What is an Exploit?

[vondrack]

Originally posted by Rommel2D
I didn't mean to question the accuracy, so much as to clarify if this was only for those 'moved' units in a stack. I guess it would have to be for this to be an exploit. Now Snotty's post is confusing me- how could you use the 'fortify all' order for a landing party? Does unloading outside a city still ends a unit's turn, or is this another undocumented rule change? Or were they a 'landed party'? ;-)

Hope my previous post gives the idea of the test in an understandable way... it actually took me well over three hours to sort all the possibilities out - you first need to stage a PBEM, then you have to look for a situation where fortification matters...

As for using 'fortify all' on a landing party - you can do that, that's the problem. ANY stack, even a stack of units that have no movement points left (like a stack of units just landing), allows the 'fortify all' command on them.

[Rommel2D]

So the graphics definitely appear before the modifier is available? A category I/II hybrid?

[Donegeal]

I still listed it as a cat II because it is a cat I in graphical terms only. Graphics really don't make much of a difference, they are just astetic. I considered it a Cat II because it would allow a unit to move (unsing all its movement points) and then fortify (which would require some movements points left to do) and it is dependant on the play order.

Can any one think of other "exploits so we can categorize them?

[vondrack]

The graphics may appear before the modifier is actually present. IIRC, I have seen it both ways. It's like that you (using 'fortify all' on your stack) will see the stack as not fortified, but the players after you in the turn order will see it as fortified (even though it will actually NOT be fortified until the end of the turn).

[Snotty]

I imagine a lot of work has gone into this but I seem to be missing something Vondrack. In the save you only used small numbers of warriors, so surely the results could be put down to the RNG doing its funny thing. I thought, say, 2 stacks of 100 warriors would be needed to produce a number of results that averaged out. I will maybe have a poke around with it myself.

You learn something new everyday, and it looks like I will stop using fort all as it is actually of detriment to me because I have confidence I have forted when I havent.

Do defensive units with bombard ability still fire if attacked and unforted?

[Donegeal]

yes

[vondrack]

Originally posted by Snotty
I imagine a lot of work has gone into this but I seem to be missing something Vondrack. In the save you only used small numbers of warriors, so surely the results could be put down to the RNG doing its funny thing. I thought, say, 2 stacks of 100 warriors would be needed to produce a number of results that averaged out. I will maybe have a poke around with it myself.

Nope, a single case is all you need to prove the theory.

If you play a game with the RNG seed preserved (all PBEMs are played that way), then every time you open a given save, the same sequence of RNG values is generated and used to resolve whatever asks for dice rolls in the game. If you apply this (all the time the same) sequence to a series of "same" combats, you can tell, under appropriate circumstances, whether there is "something" affecting the results.

In the test scenario, you perform a series of warrior vs. warrior engagements - the only thing they differ in is that some of them are fortified and some of them not.

If you leave all the Frenchies unfortified, you can verify what I say is true - leave the French warriors unfortified (or properly fortified, it does not matter, just they all need to be the same way), save/"forward" the game, load the save as Carthage, attack one of the French warriors. Then reload and attack another one. As long as both the targetted warriors are unfortified, on the same terrain, the combat progress & result will be precisely the same as in the first case (you can attack all three warriors in a sequence and you will see that the combats will be exactly the same way, no matter what warriors you attack). That's because of the way RNG values are used. The game simply takes them one by one and uses them for the combat resolution - does not matter which units are engaged.

Another test you can try... create a similar scenario, but use a bunch of swords to attack a mix of swords and spears. Swords & spears have the same defense value, no retreat ability, so as long as they will all be un/fortified and on the same type of terrain, you will get exactly the same progress/result for whatever series of engagements you chose (no matter whether you attack spears or swords), because RNG values will be used to resolve the very same series of A=3 against D=modified(2) combat rounds.

The "funniness" of the RNG is in that it can generate streaks. But with the random seed preserved, the RNG sequence is the same every time you open a given save (even if it is streaky). The only thing you have to be careful about is that there are certain non-combat actions that use RNG values up, too (some you would not suspect) - that may screw your results (that's actually why I used an extremely simple scenario/situation with nothing but the combat happening).

Sorry if I seem confusing... I am scratching this in a hurry (lacking time to do it properly). If it's not understandable, ask for clarification - I should have more time tomorrow to put together a better illustrated example.

[Snotty]

Thanks Vondrack, I get it now.