....
LovePosted by Eskil Steenberg Sun, January 17, 2010 09:55:34In the afternoon the European players start playing and when they go to bed the US players take over. I want to be awake when people are playing to collect feedback, and be able to respond to bugs and crashes. The thing is that once the US players go to bed the Australians start playing. One week has gone by since the Beta opened, and I haven't gotten a lot of sleep.
Ive heard that there is an Arab tradition of not judging a man for his failings, but his failure to mend his failures. In that sense I think I have done very well as pretty much all major bugs have be been squashed. What I worry about when I released the beta is not so much if the game will be fun or not but weather things will breakdown completely and the message boards fill up with hate that you cant recover from. Luckily I have managed to avoid that.
The response I have gotten so far can be placed roughly in two camps "I love it" and "I don't get it". Love has never been intended to be a hand holding game, but spawning players under water to immediately drown them and forcing them to trek though half the world to even get to a settlement was perhaps a unnecessarily rough first impression to give. Much of this has been fixed, and if you are in the second camp I hope you will try again, or why not write me an e-mail telling me how you got stuck because I really want you to be in the first camp.
I get a lot of feature requests and mostly they involve flying, climbing, building outside your settlement, god weapons, or some other feature that would break the game by making it easier. The features I want to add mostly revolve around forcing players to play with in the limitations of the game that are already there. I want to force players to communicate and use the full arsenal of things that are already in the game in order to succeed. I have managed to find time to add some new features, like new tools, colors, graphics and other changes, and that bodes well for the rate of development in the future, You can expect much more in the coming weeks, but my plans right now is mainly to highlight what is already in the game, and give players the opportunity to discover the many thing the game can offer.
Today I woke up for the first time without a single E-mail reporting a bug or crash and the servers have all been running for days since i fixed the last server crash bug. I went right back to sleep.
LovePosted by Eskil Steenberg Fri, January 08, 2010 04:59:57If you haven't already noticed the
Beta is live. The servers have been filling up and i have been doing my best to keep track of the games. Any game that is multiplayer or procedural will be very sensitive to changes, so now my goal is to not add anything radical but rather make sure what is in the game works perfectly. So right now what I really need is your feedback, especially on things that you don't like or reasons why you wouldn't play this over the long haul, It doesn't have to be specific, tell me what you feel and ill figure out what needs to be done. If you have stopped playing tell me why. Things like pacing over a 3-4 hour play session are very hard for me to test myself since I'm so busy so it helps a lot to get feedback on how the game progresses.
Having a beta for a game that you have no intention of stopping the development of once it gets released is a bit strange. What you see in the game now is in no way the end game, yet at some point you need to set down your flag and say that the game is done enough to be playable. Partly because unless i have players I cant really know what works and what doesn't, and partly because of the financial reality that I'm going broke quickly, and unless this turns in to a commercial game, i have to stop developing the game and get a job.
Right now I'm trying to think of the next step of development for Love, things I would like to add and change. and what the next big system I should add would be. What I keep coming back to is how nice the players are who play Love. Its not a coincidence. First I think that the game doesn't attract people who like to run around with chainsaws screaming F-yeah!, but that's not all. The game is designed to be cooperative, and there is no competition at all, the success of one is the success of everyone. People help each other because its in their self interest.
This makes it tricky to add new gameplay. For instance I have been very interested in implementing more "Social" IA, where you could become friendly with the AI. The problem is that if your settlement becomes friendly with a AI settlement it would only take one player to completely destroy that friendship. It would become incredibly frustrating for players trying to build up relations with other settlements, and player conflicts would inevitably flair up. I still want to do something like that, but I have to figure out a way of doing it that doesn't make players hate each other. Another thing that i would like in some form is persistent progress. I have written before about my dislike of "achievements", but giving people a goal like that is an effective way to get players to comeback, and to try to explore parts of the game they otherwise wouldn't. At the same time Achievements can be very anti social, and force players to do boring repetitive tasks they really don't want to. It may seam easy but one false step could very much ruin the game.
The next few days i will be paroling the servers and check for bugs and feedback. Expect many updates with small tweaks and changes. I'm trying to get a larger Voice over IP server as soon as possible so that we can all meat up and talk about the game. Because that's what i really want from you right now, feedback.
