Thanks,
@havox79 - appreciate the help. I'm quite concerned on getting the numbers right - the chances of mission success for each operation, the chance for an international reaction, the severity of the reaction, chance of attacking operator survive, are the international reactions too severe or not severe enough, etc.
Also need feedback on any/all of the features about the AI - does Alexander attack too much, does any of the AI attack too little, does the AI have too much advantage NOT having to maneuver units halfway around the world to attack, etc.
A lot of the "settings" for these issues have been fairly genericized - meaning all missions carry about the same chance of success/reaction/survival. Obviously, this is not intended for the final version, but it seemed like a good place to start - testers have to see a high chance of success for the "steal city" mission during testing. If it only works 2% of the time (the basic planned amount), then it's likely they'll never see it work - but during the attempts they'll be quite familiar with the global community declaring war on them as a result...
So- I think the numbers are set at 50% chance of mission success, 75% of operator survival, 10% chance of international reaction, and 75% operator stays in place after the mission. Should help testing nicely, but is nowhere near what the final numbers should be. I can take a guess at it, but what seems right to me may break the mod for other players. Of course, I'll probably be tweaking these numbers as long as I can support the mod, but I still want to get the best numbers possible in there.
I'm not surprise the Great Spy mod design approach for using units is hit or miss. It basically relies on flavors and other game concepts that aren't very effective at getting units to behave in a planned manner. Originally, I tried this, with a combination of LUA-driven code (the wonderfully imprecise "pushmission" method...) but even with a lot of code applied, the units still behaved in a very erratic, unconvincing way. The AI was just too easy to beat. So, if I made the units weaker to limit the human player advantage, the missions didn't do enough damage for anyone to want to assume the risk/cost of using them. So if the units are too much a pain to use, what have the mod in the first place? At one point during development - I had just started using the "simulated" AI attacks, but would spawn an AI unit just before the attack somewhere near the target city, then remove them after the attack. They popped out of nowhere (like submarines do when they attack) then disappeared a few seconds later. Was a bit confusing, which is why I was going to try to add an animation like when the Great Prophet builds a Holy Site - a couple seconds of "dancing" to get the human player's attention. Eye candy, but not much else, so eventually I dropped AI unit concept entirely.
And finally, let me say, I think your English is wonderful! You've given me 4-5 great suggestions already that I believe will make the mod much better. You've also given me a problematic suggestion (removing units entirely) which is actually quite good as well, but one the requires major adjustments to both the code and my own concept of what would make the mod even better. It's causing me to have to really THINK about how to make the mod the best it can possibly be, and challenging my fundamental assumptions about what I was trying to do, what I want to do, and what I really should to to make this mod as good as I think it can be.
if the fake AI is balanced that's not a great problem, I even want to say that's your mod is great step compared to the "great spy mod" in which AI do anything, no ops nor diplomatic reaction...
