Oolite Bulletins

Space games have a definite appeal, especially for smaller and homebrew developers. On the technical side, they're (relatively) easy to do: if you stay off-planet, there's no landscape to model, and not much in the way of obstacles to worry about. Yes, your player might plough into an asteroid or a station, but you don't have to worry about the walls, trees, rocks, furniture, barrels etc. that clutter up the average FPS. And movement is usually unconstrained in 3D: you don't need to check if there's a floor under your character, or provide them with ramps, stairs, ladders and so on. Plus, things players do encounter can be kept big, (largely) uncomplicated, and for the most part unanimated: in a fight between starships you don't need to show legs, or fur, or clothing flapping about.

And people are waking up to the possibilities of procedural generation. When it was used in the original Elite, it was mostly about cramming content into a tiny space. There aren't anything like the same constraints on memory these days, but people are realising that, if you produce the right procedure - something that could conceivably be done by one person - you don't need to employ dozens of artists, animators, and writers to create a sense of scale and diversity.

Of course, all that freedom can cause problems of its own. How do you get a game to happen, if everything is really far apart? How do you get players to involve themselves imaginatively in a procedurally generated game-world - to activate the "game-in-the-head"? From what I've seen of E:D, the first bit works well (although it does depend on multiplayer, which brings its own suite of issues ... E:D's NPCs appear to be worse than useless, openly flagged as bots, and serving only to rub "You're playing a game!" in the player's face), but falls down badly on the second. This latter failing appears to me to stem from two separate causes: 1) giving players access to 100,000,000,000+ star systems, 99.9999% of which are tediously uninteresting (and making travel across these vast distances trivially quick, totally undermining any sense of scale); and 2) the clunky Federation-and-Empire setup dragged over from Frontier, which seems to infiltrate every active system in the game, leaving players no room to make their own minds up about things (something else which multiplayer inevitably screws up: you might envisage yourself living in a baroque, gothic future, but if almost everyone else behaves as if they're 2016 WEIRDoes* with spaceships, it will spoil your party).

From which I'd conclude: that scale is great, but 100 interesting star systems, replete with cities, moons, mining camps, waystations, Oort-cloud eccentrics and so on is far better than 100,000,000,000 yawns; and that individual narrative freedom is crucial.

*WEIRD: Western, Educated, Industrialised, Rich and Democratic - see Charles Stross's Guide to World-building

Disembodied wrote:

... making travel across these vast distances trivially quick...

<grumbles> Tragic!

Disembodied wrote:

And people are waking up to the possibilities of procedural generation. When it was used in the original Elite, it was mostly about cramming content into a tiny space. There aren't anything like the same constraints on memory these days, but people are realising that, if you produce the right procedure - something that could conceivably be done by one person - you don't need to employ dozens of artists, animators, and writers to create a sense of scale and diversity.

For those interested in procedural generation and its challenges ("the right procedure"), here is a series of videos about rogue-like games that talk about it:

https://www.twitch.tv/irdc_usa/profile/highlights

The sound is often terrible, though.

E:D's NPCs appear to be worse than useless, openly flagged as bots, and serving only to rub "You're playing a game!" in the player's face)

Multiplayer solves two major problems: have the players to connect to your server so it's not too easy to pirate the game, and you don't have to program a clever AI (all you need is to refill your cannon fodder^H^H newbies tank from time to time).

Disembodied wrote:

<snip>...leaving players no room to make their own minds up about things (something else which multiplayer inevitably screws up...</snip>

See, this is something I think that Eve got right. Let people be dicks and heroes. Let the players define their own rules. This is one area where ED screwed up. By attempting to exert such tight control over the 'story' they have wound up with something bland.

I seem to recall David Braben saying that space was an obvious choice for Elite as it was essentially empty (as Disembodied explained). What was a strength of the genre is now arguably a weakness.

The sheer vastness of space and the wonders that fill it (however sparcely) have a romantic appeal that the dullness of vacuum does not. When I first played Elite, I dreamed of combining it with other games and I think this is something that may not yet have been done very successfully.

Wandering around a space station or planet is potentially just as dull as wandering around empty space but I feel there's a good space game to be made that actually spends less time in 'space'. When nearly all of the player's time is spent there then it's bound to get a bit 'samey'. If interesting things can also happen when not flying about then the game stays fresher.

There is, I think, an inherent conflict within the genre: freedom seldom makes for a good story (unless removed) but a small arena doesn't always sit well in a space game, and indeed robs space of part if it's appeal.

Astrobe wrote:

Multiplayer solves two major problems: have the players to connect to your server so it's not too easy to pirate the game, and you don't have to program a clever AI (all you need is to refill your cannon fodder^H^H newbies tank from time to time).