Edit:Two days have passed and after marathon sessions of bug fixing with me popping in and out of servers and sending out many updates, things are starting to stabilize. There are still many things to do, but a few game breaking bugs, like the AIs inability to attack that have now been fixed, or constant spawning in to water that has been annoying but are now (hopefully) fixed. I have gotten some amazing feedback, and i want to thank everyone who has helped and i hope that you now can start enjoying the game even more now as most things start to work as they should. I still have a to-do list and I'm checking thing off it as fast as i can.
Edit:I finally got my own
TeamSpeak server up, Join us at teamspeak.quelsolaar.com:9987
LovePosted by Eskil Steenberg Fri, January 01, 2010 19:36:22Today I uploaded the first version of the Beta. Putting out something that you have worked on for so long is a pretty amazing feeling. The beta will only be available for people who still have active Alpha accounts. On the 7th I will open up the Beta to everyone. The reason for this is that I want some feedback from more experienced players so that I can iron out the major kinks before I let everyone else in. The beta will still cost 3 Euros for 30 days , just like the Alpha to pay for the servers, but I will remove the 24H penalty for switching servers. If people start switching servers like crazy, I may add some new penalty to prevent it.
So what is this Beta? Well Beta means "feature complete", meaning that all functionality planed for the full release have been implemented, so its got loads of new cool stuff that I don't want to spoil. That doesn't mean everything works, or that I wont come to the conclusion that changes needs to be made. The concept of "Full Release" is a bit different for LOVE too, The idea is to keep adding things and much of what is new in the Beta is the infrastructure that lets me add more things very quickly. This new infrastructure has worked so well that i managed to get the Beta out a full 3 weeks before I thought I would get it out. Something that I have very continuously neglected is sound and graphics, since fixing that rarely introduces bugs. So expect a large number of patches and updates in the coming days and weeks. If you have any comments or bug rapports please send them to me so that I can make the game better
I know that many people use open Betas as Demos to evaluate games, and in some ways that is understandable, as long as they understand that many things will change. Love is a very different game, and not one that is easy to get in to. Its a game that's most people start to appreciate over time. There is no hand-holding, no glowing arrows, no female voice telling you what to do, and no quests. In many ways Its an old school PC game, that takes a bit of time getting used to. Evaluation of games are always tricky, and with a game so different its sometimes hard to know if something is broken, just different or something you haven't figured out yet. I guess my best advice is to take your time and talk to other players (Love has some amazingly friendly players you wont find in other games). One excellent resource for new(and old) players is the site
love in verses, go check it out!
Also join us in the new
Ventrilo server at vent.quelsolaar.com to talk to other Love players. (I try to check in frequently too)
On a side note:
I get lots of E-mails form people who wold like to learn how to make games, so I thought I would put together a 3 day intensive course in game programming (for total beginners) that would teach you the basics of programming, and let you make a simple game. Each day would be around 3-4 hours long preferably in the evening. My plan is to try to do this in San Francisco the week before or after GDC. It would be free, and everyone would be welcome.I´m still working on trying to find a venue to host the event. If you are interested in joining up, (or know of a good place to host it) send me an e-mail so that I can gauge if there is enough interest.
LovePosted by Eskil Steenberg Wed, December 30, 2009 20:22:24Funny thing about decades: There are never really where they should be. People may say 50s and they don't mean 50 to 59. They mean that time after the war where every body was happy in suburbia drove around in cars with big fins, and they listened to the beach boys, and mowtown coming out in the jukebox in diners. Its basically 1946 to 1963. Sure Mowtown started in 1959 but you cant run a 50s diner with out it, and The Kennedy family so embodied the happy 50s that they had to shoot him in order to end the decade. In reality things had been wrong all along.
It took some time to get the 60s started but In 1966 a young black man landed in London. His name was
Jimi Hendrix. If you lived in London the sixties started in 66 everyone else had to wait another year, but Like no other decade the 60s was an explosion, so strong that even 68 is a decade in it self.
Woodstock,
Sgt. Pepper,
Easy rider,
May68,
The acid test,
Kent State,
2001,
Vietnam,
Summer of love,
Led Zeppelin, Free love,
I have a dream. Its the kind of thing that couldn't last very long. It was too strange to live but too rare to die. And in 1972 it was over. Or to quote
Hunter S Thompson: "So now, less than five years later, you can go up on a steep hill in Las Vegas and look West, and with the right kind of eyes you can almost see the high-water mark — that place where the wave finally broke and rolled back." - Jimi was dead.
