How about the "Space In Between".

If you ahve ship jumping from one system to another there are two ways to do it: Wormholes, or Warp Drive. It doesn't matter what you end up calling them (jump nodes, FTL drive, or whatever) the function of them is what is important.

With Warp Drive, ships are free to travel anywhere at any time. In this case, there is no real in between spaces to explore.

However, using Wormholes would only allow you to travel between known entry and exit points. This allow for the possibility of new wormholes to be discovered that take you to new places, without having to increase the physical size of the space.

If exploring new locations (star systems, planets, etc) is either (or both) risky or expensive (in money and/or time) then this makes for a realistic reason that not all places are explored, and that places unexplored are the more dangerous/hard to do.

What you can do then is to have new locations slowly explored by nearby systems (nearby in terms of the network of wormholes) over a period of time and thus opening up new areas for the player to explore.

Or, if you allow wormhole entrances to be discovered over time, but then allow the player to volunteer to explore the other side, this give a natural and constant source of missions for the player.

As an aside: There is an interesting idea that NASA uses that they call the "Interplanetary Superhighway" (http://en.wikipedia.org/wiki/Interplanetary_Transport_Network). Although are not Wormholes in the typical sense, what if you stretched science (like all good sci-fi :D ) so that only through these Interplanetary Superhighway links a ship can achieve superluminal speeds (these links would also exist outside out solar system between stars). In reality, they are just the most efficient path between gravitational bodies and nothing to do with real wormholes, but a very cool idea anyway.

But back to the topic at hand:

As for getting the player to discover the "Unknown Space" missions, if you have a significant reward (and the risk being high) and have a procedurally generating algorithms place links to these missions in Known space (adverts for "Brave Space Explorers" or such), then this allows you to have these Unknown Space missions knowable to players in the Know Space and give them incentive to become the "frontiersmen".

If you also allow known systems to take on a different characteristic (say low paying transport jobs), then it will give the player a sense of the receding frontier, and give them a subtle psychological push towards the unknown (you could even have the player meet grizzled old spaces lamenting the lost frontier and the encroaching strip malls).

By itself, it seems it'll be problematic for planned content (the other replies did a much better job of giving thoughts on that than I could have). To solve the same issue, why not go with area of unknown space explored instead, rather than time or distance? They are still guaranteed to find your planned content but without some of the problems mentioned by others that going by time introduces. Even if it were just the events that were based on time, and not the actual planets etc., you would still have the issue of just waiting around x minutes, and then go explore and find the first planet to get your next event.

Also, this probably occurred to you, but not telling others will only work for so long. Look at every other game ever - once someone figures out a game mechanic the world quickly finds out.

I like the idea being able to explore the great unknown, and having the interesting bit be where ever I am. But I’m not keen on the time based factor.

I region based system would achieve the same effect but be more interesting I think. When exploring the frontier when entering an unexplored area for the first time it generates a region of space most likely from some preset designs with procedural generated contented and random factors to make them unique. Regions of space would have different sizes and features unique to them. Players wouldn’t know what region of space they were in until they had started exploring it. For instance I might be currently travelling through a region filled with dangerous spatial currents, energy damping fields, and rouge planetoids. The game would provide events appropriate to that areas but also decided how big it is. So it might only take me 10 minutes to fly across this small region and once I reach the other side of that region the game would generate a new region to explore. The new region might be a vast area of space controlled by race of cybernetic creatures with on host of challenges. It also means if I think it to risky I can just turn around and try my luck somewhere else.

Regions will allow you to achieve the same results as time based event generation without the time sink. And allow you to group similar content together, let the player move onto other places should they not like the content in one place, and make creation of a generated map of space and expanding known space easier.

Also one idea you could use with region is the idea of the Heechee spaceships from the book "Gateway" for exploration. They were ships discovered on an alien space station by humans who knew how to use them but not how they worked. This meant they knew how to enter coordinates into the ship and take off, but no clue as to what the coordinates meant. Teams of explorers would go off into the complete unknown with no idea where they were going, how long it would take to get there or what they would find when they got there.

If you incorporated a Heechee type drive as the means of exploring the frontier in which you can swap coordinates with friends it would add community aspect of the game. Players could then exchange stories of adventure in the same regions even though they are playing solo. It could also be used if you wanted to host contests and events.

