Post by: Elvin on April 15, 2010, 07:44:20 PM
30 systems later, and I have found no more interconnections at all. Just this one solitary connection between two systems one jump from Sol. Turning on the System ID numbers in Galaxy view, I can see I have most of the systems numbered 1-15, and then a dozen random large numbered planets. Even after finding the Number 3 system, which had 8 jump points in it, I had 0 connections to systems I already knew. And this repeated in a similar incident with the number 8 system, which had 6 JP's.
So is this a case of me being spectacularly unlucky, or is this a misunderstanding on my part of how the Jump Network maps out? If a system has no more unexplored jump points, is it possible for a new one to appear? ( I noticed a "suspected dormant jump point" box somewhere, what is this for?) If I want a very interconnected system, do I set the local generation chance to 95%, and the spread to 5 or something?
Thanks for any pointers on the subject,
Elvin
Post by: sloanjh on April 15, 2010, 11:17:39 PM
Quote from: "Elvin"
If a system has no more unexplored jump points, is it possible for a new one to appear?Yes, if a system rolls to connect back to a system with no jump point "slots" left, then it opens a "dormant" jump point. A dormant jump point is one that can't be detected until a ship jumps through it from the other end, at which point it turns into a "normal" jump point, which can be found by resurveying the system.
Quote
( I noticed a "suspected dormant jump point" box somewhere, what is this for?)This is a left-over from Starfire Assistant. In Starfire such a jump point is "closed" IIRC, which means you can't find it by survey - only by transitting it from the other side or by seeing someone who knows it's there making a transit. If you read the Rigellian Diary, you'll see several cases where NPR were attacking through such a closed WP. The checkbox allows the player to mark such a system as a memory aid.
Quote
If I want a very interconnected system, do I set the local generation chance to 95%, and the spread to 5 or something?
I think I use 80% and 20, and don't see a lot more loops than you do. Think of it this way - if it's set to 100% and 10, then you've still only got a 1 in 20 chance of hitting a particular system in the range band (+/- 10). Similarly, with a 75% "locality" chance, you've only got roughly a 50% chance of a system 2-jumps away being in the same range band. I'm too tired to work out the probabilities, but I think you're just suffering from a combination of bad luck and too loose of criteria. So yeah, try cranking it up to 95%/5 and see what happens.
John
Post by: The Shadow on April 16, 2010, 01:41:22 AM
I've actually been considering backing it off a bit for my next game, especially since the NPR's are likely to be much further away in the new version.
Post by: Brian Neumann on April 16, 2010, 05:29:08 AM
In mechanical terms what happens is when you explore a jump point the computer pulls up a system number semi-randomly. If that system already has been explored it then checks for any open jump points that have not been explored. If all the jump points in system already lead to other systems it creates a new jump point and adds it to the system. If there are any jump points that have not been explored it uses an existing jump point to enter the system.
As a recap the difference between Starfire and Aurora is that in Starfire there were warp points that were undetectable exept when a ship transited them from the other side. In Aurora all jump points are detectable, and because of the game mechanics there needed to be some fluff about why you couldn't detect the jump point when you surveyed the system.
Brian
Post by: Steve Walmsley on April 16, 2010, 09:25:43 AM
Quote from: "The Shadow"
I've actually been considering backing it off a bit for my next game, especially since the NPR's are likely to be much further away in the new version.Famous Last Words
Steve
Post by: The Shadow on April 16, 2010, 12:52:00 PM
Quote from: "Steve Walmsley"
Quote from: "The Shadow"I've actually been considering backing it off a bit for my next game, especially since the NPR's are likely to be much further away in the new version.Famous Last Words :)
Post by: Elvin on April 24, 2010, 08:18:44 AM
Post by: The Shadow on April 25, 2010, 02:12:25 AM
By the way, there's an option on the galactic map to see the system numbers - that will tell you for sure if anything's out of bounds.
Post by: Elvin on April 25, 2010, 08:57:36 AM
Quote from: "The Shadow"
Could you clarify? How many systems have you explored? Could you post a screenshot of your galactic map?
By the way, there's an option on the galactic map to see the system numbers - that will tell you for sure if anything's out of bounds.
That's precisely what i've got on at the moment, and the numbers just don't make sense with the parameters I gave them.