I think the 70s was just some sort of mutant cousin of the 60s. I have dream became Watergate, Dodge Challenger became the oil crisis, Free love became wife swapping, Freedom fighters became terrorists, and the tripped out outfits became ugly brown Manchester. Nobody knows how, but somehow the forces of old and evil stayed in power. The fun was just over, and Pink Floyd wrote the
soundtrack. "Everything under the sun is in tune, But the sun is eclipsed by the moon." The movie theater wasn't full of stars, but rather a sobering world of
Taxi drivers,
Godfathers and
Napalm in the morning. The artist grew bolder but there was nobody they could move. The punks tried, so did Warhol, but
disco won out.
I really cant tell when the 70s turned to the 80s, 1982 perhaps or 1984? But at some point I think people just embraced inequality, bad taste, sock less loafers, and stopped hoping for the 60s to come back. With enough Coke any one can do it.
Lucky for all of us I don't think the 80s lasted very long. Without even trying the wall came down and nothing would ever be that same. For the powers that be the 80s was a comfortable place, and there was a few years of nothingness but in 1991 an unknown group released the single "
Smells like teen spirit", and for almost the first time in 20 years something felt real. The 90s had begun. The flood gates opened when
Charlie made the charts, for Rave, the
Chemical brothers,
Daft punk,
Portishead,
Massive attack,
björk and so on. We went to see
Pulp fiction and had never seen anything like it! Every music video was a peace of
art. I really like the 90s, on the face of it not much happened, but it was a decade where the really great stuff was made in cellars, bedrooms and garages by people without financial backing who just really wanted to do something great. The technology was cheep enough so that more people could do it, but hard enough to use to keep the dabblers out. I have feeling the
90s ended in 97-98 and the last defining moment was Aphex Twins
Window licker.
We where all really sure new years eve 2000 would be the seminal moment with the end of days planes crashing, and cars flying. So when it dint really go that way its understandable that we where a bit confused. In hind sight its very easy to see when 00s started: 9/11 2001. In my mind the 00s where very bad, It a lazy time where everybody thought they could Stop a war without protesting, Win a war without fighting, Make movies without stories, get rich without working using the stock market, have a clothing line without designing, become a musicians by mashing other people music, and become a celebrity without doing anything at all.
1984 happened 20 years later, but it turned out that that people would subject themselves to being filmed not because of their blind belief in big brother, but their blind belief in TV celebrity.
Luckily it has already ended, it ended in the fall of 2008. The question is has the 10s started yet? I don't think so, I think we are still in the never-never land between decades. I don't know what it will be but I hope that it will show me something I have never seen before. I hope it will be people taking initiative and doing things themselves, not in haphazard YouTube kind of way, but in grand aw inspiring way. At some point its has to happen, and I'm working really hard to make it happen sooner Rather then later.
Until then,
Happy new year
LovePosted by Eskil Steenberg Fri, November 20, 2009 18:23:31I'm now back from Tokyo, DD64 in hand, and I'm happy to report that the Love Alpha has stabilized, and I have sorted out most of the problems (like re-appearing settlements, and the sudden outbreak of Gigantism) so now I will focus on getting to Beta. I have written a list of all the 26 things that I want in the Beta. I will start counting them down in my twitter updates as I get them done. I don't know how long it will take but some of the new things are major. To make the beta release more exciting I wont release any of the new things until they are all done with the exception of some changes to the AI settlements and bug fixes. All alpha accounts will carry over to the beta, and the beta will cost the same as the Alpha.
I know there are some more casual players who would like to know of good times to play so while we are so few, my suggestion is to have play events on Thursdays at 8PM CET, or 7PM GMT on the UK servers, and 7PM Pacific, 8PM Mountain, 9PM central, and 10PM East cost on the Atlanta servers.
LovePosted by Eskil Steenberg Tue, November 17, 2009 18:59:17Some Economists claim that, unless governments had intervened last year, we where days away from ATMs shutting down in the US and UK. A few days without ATMs may have done some good to us, perhaps it would have moved the financial crisis from the front pages in to our lives in a more tangible way, and then maybe then we would have had the guts to change the system that produced the crisis. If you think we have averted any future crisis, think again. Let me just say the words "high frequency trading", so that I can point to this post some day, and say I told you so.
