Intra-Region Procgen Positioning of Stars relative to each other

[celabgalactic]

Has anyone done any work to understand the progen algorithm for the 3D placement of stars within a region, eg so that the inter-stellar distances of stars within the same region can be reliably predicted.

In the AGT we have long been modelling stars in 3D space within a region using in game measurement data and mathemathics (Our code for AGT NAVI is on github). However the amount of data required to model grows exponentially... we would like to know if anyone has looked into the procgen algorithim for this ...

[monkeyman192]

I just had a bit of a look at the older game code regarding how this works and it looks quite complicated.
I am not sure if it has changed since then since that code is a few years old, but it probably hasn't changed too much at the core.
I think to get the exact algorithm may be tricky and require a bit of work to reverse engineer it all.
The code to determine the distances between two points at least looks like it's just using the usual Euclidean distance calculation.
As you say, it's the generation of the actual locations of the points that is the hard part.

Out of interest, can you link your current code which does the calculations?

[celabgalactic]

Hi, Yea I heard same from a few other sources but I just wanted to confirm no one has figured it out. At present we are doing location positioning using 4-6 measurement points from adjacent stars. The github for AGT NAVI is here, created by Max Hebrand a few years ago (not me). This does all the 3D modelling. https://github.com/mhebrard/nms-browser You can run it here: https://mhebrard.github.io/nms-browser/ I recommend looking at this wiki page where we *occasionally) update the regions as we map them: https://nomanssky.fandom.com/wiki/AGT_NAVI Our website info on this: https://www.nms-agt.com/agt-navi CEO, CELAB Galactic Industries Senior Leadership Team, AGT A 'No Man's Sky' company - PC Normal - Euclid Time Zone UTC+7:00

…

On Tue, Jan 20, 2026 at 6:44 AM monkeyman192 ***@***.***> wrote: I just had a bit of a look at the older game code regarding how this works and it looks quite complicated. I am not sure if it has changed since then since that code is a few years old, but it probably hasn't changed too much at the core. I think to get the exact algorithm may be tricky and require a bit of work to reverse engineer it all. The code to determine the distances between two points at least looks like it's just using the usual Euclidean distance calculation. As you say, it's the generation of the actual locations of the points that is the hard part. Out of interest, can you link your current code which does the calculations? — Reply to this email directly, view it on GitHub <#12 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ATEIXD52CP645AN2XO2BLQD4HVT4NAVCNFSM6AAAAACSAHBGPOVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTKNJUGMZTGMI> . You are receiving this because you authored the thread.Message ID: ***@***.***>