TAF and game build configurations

[Wildeblood]

- the stations are barely visible until planetary orbit is reached (and even then the compass change is needed to actually find them; also, there's more than one without it looking crowded)

That's my number one bugbear - the [adverb] small planets. I'd like planets to be ten times the size they are now. That would still be too small, but it wouldn't be so obvious just how very small they are. I suppose another possibility is just to place the stations well away from planet. Is there a technical requirement that witchpoint --> station --> planet rather than station <-- witchpoint --> planet?

[Disembodied]

So that gave me an unusual idea for a third option - non-linear display of space [...]

Aha! Very interesting ... with something so inconveniently large as space it's probably the only good way to simulate it in a game!

It would make things very interesting with regard to having multiple planets, and proper orbits ... it would also have implications for things like "spacelanes", and how pirates would find players (or how pirate-players would find merchants) ... maybe some sort of very long-range scanner, or in-system chart showing known and suspected positions of ships and stations, would be required too ... I know it's only hypothetical but it sounds like it could be really cool!

[Cody]

Yes, very interesting - I'm trying to visualise it (and get my head around non-linear space).

[cim]

Is there a technical requirement that witchpoint --> station --> planet rather than station <-- witchpoint --> planet?

Mainly AI-based, I think. The main trader AI goes in an attempted straight line witchpoint to planet vicinity, and then an attempted straight line from there to the station vicinity. Putting the station on the witchpoint side of the planet means that the traders don't have to then dodge around the planet when they do so, which considerably simplifies the state machine. (And also the flight in for new players, who sometimes have enough trouble finding the station as it is, of course)

There's also a bit in the aegis check code which essentially assumes that the main station will be within 3 radii of the planet centre, but because of the way the trader AI is written you'd probably get away with it if you moved it further out.

[Commander McLane]

So that gave me an unusual idea for a third option - non-linear display of space [...]

I like that idea! Really out of the box.

[cim]

The problem, of course, is that distorting space like that doesn't preserve angles. That gives some very odd effects. So for instance, this is Lave at 150km. The atmosphere hasn't been rescaled, so that gives you an idea of where the planet really is. If you set a course between the apparent surface and the real surface, and fly in a straight line, you'll end up crashing into the planet (though, when you get closer, you'll be able to see it coming)


Similarly objects near the planet look a bit oddly placed - this is about at 100km from the station and planet (station highlighted). It appears to be in far higher orbit than it is - from the witchpoint or sun with an ASC you wouldn't be able to see it, but it would appear to be several radii away from the planet.

It works for dropping distant ships out of sight smoothly, but planets/suns are just too big - shrunken as they are - for it to look plausible. Oh well...

[Disembodied]

Similarly objects near the planet look a bit oddly placed - this is about at 100km from the station and planet (station highlighted). It appears to be in far higher orbit than it is - from the witchpoint or sun with an ASC you wouldn't be able to see it, but it would appear to be several radii away from the planet.

That screenshot looks good to me ... it makes the planet look suitably large and the station look suitably small in comparison! If you can only see the station like this when you get close to it, I don't think it's a problem (at least not from an aesthetic point of view). It might cause some problems with the compass, perhaps, but we could handwave away any long-range inaccuracies ...

[cim]

Oh, that half of it does work. The objects only start getting shrunk when they're small enough already that the angular distortion is unnoticeable. Very nice for making orbital space look less crowded, and making it harder to see everything coming on the spacelanes (which should also speed the game up by drawing less stuff - or at least compensate for the extra calculations to shrink them...)

It's doing the same thing to the planet or sun which noticeably doesn't work.

[Disembodied]

It's doing the same thing to the planet or sun which noticeably doesn't work.

Could it be fixed with smoke and mirrors, perhaps? At certain distances and beyond, planets are displayed as points of light or just vaguely-coloured mini spheres? Stars could be covered up with essentially dimensionless but bright light sources. It might be fiddly to work the transition from mask to real object, of course ... am I misunderstanding something fundamental? (The answer to this one is probably "yes", I know ...)

[cim]

The misunderstanding is in "certain distances", I think. *hunts around for vaguely round objects*


The pepper (it escaped from a Giant Space Pizza) in the middle is the real position of the object. The one on the right is its current apparent position, as viewed by the observer in the ship on the left.