The ED AI is not that bad in tactical combat (at the current top level it would beat a top Oolite AI, and they plan further improvements) - and the scale of the ED universe (even just the populated part) means that meeting a player in combat outside of the simulator is rare without prior arrangement, so it's definitely necessary. NPC Combat is certainly easier to survive than combat in Oolite - but that's because NPCs generally attack alone or in small groups of weak ships, tougher fights must be explicitly sought, and running away is much easier (the running away bit also applying to fighting players: I've died once to a player and that was in my first week; the other two were in ships that outclassed mine by a bigger margin and I survived both times).

I think the core problem with ED is the multiplayer - in that people have really different expectations for an "Elite"-like game. Some people want a power fantasy where they rise rapidly from a pilot of a weak ship to god-emperor of the universe. Some people want a slightly more realistic power fantasy where they rise from a pilot of a weak ship to a pilot of a slightly tougher ship with bigger guns. Some people want to be able to just try out stuff "today I'll see how many Asps I can stack on top of each other". Some people want to tell a story, or a set of stories. (And many people want to do more than one, of course) In a single-player game, especially a moddable one like Oolite, you can do any of that - in so far as the game engine allows it - without difficulty. You can still do all of that in ED, but the game design means that you may keep running into people - or the consequences of people on the game design, more likely - that conflict. The only conflict that's not avoidable is the one between the lifestyles where getting the money is the fun and the lifestyles where getting the money lets you have the fun, I think - and that's handled about as well as it could be.

(ED definitely got better with planetary landings - much more varied terrain, the buggy is great fun to drive, and the scanner is a skill-based tool rather than a point-and-win)

Disembodied wrote:

From which I'd conclude: that scale is great, but 100 interesting star systems, replete with cities, moons, mining camps, waystations, Oort-cloud eccentrics and so on is far better than 100,000,000,000 yawns; and that individual narrative freedom is crucial.

I think so, yes - but I'd split the 100 systems up into clusters of 10-20 separated by (mostly...) boring stuff: that sort of distance can be important too.

In-system scale is the other problem: I think Oolite is closer to "right" than ED's realism here (though ED's supercruise is mostly better than Oolite's torus drive), but it's still a massive compromise whatever choice you make. Dense orbital complexes traversable at conventional speeds separated by relatively quick but mostly empty interplanetary travel is probably the way to go here (though a lot of detail-building)

Oh, and fixed and low-ish jump range. ED has the problem that any jump range low enough to make in-bubble travel interesting rather than point-and-go for a few jumps would be far too low to reach even nearby external points of interest in any reasonable time. Though I think the excessive speed is exaggerated, in some ways - it may be possible in a racing ship with practice to cross the galaxy in 24 hours, but there was a big collective expedition to reach the opposite side recently, taking 3 months and losing half the fleet, so it's not easy.

cim wrote:

I think the core problem with ED is the multiplayer [...] the game design means that you may keep running into people - or the consequences of people on the game design, more likely - that conflict.

Definitely. And it's something which I hope other games designers pick up on: the commercial pressure/on-line yammer demanding Multiplayer!!1!!1!eleventy! must be very hard to ignore.

cim wrote:

I'd split the 100 systems up into clusters of 10-20 separated by (mostly...) boring stuff: that sort of distance can be important too.

In-system scale is the other problem: I think Oolite is closer to "right" than ED's realism here (though ED's supercruise is mostly better than Oolite's torus drive), but it's still a massive compromise whatever choice you make. Dense orbital complexes traversable at conventional speeds separated by relatively quick but mostly empty interplanetary travel is probably the way to go here (though a lot of detail-building).

Again, here, too, is an argument that must be difficult for commercial designers to swallow: that the genre requires "boring bits" - long (but not too long) stretches of routine flight, to create the sense of scale (and, indeed, to make the exciting bits stand out). I think there's a role here for intelligently created mini-games, of the timewaster variety: non-compulsory things to fiddle with during the quiet patches ("calibrating scanners", or "adjusting inertial loading", or "reversing the polarity of the neutron flow", or whatever), which can have minor temporary benefits. It would likely be difficult to do well, but small, skill-based tasks, which can be started and abandoned easily, would do a lot to pad out these necessary sections of the game. The trick would be to make them feel like an organic part of the main game.

Disembodied wrote:

Again, here, too, is an argument that must be difficult for commercial designers to swallow: that the genre requires "boring bits" - long (but not too long) stretches of routine flight, to create the sense of scale (and, indeed, to make the exciting bits stand out). I think there's a role here for intelligently created mini-games, of the timewaster variety: non-compulsory things to fiddle with during the quiet patches ("calibrating scanners", or "adjusting inertial loading", or "reversing the polarity of the neutron flow", or whatever), which can have minor temporary benefits. It would likely be difficult to do well, but small, skill-based tasks, which can be started and abandoned easily, would do a lot to pad out these necessary sections of the game. The trick would be to make them feel like an organic part of the main game.