Maybe if it was something like reading local newspaper headlines and you could choose to investigate a situation. But in general, it doesn't really strike me as being much different than a random combat encounter generator. I think having a situation suddenly pop into existance a couple feet from you with no explination of causality will quickly give the same feel.

Quote:

Original post by kseh Maybe if it was something like reading local newspaper headlines and you could choose to investigate a situation. But in general, it doesn't really strike me as being much different than a random combat encounter generator. I think having a situation suddenly pop into existance a couple feet from you with no explination of causality will quickly give the same feel.

You don't need to have them as completely random and unconnected. The game can keep track of various things and NPCs you have dealt with in the past (the amount of memory is not as restrictive as it once was), even if these were initially randomly created.

Using these are the basis of a procedural generation algorithm you can then ahve the players past actions and associations "cause" (well appear to cause) these new events.

For example:

Say the player once did a delivery to a Guy that worked for Space Adventure Inc. (an exploration company). Later when the game states that a "random" mission to unknown space is available, it looks at the NPCs in your past and sees that you had a connection to this company. It then creates an event where by the same NPC calls you up and asks if you are interested in a job (and makes mention that you did a job for him once).

Of course, not every random mission would be generated like this, many could just be random NPCs (which might be used later). If the player has repeated interactions with a certain NPC (because they keep taking jobs from them) then you can increase the frequency that they are involved in these random events.

If your algorithms are smart enough, then you can actually weave a story like this through procedurally generated content. You can also do it with scripted events.

There was a party game that went around a few year ago (and is still fairly popular in some circles), called: Mad Libbing.

The idea of this is that you ahve sentences where some words are mission. You are given a few blank spaces (indexed with a number and what kind of work should go there) and are not allowed to know the sentence (until later). You fill in the words as directed and then place them in the sentence which is then read out (supposedly to everyone hilarity).

But this could be used for games too. If you ahve "Roles" within the story that you would normally construct a specific NPC for, you could instead leave it blank and fill it with an appropriate NPC that had been created by the procedural generation.

So if the plot called for a friend of the player to be killed, then the game could keep track of such an NPC, and then when the appropriate time is right for this event to occur, the NPC the player has developed a friendship with is the one that is killed.

Basically you would need 3 things: Names, Time s and Places to be filled by procedurally generated content. These would ahve tags (or such) associated with them that would guide the game to assign the right NPC, Time or place into the needed slot.

It would take a bit of work to make this. It is not simple. But if done right, it would make truly dynamic games possible, with both scripted (semi scripted really) events and completely constructed events from the actions (and procedurally generated events) that the player take and the decisions they make.

It is one thing to rob a guy, but then another thing to see him later begging in the street (the game takes note that you robbed that particular NPC, and then when it calls for a beggar, that NPC's name fits the needs and so it is selected).

Quote:

Original post by ID Merlin If players can't tell the difference, it will make no difference, correct?

Technically you're right, but as popsoftheyear points out players are eventually going to find out, either through replaying or by comparing notes with friends. At that point I'm not sure if the thrill of the discovery is a letdown or not. It could be undermining, sort of like combat systems that automatically scale difficulty which leave you asking if you won the fight because you were lucky or because the game took pity on you.

Quote:

Original post by Atrix256 mystery is good, players fill in the details with their own imaginings (: A lil off topic, i think this is why old games like zelda 1 and metroid were so fun. They didnt explain anything, you appeared in a world of unknown size and unknown possibilities and "went to it", figuring it out as you went, never knowing when the next new thing would come around and open up whole new places to explore.

I agree with you there. People will fill in explanations and will likely enjoy mastering the system as they uncover it. Speaking of old games, I'm using Starflight as inspiration and that's what brought up this thread. In Starflight there's coordinate based treasure hunting gameplay, where you search about for artifacts, receiving clues from aliens or ruins you stumble upon. But I don't really see it working in today's walkthrough age, where it's all too tempting to just pull up a list that spoils the surprise of where everything is.

In considering exploration, I've been mulling over what exactly makes it fun. Wandering aimlessly isn't fun, nor is uninterrupted repetition. But if content appeared near you, it would make both your universe unique and your experience controllable from a design perspective.

Quote:

Original post by Hodgman In my degree, we had to build this kind of sim using a poisson distribution to determine the amount of time until the next event. It worked very well and is tweakable.

I knew one day I'd kick myself for not paying closer attention in my Stats class. [smile] This is very interesting. I've not thought about making the analysis this sophisticated but it at least gives me a direction to consider.

