Combat skills formulas

... A place to ask for help on any topic whether it be starting out to player killing to IT issues.
Mhaliah
Posts: 335
Joined: Wed Feb 25, 2015 11:48 am

Re: Combat skills formulas

Post by Mhaliah » Mon Apr 15, 2019 5:11 pm

Fermin wrote:Think either way it means 7d3 is actually better damage then 3d7 which is what I had been thinking...
Damage is always improved by more dice when dealing with the same max roll. I could do the math for a 7d3 vs 3d7, but I have a 15d2 vs 5d6 already memorized so I will use it as an example. (numbers are rounded to the nearest integer)

For any random roll of a 15d2 vs a 5d6:
15d2 has a 85% chance of being greater than 5d6 roll
15d2 has a 5% chance of a tie with a 5d6 roll
15d2 has a 10% chance of being less than a 5d6 roll

5d6 has a 23% chance of a roll that is less than the minimum 15 for the 15d2

The only possible downside of the higher number of dice is that the more dice for the same max the more narrow the distribution becomes and the less likely a roll is to get "max" damage. The odds for a max roll for 5d6 is 0.01% but for 15d2 it is 0.003%. So this might mean a "blick" from full is less likely. I say might because that depends on how far from the maximum the roll has to be for the modifiers to not exceed the hp of the victim.

Thuvia
Posts: 143
Joined: Sun Feb 22, 2015 12:22 pm

Re: Combat skills formulas

Post by Thuvia » Thu Aug 22, 2019 12:19 pm

I've finally gotten around to reading this (hopefully I didn't miss anything obvious) and wanted to add some thoughts about bash.

1. While Step 2 in the bash code deals with relative BMI differences only when they are greater than 1, the "stockiness" messages seen when conning a player or mob change for relative BMI differences greater than 0.1, instead. This naturally makes me wonder if the bash code is accidentally off by a factor of 10, since BMI would play a limited but noticeable role in many everyday PvP combats otherwise. Conversely, changing the BMI formula to 100*Total_Weight/(Height*Height) would similarly make the current bash code impact PvP. You'd want to change the stockiness messages to operate off of relative BMI differences greater than 1, as well, in that case.

2. I seem to remember talk way back that the game was originally designed for a max Defense of about 200 or so. If so, it makes me wonder if Step 3 in the code originally subtracted the Defender's PB/4, rather than adding the Attacker's PB/4, the change possibly having been a simple way to "fix" the problem of higher-than-intended Defense instead of tweaking stats on multiple pieces of equipment. Or the opposite may have been true, where the Attacker-PB code was always (accidentally) present, and Defense was upped to compensate for Bash being too strong.

3. For Step 7 favoring Clubs, I wonder if some Circlemud code is being used wrongly by accident. If the bash code is reading a type flag number, rather than the "club" type being spelled out, it is possible that this bonus was meant for All weapon types. We know that either a Weapon or Shield has to be equipped to bash, and shields themselves confer no bonus (other than the potentially-accidental PB-boost, see Point 2 above). In Circlemud, the Weapon type is flagged 5, while some Imm I believe made a post showing the relative positions of the various weapon flags (a different list), with "Clubs" being 6. Not a match, but type 4 in that case is "Sickle", so if the game ignores that then Clubs would be slot 5 and would get the prac% bonus. As an alternative, the same general reasoning above may still be true, but a "6" was targeted instead of "5" by mistake. Or, maybe I'm just wrong (quite likely) and Clubs really are meant to be special.

4. Regarding the 95% max chance in Step 8, I wonder if this is in the spirit of old DnD, where a roll of "1" on a d20 would always "fail" for certain actions, regardless of other modifiers.

5. Finally, I believe someone asked about weapon condition. While it does not impact OB, it does reduce the dmg roll.
Last edited by Thuvia on Tue Aug 27, 2019 5:23 pm, edited 1 time in total.

Taziar
Posts: 685
Joined: Sat Mar 21, 2015 10:28 pm
Location: Seattle

Re: Combat skills formulas

Post by Taziar » Sat Aug 24, 2019 2:09 am

One question that just crossed my mind... is the parry added to bash calculation the constant value seen on stat command or does it take into account when split via multiple opponents?

stark
Posts: 47
Joined: Sat Aug 26, 2017 4:03 pm

Re: Combat skills formulas

Post by stark » Fri Aug 30, 2019 6:35 pm

Thuvia:

For point 1. I personally agree that the bash code seems accidentally off. There are two different clauses for small BMI differences (one for < 0.2 and one for 0.2 to 0.5) that do some math but they each always evaluates to 0 due to integer truncation. So it seems like an oversight to me, why write two branches that never do anything? But it's hard to be sure.

For point 3. the flag specifically looks for 'club', it's
obj_flags.type_flag == WPN_CLUB
Taziar:

Hard to say. The code invokes a 'GET_PB(ch)' function for which I don't have the source. My hunch is that it's just the PB value as seen on stats, but I also can't quite rule out your theory because who knows what that function does.

Estindn
Posts: 30
Joined: Sun Jan 29, 2017 3:50 am

Re: Combat skills formulas

Post by Estindn » Fri Sep 13, 2019 4:19 pm

Do str have an effect on stab damage?

Merrick
Posts: 83
Joined: Mon Feb 05, 2018 11:44 am

Re: Combat skills formulas

Post by Merrick » Fri Sep 13, 2019 4:54 pm

No. Weapon, hone, master, backstab prac %.

Lisette
Posts: 3
Joined: Sat Oct 03, 2015 3:19 pm

Re: Combat skills formulas

Post by Lisette » Fri Sep 13, 2019 11:51 pm

Didn't read through the whole thread, but is there a known impact on the actual calculations based on enemy type? For example, stab and charge both do next to nothing to undead. Is this true of other types? Does anyone know if weaves are also impacted against undead?

Asandra
Posts: 11
Joined: Mon May 13, 2019 11:30 am

Re: Combat skills formulas

Post by Asandra » Sat Sep 14, 2019 4:27 am

Nothing factual known about weaves yet, but might be good to test it?

Post Reply