Howdy, Stranger!
It looks like you're new here. If you want to get involved, click one of these buttons!
Quick Links
How SWG's dynamic world worked
Raph
MMO DesignerMember RarePosts: 248
New blog post for you all. 
http://www.raphkoster.com/2015/04/20/swgs-dynamic-world/
http://www.raphkoster.com/2015/04/20/swgs-dynamic-world/ 
Comments
Herald of innovation, Vanquisher of the old! - Awake a few hours almost everyday!
Great tips on building a world without the pain of doing it by hand. I truly believe the future of large scale mmos is dynamic content generation. We MUST go away from static content and world building take too long manually as well.
I actually didn't like how SWG made each of their worlds. Ultimately they were a little too small and square to make you feel like each was their own planet. Some sci-fi games right now are allowing space to planet transition and round worlds but for a fantasy world mmo we need far greater detail but the same principles should apply: bigger is better!
Once developers get to a point where a world can evolve nearly by itself we will finally start seeing mmorpgs return to the form and concept they were originally supposed to be instead of shrinking people into tiny themepark versions of reality. This is actually what the term themepark means for an mmo ... it is getting Disneyworld parks with rides and shows instead of a vitual world that rivals our real one in size and scope. Most mmos today were not even attempted to be made into actual mmorpgs. I'd argue Wow started out making one but then shifted very quickly toward themepark shrink.
New developers will only be able to make virtual worlds with entirely new tool sets that solve many of the issues inherent to this genre. I believe at this point that the next major frontier for mmorpg advancement is Geographic mapping technologies (GIS, etc) and superior procedural content development. Developers basically have to develop world emulators and from there adapt it to the theme they want.
Mmorpgs need to be redefined BACK to their original concept and this requires new tech and new development approaches. It isn't just current market setup driving development toward such limited scope as it is today but also the development style and process most studios are pigeon holed into.
If a developer can meld GIS type technologies with a physics engine designed for video game interaction ... OMG, and I'll leave it that.
You stay sassy!
Heh, too small and square? I am pretty sure SWG's scale is still among the bigger MMOs, to this day.
Bigger ISN'T always better, Content density matters, so does player density. A big giant world where players spread out too much will basically feel like a single player game and lose a lot of MMO qualities. Travel time has huge impacts on the perception of size as well.
We can do that now, it's more that no one does. Heck, in 2006 at SOE we built a system with full water cycles, season capabilities, snow accumulation, and more.
Personally I think huge worlds that otherwise would be too big can still function well (even better) by having social hubs/centers of activity (economic and otherwise).
I'm guessing that most developers would see that as waste, but in my mind more options is never a waste even if they are not used to the satisfaction of the developers. It's the satisfaction of the customers that counts.
Once upon a time....
Nothing is too big. It is all about how it is developed. Too big for transportation? Well magic and technology has managed to solve this in most games and certainly technology in our real world is doing the same.
It depends on the who the game is targeting. SWG required players to head out and create towns and obtain resources. When the servers were full the size of their maps still caused issues as space became limited. Maps were indeed larger than typical zones in most mmos but the square shape when trying to emulate a planet was distracting. Beyond that it was their terrible physics engine that ruined it for me. They for some reason even got rid of elevation affecting movement which made the game silly when speeders rocketed up cliffs like flying over flat ground. Not sure what drove this choice but the game was built on tech very aged by today's standards.
Virtually all games today start out with the game designer taking their concept then "shrinking it" to what tech and development limitations they will face. This has to change or else concepts that go beyond what tech and development scope can handle will never be produced. Imagination cannot be limited but the current industry limits it far beyond other genre and media that video games pull ideas from.
If developers and players cannot gain a vision far grander in scope than say what Lotro has done for Middle Earth then I pray younger and more visionary developers and players are born some day that can.
You stay sassy!
The absence of a Z axis was a bummer. But the planets were quite expansive.
