Oolite essay: game lore, features and mechanics

[Redspear]

while your Masslock Reimagined is seriously cool (and I'll probably grab it for bug testing)

Do. It's a lot of fun I think but not particularly balanced at this stage.

[szaumix]

Redspears' frequent mods to tackle masslock reminds me that I have never considered Oolite primarily a hard and fast game. When I first got it, I honestly assumed there were supposed to be long periods of serenity, like a train ride through the space garden. And I don't remember the last time I felt bored playing Oolite. Then again, i may be a harder than average man to bore generally for reasons that are my fault.

However, I don't know if I said this to you in another thread or just thought it... the solution to filling the emptiness of open worlds is with crowdsourcing. I got the idea one time when I was adding to Broadcast Comms, I remember thinking, "why are there only ten responses here? There are hundreds of ways that the various flavors of person might express this!" -- which gave me the idea that developers should be making threads asking for mass contributions to fill the cavities of flavor and imagination in their projects. The lone developer is crippled by the fact that he must do everything himself, but as soon as it becomes a 5 man, 10 man, 50 man effort the increases are probably exponential.

Take any mod with a description, anything with goal or purpose or variety that is easily added to already existing code. Then imagine that not one (or two or three) developers, but FIFTY+ aegidian users had each thought up tens of variations of each subset of thing in that mod... suddenly the richness of the world explodes.

[Redspear]

By the way, I do have plans to make fuel more expensive/rare in some instances (e.g. if it's aquired from the local star then it's harvesting should be more dangerous in some systems and that might be reflected in the cost).

A whole lot of programming trouble for more realism...but I'd use it if it was made.

I was forgetting that I'd already made something similar

[Redspear]

the solution to filling the emptiness of open worlds is with crowdsourcing.

For something like  Your Ad Here that can work great but for something like ships, much less so I think.
Consider how much they vary in style, quality and power, then think of the gameplay balance issues; would it be wise to install even a quarter of them?

Many hands make light work may be true but then I think many heads tend to look in many different directions.

For the likes of Broadcast Comms however, why not?

[phkb]

For the likes of Broadcast Comms however, why not?

And BCC can be expanded in a couple of ways. One way is just to add additional responses to the existing ones. PM me your suggestions and I'll add them in.
Or, you can go the route Cholmondely and I did with the "Broadcast Comms - Digebiti Variations" and replace the existing messages with culturally appropriate ones.

It should be noted, that BCC doesn't impact on anything without sending a message in the first place. If you want to expand general gameplay messages, "Communications Pack A" is your best bet.

[szaumix]

So Broadcast Comms, Death Comms, CommsPack... comms generally. Mission descriptions, personality descriptions, actually just descriptions generally. Any text or image that does not require parametric definition/ relation to parameters generally. That sort of stuff, is what I meant. And I should have said a solution, not the solution. I mean if people are sticking to flavor-based contributions and not really crowdsourcing story/lore/ooniverse specifics that create constant heretical or directional divergence, what's the harm.

Ships are great but leaving their specs leaves balance a mess, I agree. Nevertheless, as far as I've been able to figure it; procedural generation and crowdsourcing are pretty much the only rational answers to attempting to fill a universe.

[szaumix]

Also I never thought of Oolite as primarily an eye-candy thing, and eye-candy seems like the slowest and least effective way to fill/enrich a universe. Having sights to see and things to explore is cool, MMORPGs have the dedicated development companies/teams to make an eye-candy world, I despair of even the possibility of Oolite doing so -- ever.

But that's OK. We skip from station to station as text menus, market to market, mission screens and clients etc. And we interpret simple models -- actually often they are archaic models (by even classic 3D standards, now) -- entirely according to their description / lore / specs / roles / dot type on our HUD/scanner. Eye candy is awesome and I pump it as hard as I can, but I figure Oolite's richness is text and imagination based first and foremost, eyecandy is secondary. Not sure if everyone feels the same way but that's always been my impression!

[Redspear]

Also I never thought of Oolite as primarily an eye-candy thing, and eye-candy seems like the slowest and least effective way to fill/enrich a universe.

Increasingly I find myself favouring low-res, hide the details visuals. I think that theres a level of detail that makes it obvious where the gaps are and sometimes 'dialing it down' gives less jarring results. For example, when watching a cartoon I no longer expect it to be a film and can start to forget it's limitations but when I'm watching a film I need a higher level of presentation in order to suspend disbelief.

