Better Exploration

[FluffyFreak]

With lwho's System exploration #3201 we've finally got the possibility to explore star systems.

The cheeky note: "Maybe, someone at the Lua front will soon invent a new ship component that will be required to explore a system..." is well timed as I just started a branch the other day to add exploration probes.

My plan was to filter the SystemView and SystemInfoView screens to show very little data except for the presence of Gas/Ice Giants and Stars in a system.
When you entered a you'd get find out if there are rocky worlds, perhaps some of the larger Moons but to actually explore the system you would have to visit each world.

Obviously visiting them all would be extremely tedious, not to mention difficult with our crazy refuelling system. So you fire off probes to each world to do the scanning for you, these occupy missile rack space but they can identify the Moon systems around Gas/Ice Giants and do all the identification work you would otherwise need to, as well as get there a lot faster. You get paid the same for identifying a world by probe as you would to turning up their yourself.

Does that sound like a good idea?

[impaktor]

Seeing gas giants in unexplored systems might be too kind, since to actually make those observations you need to study a star system over a long period of time to observe the wobbling of the star. The way it is now, it just says "unexplored" (I think), and I have no problem with that. If showing some information it needs to be vague, I think: "Unexplored system. Observations indicate/hint this and that".

If the player sends out probes once they've entered the "unexplored system", then will this not be just boring and automated? I guess there will be a manual way to explore the planets as well? So a rich player can afford to use probes, but when starting out in a ship with a decent range, the player actually has to go to each system. I think it would be good if the probes actually look for something, so once a probe report a positive finding the player needs to go there.

Will we require an atmospheric sample of earth like planets, and the system star? The goal of the mission is to find a habitable world? When that happens the player gets a bigger reward?

Hmm. Problem with all probe exploration vs. manual exploration is that there is a massive difference in work put in, so you would like manually explored system, and probe explored to give different rewards, perhaps?

I think going to each world by probe would require there to be a manual way as well, in case a probe is lost, or you didn't bring enough probes.

Would there also have to be some "half" explored state of a system, or what happens if the player only explores a subset of the planetary bodies present in the system?

Different faction might compete for system, to establish outposts? Feels like it's getting a bit too advanced now... I should stop.

[FluffyFreak]

I was just trying to think of ways it might work in the future, something interesting for the player rather than being 100% realistic.

The idea for the probes was actually to avoid the "boring and automated" part so I imaged it this way;
You jump into a system and have two choices:
A) manually visit every single planet in turn,
B) launch probes to each planet - then visit the interesting one(s).

Which is the boring one? I just want to remove the part where you visit 4 out of 37 bodies in a system before you run out of fuel and have to return 9 more times to complete the survey and then only find the interesting planet with indigenous life which gets you the big payout at the very end of the last visit...

So the probes would just go into orbit around a target planet and "scan" for a while finding Moons and identifying the basics about the planet.
They can't land, or sample return or anything else, so if they're more to find out the basics then the player would have to do any landing etc, I was just going to try it and then decide what to do about that once I'd written the probe part.

There's different amounts of effort required for manual vs probe exploration, but the player has to pay for the probe in advance and might not find any interesting star systems for a while so punishing them for simply being prudent and using a game mechanic seems mean. It doesn't matter how you investigated a planet (notwithstanding landing on it of course), you still know the same things about it so the payout should be the same.

Systems would be explored or unexplored because even in the Solar System there are still new bodies being discovered, comets, asteroids, small moons around the outermost bodies etc.
So I imagined that you'd simply visit, scan a very small percentage of the bodies in the system and it becomes "Explored".
Of course you only get the money for actually doing the scanning of the major bodies so that would need tracking somehow.

I'm not thinking beyond the exploring stuff, if someone wanted to extend it later on to what factions gets control of the explored system by putting outposts etc there, well that's work they can do :D

[impaktor]

OK. Sounds good. I was just transcribing my initial thoughts while I was reading your first post.

Slightly related: only half of 2380 is fixed, so zooming out treats unexplored as having a faction, if I remember the issue correctly.

[Tichy]

That raises a question: "Who lives in those unexplored but inhabitated systems?" :)
Which leads to "How do we behave with aliens?"

Shouldn't we have something like Star Trek's "first directive"? Or some kind of non-interference rules?
What should happen when we land on an unexplored but inhabitated planet, and what when we send probes to it?

Some time ago i wrote some thoughts about that:
http://pioneerspacesim.net/forum/viewtopic.php?f=3&t=8
(Scenario 4)

[lwho]

That raises a question: "Who lives in those unexplored but inhabitated systems?" :)
Which leads to "How do we behave with aliens?"

It might be just a pirate base. The auto-exploration code I put into the pull request already deals with the possibility that there are spaceports in an unexplored system (and thus triggers a "found x hidden bases" message), though I'm well aware that the current generation code generates unexplored systems with zero population and thus no bases.

Slightly related: only half of 2380 is fixed, so zooming out treats unexplored as having a faction, if I remember the issue correctly.

Ohh, since I mangled the factions code several time since then, I probably should give it another look. Might be easier to fix now than before.