Since I acquired a Manifest MFD, I usually spend the first few minutes checking the prices at different stations, and to plan my route in-system to maximize profit. Other activities are: checking that a non-valuable is readied for dumping in case some pirates are really convincing, checking that a destination is ready for an emergency jump, checking that weapons are online (because I put them offline in Aegis), checking that the right kind of missile is ready for launch... this could really evolve into a checklist. Lot's of little things that you don't have to do but you'd be better to do. It might just be my former-flightsim-pilot background, but I think it's the kind of mechanics to look for.

Perhaps a simple thing to do is to make the F4 screen available in-flight. There's a lot of information there that can keep you busy for a while (checking email, news, available contracts, etc.).

If, as Redspear suggested, one wants more diversity in the game, I remember how Vega Strike handles station interfaces: it's almost a point-and-click game, because it's not obvious where the hotspots are (I don't think it was intended, though). One could go further and throw a Captain Blood-like in the mix. Something like this exists already in Oolite (as an OXZ). It'd be interesting if it could evolve into a framework in which contributors could "plug in" stories. Some parts could be made available in-flight as a simulated IM system (or just Email if the F4 screen can be made available).

About distances: We also have an OXZ that already solves the problem: Gates OXZ. One could generalize its usage to shorten the distances when necessary while preserving the sense of scale. I believe it could even replace inter-system jump (replace fuel cost with credit cost and keep the fuel for injectors or even maybe in-system cruise) (yes, it really looks like Stargate). Our Captain Blood could also collect unregistered gate coordinates.

Astrobe wrote:

About distances: We also have an OXZ that already solves the problem: Gates OXZ. One could generalize its usage to shorten the distances when necessary while preserving the sense of scale. I believe it could even replace inter-system jump (replace fuel cost with credit cost and keep the fuel for injectors or even maybe in-system cruise) (yes, it really looks like Stargate). Our Captain Blood could also collect unregistered gate coordinates.

You successfully ignited my mind, here are my thoughts.

Astrobe wrote:

Since I acquired a Manifest MFD, I usually spend the first few minutes checking the prices at different stations, and to plan my route in-system to maximize profit. Other activities are: checking that a non-valuable is readied for dumping in case some pirates are really convincing, checking that a destination is ready for an emergency jump, checking that weapons are online (because I put them offline in Aegis), checking that the right kind of missile is ready for launch... this could really evolve into a checklist. Lot's of little things that you don't have to do but you'd be better to do. It might just be my former-flightsim-pilot background, but I think it's the kind of mechanics to look for.

Perhaps a simple thing to do is to make the F4 screen available in-flight. There's a lot of information there that can keep you busy for a while (checking email, news, available contracts, etc.).

Yes, definite possibilities there! Perhaps, too, if there was more information available about the state of the ship - if it was possible to check (and adjust) the engines, fine-tuning them to the curvature of local space or whatever, which would provide a small speed boost (and possibly slow down system degradation, increasing the amount of time until the next service is required). Similarly one could fiddle with the cooling coils on the laser to slightly speed up laser cooling, or optimise the energy grid to slightly increase recharge rate, or balance the shields to slightly strengthen them ...

Of course, these tasks would need dedicated screens of their own, with gauges, network diagrams, and so on, and would need to be interactive: a lot of GUI work would be needed!

Disembodied wrote:

Of course, these tasks would need dedicated screens of their own, with gauges, network diagrams, and so on, and would need to be interactive: a lot of GUI work would be needed!

Interfaces I think is the tricky bit with this.

If the fast travel between places doesn't itself need some control work and watching for danger - things which could intercept you half-way - then it's probably better to keep the time taken by it short, and give the sense of scale a different way.

If it does need that, then you've got to be able to fit this work into a small instrument panel or HUD which can be watched at the same time as the usual sensors and visuals, and make it something that can very quickly be suspended.

cim wrote:

Disembodied wrote:

Of course, these tasks would need dedicated screens of their own, with gauges, network diagrams, and so on, and would need to be interactive: a lot of GUI work would be needed!

Interfaces I think is the tricky bit with this.

If the fast travel between places doesn't itself need some control work and watching for danger - things which could intercept you half-way - then it's probably better to keep the time taken by it short, and give the sense of scale a different way.

If one follows my suggestion about scanner range and visibility range, and combine it with the new camera features of the upcoming version, "watching for danger" could be a nice activity: I can imagine we could use some kind of periscope to visually check for nearby ships.

An other take on this is that Oolite specifically "slows down" when one is mass-locked and one cannot escape it. Other traders become an annoyance, which is sad. I think one could work on fixing that by turning those encounters into opportunities. For instance: buy the market data of the system the trader is coming from, or buy an import permit (with the Smugglers OXP), or offer/obtain in flight refueling...

If it does need that, then you've got to be able to fit this work into a small instrument panel or HUD which can be watched at the same time as the usual sensors and visuals, and make it something that can very quickly be suspended.

It sounds like "moar MFDs" to me (or that would be the simple answer to it). The display and interaction possibilities are limited, but that could be good enough. One way to make it work better could be: it seems to me that MFDs often come as a part of some equipment, so if one could have a key that primes directly the equipment to which the MFD is attached (one downside though, is that it sometimes is twice the work for OXP authors: they have to program the station interfaces and the MFD interfaces - perhaps rather than making station screens available in flight, make MFDs available when docked?).

The limited interactivity of MFDs maybe is not a bad thing by itself. A space game programmer rejected the idea to include mini-games because that would mean his game is boring. Oolite and others are trading and fighting games first, people don't come here to play Tetris. If anything, nowadays one can alt-tab or use a second monitor for that.

Astrobe wrote:

If one follows my suggestion about scanner range and visibility range, and combine it with the new camera features of the upcoming version, "watching for danger" could be a nice activity: I can imagine we could use some kind of periscope to visually check for nearby ships.

That could be a possibility ... if, instead of a torus drive, the player had to plan out a series of mini-jumps, in-system, that could involve scanning and checking before making each jump.

Astrobe wrote:

An other take on this is that Oolite specifically "slows down" when one is mass-locked and one cannot escape it. Other traders become an annoyance, which is sad. I think one could work on fixing that by turning those encounters into opportunities. For instance: buy the market data of the system the trader is coming from, or buy an import permit (with the Smugglers OXP), or offer/obtain in flight refueling...

This was where I saw the tinkering fitting in - during the slow overhaul of an Anaconvoy, when things are (relatively) safe. Ship-to-ship interactions would also be a good thing to do - especially if there was a dynamic galaxy, with events of note, and a slow percolation of news from system to system (i.e. no pan-galactic "broadcast network" - the only way for news to travel faster than light was for it to be carried by ships). Swapping news stories with non-hostile NPCs could give the player an important update that they might otherwise miss.

Astrobe wrote:

The limited interactivity of MFDs maybe is not a bad thing by itself. A space game programmer rejected the idea to include mini-games because that would mean his game is boring. Oolite and others are trading and fighting games first, people don't come here to play Tetris. If anything, nowadays one can alt-tab or use a second monitor for that.

I think it depends on what is meant by a mini-game ... to a large extent, the trading in the original Elite was a "mini-game", to provide a break from the dogfighting. And of course it's possible to do other things during quiet patches in the game - but flipping to another window to play Tetris or whatever is a bit of an immersion-breaker.

cim wrote:

give the sense of scale a different way.

If the ship computer place a series of waypoints into the predicted path at the same distance from each other then the current speed would be sensible by the time between passing waypoints. In this way the travel would be a bit more filled and longer tolerated imho. An idea for an OXP.

Disembodied wrote:

the only way for news to travel faster than light was for it to be carried by ships

Yes, gettig news from NPCs is a good idea, give a reason to stay on the space lanes. Moreover a new type of contracts could be born, where only the first who deliver the info will get reward. You can give away news in echange which is not in your planned path. [wiki]BroadcastComms MFD[/wiki] give ideal frame for this.

Astrobe wrote:

make MFDs available when docked?

In hud.plist could be a property of multi_function_displays to show when docked, like alert_conditions. In this way an MFD in the bottom part of the HUD could be usable in dock, and even could support different lookout in red alert.

The content of some MFDs are controlled by equipment activation (like scrolling in [wiki]CommsLogMFD[/wiki]) which is disabled in dock, but could be a new equipment event pair: activatedInDock and modeInDock so MFDs planned to be controllable in dock could be able to run the same code than in normal activated and mode events.

Norby wrote:

Yes, gettig news from NPCs is a good idea, give a reason to stay on the space lanes.

That's an excellent point ... well worth thinking about some more! What sort of information could players expect to get? Trading contracts, news on (short-lived) price fluctuations (a glut in Radioactives on such-and-such a planet, or a high demand for Liquor and Wines on another) ... maybe information about pirate activity in the system/local systems - specific ships to watch out for? There would need to be some way to make this advantageous to know, though - perhaps via Astrobe's idea for a "periscope"/long-range scanner? As well as letting traders try to avoid particular pirate hunting-grounds, it would be useful for bounty hunters pursuing named targets, if we could get even a few "persistent NPCs".