Chance to hit (accuracy) - does not sense in 3.1 (?)

[Reg312]

what about splash damage? if medium cannon missed does it means what all splash damage will be equal to 0?
then for easier balancing we have to remove splash damage, incendiary damage

@Iluvalar: chanceTohit is only required FOR splash effects and actions with masses of tanks
if we will set missed shots damage to 0 then say, why we need accuracy ? it that case accuracy does not required (just reduce % of damage for unaccurate weapons)

[Shadow Wolf TJC]

If weapon accuracy seems to be too big a deal to balance out without having to switch to, say, the Supreme Commander method, then maybe we could just set all weapons' accuracy to 100%, and remove all the accuracy upgrades from the research tree? Not all rts games out there, including the Command & Conquer series, or Starcraft 1 and 2, even have an accuracy model, you know.

Sorry if my tone seemed offensive there, though it seems to me that you're getting fed up with Warzone 2100's accuracy system, and that you're having a difficult time trying to come up with a workable fix.

[Iluvalar]

You keep the splash damage as is. I don't understand that question.

Accuracy is not canceled by the armor, it's not the same as the damage. it's useful to balance with the ROF. To not change the damage ratio vs armor thingy.

It's also a very potent mechanism (like it is now). Even if we dont use it correctly now. We could use it better by creating some weapons with very high accuracy and low accuracy in the same line. So the player could specialize in some smaller weapon set.

We could also create "heroic" weapons, that are bad until an experienced units hold it.

In futur : some upgrades that reduce the enemy accuracy. To work like the thermal armor (specialized against some weapons). etc.

[Reg312]

In futur : some upgrades that reduce the enemy accuracy. To work like the thermal armor (specialized against some weapons). etc.

oh, evasion? stuff from RPG games, i think its just... dont fit warzone

In other words : don't care about hitbox for misses at all !

Iluvalar, if we have splash damage then we should use accuracy formula to determite coordiantes of splash effect? ?
you suggested just forgot about hitbox and set missed shot damage to 0

@ShadowWolf: sorry i dont undestand, i did not played SupremeCommander

i just know here is people who real likes mathematical stuff, but they dont want help with good accuracy formula!

[Giani]

In futur : some upgrades that reduce the enemy accuracy. To work like the thermal armor (specialized against some weapons), etc.

oh, evasion? stuff from RPG games, i think its just... doesn't fit warzone

It could fit for vtols

[Iluvalar]

In other words : don't care about hitbox for misses at all !

Iluvalar, if we have splash damage then we should use accuracy formula to determite coordiantes of splash effect? ?
you suggested just forgot about hitbox and set missed shot damage to 0

Sorry i tough it was obvious but maybe not... I meant : forget structures and droid hitbox. Not "planified" ground hit.

Here is a nice exemple I just tough about : We could make the HMG upgrade poorly on accuracy, TMG upgrade fine and single mg upgrade very nicely to the same accuracy upgrade. As a result, the TMG would keep competitive if you have enough accuracy and the smg would even make a comeback if you focus on that. Result :

ROF focus: smg => tmg => hmg => ac => tac
Mixed focus : smg => tmg => hmg => tmg =>tmg => riffles
Accuracy focus : smg => tmg => tmg => smg => sniper

[Deus Siddis]

If there are problems with accuracy, its because there is too little realism and simulation, not too much.

Dice rolls should only be used for random trajectory adjustment due to shifting winds and the 'accuracy' level of the weapon system and gunner. After that, the collision detection decides what happens.

[Iluvalar]

Deus Siddis : So i guess you have the formula for the quantic density of the units depending on their price and their gaussian distribution depending on their propulsion. And that you are volunteer to make the appropriate changes in the stats without affecting game play (as if it was possible) ?

Or are you just asking for "realism" without having any clue at all if it is only possible ?

[Deus Siddis]

Deus Siddis : So i guess you have the formula for the quantic density of the units depending on their price and their gaussian distribution depending on their propulsion.

So you want me to tell you how to mathematically relate totally unrelated concepts. And the point of this is?

And that you are volunteer to make the appropriate changes in the stats without affecting game play (as if it was possible) ?

Any change to the fundamental game mechanics will affect game balance. If that is a problem for you, then so is making any changes to the accuracy code. That's fine, but it isn't an argument against what I suggested.

Or are you just asking for "realism" without having any clue at all if it is only possible ?

Nope.

[Iluvalar]

Deus Siddis : So i guess you have the formula for the quantic density of the units depending on their price and their gaussian distribution depending on their propulsion.

So you want me to tell you how to mathematically relate totally unrelated concepts. And the point of this is?

My point is clear, the mean density of the opponent units and the size of his units would become you principal concern.
In a gaussian accuracy world, your chance to hit will double depending if you sit at 7 or 5 tile of the enemy and depending if it's a cobra or a python. moving your units half a tile away would clearly become you main preoccupation. And it would not even look like the game we have now.

[Deus Siddis]

You are assuming:

Fast units.
Slow projectiles.
Evasive maneuvers requiring constant and direct player input.
Weapon accuracy upgrades and units experience wouldn't play any part in trajectory calculations.
Having different sized units is somehow undesirable?

If we don't make those assumptions then your fears seem unfounded.

[Iluvalar]

Ok sorry, I just understood what you really meant to say.

You meant, rolling the actual dice roll, and then shooting the projectile, either on the target or beside the target if it's a miss.

It's exactly like that now. Except, we often try to miss behind the target. Which cause, of course, a false hit.

[Deus Siddis]

You meant, rolling the actual dice roll, and then shooting the projectile, either on the target or beside the target if it's a miss.

No, I meant shooting a projectile directly at a target, except with some deviation in the trajectory derived from a dice roll. Whether the projectile hits or misses a target is determined entirely by collision detection.

It's exactly like that now. Except, we often try to miss behind the target. Which cause, of course, a false hit.

This problem is caused by poor collision detection.

A similar problem with the same cause is the projectile traveling straight through a hillside to hit a target.

[Jorzi]

The problem with simulating actual trajectories is the computing requirements; it is not a trivial thing to do efficiently and its cost is always proportional to the amount of units on the map.
Also, collision with the actual geometry is, as said, not an option. We would have to use some kind of proxy object / bounding box system for collision, so that the graphical appearance of units doesn't affect hit probability.

[Iluvalar]

You meant, rolling the actual dice roll, and then shooting the projectile, either on the target or beside the target if it's a miss.

No, I meant shooting a projectile directly at a target, except with some deviation in the trajectory derived from a dice roll. Whether the projectile hits or misses a target is determined entirely by collision detection.

than it would be a different game where the size of you opponent would be a constant preoccupation.

It's exactly like that now. Except, we often try to miss behind the target. Which cause, of course, a false hit.

This problem is caused by poor collision detection.

A similar problem with the same cause is the projectile traveling straight through a hillside to hit a target.

The projecticle trough cliff is a 2.3 problem. Are you sure you talk about the good stuff ?