As for the exploration topic: I'm more concentrated on the technical aspects, how to realize such features.I assume the needed features in core would be appropriately exposed to Lua and it wouldn't be hard to experiment with different policies with a few lines of Lua. (Yes, I know, I'm a lazy dog leaving all hard work for the "Lua Front", or was it the "Front of Lua" ;)).

Let me brainstorm a bit about possible core features I see here:

1. The "exploration detail level" of a system. So, instead of completely unexplored or completely explored, you have several levels inbetween
   • gas giants visible (I assume this would be possible with standard instruments of a ship when entering the system)
   • rock planets visible
   • moons of gas giants visible
   • moons of rock planets visible
   • independent minor bodies visible
   • spacestations (maybe separate for orbital/ground, maybe some technical equipment sensing for artificial radiation).
   • Kuiper-belt objects visible
   This would be a natural and not so hard to implement followup for my current pull request.
2. Selectively exploring bodies of a system (e.g. by Andrew's exploration probes). This would require giving each body its own exploration flag (or levels). More complex than #1, but probably not so much. Actually, having this, we don't need #1 anymore, as Lua could automatically mark every gas giant in the system explored etc.
3. Selectively exploring properties of planets. For example, you would know that there is a "medium size gas giant", but not which atmosphere it has (until you send it a spectral analyzer probe for example), the exact mass etc. This would require some kind of filtering and/or fuzzing infrastructure for body properties we currently do not have at all. So, I would consider this feature more complex by at least an order of magnitude.
4. One orthogonal feature helpful for when to trigger exploration of a body would be Lua callbacks when the player ship comes "close enough" to a body.

[pbar1469]

This is surely something I would love to see implemented!

It could be something as simple as:

- Player needs specialized equipment:
- Long Range System Scanner (that would return the number bodies in the system and their "probable" type - As in, slightly random. Maybe sometimes a planet gets ID'd as rocky but it's actually earth like)
- Detailed Scanner (Player needs to be at a certain distance (high orbit) from the body in order to fully understand it's type and composition - Used based on results from previous scanner)

- As incentive for exploration, there could be two types:
- BBS post, someone asking for that info. Request could be to deliver simple long range scan (not much money), or to deliver full scan of some/all bodies ($$$!)
- Some type of official entity (Galactic Geographic) that purchases that data for fixed values (depending on type of scan, but always a very small value)

I understand the argument for probes, but what could be more boring than waiting in the same spot for a month or two (depending on system size) for the probes to report the data back, given that they use the same propulsion system to get to the body?

The idea for the scanners is for them to report the type of data that we already have displayed in the system map for uninhabited systems. Difference would be that without the scanning action, we could only see the star type on that system. The data that scanners gather for that system could still be created using the same procedural algorithm that is used now to populate it.

What do you reckon?

[laarmen]

There is already somebody working on the Detailed Scanner, but you will have to be patient, as this would be his first contribution to Pioneer.

I'd be willing to do a bit of brainstorming on the long range issue, but I'd need somebody with better understanding of all things sysgen (I'm more of an equipment guy)

EDIT: Oh, I didn't see #3201 ! That should make it fairly straightforward to imagine a long range scanner. Impaktor, do you think this could leverage the future "sensors" interface ?

[FluffyFreak]

I understand the argument for probes, but what could be more boring than waiting in the same spot for a month or two (depending on system size) for the probes to report the data back, given that they use the same propulsion system to get to the body?

You've misunderstood the reason for probes :)
Or to put it another way: "What could be more boring than having to visit all 181 moons + 8 planets + asteroid belt(s) and rings of the Sol system to scan each of the individually?".

When you arrive in system you would do a scan and get better information than the inter-system scan from another star system can provide (which in the year 3200 is I'm sure already pretty good), then you pick the planet / moon most promising and start heading for that... but that could leaves HUNDREDS of other planets and moons to explore. Do you go to all of them yourself, arrive in orbit, scan it, then move on?
No, you pick the best 4 or 5 candidate and fire off some probes to do the boring work for you whilst you personally go and explore the most interesting and rewarding looking planet.

It saves time and fuel too since you will need to harvest some more somewhere and somehow in the system (this is an open problem we need to solve, fuel scooping isn't good enough).

So the purpose of the probes isn't to fire them off and then wait for the month of two, it's to avoid having to wait for a month or two, or indeed to avoid having to hop from planet to planet until your fuel runs out :)

Andy

[impaktor]

I like it that you have to return to inhabited space and sell the information you have about your discovered system. Perhaps the "travel report" will say how many bodies you (and your probes) visited and how many were not.

EDIT: Oh, I didn't see #3201 ! That should make it fairly straightforward to imagine a long range scanner. Impaktor, do you think this could leverage the future "sensors" interface

Not sure I follow, so I'll just make a general statement: The sensor I need for the re-written scout mission is just to check that the player has been flying close to a planet long enough. Point of that mission is to "motivate" player to see some planetary scenery (doesn't matter if explored or unexplored system). ... so that would be the "Detailed scanner" mentioned above.