I really don't think the AI uses queuing. If so... then it is a problem. A well written AI, however, would adapt to every moment of the game, not deciding on the next build until the current build is done; so I don't think this will be an issue. Then again... Stellaris did write a pretty bad AI so anything is possible. My assumption though is that I don't have to worry about it.What stops the AI from doing exactly this?
EDIT: I was a primary programmer for a Civ 4 mod that modified the DLL. I know from first hand experience that Civ4 never queued anything for building. We do have evidence that the Stellaris AI doesn't queue either. I was testing the AI at one point and the AI kept switching where a building was being built, implying that it makes a decision every time a buildling is done. The 1.8 patch notes mention that it fixed this issue (meaning it won't constatnly re-decide where a building should be... hopefully they didn't just start using the queue as a quick-fix but intelligently made the AI not do this).
LOL. Well I already told you that and you offered the suggestion to do it anyway. Unless by your question you are asking, what would work best scientifically. If that is your question. they both make sense... a building would certainly represent the number of people that might be involved. On the other hand, it would probably be a global network of smaller buildings (and hardware) so.... what do you think?How would these work scientifically?
Last edited: