Newtonian Aurora - Galactic Map Question

[Steve Walmsley (OP)]

Well, that's true on the solar system level, because those are mostly 2D, however if you did add a full 3D galactic map you would have some big differences. The main one being that you could have invasion forces coming from stars directly 'above' or 'below' the ecliptic plane, completely bypassing listening posts or fleets based out on planets, moons, or asteroids.

That was another reason I was considering sticking with 2D. With a 3D galactic map, ships would still have to transition to a 2D system map, so ships coming from above and below might enter the system map from the same direction, if their systems of origin had the same xy coordinates but different z coordinates. There is definitely no way I am considring a 3D system map

Steve

[Steve Walmsley (OP)]

I'm having an "adding real stars" blitz at the moment . There are now 2850 real stars, including every star within 72 LY of Earth that is M9 class (small red dwarf) or larger.

Steve

[chuckles73]

I believe this was already mentioned, but being able to select the reference from which to base the flat projection would be my preferred method, if recalculating the projection wasn't a costly operation.

Otherwise, the point about how two stars far from Earth being the wrong distance apart won't be noticed is a good one.

[UnLimiTeD]

Can't you offer options at game generation?
Like one that flattens and distorts completely, one that flatts by 80% and then clusters the result, etc?
Given that you're already experimenting with it, the cost must be there. 
Though I certainly think full 3D would distract me from the game.
Reducing all heights by 80% and clustering would be my favourite, that way, one could still watch it from "above" for simplicity, only the jump distances wouldn't be properly represented.
Which I wouldn't care for in the first place, it should be possible to select a task force, then select a system on the map, and it tells you how long a jump would last at, say, 1000kms.
Or daily acceleration, or whatever.
What do I care if the jump duration matches the shown distance, if I can actually see both if I have to?

[Mel Vixen]

I guess the table for the new stars looks a bit different then the one for the old by having some new attributes  . So my question is if the procedural generation could also be manipulated a bit more on Game-start? I would like to have some way to manipulate the density of clusters, average number of stars in a cluster, average distances between clusters, Startype distribuitions etc. .

Densely packed clustes which are far apart from each other could lead to a easy and fast start with strong empires clashing later on for example.

[chrislocke2000]

As Steve had mentioned, the idea of trying to mix a 3d star map with a 2d system map seems like a bad idea.

Given the issues with trying to turn a 3d solar system into a 2d map I also agree that once you get the most common stars in place you are probably better off with what is close to a random distribution of stars further out. If you are just packing down stars from their 3d locations I would think you would need a system to spread them out where they would otherwise be pretty much on top of one another.

Oh and Steve great going on adding all the new stars out to 72 light years, although I'm sure I read in the press recently that they had spotted a nice new planet in the goldilocks zone at around the 600 light year mark.....!

[Steve Walmsley (OP)]

As Steve had mentioned, the idea of trying to mix a 3d star map with a 2d system map seems like a bad idea.

Given the issues with trying to turn a 3d solar system into a 2d map I also agree that once you get the most common stars in place you are probably better off with what is close to a random distribution of stars further out. If you are just packing down stars from their 3d locations I would think you would need a system to spread them out where they would otherwise be pretty much on top of one another.

I'm not using a random distribution. I am keeping their bearing and their distance but moving everything down to a 2D plane. Imagine every star connected to Sol by a string. I am keeping the string taut and moving the star down to the ground (or up to the ground) without changing its direction. They aren't on top of one another because they are moving outwards as well as down. The distribution of stars remains constant because although there are more stars as you move further out, they are spread over a wider area. This isn't a guess - this is based on generating the maps. Occasionally 2 stars will be too close to one another so I devised a way of overridng their bearing from Sol and moving them a degree or two left or right while maintaining the same distance. Within 25 LY of Earth I have had to adjust about 6 stars from 2-3 degrees so it isn't a major change.

Oh and Steve great going on adding all the new stars out to 72 light years, although I'm sure I read in the press recently that they had spotted a nice new planet in the goldilocks zone at around the 600 light year mark.....!

That might take a while . Every LY out from Sol has an increasing number of stars so it becomes exponentially more work for each extra LY.

Steve

[UnLimiTeD]

Do we really need all of them?
Because, that'll limit it to a very small distance if you take a small amount of systems...
Taht's primarily why I asked to be able to shut some off.

[Steve Walmsley (OP)]

Do we really need all of them?
Because, that'll limit it to a very small distance if you take a small amount of systems...
Taht's primarily why I asked to be able to shut some off.

