numbWing - xwing numerical analyses

By prauxim, in X-Wing

Crossposted on reddit, fwiw.

Hey guys, been working on a project. Mostly for fun, but want to put a feeler out there for community interest.

Basically, it's like xwing probability calculator on steroids. It simulates a whole game between one or multiple ships, state-by-state, complete with die rolling, token retention, spending logic, ship destruction, etc, and can repeat the simulation however many times required to iron out the average results. It's in early stages but basic functionality is working.

The idea is to create a tool for ship/upgrade/strategy analysis . Efficiency analysis is a possible use, but not the sole one. I'm personally more interested in upgrade value analysis and target priority analysis. That being said, the initial output is an efficiency analysis of sorts, mainly because it serves as a good stress test and ability/upgrade implementation is still in the works.

If you want a peek at a single graphic, here is the generic 1v1 benchmark result plot

Here is more info including a full set of current output: numbWing Output Spreadsheet

Link to source code in the spreadsheet.

Let me know if you think this is something the community would be interested in or if you may be interested in contributing.

Wow! This is amazing! Congratulations, that's really cool!

Edit: it looks like line 76 in numbWing.py has a fixed reference to your private working directory?

Also, is there a particular reason for 2.7 instead of 3.6+? I just recently got into python, work related.

Edited by GreenDragoon
24 minutes ago, GreenDragoon said:

Wow! This is amazing! Congratulations, that's really cool!

Edit: it looks like line 76 in numbWing.py has a fixed reference to your private working directory?

Also, is there a particular reason for 2.7 instead of 3.6+? I just recently got into python, work related.

Should be fixed, try pulling.

Yeah I should have used 3 but my work uses 2.7 so force of habit. They're pretty similar for what it's worth. I may consider converting it into 3 before it gets too big.

2 minutes ago, prauxim said:

Should be fixed, try pulling.

Yeah I should have used 3 but my work uses 2.7 so force of habit. They're pretty similar for what it's worth. I may consider converting it into 3 before it gets too big.

Will do if I find other things. I don't have time now, but eventually I'll have a look and play around with it. Really cool though!