Having sights to see and things to explore is cool

Indeed and I've got a few, very modest, plans for that.

[Switeck]

I took the very simple geometric shapes of ships to be a "necessary evil" of the complexity of projecting a complex force-field shield around them.

Also, looking at the evolution of planes and ships shows they have become smoother in appearance from the outside...so pushing that to the extreme would result in flat panel surfaces for spaceships as well.

[Cholmondely]

(from Stranger's essay, above, but with my emphases):

Three decades ago, a procedural generator was used to dynamically fill a memory space with content that was too small for manually created and customized game worlds. Now game universes created on the server side can already take up terabytes of memory. Filling such colossal amounts of memory with meaningful content manually is a hopeless task. The current procedural generation algorithms no longer just create unique configurations of solar systems in general terms. They are used to generate unique planets with their own relief, climate and unique biota. Amazing progress ... and mostly pointless. In the truest sense of the word. No Man's Sky. Space, in which a person, by and large, has nothing to do.

I don’t mean to say that procedural generation is a waste of time. It creates the foundation of the world, freeing the game coders from the unbearable volumes of monotonous technical work. But procedural generation only sets the stage for meaningful content. Plot, history, cultural layer, lore - call it what you want, without this there is no game. And it is precisely the persistent misunderstanding of this fact that is the reason for the fiasco of the Parallel Reality 002 project. Some kind of motivation is needed to stimulate the gamer to wander purposefully, not just at random no matter where...

...The game engine technically allows these systems to be as spacious as you like, and the procedural generator is unique. The question is still how to fill this procedurally generated kaleidoscope of locations with interesting meaningful activity. The procedural generator by itself, as we noted above, cannot do this.

Designing all these 2048 systems by hand, customizing them individually, saturating them with meaning and plot - well, you know, this is far too ambitious not only for a lone amateur, but also for a team of game developers. But the good news is that you don't need to meticulously design all 2048 systems. Cosmic wonders do not have to come across at every step. Interesting hand-sculpted locations, separated by routine procedurally generated intermediate points - why not? Let's drop it offhand. 16 individually configured systems on the map, maybe even only 8 systems - this is already enough to stimulate long-distance flights within the sector. And it will be a completely meaningful game goal as opposed to the meaningless infinity of No Man's Sky, where it doesn't matter where one flies.

Let's face it, the potential of this open world is poorly realized. And here the criticisms are not directed against the developers of the game, but against the community of addon oxp developers. At first, addon developers drew inspiration from the lore that grew out of Holdstock's Dark Wheel. In Ooniversum, in addition to the hermit asteroids, which are in the default game, there are deep space dredgers, generation ships, thargoid craft - all these legends of the old Elite have been brought to life. If anything, now, the legendary planet RAXXLA can be technically created - of course, linking its search with a non-trivial plot. There is a legendary space graveyard in the Tionisla system, there are three more systems with individual settings, and finally, there is a promising, but alas, abandoned project The Famous Planets. There are finally some epic missions like Trident Down. Alas, this is practically everything that now exists and almost all of this has become so outdated that it urgently needs at least a cosmetic update. Ooniversum is not attracted by the concept of world-exploration. The first meeting with the colossal Generation Ship, of course, is impressive, but only just - I saw it, took a screenshot as a souvenir, unloaded it into the gallery and forgot. This meeting gives neither answers to old secrets, nor ties to new plots. The pulsar in the Tianve system, the orbital cemetery in the Tionisla system - the same issue. The first time one looks it is interesting, but nothing more. There are no storylines for these locations.

So LitF is getting there. There will be flavourful stations for several of the Rough Guide Planets in v.1.0

I'm hoping that at least Lave & Digebiti will be systems which stick out as special.