It's called Real Stars for a reason . I won't be randomly removing some as that would then be random stars and there is already a random stars option. One possibility might be to exclude stars under a certain mass but that would take away a lot of the nearby stars that players will have heard of, which sort of defeats the object of a real stars map. For example, Proxima Centauri, Barnards Star, Wolf 359, Teegardens Star, Lalande 21185, Ross 248, Lacaille 9352, etc, are all low mass red dwarf stars.

Besides, there are no performance reasons not to have a large number of systems as the planets are only generated when someone first visits the system. The only difference will be the loading time of the galactic map. There is always random stars if you want a less realistic option.

Steve

[chrislocke2000]

Steve

Apologies for the tounge in cheek comment on getting further out on mapping all the stars, I realise what a huge laborious task task that is. Regarding your point on system generation and keeping as is with current aurora I was wondering if in fact players should have an idea of the planets in another system ahead of visiting them given our existing ability to detect planets at huge ranges even today. I guess this would mean a significant load at start up but systems could be thought of as fixed until visited and details around minerals, asteroids, moons etc plus aliens cold be left for generation once actually visited?

[PTTG]

Perhaps "Deep Space Tracking Stations" might have a rating where as you add more stations, you can detect smaller planets at greater distances. Of course, you have to visit the system to know anything other than the mass of these planets.

I'm getting the feeling that the player will explore far more star systems in Newtonian, since you could build a very crude ship that might burn through fuel like mad, but could effectively travel through several systems in only a few months. As it is, you need to have a small squadron on ships to jump to a system, scan, scan, scan, scan and jump to the next system.

[UnLimiTeD]

We can assume that this would only include superjovians being visible on distances above 10 LY, as "reasonable timeframe" is a fitting keyword here.
That that'd alone be interesting.

Most interesting planets we discover have a specific characteristic that makes them visible to us, like being very massive, very close to the star, traveling right in front of it....
Hard to just simulate this, though I suppose that entering the info we have would do no harm.
It does, however, include serious work.

[Yonder]

We can assume that this would only include superjovians being visible on distances above 10 LY, as "reasonable timeframe" is a fitting keyword here.
That that'd alone be interesting.

Most interesting planets we discover have a specific characteristic that makes them visible to us, like being very massive, very close to the star, traveling right in front of it....
Hard to just simulate this, though I suppose that entering the info we have would do no harm.
It does, however, include serious work.

Actually it seems like it would be really easy to simulate that. If you know the mass of the star and the planet (and the game well) you can make some sort of metric for the probability of detecting it every month or so. The game also knows when big planets are in between the Sun and their star.

The real problem is that would involve generating the planets (at least) of a lot of systems before any ships ever actually went there, which could be a memory issue.

[Steve Walmsley (OP)]

Steve

Apologies for the tounge in cheek comment on getting further out on mapping all the stars, I realise what a huge laborious task task that is. Regarding your point on system generation and keeping as is with current aurora I was wondering if in fact players should have an idea of the planets in another system ahead of visiting them given our existing ability to detect planets at huge ranges even today. I guess this would mean a significant load at start up but systems could be thought of as fixed until visited and details around minerals, asteroids, moons etc plus aliens cold be left for generation once actually visited?

Something along those lines might be possible but there are some issues. The first is that in order to be able to detect such planets I would have to generate them first. At the moment system bodies are only generated when someone enters the system for the first time. Having several hundred or even several thousand pre-generated systems would slow the game down considerably and it would take a long time to create the initial game. I couldn't simply leave out minerals, moons and asteroids, etc. without significant changes to system generation, not to mention that some moons are larger than some planets. A second lesser issue is that I would have to track which races know about which planets in which systems.

The major question though is whether knowing that system x had a large planet but you couldn't detect one in system y would significantly change your exploration of the galaxy. I would imagine players will generally visit all nearby systems using scout ships, regardless of whether a large planet had been detected, just to see for sure what was in the system. If nothing significant changes as a result of being able to detect planets at interstellar distances then it seems like a lot of work and a performance hit for no real gain in gameplay terms.

Steve

[Steve Walmsley (OP)]

I've reached my target of all stars within 100 LY of Earth entered into the Newtonian Aurora DB. There are now 4250 real star systems, 3850 of which are within 100 LY of Earth. Outside 75 LY or so there are far fewer red dwarf stars, presumably because they are hard to detect at that distance with our current technology.

Steve