There's quite a lot of evidence to say that packing players more tightly that is reasonable actually greatly improves a lot of the social play.
Players tend to pack themselves that way, for that matter. They clump and cluster quite a lot.
If you read the article, it explains why this happened.
I am all for vision and ambition. But saying that designers should not take tech and dev scope into account is just a recipe for never shipping at all.
Ryzom is indeed very dynamic. It was pretty strongly inspired by both UO and SWG though.
I know this is nit picky, but you know what always bugged the crap out of me? Looking up at the sky at night and seeing the stars in front of the moon.
It doesn't matter how large the game is. Social hubs are called hubs for a reason. In worlds they are called cities. They do not have to be massive. Most space games have social hubs even though the games are truly massive in size. Your response suggest that these games lack social interaction when in fact they do not. These are niche games we are talking about here. An exploration game targets explorers and their social needs are not the same as socialites. In fact a game truly massive in scale (effectively unlimited such as games like EVE and other procedurally generated worlds) can have any type of game play within it. This is where imagination comes into play as I suggested. You could have one single planet with every game play component a social gamer wants without ever leaving it.
The very fact a planet is so large allows players to be spread out enough to form the player driven communities want or factions developers to form. When talking games that host 10-60k you have to have some spread and players would simply be brought together in logical and manageable groups. Social games break down then they social units are too large. Sociology is a well defined science far within the scope for adoption to gaming. No themepark mmo even comes close to these limits and one would likely look more toward Second Life and the more variety added to a game clearly reveals players actively limiting the scope of their social interaction than increasing it. Humans frankly don't like many other humans. They only want convenience and this world even better in games massive in scope because virtual space can be developed easily within the "real" virtual space. Wow does it already but instead of pulling from from all servers you only have one true play area which makes a hell of a lot more logical sense. The bigger the game means the better you can maintain immersion and concept and put as many sub-systems within as you like.
These are not games with mini-games. They are games within games within games, etc. A developer could launch an mmo on a single world and later add endless space to explore or open space to explore and a world within it that is a fully developed mmo in itself.
As for reading the article to explain what I said: I did read the article and I explained the ultimate limitations and is why I went beyond it in scope to explore where it could ultimately lead.
You last comment reveals you didn't understand what I was saying. I said current tech and development process limits scope. Why? Because it HAS TO be taken into account. It is not an option or your project management will lead to failure. I suggest that the current limited scope should be explored and new ways to develop needs to be explored. This means new tech must be explored, adopted or developed so that scope isn't as limited as it currently is.
You stay sassy!
I think Raph was saying that the tech has already been developed. Bu I agree with you that it would behoove a producer to put both the social hubs and large amounts of free ranging exploration into their games. As you point out, why shoot for a greater chance at failure? The costs, of course, limits what can be done but the best possible build should be the goal. Hell, if you run out of time you can turn the extra space into waste lands and give it lore to account for it and some basic playable content. Polish it up later as you go.
Players like to gather, but they also like to go out there and explore and find things. Options are good.
Once upon a time....
If you're setting the height by formulas, then now you could use tessellation to do it on the GPU. That way, you can dynamically generate however many points you need to in order to look as smooth as you like (many up front, few far away--and nearly everything is far away). Different "rules" can use different programs, and so long as each "rule" governs the height in a large enough area, you get plenty of parallelism to take advantage of the GPU properly. The sort of "rules" that will fit a GPU will restrict you somewhat (e.g., recursion is very, very bad), but it does make many computations nearly free because you can exploit the massive computational power of a GPU.
I've been working on a procedurally generated game world for the game I've been programming. I haven't put too much work into terrain yet; for now, I'm working on a character creator, in the expectation that animations will be the harder part. At the moment, ground height is just the graph of a function, which is something that I'll presumably change at some point.
Thats the problem, everyone is harping about "vast/endless" world.
But what use it is if theres noone ever there because all people hang in just few hubs and its mostly empty space with few random mobs scattered around.
On the second point, more like intelligent-ish evolution where players drive some part of the change. I woudnt call water cycles, season capabilites and snow accumulation evolution, sure nice things to have, but in the end they will repeat themselves almost identically forever
To give an example, lets say MMO starts at sone age and you only have stone/wood/mud as materials, and at some point game decides its time to discover metal (bronze age) and there is (hidden way) to "discver" bronze ingame as well as how to use it for - everything, from decorations to weapons, along with new discoverable schematics.
Not "we will have bronze age in new expanasion along with new "node" and 150 new recipes" regardless is it world ready or way overdue for something like that but just because it was on your timetable to launch 18 mothns after launch ;P
That is the themepark instance hub sitter talking in you.
I like(d) MMO's because they were open worlds. I did not log in and sit in hubs waiting for a instance popup to appear, like 99% of the MMO players out there currently do.
I liked SWG back then because I could go OUT in the open world, away from everyone else, do my business, whether it was exploring, hunting, find stuff to set up a business (minerals harvesting etc). I did not follow the mass on purpose. I visited trade hubs to trade and went back to my area again, actually playing the game.
I liked the idea of a big dangerous wasteland out there and travelling there with 1-2 friends or solo without seeing anyone else for hours. It created immersion for me.
For a immersion MMO player who could care less about linear progression it is very important to have open space, seasons, day/night cycles and weather, even if there hasn't been any player in the area for hours/days.
You guessed it, I am a sandbox player, I despise themeparks as I want to play the game Iwant to, and thus SWG -pre NGE- was a good game for me.
"going into arguments with idiots is a lost cause, it requires you to stoop down to their level and you can't win"
You claim people didnt/dont cluster in sandbox game.
Game should be large enough, but as i said, excess of nothingness is just useless.
And not many people dream of being a hermit in a MMO ;P
Makes me sad to think we will never see a gaming world (and crafting/resource system) like SWG ever again.
thanks for the (somewhat sombre) trip down memory lane Raph.
PAST: UO-SWG-DAOC-WOW-DDO-VG-AOC-WAR-FE-DFO-LOTRO-RIFT-GW2
PRESENT: Nothing
FUTURE: ESO
Executive Editor (Games) http://www.wccftech.com
Raph,
As a SWG player who played for almost 2 years before WOW came out. I felt SWG was a very good game in a sense. However I would like you explain to you why I left and it was not to go to WOW, I ended up playing a lot of FFXI at the time.
My core problem was time spent in game doing things that were not fun. For example if my friends were off on Dantoonie or Dathomir which we were often doing stuff and I was sitting on Rori where my extractors were it could take me an hour to get to my friends who are doing things. I would have to wait about 5 minutes for a shuttle to a city where I could get off world, if unlucky I would have to wait 15 mins there, then if unlucky again 15 mins from naboo to corellia, then another 15 to Dantooine or Dathomir. It was a lot of sitting around wasting my time. Once I got my full time IT job as a IT consultant and was working 60+ hours. I quit because I could not stand taking an hour to get to where my friends were. It was a waste of time for me and one of the core reasons I quit to never come back.
Now lets fast forward to today and why I do not feel Crowfall will not work. First PVP full looting. I loved it back in 1998 when I played UO however today I hate it because the amount of time I have for MMOs is no longer 40+ hours a week. I might get on the very high end and only when I take vacation time 15 hours a week. Average between 8 to 10 hours a week. Full looting and spending all the time dealing with corpse campers which I did back in the day and all that I have no time for. The Average gamer today will not either sorry but they just dont.
Lets add to that your trying to create an open world sandbox MMO. Sorry this will not work today. I cannot take my wife and many of my other friends who never played an MMO before WOW. I tried to take my friends who only played WOW or other Themepark MMOs to UO and SWG. They get lost because there is no bread crumb train. Heck a few tried AA and got to 50 and said, ok now what do I do? People today do not nor will they understand MMOs pre WOW. While there is a Major forum presence for a Sandbox MMO, many of todays MMO gamers dont ever post on a forum ever. I am the only one out of a few dozen of my friends who I either still play MMOs with or I still keep in touch with that ever post on a forum. The Majority of players will never post on a forum. Sorry but its true. Thats why WOW is in shitty shape, WOW went from Hardmode to Easymode and subs kept dropping however the forum people keep asking for things to be easier, and faster to get through. Yet the majority voted with their wallets and quit.
I do think there might be room for a few sandbox MMOs however not until the over saturated shrinks. There are too many MMOs on the market today, many just a clone of WOW. Until the market shrinks and there are fewer MMOs out there I think any new MMO even Indy driven will end up having a very hard time with population. Thats why worlds are small today to make players think there are more players in the world than there really is, and why MMOs go F2P because they are low quality and because they need more players to not feel dead.
Extremely rare for me to post here, though I read the forums daily.
Just wanted to say what a fascinating insight this was to read, and also to say 'thank you' to Raph for taking the time to post here.
Though I am nothing more than a hobbyist designer - I have enjoyed fantasy, imagination-driven gaming since the early days of Runequest and D&D - my first MMORPG being Everquest - I was utterly floored and delighted by the experience of meeting a real person in a convincing, imaginery world and my best gaming times have been in MMO's ever since.
Mind you, it struck me whilst reading the article that - like many things - limitation is the mother of innovation - and reading about the ingenious ways that Raph and SWG team coded and designed in order to make a hugely flexible world despite the limitations of the tech of their day made me worry that todays developers are perhaps a little 'spoilt' by the processing power, especially the HDD drive and the huge bandwidth available these days. A little like a generation of mechanics who've seldom seen inside an engine, or developers who never studied machine code.
Also, I wonder if any of you (or even Raph, if he reads this) would agree with the notion that tomorrows developers are growing up playing rather dumbed down mechanics with low expectations of 'realism' or 'immersion' - and that this might lead us into the slow death of endless creavity and innovation and a continuation of worryingly safe game design for the future?
*edit to add this:*
I read the post above, and whilst I respect the authors opinion, I feel strongly that it is PRECISELY the lack of downtime in modern MMO's that makes them feel shallow and far less immersive. For me, it was precisely the mundane elements, such as waiting for a shuttle, or resting in a camp - that made games like SWG and the 1st generation MMO's so amazingly sociable (because there was time to actually talk) and made the times when you achieved something major that made it so exciting. Todays instant gratification games and the 'everyone gets a prize' mentality - for me at least - just waters down experience.
Cool stuff. My idea for a space-themed MMO is to allow players to land their spacecraft at any point on a planet by entering the latitude, longitude coordinates X, Y. These would then serve as seed values for a procedural generator that would carve out a small patch of earth, say 10 x 10 miles. The zone would not connect to any other zone, but wrap on itself Asteroids-style: topologically, it would be a torus, but it would render as flat (with exaggerated Z dimension relative to the toroidal projection).
So, every time I land at 103,552 I come to the exact same spot, same landscape, same types of enemy encounters. But if I change the coordinates just slightly to 103.0001,552 -- a completely different place. A planet could then support near-infinite exploration or player housing, at the cost of not being seamless.
Is this possible? Has this been done? I have always wondered more games do not use maps that wrap.
Excellent point about limitation being the mother of innovation.
As far as what tomorrow's developers turn out to be, I don't see today's lackluster products affecting that much. What will drive it is the producers and what they decide to make, just like today. There's always the cream that rises in any generation, it's more a matter of how the stars are aligning.
I enjoyed your post.
Once upon a time....
Thank you - your words reassure me somewhat! Let's hope human nature continues to strive and try to outdo itself!
Also, kind of you to say you enjoyed my contribution. As I said, I very rarely post - so that's a nice thing to see.