Appendix
Looking at systems in the 8, a number come over as special from the perspectives of Lore and of oxp's:

Lave (Lave.oxp, Lave Academy, Monument)
Tionisla (TOGY, TCA) - neither really do anything other than provide eye-candy
Lerelace (Taranis HQ)
Riredi (Sodalite station - just eye-candy)
Aruszati (RRS HQ)
Tianve (Pulsar etc.)
Teorge (2 different Lores - the Elite/vanilla Oolite description and the conflicting Dark Wheel/Clym Angus description)
Ceesxe, (Ascension,) Xexedi, Onrira & Inera (+ Tianve) - TL14+
The Rough Guide planets

[Redspear]

(from Stranger's essay, above, but with my emphases):

Three decades ago, a procedural generator was used to dynamically fill a memory space with content that was too small for manually created and customized game worlds. Now game universes created on the server side can already take up terabytes of memory. Filling such colossal amounts of memory with meaningful content manually is a hopeless task. The current procedural generation algorithms no longer just create unique configurations of solar systems in general terms. They are used to generate unique planets with their own relief, climate and unique biota. Amazing progress ... and mostly pointless. In the truest sense of the word. No Man's Sky. Space, in which a person, by and large, has nothing to do.

I don’t mean to say that procedural generation is a waste of time. It creates the foundation of the world, freeing the game coders from the unbearable volumes of monotonous technical work. But procedural generation only sets the stage for meaningful content. Plot, history, cultural layer, lore - call it what you want, without this there is no game. And it is precisely the persistent misunderstanding of this fact that is the reason for the fiasco of the Parallel Reality 002 project. Some kind of motivation is needed to stimulate the gamer to wander purposefully, not just at random no matter where...

OK, so (simplifying) the limits of procedural generation and the lack of plot?

I'll address these one at a time.


Procedural Generation (PG)

Amazing achievement in 1984, groundbreaking etc. I think most are agreed but what was it used for exactly? To generate the 'galaxies' most of the info for which was essentially 'fluff'.

Government, economy and tech level were all highly relevant in game but also displayed a correllation which, however reasonable, reduced variety. Population, radius, inhabitants, and descriptor blurb meanwhile, all meant nothing once you'd read them. So the PG had actually 'created' much more materuial than was really being used in any significant way. Might it not be more interesting if this info was used? Further, might this not doubly be the case if each additional info category was not correlated with some other? (I'm working on it, I reallly am...)

PG was used within elite to create random danger level and random economy type, combined with a correllated tech level this was entirely sufficient to promote trading navigation strategy but sadly little else. Oolite's contracts further exploit this content but trading aside it offers little.


Plot

The original plot within elite was, I would argue, the missions. Impressive as some of them might have been, they weren't exactly many. I think they all added story of some sort and nerarly all of them added gameplay or at least goal-oriented motivation for the same. Stories have something in common with trading however: they each reach a threshold beyond which they can quickly become meaningless.

The first meeting with the colossal Generation Ship, of course, is impressive, but only just - I saw it, took a screenshot as a souvenir, unloaded it into the gallery and forgot. This meeting gives neither answers to old secrets, nor ties to new plots. The pulsar in the Tianve system, the orbital cemetery in the Tionisla system - the same issue. The first time one looks it is interesting, but nothing more. There are no storylines for these locations.

This is visual 'fluff': no less great than the planet descriptions were within elite but afterwards entirely forgettable.

So, what to do about it?
Adding significantly more missions is often either labour intensive or requires a somewhat generic methodology e.g. Random Hits. Meanwhile, the most variable thing about the approx 2,000 star systems is currently the planet texture. But then the planet texture is procedural too...


Using Existing Procedural Generation to create Plot (sort of...)

If, outside of the missions, the early/beginner's game is the most engrossing part (new experiences, upgrades) then how to maintain the thrill?

Upgrades should likely run out or become meaningless by their very availability. Weapon Laws tried to address making their aquisition more interesting (effectively a mini-mission to get one) while New Lasers tried to do something similar with oxp equipment. Weapon Laws actually reversed the inherited trend of finding the more powerful items more often in the safer systems (a PG element that wasn't sufficiently checked for gameplay value IMHO).

New expeiences? Well, there aren't many are there? Lots of ships, yes, lots and lots of them if you want but new experience? There are the capital ships I suppose and there are some oddities to fond but little to mark a system as rare or unique that actually adds gameplay.

What if a feline system contained more hunters or a rodent system more scavengers?
What if a legal rating was inversely useful when purchasing missiles, mines and bombs?
What if populous planets meant (much) busier systems, greater demand and higher quantities for sale?
What if radius influenced viper patrol frequency (shorter distance to witchpoint)?
What if civil war systems regularly had aegis raiders?
What if a love of tourists meant warnings more often than fines?
What if almost all of the above was not correllated with each other?

