More control over AI cloning "cheat"
AI's little cheat which makes it clone units and infantry when two FactoryTypes of specific type are present is something that modders often use to make their AI more difficult. However, AI cloning also applies to BuildingTypes, which rarely is something that modders desire. Structures with BuildLimit=1 are copied, AI gets more and more factories and the whole thing is just out of control.
ATM there is a way to disable AI cloning whatsoever. Since there aren't too many various object types (buildings, aircraft, infantry, units), a seperate tags for cloning of these four types would be useful.
Blueprint information
- Status:
- Not started
- Approver:
- None
- Priority:
- Undefined
- Drafter:
- None
- Direction:
- Needs approval
- Assignee:
- None
- Definition:
- New
- Series goal:
- None
- Implementation:
- Unknown
- Milestone target:
- None
- Started by
- Completed by
Related branches
Related bugs
Sprints
Whiteboard
I disapprove of the direction this blueprint would be taking Ares. Cloning objects has always been a kludge at best - something in the middle between a bug and a hack to make the AI "harder". Making this kludge customizable, spending development time to improve it, to document it, etc. would legitimize it. Instead of an annoying semi-bug or an expression of Westwood's inability to code a smart AI, it would become a feature. And not just that, it would become a feature we'd have to support. If, somehow, one day, the AI suddenly behaved fairly and *didn't* clone its shit, people would come, report a bug, and expect us to "fix" this fair behavior, because they relied on customizable cheats.
Imo, it's bad enough the original issue existed at all. Let's not make it worse by turning it into a customizable feature.
More importantly, however, I have an issue with feature requests/blueprints like this in general: You're deliberately proposing a kludge instead of what you actually want.
You don't *actually* want the AI to clone its non-buildings. What you actually want is making the AI harder. I posit that you'd be just fine with the AI *not* cloning anything if you had an alternative way of making it harder/more sophisticated. You've just been conditioned to see cheating/cloning as the default way of "achieving" that goal. So ultimately, we'd spend development time to legitimize a cheap hack-bug, to "improve" undesirable behavior you don't want in the first place. Sure, you'd be "happy" for a little while, because your AI would seem harder, but you'd always have the sour aftertaste of knowing that it's not *actually* harder, its cheating is just more sophisticated/less obvious.
So how about, instead of making the AI *cheat* better, we focus on just plain making the AI better?
How about you sit down with other interest groups and write a specification of what features for AI scripting and what automatic reactions/patches would help making the AI better as a whole, without it needing cheats?
We can't promise we can implement all of it, but we'd at least have a foundation to discuss.
And that's something I'd much rather support than Cheap Hack 2.0.
~ Ren