If the pepper were really at its apparent position, then a player flying on the course marked with the red arrow would fly past the edge of it. If it were displayed in its real position, the player would see themselves flying towards a landing on it.

We need the planet (as with ships) to be displayed at its real size when the player is at short range (for the purposes of ships, let's say this means "within military laser range"; for planets probably this has to be at least "within the planet aegis at 3*radius"). So, as the player moves towards the real pepper, the apparent pepper on the right also has to be moved towards the real pepper (until, at the edge of "short" range, they merge into one object)

So the player sees the planet rising from the side of their viewscreen into their path. The real weirdness starts if you have a binary planet system. Let's say that there's a moon in lowish orbit. From a long distance, the two appear to be independent objects, separated by tens of times their own radii. Up close, though, they're nearly touching. That's not what you'd expect to see as you get closer. The angular distortion isn't between the centre of the objects - it's between their edges.

Ships have the same effect, but it's not a problem there. A Coriolis may be a cubic kilometre of advanced alloys, but at 30km that still only subtends a couple of degrees on the view screen. By the time you're at 60km and the effect is becoming noticeable, it would be only a degree across anyway, so the fact that it's only half a degree across is barely perceptible. (It might need a little calibration to work well with a Torus station, or one of the big Kiotas, but it can be done)

So we can do this to planets and suns too - but we have to start when the planet is only a few degrees across on the viewscreen, which means the "certain distance" is a couple of thousand kilometres - too far for this to be useful.

[Disembodied]

OK, I think I see ... so if it's problematic to make distances non-linear, is it possible to make speeds non-linear instead? Make the real distances bigger, but have the maximum speeds much higher the further away the player is? I remember making some rough calculations based on real solar system distances: an in-game speed which would take the player from Earth to Venus in 30 seconds (which is too quick) would also mean a trip of 30 minutes (too long) from Earth the Neptune. So what would be required would be a system whereby maximum speeds increased over long distances. (Of course, Earth to Neptune should take longer than Earth to Venus, but both should feel entertaining - not too short or too long.)

How this would be done, I don't know ... maybe it could be based on the distance from the sun? Make top speeds in the inner system lower than in the outer, for handwavium reasons? It would mean that a modelled solar system would be required, of course.

[DaddyHoggy]

Ah, so the speed is a ratio of the curvature that the solar body impresses upon spacetime?

So a ship's maximum speed is true effectively in "interstellar space" (minimal curvature) but as you approach the sun (and later planet) the gravitational gradient of the spacetime bend forces the engines to expend more of their energy overcoming the gradient - which, given they have a maximum power output when expending quirium, means that inevitably their forward velocity is reduced.

[Disembodied]

Ah, so the speed is a ratio of the curvature that the solar body impresses upon spacetime?

So a ship's maximum speed is true effectively in "interstellar space" (minimal curvature) but as you approach the sun (and later planet) the gravitational gradient of the spacetime bend forces the engines to expend more of their energy overcoming the gradient - which, given they have a maximum power output when expending quirium, means that inevitably their forward velocity is reduced.

That would be an excellent piece of handwavium! How/whether it would work in a game would be the acid test, but your explanation sounds convincing to me ...

[Cody]

Ah, so the speed is a ratio of the curvature that the solar body impresses upon spacetime?
So a ship's maximum speed is true effectively in "interstellar space" (minimal curvature) but as you approach the sun (and later planet) the gravitational gradient of the spacetime bend forces the engines to expend more of their energy overcoming the gradient - which, given they have a maximum power output when expending quirium, means that inevitably their forward velocity is reduced.

<chortles> Music to my ears - and I only have two!

[Commander McLane]

How this would be done, I don't know ... maybe it could be based on the distance from the sun? Make top speeds in the inner system lower than in the outer, for handwavium reasons? It would mean that a modelled solar system would be required, of course.

The question that springs to mind immediately is what would happen to stations and other OXP-related stuff that is deliberately placed in deep space. Suddenly you would for instance fly through a Hacker Outpost's asteroid field with mad speeds. Or what about the Tianve Pulsar? You'd get there considerably faster than now. So, basically accelerating speeds towards the outer system would partially defeat the purpose of spawning things in the outer system in the first place.