Jump Point exploration is very clearly not creating the expected number of links to existing systems.
I made a non-known stars game with max systems = 1000, local chance = 95% and spread = 5.
I used SM to perform full grav surveys, then explored each new jump point with a fleet.
I continued until I had explored to a distance of 4 jumps from Sol.
The results:
41 systems explored.
Exactly two of those had links to existing systems.
Both of those links led back to the system that had originally led into that system.
Also of note:
After the naming theme runs out of names, the usage of the "System #X" template has problems.
The value of X is always either very close to 0, or very close to 1000, and in many cases the same X is used for multiple systems.
Of the 24 systems with this name template, there are only 15 distinct names. Here are the values of X:
0 0 0
1
2
3 3
4
5 5 5
6
7
987
990 990
994
996
997 997 997
998 998
1000
A cursory examination of the database indicates that these values of X match the system number for these systems.
From this data, I think that this is what is happening:
When a link to a local system is generated, the SystemNumber is correctly selected at random from within the specified local range.
If that SystemNumber matches an existing number, the code makes a new system with the same number instead of linking to the existing system.
The exception to this is when the SystemNumber matches a system that the current system already has a link to, in which case a new link is created, resulting in a double link between the systems.
DB is attached.
TN start
Random Stars
Is your decimal separator a comma? No
Is the bug is easy to reproduce, intermittent or a one-off? Easy