None of these provide missions and none of them necessarily promote provide plot but if you try implemeting all of them then contarct runs alone have just become a lot more interesting. Now, with sufficient reputation, change the contracts offered from the mundane and generic (e.g. zero-G hockey holo-vids) to the highly sensitive and specific (e.g. plans of pirate network hideout - no loss if player blown up and prevents hacker access) and you now do have plot.

I appreciate that the above isn't wildly different tha the preexisting contract system but do you see the potential for not only sequential missions but also how it could make story elements out of all the the suggestions above?

[Cholmondely]

Using Existing Procedural Generation to create Plot (sort of...)

I'll need to go back over what you have written and have a ponder. But, personally speaking, I'm not really into the contracts. On the other hand, I'd love to see more made of TOGY & the TCA for example. It would be super if they were real destinations rather than eye-candy. Ditto with the Generation Ships. Something to explore. Things to find. Things to do.

As I mentioned just above, we have Lave, Tionisla, Riredi, Lerelace, Tianve & Aruszati. In each case, the oxp's just add eye-candy with the exceptions of Lave Academy, the Lerelace/Taranis mission and the RRS missions at Aruszati (which I presume are identical to those available at any RRS station). Again, I'd rather see things fleshed out with meaningful encounters, that sort of thing.

[Redspear]

personally speaking, I'm not really into the contracts

Fair enough but rather than deliver package x to destination y while dodging assassins z it's essentially still a contract if it's framed as for reason x head to destination y facing challenge z.

e.g. Head to Quitiri where our agent holds information that may be crucial to disrupting the thargon drone fighters and could help us to turn the tide of the war. Be careful to travel in utmost secrecy yet with some haste however as there are other organisations that may wish to reach him before you can do so.

That would essentially be a parcel contract with assassins.
It follows the same basic format (destination/reason/time limit/assassins) and only the description is radically different. The 'parcel' in this case is the player themselves, either as an 'agent' or messenger. Connecting it to follow up missions (preferably some distance apart, with consequence beyond financial reward) and plot you'd have I think, however crude it might be.

As I mentioned just above, we have Lave, Tionisla, Riredi, Lerelace, Tianve & Aruszati. In each case, the oxp's just add eye-candy with the exceptions of Lave Academy, the Lerelace/Taranis mission and the RRS missions at Aruszati (which I presume are identical to those available at any RRS station). Again, I'd rather see things fleshed out with meaningful encounters, that sort of thing.

Understood but then each is a one-off. It's a lot of work for minimal benefit i think. Imagine if we were to achieve "meaningful encounters" in each one of those systems, the game would still have the same problem most of the time. Generic encounters could work but then 'generic' and 'meaningful' don't always combine easily. Don't let me stop you, you just might have to show me how it's done without leaving every other system dull as dishwater.

Acknowledgement: there was a good attempt at what basically amounts to a cool story. But Acornsoft was no J.R.Tolkien, and the Elite game was no middle earth. You can't even say it was just limited by 80s tech -- it was also limited by the fact that games just weren't that serious a thing back then. Games were a casual hobby for kids and nerds and Elite was already pushing the boundaries of both of those with its story and scope.

I quote the above to borrow the analogy as I think it might be a quite servicable one.

The procedural generation was certainly no Tolkien and it's output was no Middle Earth. However, whilst the former was all but inevitable, the latter might just have been a little closer than is immediately obvious.

Middle Earth (by popular concensus at least) is rich in both detail and flavour but also remarkably cohesive and distinctive. Note that excessive detail can actually hinder cohesion (like carelessly installing every oxp) and too strong a flavour can render something monotonous (e.g. if Thargoids were everywhere, all of the time).

Detail can be a PITA but is more time consuming than difficult. Elite shipped with a fair amount of detail compared to most games (even post 1984) and oolite expanded on that somewhat further; so I'd argue that we have at least a servicable quantity to work with. Flavour is essentially the nature of the game I think and what attracts so many (if many is the word these days) here; so again, I'm going to say that we already have at least enough for it not to be a problem.

Cohesive and distinctive? These two haven't aged so well by my estimation.

