Posted May 31, 2016
In a multi-player game with a friend last night in a pre-warp game I used a quick tech advantage to build a colony in the only system the Sakkra could exit their home system. I also killed their first scout ship in the system I colonized and built a base on the node effectively locking them into their starting system. The ONLY planets Sakkra could possibly see were those in this one SINGLE system that I colonized.
I am not angry with what happened, but I just want to make clear that the Sakkra could only possibly spy on planets in the one system outside of their own, but yet I ended up killing a dozen or so agents who were performing missions on my other colonies. This would be impossible for a live player to do and should be impossible for the computer Ai to do.
Also the Sakkra agents failed all of their actions except 2 when on the same exact turn they got 2 of my planets to revolt. While the revolt result is excessively brutal for an espionage mission ( I still strongly believe that a high morale on a planet should render it immune to such missions) that is not the issue here. The issue is that it seems that the code is only doing one check per turn for espionage (or the specific mission type) success or failure. I strongly believe that the coders need to make sure the espionage success/failure function's RNG needs to be reset after every use to prevent excessive clumping.
I am not angry with what happened, but I just want to make clear that the Sakkra could only possibly spy on planets in the one system outside of their own, but yet I ended up killing a dozen or so agents who were performing missions on my other colonies. This would be impossible for a live player to do and should be impossible for the computer Ai to do.
Also the Sakkra agents failed all of their actions except 2 when on the same exact turn they got 2 of my planets to revolt. While the revolt result is excessively brutal for an espionage mission ( I still strongly believe that a high morale on a planet should render it immune to such missions) that is not the issue here. The issue is that it seems that the code is only doing one check per turn for espionage (or the specific mission type) success or failure. I strongly believe that the coders need to make sure the espionage success/failure function's RNG needs to be reset after every use to prevent excessive clumping.