User:Rhomphaia/Experiments

Aims

The purpose of this experiment is to determine if any reliable formula for determining the chance of a knockdown with the expert marksman perk can be extrapolated from empirical evidence.

Factors

These are the factors which ‘’may’’ influence the chance of a knockdown listed in the initial order of suspected relevance and in which I propose to test them

In game Factors

• Target Agility
• Player Agility
• Player Marksman
• Player Luck
• Damage Dealt

Notes

If marksman Skill is a factor then this would have to be taken into account when investigating luck since in other cases Luck has a modifying effect on skills, it would be necessary to test whether luck is only implemented in this way or if it has an effect of it’s own. If Marksman is not a factor then of course Lucks effect on it would be irrelevant.

Damage Dealt is also apparently a factor, since early experiments with 0 damage weapons showed 0% chance of a knock down, however is it simply a factor of Weapon Rating or is it based on damage through the Complete Damage Formula

CS Factors

• iPerkMarksmanKnockdownChance
• fKnockbackTime
• fKnockdownAgilBase
• fKnockdownAgilMult
• fKnockdownChance
• fKnockdownDamageBase
• fKnockdownDamageMult

Notes

A question arises whether or not there is a single formula for knockdowns through any means, which is then modified by iPerkMarksmanKnockdownChance, or are there separate formulae for Marksman and Melee attacks.

Method

Initially this investigation will focus on the “In game” factors since these are the ones that would have most impact on normal play. Experimentation with the Settings in the CS would come later, however since little is known about the actual effect they have, no reliable method for testing can be determined other than change them a lot and see what happens.

The first run of experiments will focus on a single attribute (or skill in the case of marksman) leaving the others fixed (marksman will have to be reset if it increases and if it proves to be a determining factor then it will have to be adjusted alongside luck when investigating that attribute to compensate for the skill modification factor).

Initially I intend to raise each factor from 20 to 120, in increments of 20, testing 100 arrows at each increment, this should clearly determine whether or not that factor has an effect, and provide a rough approximation of the magnitude of that effect.