After testing the core myself, it's clear that the core isn't working the way it should based on how most people would interpret the description. What I've found is actually quite bizarre and I don't have a good explanation for it in terms of the game code. Hopefully the problem isn't as complex as it seems and won't be too hard a fix.
The first thing I did to try to test the core was get one guy with 36 dex and another guy with 200 dex. Based on the block formula, the 36 dex guy should have a 0% chance to block and the 200 dex guy should have a 30% chance to block (maxed). Neither player used any type of core or skill to boost block chance.
I made the 36 dex guy apply Hunter's Mark on the other guy and then had both of them strike until the effect expired. I collected data on how many times each character was blocked or connected while the core was in effect:
In 155 strikes, the 36 dex guy (who applied the core) got 47 blocks, which was a 30.3% block rate.
In 177 strikes, the 200 dex guy (who was being affected by the core) got 26 blocks, which was a 14.7% block rate.
That's absurd. If the core was working properly, the expected block rate of the 36 dex guy should be one half of the base block rate (30% in this case), or 15%. It was STILL 30%, so no, the core didn't do squat in this case to reduce the chance that the user gets blocked. What's even more absurd is that somehow the 200 dex guy started getting blocks, about 15% of time. The moment I witnessed him getting a block, I knew something was really wrong because the chance should be zero.
Just to confirm that the block formula was working properly, I also collected data on completely ordinary strikes with no core applied on either player:
In 145 strikes, the 36 dex guy got 43 blocks, which was a 29.7% block rate.
In 145 strikes, the 200 dex guy got 0 blocks, which was a 0% block rate.
Phew... These numbers look exactly as expected based on the formula. There's no way in heck that the 200 dex guy should be getting blocks with such a large dex advantage. Yet when Hunter's Mark was applied to him, he was getting them at about a 15% rate, even though the core is not supposed to be affecting this number at all. Now, based on these numbers, the logical conclusion would be to assume that the core was incorrectly coded to lower the target's connect chance rather than their block chance.
However, just to make sure, I tested another scenario to see whether the above numbers were coincidental or not and if the core was actually bugged in some other way than what I'd assumed. This time I got both guys to have the exact same dex, 83 each. Obviously, based on the block formula, neither player would have any block advantage of disadvantage. Therefore, they should've both had the base block chance, which is 5%. With this setup, I got the following numbers:
In 100 strikes, the player who applied the core got 3 blocks, which was a 3% block rate.
In 100 strikes, the player who was being affected by the core got an astounding 33 blocks, which was a 33% block rate.
It's hard to say whether the 3% indicates that the core was working or not (if it was working, it should be 2.5%), but if we were to give it the benefit of the doubt and say that for some weird reason it was actually working in this case, that'd contradict what I'd just found in the other scenario. What's even more freakish is that the affected players block rate skyrocketed to the max block chance when it should have only been 5%. I don't know what to make of that at all.
Either way, it would appear that (1) at least in some cases, the core is not doing what it's supposed to do which is to halve the opponent's chance of blocking you and (2) it's doing this bizarre thing of increasing your own chance of blocking by a substantial amount when it shouldn't be affecting it at all. Yikes.
If I wanted to, I could go test a whole bunch of other scenarios to get a better understanding but I don't know if that'd be helpful for fixing the problem or if simply knowing that the core is most definitely bugged is enough.
< Message edited by NDB -- 3/17/2020 19:48:52 >