@gabrielbyrnei, I'm sorry it didn't work for you. I'm still at a bit of a loss as to what might be triggering the crash. I spent a bit of time trying to work out which parts of the DLL get triggered when you click the Single Player button, but I did not find anything suspicious. Judging by all the "fixme:" lines in the logs you posted, my guess is that it is a WINE problem; and I don't really know anything about WINE. I looked briefly at
this site; the people there are talking about which system dlls to use and stuff like that. I don't know if you're already on top of that stuff or not. I looked at a bug report on the forums for something unrelated, and they seemed to be focused on video drivers and stuff like that... So I don't know what the problem is, but I suggest you try asking about it on the WINE forum to see if someone there can help us work it out.
@Tickler66, I think most of the gameplay changes are pretty intuitive. I can't think of any pitfalls which would make you want to load a save from many turns back - except that the AI might do something you didn't expect.
At the very bottom of the
change log I've kept a list of some of the less obvious changes which might affect the way you play. (The full changelog doesn't fit in that post. So when I add a new version, I delete most of the logs from the oldest entries but keep things which I think are useful to know about for gameplay.)
Egypt and India, massive powers invade Inca lands and steal all of their cities except their capital in about 20 turns. 80 turns later they still cant/wont take the Inca Capital, Cuzco- which I admit looks like it is in the perfect spot to defend, but with over 90 units between the two of them (I took 6+ egypt cities before their main 45 unit stack got back)
The save attached is where I could actually prove it...India forces outside the Inca capital could crush it...and Im guessing theyve stood outside the capital for 75+ turns. The spy I use for vision arrives in vision of the Inca capital one more turn in. I wish I knew what the ai was thinking at that point, maybe always waiting for reinforcements when the capital makes another defender...
That save is a good fringe-case example of some of the rules the AI uses.
Gandi estimates that his attack is stronger than the city defence, but not by a big enough margin to attack. (The attack threshold depends on the state of the war, the personality of the leader, and other stuff like that.) In this case, Gandi wants around 135% before he attacks, and his first stack only has 110%. As a test, I took control of Gandi's team and gave the attack order (after bombarding). In the first turn I lost basically all the cannons, trebuchets and catapults; but most of the other units won - so that was a massive dent in the enemy defense. Inca then attacked with their catapults. On the next turn, I just promoted everyone and attacked again, and basically lost everything. So I think Gandi is wise to not try to attack with only the first stack. He waits for the reinforcements.
But when the reinforcements arrive, he still doesn't attack because he still doesn't think his stack is strong enough. -- This time I disagree with him.
So.. I've just spent some time running those few turns over and over while I tweak some numbers, and I think I've made some good improvments. I've increased the AI's estimates for collateral damage, and I've decreased the estimate of how much good will come from bombarding the city defences, and I've tweaked the threshold the the spies use to decide if they should cause a city revolt, and a bunch of other stuff. (Gandi has a spy in the city, ready to cause a revolt to drop the defences. He doesn't cause the revolt because the army outside the city isn't strong enough anyway - which is correct.. but with some of the changes I made, he caused the revolt and then still didn't attack - so I adjusted the revolt threshold a little bit.)
Also, there was a problem caused by some old BBAI code which was messing up Gandi's understanding of the river outside the city. In some cases, he kind of forgot that he could get better odds by moving from the uranium to the hill.
Anyway, that was a fringe case. At first he didn't quite enough guys to take the city, and then later he had barely enough. So it was a helpful test-case for fine-tuning some of the AI's calculations. Gandi is now able to take that city.
It's worth noting though that if Gandi still didn't have enough strength when his reinforcments arrived, the he probably would have just left the entire stack outside the city waiting for more reinforcments, even after the rest of the world declared war on him. That would be a pretty big tactical mistake but I'm not going to try to fix that paricularly problem in the near future, because it's too complex.