http://forum.worldofwarships.eu/ind...me-interesting-info-from-ru/page__st__800giải
The internal client data is held in a file called GameParams.dat. It's standard python pickle, gzipped and then byte-swapped. Someone wrote a set of (pretty simple) utils to transform it to JSON, which you can dig up with a search.
Deconstructing the accuracy system is a work in progress, and as far as I know, I'm the only person working on it. This is the current list of important parameters:
Parameter--------------Main battery values --------------------------------- Likely meaning
minRadius------------------- 0.5-2.8------------------------- Horizontal dispersion at zero range, in units of 30m (also used for torp range).
idealRadius --------------------8-12 ---------------------------Horizontal dispersion at idealDistance, in units of 30m.
idealDistance----------------- 1000 ---------------------------Distance at which idealRadius applies, in units of 30m.
radiusAtZero ------------------0.2 -----------------------------Ratio of vertical to horizontal dispersion at zero range.
radiusAtDelim------------- 0.5 or 0.6 ------------------------Ratio of vertical to horizontal dispersion at half maximum range (see delim).
radiusAtMax--------------- 0.6 or 0.8 ------------------------Ratio of vertical to horizontal dispersion at maximum range.
delim---------------------------- 0.5----------------------------- Sets the range multiplier at which radiusAtDelim applies.
maxDist------------------------- N/A--------------------------- Base maximum range in metres.
sigma ---------------------------1.5-2.1-------------------------- Sets normal distribution spread within the dispersion radius. Higher is better.
taperDist--------------------- 2000-7000 ---------------------Below this distance (in metres?), dispersion tapers linearly to zero. Possibly added in patch 4.1.
The horizontal dispersion radius at any range above taperDist can be calculated by linearly interpolating between minRadius and idealRadius. You can replicate the in-game "maximum dispersion" values like this. Below taperDist, horizontal dispersion radius is probably linearly interpolated towards 0.
The vertical dispersion formula is very difficult to test because the visual effect of vertical dispersion is heavily dependent on range. The main evidence here is that vertical dispersion is relatively small (compared to horizontal dispersion) at shorter ranges. It's particularly visible for secondary batteries. Secondary batteries probably prove that vertical dispersion is a multiplier to horizontal dispersion too.
Putting that together, the Wyoming's maximum dispersion vs range graph would look like this: