How will leveling work?

[x4000]

Well, to some extent there is no "doing it right."  And part of the cool thing about this is having the history of everything that happened.  I think that's really interesting with Dwarf Fortress, how histories like Boatmurdered developed.

In terms of how you want to play, do what you will.  But the game will certainly encourage you to stick around.  In terms of the comparisons to "having everything" in the game, that simply won't ever happen here.  You max out at level 9999, and that's true for all the items and stats and such, too. 

There is ALWAYS more to do, unless you've logged the equivalent of about 208 straight days of gametime (~4500 game hours).  If that happens, yeah, you'll probably want to start a new world unless we up the level cap.  That's assuming a new level roughly every 30 minutes of gameplay, or thereabouts.

In terms of making every savegame backwards compatible: that's no challenge at all if you do it from the start.  AI War has changed massively, including three complete savegame format revamps.   However, it is still compatible with the pre-release alpha savegames (which I consistently use for testing, actually).  It doesn't even cause much in the way of code bloat (one extra method for the REALLY old versions), it just requires good programming practices for what we do have.

[snrub_guy]

Oh, I fully intend to try to play the way you are intending. I think this game has a better chance of it than most!

In terms of motivation behind that sort of thing, I think my main motivations behind it are:

1) Coyotetheclever was bang on with one of them- sometimes I just like going back to the early stage of a game, where I was powerless, and everything was new and pristine- a blank canvas of sorts.

2) Getting bored of a playstyle/ wanting to try a new character build/ thought of some things that I would have done differently that are niggling me.

3) In the case of a randomly generated world (like minecraft), seeing if I can get a more interesting map seed, discovering new vistas and so on.

4) And this is probably the one that gets me the most, dropping a game for a month or two, going back to it and losing connection to the game/character in progress, or forgetting some aspects of the story, or just planning on experiencing the whole thing in a proper run through (however unlikely a full run through might be).

5) Just on a whim.

There may be others, but after having a think, these are the main ones that come to mind. I am aware that not all (maybe none) of the apply to AVWW, but I figured seeing as you asked, a little insight into my motivations might be interesting or helpful to you. I should make it clear- these aren't points I want you to "MAKE NOT HAPPEN" in AVWW, I'm just chipping in on the discussion.

 

[x4000]

#1 is a very good point.  From that angle I probably would start a new world myself at some point, though I'd probably always keep my "main" world, too.

For #2, in this particular game there's nothing that should cause you to do that.  With other games -- Neverwinter Nights comes to mind -- I can definitely sympathize on this point, though.  That's definitely where a lot of the replay value.  In AVWW, it wraps all that right into the main game, though.

For #3, that's fair enough.  Though in both Minecraft and AVWW you can just explore more and more to find those interesting vistas.  They just might not be close to your original start or whatever.  In Minecraft that can be a problem unless you use one of the third party servers to let you set a new start position, so you're not having to truck it back from world center every time you die.  In AVWW, that will be part of the game itself to take care of making sure you don't spawn ten lightyears away from your current area of interest.

For #4, fair enough.  I tend to devalue this one because time is so limited for me that I'd never finish a game if I restarted it every time this happened to me, so I just wind up exploring around lost for a bit when I come back.  It took me six years to finish FFX, for instance, with multi-year gaps between some of my play sessions.  That took... some adjusting to remember what was going on.   When I was a kid, I played every game many times, though, so I wouldn't have thought twice of it then.  I must have started 30+ games of FF1, and logged 10+ hours on each of them, before ever beating the game.  Not long after getting the upgrades to the improved classes from Bahamut, I'd lose interest and drift off to Zelda or something, and then I'd come back days or weeks or months later and decide I liked the earlier part of the game better, so I'd play it again with a different party.  So the more I think about it, the more I can relate; I just don't think of AVWW as that sort of game.

For #5, definitely can't argue.

All good, and thanks for providing the details and clarity.  In reality I don't really care how many worlds people create -- if you want to create a new world every week or every day, knock yourself out.  Maybe for some strange reason, like me with FF1, you just like playing the first ten hours when everything is relatively low-level and the world seems broader and more tame.  Maybe that's just so compelling you want to play that over and over again instead of getting things more built up, or even building things up and going off for a new adventure in some new wild area.  That's cool with me.

But for those who want a sense of constancy, to have a really long-term amount of play without restarting and losing progress, you can do it, and that's what is different and nifty here.  In the most extreme of completionist circumstances, I might get 120 hours out of the largest Final Fantasy game.  On the shorter end, some of the older ones are closer to 20 or even less.  That's a long, long time in any stretch, and those are all great games that I love, but once you pass a certain point... that's it.  Your involvement with the world ends, unless you want to relive the same experience again.  It's the same with books and movies, of course.  They are a finite experience.

In the case of AVWW, the goal is to make that not be so.  That there's never a point where you have to just be done, because everything is static now and nothing is new or interesting.  We'll see how we do.

[syndicatedragon]

I guess I'm not used to the idea that a game won't run out of content. In most games, when you (or the world) reach a certain level plateau, even though that might not be the absolute highest level, there's nothing new to see. Things might still get more difficult, but there are no more unique abilities, items, or creatures to find. In a game like Titan Quest or Diablo 2, this is when you start over with a new character; the many different character types you can build makes each a brand new experience. So, I'm curious to see how AVWW keeps the experience fresh aside from generating new terrain. Sure it has 9999 levels, but will there be a huge difference say for example between level 1000 and 2000? How is that different than the first 1000 levels? I guess what I'm saying is that even though a game has a level cap, there is usually some point much lower than that where the game actually stops revealing new content.

Or maybe not. I'm really interested to see how it turns out. I'm just a little skeptical at this point.

[x4000]

Well, assuming that the game hits it off reasonably well with players, we intend to keep adding content like we have with AI War.  Free DLC and paid expansions, the works.

But in terms of the actual core question, assuming that we just hit 1.0 and then for some reason utterly abandoned the game (for the sake of argument, not that we at all plan to do that), I think the game would still have an edge on a lot of other similar ones.  There will be no central storyline, for instance.  Nothing hard-coded that is "the one story that when it's done you've won."  It's a lot more open ended than that, with various storylines that can pop up in various regions, with different characters and circumstances and all that.

Does that mean that after some point, assuming we for some reason stopped adding content, that you'd get into a repeating cycle?  Sure, in that case.  But that's no different from starting the game over and then being level 1, and seeing that repeating cycle from lower levels rather than higher ones.

Given the players' ability to shape the world, the mere fact of having been there and created this legacy and all these settlements, etc, may help to create new and meaningful circumstances at the higher levels compared to just starting over.  We'll see how much we are able to do with that, but given enough time and player support I think it's something Keith and I both want to focus on.

Will all that happen?  Who knows.  It's dependent on player involvement and support, and if we can keep ourselves in business developing out content like that.  What I do know is that this will be a damn good adventure game with a ton of content well beyond its price point (as with AI War) even if we don't make it substantially past 1.0 for some reason.  But if it really takes off... we're doing everything we can to lay the groundwork for making this even larger than AI War.  Immensely larger than AI War.  It just depends on if that's something that people really are interested in, when it comes down to it.  And that's something we have yet to fully find out.  There certainly seems to be a heaping load of interest in this judging by comments and forums posts, though.

[Flatfingers]

The "content problem" will probably deserve its own thread at some point, but to tie it into leveling....

There really are only about three distinct ways to keep adding content to a published game:

1. Developer-created. Because there are only a handful of developers, but they know all the tricks of the content editor, this approach usually results in a small number of large and high-quality expansions to the base game. The benefit here is that the quality of the overall game is perceived to remain high; the downside is that users will always, always consume new content far faster than it can be created.

2. Procedurally-generated. ("Randomly-generated" is more accurate, but never mind.) This approach tries to create a system whereby new content can be generated by code, either on-the-fly and complete on the client or as a starting point for developer hand-tweaking. Random content generation has the advantage of being able to produce lots of new things to do very quickly, which helps to keep up with the insatiable craving of many users for new things to do in a game. On the other hand, it's insanely difficult to build a content generator that produces good content. If it was easy, somebody would have written a general-purpose storytelling and asset management program, and human content creators would be out of a job... but that hasn't exactly happened yet. Most random content generators can only produce very simple one-task activities -- to tell a good story, humans are still required.

3. User-generated content (UGC). Pros: since there are a lot more players of a game than developers, this creates the opportunity for a lot more content creation than what the devs alone can generate. Done smartly, this can also be a good PR tool. And if you're very lucky, you may find a handful of users whose content is absurdly high-quality -- this is a great way to help bring new talent and fresh ideas into the industry. Cons: Only about 10% of the total user base will ever take the time to develop significant content (i.e., more detailed than what a random content generator could produce). And of that 10%, 90% will probably fall victim to Sturgeon's Law, but it's bad PR to tell people that their creative work is not of sufficient quality to become an official part of the game. In this case, it may work (as some MMORPGs are currently attempting) to let every user upload their content to a central server and allow any user to try that content and rate it for quality. This way the good stuff filters to the top without the developer having to be the "bad guy" deciding whose stuff is good and whose isn't.

With respect to leveling, it should be possible to design the content creation system such that objects scale to a "level" number (whether that's a character level, a world level, or what-have-you). There'd need to be a way to select an object type, rather than a specific object, and the run-time content interpreter would know to dynamically replace that type designator with a randomly-selected object of the appropriate type (usually meaning an enemy mob whose level is within a reasonable range).

That's relatively straightforward. The real question is, who gets to create that content? The developer? Some chunk of code? Or J. Random User? (Or a combination, perhaps?)

[Zhaine]

Sounds like they're going for a smart mix of all three. And the big advantage here is that these three areas are multipliers to each other rather than just adding up:

If Arcen create a bunch of new content for AVWW, then this gets munched up and spat out as many times that level of new content (mechanics, environments, experiences) by the procedural generation the game uses. And then the players take the tools given to them and interpret the new content in all sorts of creative and unforeseen ways and "multiply" this chunk of content again.

So, speaking in some kind of abstract terms, if Arcen put n amount of "work" into new content, the procedural generation stuff does n amount of "work" also and the community contributes n amount of "work", then players end up playing with n^3 amount of stuff rather than just 3n. A beautiful thing!

[x4000]

Yeah, that's more or less what happens with ai war to a lesser degree. With avww, we're going whole hog for that.

[BobTheJanitor]

Well, to some extent there is no "doing it right."  And part of the cool thing about this is having the history of everything that happened.  I think that's really interesting with Dwarf Fortress, how histories like Boatmurdered developed.

By the way, thanks for mentioning Boatmurdered. It prompted me to go look it up. I'd heard it vaguely referenced before, and I knew it had something to do with dwarf fortress, but that was it. Now I've read through the whole thing on lparchive and it was great. I'll probably have to go try to play dwarf fortress now. As if I needed another game with ridiculously complex mechanics to learn my way through.

For more of that fascinating quality of people finding their own fun in games, I would recommend the yogscast minecraft videos, for the three people on the internet who haven't heard of them. They took the idea of a sandbox game and went all-in on it. Fantastic stuff.

[x4000]

By the way, thanks for mentioning Boatmurdered. It prompted me to go look it up. I'd heard it vaguely referenced before, and I knew it had something to do with dwarf fortress, but that was it.

My pleasure.  That was basically where I was a few months ago, and Keith kept talking about it and giving little references in AI War like the "pull the level" cheat.  Eventually I was compelled to seek it out.

Now I've read through the whole thing on lparchive and it was great. I'll probably have to go try to play dwarf fortress now. As if I needed another game with ridiculously complex mechanics to learn my way through.

Basically my reaction when I finished reading it in December.  I spent a few days figuring out the basics, and getting my first fortress set up.  It's quite a game.  It turned out that I liked reading about it better than playing it, but that's true of a lot of simulation games and me.  If you at all like simulation games, though, DF is basically second to none. 

Check out their wiki for some starting tutorials, though -- you'll need them.  I figured that after reading Boatmurdered I'd have a pretty good understanding of the basic gameplay at least, but that is now quite old by the standards of DF, it turns out.  Back then DF was only 2D and side-view, but now it has a Z-axis that makes for a 3D world shown through a series of 2D slices.  That, plus a ton of the gameplay mechanics clearly changed.

It took me literally 6 hours to select a starting location in the game.  I guess I'm just kind of anal about stuff like that, but man -- once I finally found a region without an aquifer and with this and that other attributes, things proceeded a lot more smoothly.

For more of that fascinating quality of people finding their own fun in games, I would recommend the yogscast minecraft videos, for the three people on the internet who haven't heard of them. They took the idea of a sandbox game and went all-in on it. Fantastic stuff.

Oh, neat -- I've not heard of that, I'll have to look it up.

[ShadowOTE]

I believe this is the link, for your convenience: http://www.youtube.com/watch?v=4UdEFmxRmNE&list=SL

[BobTheJanitor]

That would be the very first one, yes. There are quite a few videos now. I wouldn't suggest trying to watch them all in one go.  Something to watch on your down-time. But I wouldn't suggest skipping ahead either, you'd miss half the fun.

[ShadowOTE]

Yeah, i'm watching them in the background while doing other stuff (autoplay = awesome). They're pretty funny - the initial attempt to fight skeleton archers was great! Heh, and by "great" i mean "disasterous"

Remember folks: Losing is Fun!

And now for something completely different - http://www.youtube.com/watch?v=cglLJJ0Czo8