Quote:

Original post by hiigara Have you never been in a situation were you felt "I just want to go out today, but don't know where", and then you get outside your door and think "left or right?". And it really doesn't matter if you go left or right, you just want to get out of your safety zone. The decision of which direction you leave the known universe is not a matter of skill. If they find a new planet based on how long they have been traveling, and the new planet is made persistent from then on, then they will be more than happy.

I like your example, but there's one point I think I'd better clarify: The player is going to have some overall understanding of the probability space before they start going left or right. Large scale territory is relatively fixed, but what you find in it varies within a range. Finding a habitable planet which would net the player a handsome reward, for instance, would be based on investigating G type stars. But what's on the planet or hanging out in the system, facts which the player has no logical reason to believe they should know beforehand, would be random and potentially time based.

So just as you might expect a certain range of things to happen based on what neighborhood you live in, certain locations should clue the player in to possible encounters in advance.

Quote:

Original post by Robert-Glen If your game is tracking 'play time' will there be scripted events between random time frames? Example being say between 10-15 hours of total play time a scripted event plays out but how something happens depends on what you've accomplished within that time.

Yes, more or less in the form of large scale events like wars, discoveries and famines.

Quote:

One scripted event could have multiple branches and because everyone plays differently would offer a different game experience for each, they also offer replay value. I'm just curious ^^

You're right. I've been working toward making different outcomes based on a series of dependent stats. In some cases, it's as simple as a change in the text that describes an outcome, but in others it should (hopefully) generate different entities in the play area or changes to game rules (as would be the case of a technology discovery).

Quote:

Original post by Trapper Zoid The main danger I see is if the reward system provides a net benefit to avoid exploring. If your system chooses interesting events to be the discovery of useful planet types, for example, then it would be prudent to game the system.

I hadn't thought of that, although I was thinking that there'd have to be some activity that would drive what I'm thinking of at the moment as the "discovery timer." Maybe it's that you have to be peeling back the map borders, or you have to actively be using some skill.

Quote:

If I'm wanting to scout out mining deposits and I know that a rich mineral planet crops up far more frequently every half hour, I might as well go and do something else while the timer ticks down then go find a series of boring useless planets, increasing the distance my ships have to fly than if I wait and create the next juicy planet the next system across.

Good point. Those mineral deposits would have to depend on detecting that the player is actually looking.

Quote:

Play balancing would counter that, however. I could see it working well if you had separate pools of events for exploring and those on a timer for "interesting events". Things like bounty hunter attacks are far more suited to timers, while finding "special planets" are not.

I had been thinking of artifact sites and what's in them, sort of akin to the chests in a fantasy game. Whether or not you run across an interesting ruin could be time based, but I'd have to be certain that the time was somewhat irregular if I did do this.

Just to clarify, take this example: You can choose to land anywhere on a planet. If I generate artifact sites, I've got to tweak the generator algorithm until the density is appropriate (not to mention what you find). If I make it time based, then where you land isn't really as important. I can then place emphasis on gear (detection equipment) and staff (trained archeologists) and generate according to how you've chosen to use limited resources.

Quote:

Original post by Drethon You might want to consider having this system generate exploration quests. The actual area to explore is static but the trigger for the quest to be given (a simple NPC wants something over thattaway perhaps) could be generated based on time.

That's possible, and maybe far more controllable than what I was thinking. My misgiving would be that you're in an open ended game world, so even if the quest trigger is timed nothing prevents you from going there ahead of time.

Quote:

Original post by Feldrin Why not make the border regions have both fixed and time-variable encounters? It should help some way in disguising the time dependency of time-variable encounters.

I like that idea. I was thinking you'd get this anyway with the encroachment of civilization (frequency of prospectors, bandits or explorers, for instance). It might work to have procedurally generated and time-based finds, maybe with the proximity of the procedurally generated ones modifying the time-based ones. So if you land on a planet and already know there are n locations of interest, and that's a threshold for discoveries at that location, then maybe the time-dependent stuff doesn't even kick in because you're expected to go where you know stuff is.

Quote:

By placing some fixed locations with more fleshed out and complex missions you strengthen the spirit of exploration and a feeling of diversity. A time dependent could do this as well by simply having missions which differ in complexity and such pop up, but a fixed situation creates more of an anchorage for you and lets you tie it in with other fixed quests.

Good point.