Here's the picture:
(http://i278.photobucket.com/albums/kk105/ElvinDrude/galacticmap.jpg)
Post by: sloanjh on April 25, 2010, 10:43:00 AM
Quote from: "Elvin"
That's precisely what i've got on at the moment, and the numbers just don't make sense with the parameters I gave them.
You said 95% and spread of 6. What's your max system number? Something like 250?
If you really want to do an easily falsifiable experiment, you should use 100% and something like 6. That way you'll know that all links must be within the spread.
John
Post by: Elvin on April 25, 2010, 10:47:12 AM
If/when I start a new game, i'll try 100% chance of local and see how it goes.
Post by: The Shadow on April 25, 2010, 11:52:30 AM
Post by: Father Tim on April 25, 2010, 12:32:04 PM
It's as if instead of a 95% chance of +or- 6 from the system you're exploring out from, it's a 95% chance of +or- 6 from the last system you explored to.
Post by: Steve Walmsley on April 28, 2010, 03:30:02 AM
Quote from: "Elvin"
Humm...Something is almost certainly bugged with this feature. Despite starting a new game in 5.13, with the settings of 95% local chance and a spread of 6, I'm getting galaxies that seem to be being generated by the default settings, despite still showing the 95% and 6 spread in the Game Info box. Any ideas what might be causing it, as I don't think the probability of the result I'm getting is greater than .1%If this is a real stars game, then the local chance and local spread don't apply. Those are only used for random games. In a real stars game, systems are linked based on their proximity in real space, which is based on their celestial coordinates in reality.
Steve
Post by: ndkid on April 29, 2010, 02:22:07 PM
Quote from: "Steve Walmsley"
If this is a real stars game, then the local chance and local spread don't apply. Those are only used for random games. In a real stars game, systems are linked based on their proximity in real space, which is based on their celestial coordinates in reality.Steve, could you make a change to the new game setup that disables the local chance and local spread if you choose real stars? I never realized what you just said was the case, and never would've guessed it, either, and having a graphical representation may helps those who come after me.
Steve
Post by: Elvin on April 29, 2010, 02:23:19 PM
Quote from: "Steve Walmsley"
If this is a real stars game, then the local chance and local spread don't apply. Those are only used for random games. In a real stars game, systems are linked based on their proximity in real space, which is based on their celestial coordinates in reality.
Steve
Aha, thanks a bunch, I'll remember that for the next game.
Post by: Steve Walmsley on May 06, 2010, 04:37:44 AM
Quote from: "ndkid"
That is already how it works. Local chance and local spread are disabled in a real stars game. The local chance and local spread are an abstract way of generating star locations in a random galaxy. In a real stars game, the exact location of every star is already known. The method used is different though.Quote from: "Steve Walmsley"If this is a real stars game, then the local chance and local spread don't apply. Those are only used for random games. In a real stars game, systems are linked based on their proximity in real space, which is based on their celestial coordinates in reality.Steve, could you make a change to the new game setup that disables the local chance and local spread if you choose real stars? I never realized what you just said was the case, and never would've guessed it, either, and having a graphical representation may helps those who come after me.
Steve
When a jump point exploration takes place in a real stars game, the program chooses to generate a new system (95% chance) or link to an existing system (5%). In the former case, there is a 90% chance of a close link and a 10% chance of a distant link. For a close link, the program creates a list of all possible stars in ascending order of distance and then checks each one in turn. There is a 20% chance that the program will stop and select a particular system. For a distant link, the program does the same but with only a 4% chance of choosing a particular system. When linking to a known system, the program creates a list of existing systems in ascending order of distance and uses a 20% chance for each one.
The reason I chose this particular method, rather than a simpler method of simply arranging all system in ascending order of distance regardless of whether they have yet been generated and just picking one, is that the simpler method could easily lead to a tangled mass of internal connections. One option might be to allow user parameters for the chance of linking to an existing system and also for the chances of close and distant links.
Steve
Post by: UnLimiTeD on May 06, 2010, 09:12:00 AM
Post by: Steve Walmsley on May 06, 2010, 10:03:13 AM
Quote from: "UnLimiTeD"
I think he means that players should not be able to set the local chance if they ticked real star systems, as an additional hint that it won't actually do anything.Yes, that it is a good idea. I'll add that for v5.20
Steve