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

The place to discuss balance changes for future versions of the game.
(Master releases & 3.X)
User avatar
NoQ
Special
Special
Posts: 6226
Joined: 24 Dec 2009, 11:35
Location: /var/zone

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

Post by NoQ » 03 Sep 2012, 21:10

Check this out: considering the accuracy bug, it's perfectly ok for those ripples to one-shot your tanks. Also, you should normally have at least elite commanders on these levels, buffing all your units accordingly. There is no bug in the game engine damage calculations and there were no changes in the campaign stats ever. CAM_2C is still easily beatable in under 40 minutes, which is way too easy for a 2-hour time limit level (you can see how exactly it works in 2.3 in my walkthrough videos; ripples do actually one-shot your tanks unless they are making unpredictable moves). I don't see how making this level even easier is supposed to improve the game.

What i'm essentially saying is that "omg campaign is too difficult make it easier" whine posts do not belong to this forum section, and i'm suggesting users to try learning a bit more about the game before posting misleading stuff into the developer talks.

manictiger
Rookie
Rookie
Posts: 19
Joined: 29 May 2009, 05:45

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

Post by manictiger » 03 Sep 2012, 22:06

NoQ wrote:Check this out: considering the accuracy bug, it's perfectly ok for those ripples to one-shot your tanks. Also, you should normally have at least elite commanders on these levels, buffing all your units accordingly. There is no bug in the game engine damage calculations and there were no changes in the campaign stats ever. CAM_2C is still easily beatable in under 40 minutes, which is way too easy for a 2-hour time limit level (you can see how exactly it works in 2.3 in my walkthrough videos; ripples do actually one-shot your tanks unless they are making unpredictable moves). I don't see how making this level even easier is supposed to improve the game.

What i'm essentially saying is that "omg campaign is too difficult make it easier" whine posts do not belong to this forum section, and i'm suggesting users to try learning a bit more about the game before posting misleading stuff into the developer talks.
85 range one-shotting elite tanks...
Yeah, I'm sooo casual for not finding that to be remotely fair or enjoyable.
Dude, it needs to be nerfed. I'm not going to get into an eloquent debate about this.
The devs can feel free to ignore the request if they so desire, but I'm not going to retract my statement that I think there is something seriously wrong with ripple rockets in the campaign. Period.

"What I'm essentially saying" is that your attack on my post doesn't belong in this entire forum. I interpret your misrepresentation of my issues with campaign balance as whining, so you can just keep all your "l33t" retorts to yourself from here on out. No one wants to hear them. If the devs or some mod wants to get on my case, fine, but some l33t kiddie being "edgy" does not impress me. Over and out. Not cordially, bye.

zydonk
Trained
Trained
Posts: 453
Joined: 12 Jun 2008, 18:31
Location: Dublin, Ireland

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

Post by zydonk » 04 Sep 2012, 12:04

manictiger wrote:
NoQ wrote:Btw, maybe you post a savegame? Cause i srsly want to see those deadly ripples.
I'd love to, but there'd be about a couple dozen .ini files that you'd be missing which could be part of the issue. In fact, I don't even know if the game would load with out them and if it did, whether it would load properly or not.
I suppose I could .zip them, but I'd rather go through the motions for a dev that can actually do something about it.
Post a savegame anyway, will you. Ripples are very useful, but long reloading balances range. Use faster droids to nip them?

There has been so much bitching about how powerful this and that asset is, that WZ has been reduced practically to a peashooter for kids.

User avatar
Stratadrake
Trained
Trained
Posts: 197
Joined: 07 Sep 2008, 09:43
Location: Pacific NW
Contact:

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

Post by Stratadrake » 07 Sep 2012, 19:26

So none of that stats are changed? I could have sworn that when they did the 2.x update, stats were changed, namely, the Lancer missile (I haven't really ever gotten past campaign 2, so my memory is a bit flimsy about the changes with the ripple rockets).
I can confirm there are no pertinent changes between the v1.10 and 3.x campaigns - the stats files show a few bugfixes and graphical tweaks here and there (plus a few tweaks to the file format) but nothing that would actually affect gameplay.
Jorzi wrote:What NoQ is essentially saying is: Campaign is what it is, a legacy from the original game. If the campaign is ever changed, it will be through a mod, out of respect for the original developers. Also, new campaigns are in the making, using the new units & balance.
It's the same thing in games like supreme commander, tiberian sun or red alert.
The campaign has ever been intentionally changed (see above comment), but it definitely feels different from the original game in subtle ways that are not campaign-specific, but more like game-engine-specific:

- Weapon ROF now maxes out at 600 (10 attacks per second), used to be 240 (4 attacks per second). This alone doubles the effectiveness of your Assault MG and Whirlwind AA (their stats are actually unchanged).
- Projectile hit/miss determined by an actual hit/miss check, not the initial diceroll. Improves realism, but also increases the overall effectiveness / damage-per-second of a weapon in combat, sometimes significantly (e.g. weapon with 50% base accuracy can be expected to hit a Vengeance Tracks nearly 100% of the time, a 2x increase).
- Omniscient artillery bug: Artillery pits/emplacements can acquire ANY target visible on the map even WITHOUT an actual sensor turret/tower to spot the target (see screenshot comparing 1.10 to 3.1).
- Reloading bug fixed: Previously, if a weapon fired off a partial salvo it had to use up the remainder of shots before it could reload. Now it can reload from partial to full salvo again.
- Prophetic accuracy bug: Units/structures split their salvos between multiple targets based on how many projectiles are expected to hit/kill the current target.