When it has become as easy as a button press to move all your money to the most successful company in the world, why would anyone in their right mind put their money in a new one? The value of companies are not in their output but in the value itself.
This way of thinking is spreading. Houses main function is not to live in but to retain value. When the value of peoples homes drop, people default and have to move out, even if the house has just as much value as before for the people who live in it. The same goes for consumers, we are more and more valuing things by preconceived notions then the cost of their making, or their use. Why is a bag from Prada worth, paying 1000$ when a no-name bag isn't worth 10? Why do everybody talk about what consoles should cost to buy, and never what they cost to make? Why must a online service be free even if it costs money to run? If a musician is entitled to money every time someone listens to her beautiful music, why isn't an architect entitled to money every time someone looks at her beautiful building?
Companies and things main use have become storage for our investment, not any actual use or meaning. In this mindset any change to anything, is a threat to the value of our investments, and it will mean that it becomes our main imperative to stop any change from happening. But change happens.
While US car manufacturers have successfully fought any emissions or safety regulations to cut costs, Japanese and European manufacturers have tried to make safer and less polluting cars. For Japanese and European manufacturers the development of those features weren't free, but once made, the have put US car makers hopelessly behind.
People complain about bailouts, but the bailouts probably make sense, what doesn't make sense is the inability to use the bailout to make the changes we need. Ownership shouldn't be about value, it should be about how you can use what you own, and so far the none of the bailed out companies have been used for anything.
Recently a whistle blowers at IEA (International Energy Agency) claims that the predictions of future oil production have, for political reasons, been overly positive and that we are likely to reach peek oil production very soon if we haven't already. If we know oil is running out and that we are destroying out planet by using it why not just stop? If a country would simply raise the price of oil by 50% using taxes every year for the next ten years, they would create a oil crisis, but a very predictable one. Would it hurt? Sure the ATMs would stop working every now and then because of power outages, but ten years later when oil is running out for the rest of the world, you would be world leader at selling the energy solutions the rest of the world will desperately need. Its not that hard to do, if only you can think that far ahead. When profitability becomes more important then long-term sustainability you get neither.
LovePosted by Eskil Steenberg Fri, October 30, 2009 13:26:04Everyone wants to be loved, so why are we so drawn to abuse?
Playing an MMO is like having an abusive relationships. They kick you and they slap you, and they make you beg for every scrap of appreciation. Lets face it, if you are grinding for hours to get a mount, then you are not that far away from the girls who say "He is really a great guy when he doesn't drink!" ... yeah, sure he is.
Its not the kind of relationship I want, I have no time for fights, no matter how good the makeup sex is. I'm not going to insult someone even if that's what will make them want me. I want to treat my players right, but what if they really want abuse? Everybody says they love Shenmue, Okami and Another world, pinnacle of "interactive story telling", but then they go back to their desks to kill 15 scorpions or collect 30 bore pelts. Clearly something is rotten in Denmark.
The less you get out of games the more you want out of them. The more time wasted the higher the honor of its accomplishment. Gamers will put up with the most punishing games, just as long as the game every now and then stop by that store at the airport that only sells poor relationship band-aids, to pick up chocolate, alcohol or what ever else they can use to instill hope that the relationship is not a waist of time.
Ask yourself, would you rather play an hour of grind to level up, or and hour of great gameplay who's progress you couldn't save?
I think many gamers would rather take a beating if they think that they are accomplishing something, then have some pointless fun. That's a real shame because, lets be clear, you don't accomplish anything when you play games. There is nothing wrong with that, we don't have to accomplish things all the time, it should be OK to just have fun every now and then. That's why I'm not so sure I'm OK with is trying to fooling you in to thinking you did accomplish something when you played my game.
I'm not the beating kind, and i don't waste time, yours or mine. By now you shoul have figured out that I'm a romantic but also that I speak my mind. But if love is not what you want, perhaps it is me I end up fooling? It begs the question: Is game design by its very nature to be a con artist of the seducing kind? Everyone know that the best con is the one where the one being conned enjoys being conned.
