I was thinking the other day that the Galaxy map leads us to misunderstand how travel works in Aurora. It shows the systems as dots, and then big long connections between them. That puts you in the mindset that the JPs are the long travel part, and the systems are "small" local areas. In reality travel between systems is instantaneous, and it's travel within the system that takes a long time.

With that in mind I tried to envision what a map would look like that showed the travel distance between JP within systems, i.e. flip the script and make the distance between systems zero and within systems large, as it really is. The only way I could make it work from a purely geometry perspective was to put it into 3D. I used Matlab's graph object, with the JP as nodes (one single node for say the "Sol-Vega" JP, not one for each direction), with the 'force' distribution layout on so the edges are 'direct' weighted by distance between JP. In theory this should result in the edges length as displayed being proportional to the distances between JP, although I'm skeptical it's strictly proportional. That said, here's my first attempt.

It's definitely easier to interpret when you can spin it real-time, but I do feel like it does a better job of representing how long it takes to get from one place to another. I only represented JP from 6 systems (slightly different colors on each system, used a random RGB color and they weren't as far apart as you'd hope) since I was having to enter the JP positions manually from the game screen. I'm in the process of writing a script to import a csv file extracted from the Aurora DB to create the entire galaxy in one go, since what you see here was manually inputted. If I can get that working I would like to add nodes and edges for key bodies, i.e. player and known NPC colonies, since that's pretty important for understanding transit distance/time. For example in this map, Earth is very close to the Sol-Chara gate, meaning it takes a long time to get to the Vulcar-Rana gate even though it's technically only two jumps.

Coding has been slow since I'm pretty rusty. I found that AI has been a big help with that coding though, which is interesting and echo's Steve's point recently about how he's using it in his work. I'll post an update if/when I get the automated version working. I was able to take some time off today and this is how I spent it, instead of being traditionally "productive"

   

Quote from: Steve Walmsley on October 01, 2024, 03:02:36 PM

Quote from: Steve Walmsley on October 01, 2024, 07:38:19 AM

Quote from: Ghostly on September 30, 2024, 02:13:09 PM

Having unmanned ships with 100% morale definitely seems like a bug to me.

Yes, I agree on this point. I need to find a mechanic that creates a penalty but without any fiddly micromanagement and without players ending up designing ships that carry extra crew around. I have an idea that I am going to try out.

I've added the following new posts on the Changes List:

Under-manning Penalties: http://aurora2.pentarch.org/index.php?topic=13463.msg171631#msg171631
Captured Ship Penalties: http://aurora2.pentarch.org/index.php?topic=13463.msg171632#msg171632

Thank you, this is excellent! Being able to load conscript crews is an especially nice touch.

I assume there's no intention to make under-manning impact refueling/ordnance transfer rates? This would mean captured enemy tankers could be pressed into service without crew replenishment, but this is a non-issue compared to losing the ability to use their fuel in the middle of an offensive campaign.

Yes, I considered that but decided it wasn't important enough for the amount of effort required to adjust all fuel, supply and ordnance-related code.

skoormit, you're absolutely correct, it was a frame of reference issue. Matlab treats polar angles as counterclockwise from the positive x axis, i.e. "math notation". Based on the Galaxy map Aurora appears to use clockwise from the positive y axis, i.e. "navigation notation". I had considered that the x vs y axis might be the zero point, but didn't consider that the "handedness" might be flipped. I'll adjust accordingly. Thanks for pointing it out.

Will the "Add replacement crew" command be an option between ships? Or maybe an equivalent "Transfer prize crew"? Otherwise you have no way to add a prize crew, unless perhaps the boarding crew counts toward the manning requirement?

Steve, in exploring some mapping alternatives, I ran into what seems to be a programming error. When I take the AU and Bearing values from the Galactic Map Overview tab for each jump point, and convert them from polar to Cartesian, I end up with a map that looks inverted from what the game displays, i.e. X and Y for each coordinate are swapped. I just used the standard Matlab function pol2cart and plugged in the theta matrix (bearing, converted from deg to radians using deg2rad function) and rho matrix (AU values) straight off the galactic map overview.

Whether the error is yours or mine remains to be seen. Is it possible you transposed the X and Y results when converting from polar to Cartesian? If I intentionally plot x as y and vice versa, I get a map that looks like what's in the game.

Not sure this matters, as long as it's consistent within the game, but just curious. Granted I'm pretty crap at coding, but I can't seem to find an error in my (3 lines of) code.

The X,Y coordinates are generated from the distance and bearing and the locations on the map match that distance and bearing. If there was a problem with the conversion, they wouldn't be in the right places.