Anyone know of any breakdown of the mechanics in AoW 1? Specifically the actual numbers behind the Attack and Defense stats and how the game calculates whether an attack hits or misses.

I'm also wondering if the 9 DEF vs 10 DEF discussion was ever sorted out - the game seems to have a bug where 9 DEF is actually more effective than 10 DEF (easily tested via a custom map where one hero with 9 DEF and one otherwise identical but with 10 DEF attack a full stack of Ballistae - 9 DEF hero will win more often than 10 DEF hero). Were the numbers ever crunched on that?

I feel like even after 20 years there's so much about this game that's still pretty opaque.

I just conducted the following test using modified units.

-Slingers with 1 DEF and 1 HP. So they nearly always die in 1 round of being attacked.

-Cavaliers with 10 ATK, 10 DAM, 50 HP, and 48 movement points. Four varieties - 7 DEF, 8 DEF, 9 DEF, 10 DEF.
So that should be a 10% chance to hit for 1 damage, regardless of Cavalier variety.

Each variety of cavalier has three tests where they fight little groups of slingers, and then I totalled the remaining HP of the three cavaliers for each of the four varieties. This is shown in the imgur link at the end of this post.

Totalling their HP, I got 100, 96, 88, 102 HP left over for 7 DEF, 8 DEF, 9 DEF, 10 DEF.

(the slingers use goblin graphics but they are using Hurl Stones)
https://i.imgur.com/N0DZBtz.png

I think my results are basically the same for the four DEF categories with minor statistical variance.
I don't think there is a difference between 9 and 10 DEF, unless you're against something with 6 or more ATK of course.
But this only covers chance to hit, not damage roll. Will test vs ballistae

Just did a smaller version of the test, but with Shoot Ballista.
10 DEF is for some reason slightly better than 9 DEF, when they should be the same against ballista (5 ATK)
But I don't think it's statistically significant
Did several more tests and I don't think there is any special status for 9 DEF
10 and 9 DEF seems to perform equally

Hmmm, so the deal is that with equal ATT and DEF an attack has a 50% chance to hit, and every point of difference increases or reduces that chance by 10% to a minimum of 10 or a maximum of 90?

Yep, to the best of my knowledge that's correct.

Yes, but don't forget the ''critical hit'' mechanic, where a hit deals max damage instead of rolling damage.

If you have a 90% chance to hit, then only 1 in 9 hits will be a critical, but if you have a 10% chance to hit, then all of those will be max damage.

That definitely seems to be the case, but I had never confirmed it to always be the case. Good to know!

You may notice that combat sometimes feels predetermined. You can reload a file and get identical results repeatedly until you decide to move a unit or change some other variable.

This may indicate that combat "rolls" are drawn from a prerendered string (e.g. using a random segment of Pi to make "random" chain of numbers).

Really! I actually didn't know about that. I knew there was a crit system in the sense that rolling a 10 is always a hit and rolling a 1 is always a miss irrespective of the stats involved, I had no idea it impacted damage as well. That's really interesting!

(I'm assuming the base roll to hit is just a straight d10 roll or something similar).