• A bunch of PG fluff that is irrelevant/unused without oxp intervention (inhabitants, planet descriptions)
• A highly influential but now archaic game model (questionable markets, masslock designed for player-centric traffic, rinse and repeat gameplay)
• Poor planning re long play despite the apparent desire for such (best equipment in safest systems, extreme laser disparity)
• Lack of variety contrary to the lore and/or as a consequnce of 1984 memory restrictions (star + planet + station + traffic, only one of which varies significantly and even that is debatablte at times)

Rather than think of 1984 elite as purely 'of it's time', I prefer to consider it as a proof of concept for Oolite (of course, it was never intended as such but that's hardly my point).

• So...
  • Texture those ships and planets - the former did wonders for variety of systems but could perhaps use a little help
  • Expand the use of commodites rather than the length of the list
  • Adjust prices so that progression doesn't jump from slow straight to fast
  • Consider where non-player centricity improves gameplay and also where it doesn't - it's the illusion that's desirable more than the consequences I'd suggest
  • Have inhabitant type (and occasionally planet description) mean something, even if it's only system cosmetics or ship chatter
  • Add a little non-player centricity (already done) so that the player can feel like they're part of a galaxy that doesn't revolve around them (it does of course but we can do much better than elite in terms of hiding the fact)

Note that at least two of those have long since been done and, I would suggest, add quite a lot to the experience of the game. The others can in many cases be tackled by oxp and I've not been the only one to do so.

So perhaps it's not so much a case of freeing ourselves from elite's design shackles but rather exploiting modern computers' abilities to finish the job.

Maybe I'm just reframing what's already been said but when it comes to solving a problem that can sometimes radically change where you head next.

[Switeck]

Middle Earth (by popular concensus at least) is rich in both detail and flavour but also remarkably cohesive and distinctive. Note that excessive detail can actually hinder cohesion (like carelessly installing every oxp) and too strong a flavour can render something monotonous (e.g. if Thargoids were everywhere, all of the time).

I can second the Thargoids end up everywhere if you have lots of OXPs installed! Lots of OXPs add more, basically none remove them.

Galactic Navy OXP (with its broken license issue) is one of the worst offenders of too strong a flavour rendering something monotonous -- if it's installed, Galactic Navy ships are nearly everywhere. This can screw up campaigns/missions, as these Galactic Navy ships act more like bounty hunters/police than unconcerned scenery.

Anarchies OXP is a bit too much in how much it changes Anarchy and some other systems.
Commies OXP is a bit annoying, with their big+slow cargo haulers and signs that mass-lock you all the time.
Dictators or its subset OXP - Imperial Astrofactory, is comparatively minimalist.
Feudal is likewise limited in changes.

This leaves Confederacies, Democracies, and Corporate States "wanting" for their own flavor OXP.
...But some other OXPs directly/indirectly affect them. I think Democracies are where slaves can be freed?

Destinations at various system types are likewise overwhelming if lots of station OXPs are installed, especially at highest tech levels.

It is a LOT of effort to stylize a particular system beyond adding a station and/or ships based on tech level, gov. type, and/or economic type...plus if only 1 system is done that way, that's a lot of work for a place few players will ever visit.

[Cholmondely]

I find that certain systems get my imagination going: and with the FPO planet textures, this is now further accentuated!

FPO Teraed
Teraed became one of the largest hubs in the north-east quadrant of Galaxy 1. Travellers to Tianve often use it as a stop-over and trading point. Nowadays the planet is home of the largest aqua city in Galaxy 1, easily visible from space. The Galaxy Dome with 3.2 million seats (estimated) is one of the biggest Zero-G Cricket venues in the sector. (Famous Planets OXP)


Teonan (Galaxy 2)
Teonan is a tourism hub and the capital planet of the Teonan web region. 'Teonan Prime' is the enormous capital city built over the site of an ancient impact crater. The city itself is built from the light material 'Plumium' which is five times stronger than concrete, yet remarkably elastic. The city spans a radius of over three hundred kilometres, encompassing both land and sea like a spider web. In the center of the web is the 'Teonan Forum', an enormous building housing literally everything from the spaceport itself, local government offices and the hoopy nightclubs which Teonan is also famous for. The climate is pleasant and calm. (Famous Planets OXP).


Looking specifically at Teraed, to what extent is it possible to replicate the underwater aspect within Oolite? Or to restrict landing to water-capable ships such as the Moray? Or to create non-text-based adventures on the surface?

I understand that one can easily include snap shots of the landing sites (a la PlanetFall).