Numbers 3 and 5 in particular are what make Ripple Rockets so much more deadly now than they ever were before. In the original 1.10 campaign, when you wandered into Ripple Rocket territory you were greeted with eight rockets all coming down on a single unit, certainly enough to KO it but at least it was only one unit, right? But add in point #5 and now you have eight rockets coming down on 2-3 different units because each ripple rocket "knows" how many rockets-per-target it should be firing, and you end up with double or even triple the casualties.

Now add in point #3 for campaign mission Beta 7. Even if all you do is fortify your LZ with a repair facility and defenses, and never ever wander out into enemy sensor range (which is like 6 squares away from your LZ, btw), no sooner does the Collective make their first air or ground strike against you, you inexplicably have their Ripple Rockets bearing down on your LZ targetting your units. This is something NOT allowed with mobile artillery - only fixed artillery pits can do it. It's a bug + it's exploitable = it's a game balance issue.
Strata @dA, @FAC

Reg312
Regular
Regular
Posts: 681
Joined: 25 Mar 2011, 18:36

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

Post by Reg312 » 11 Sep 2012, 12:08

Stratadrake wrote:Weapon ROF now maxes out at 600 (10 attacks per second), used to be 240 (4 attacks per second). This alone doubles the effectiveness of your Assault MG and Whirlwind AA (their stats are actually unchanged).
are you sure game cannot process 2 hits in 1 update-iteration?

Stratadrake wrote: no sooner does the Collective make their first air or ground strike against you, you inexplicably have their Ripple Rockets bearing down on your LZ targetting your units. This is something NOT allowed with mobile artillery - only fixed artillery pits can do it. It's a bug + it's exploitable = it's a game balance issue.
previous versions of Warzone (2.3x) had same issues with mobile artillery

ripple rockets gets their targets, it looks real, is'nt it?
why vtol units should not be able to mark targets on ground for rocket batteries?

Cyp
Evitcani
Evitcani
Posts: 767
Joined: 17 Jan 2010, 23:35

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

Post by Cyp » 11 Sep 2012, 12:32

Reg312 wrote:
Stratadrake wrote:Weapon ROF now maxes out at 600 (10 attacks per second), used to be 240 (4 attacks per second). This alone doubles the effectiveness of your Assault MG and Whirlwind AA (their stats are actually unchanged).
are you sure game cannot process 2 hits in 1 update-iteration?

...
As far as I remember, if it fires, it doesn't check again when it's time to fire next, until the next update, so it's limited to 10/s (but 10.1/s still means firing about 99 out of every 100 updates).

User avatar
Stratadrake
Trained
Trained
Posts: 197
Joined: 07 Sep 2008, 09:43
Location: Pacific NW
Contact:

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

Post by Stratadrake » 11 Sep 2012, 17:51

Reg312 wrote:
Stratadrake wrote:Weapon ROF now maxes out at 600 (10 attacks per second), used to be 240 (4 attacks per second). This alone doubles the effectiveness of your Assault MG and Whirlwind AA (their stats are actually unchanged).
are you sure game cannot process 2 hits in 1 update-iteration?
Yes. Otherwise you'd see odd cases of Lancers appearing to fire off only one rocket per salvo (when it's actually two rockets simultaneously aimed at the same target). That never happened. I remember seeing some of the original coding at its GPL release and am pretty sure (but can't actually verify) that a weapon's attack round was internally handled using an:

Code: Select all

if (time - lastFired > weaponReloadRate)
{
// Make an attack

// Reload
lastFired = time;
}
And not a:

Code: Select all

while (time - lastFired > weaponReloadRate)
{
// Make an attack

// Reload and check for another shot
lastFired += weaponReloadRate;
}
Stratadrake wrote:...no sooner does the Collective make their first air or ground strike against you, you inexplicably have their Ripple Rockets bearing down on your LZ targetting your units. This is something NOT allowed with mobile artillery - only fixed artillery pits can do it. It's a bug + it's exploitable = it's a game balance issue.
previous versions of Warzone (2.3x) had same issues with mobile artillery
ripple rockets gets their targets, it looks real, is'nt it?
why vtol units should not be able to mark targets on ground for rocket batteries?
Because (1) marking targets is the role of a sensor unit, and (2) yes you can issue attack orders against any discovered enemy structure on the map, but the unit in question can only start lobbing shells when they actually SEE it (in "real time"). Even mobile artillery plays by this rule ... but fixed artillery pits are exempt?

Also interesting to know this goes back to the 2.x series.
Strata @dA, @FAC

Reg312
Regular
Regular
Posts: 681
Joined: 25 Mar 2011, 18:36

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

Post by Reg312 » 12 Sep 2012, 11:43

Stratadrake wrote:
Because (1) marking targets is the role of a sensor unit, and (2) yes you can issue attack orders against any discovered enemy structure on the map, but the unit in question can only start lobbing shells when they actually SEE it (in "real time"). Even mobile artillery plays by this rule ... but fixed artillery pits are exempt?

Also interesting to know this goes back to the 2.x series.
role of sensor units can be changed, who setted them to that role?
i remember games in 2.3x where you could make many artillery emplacements in invisible mode
(they just did not attaced until you build first sensors)

also in 2.3x was common situation when artillery draw all fire on 1 unit
i think it was also wrong as prophetic selective fire

i think we need special buttons/commands for artillery
- fire on ground (why not?)
- hold fire (for ground emplacements)
- CB-fire only
......
but to make this we need review/remake whole artillery system
it is hard to simplify all this

User avatar
Stratadrake
Trained
Trained
Posts: 197
Joined: 07 Sep 2008, 09:43
Location: Pacific NW
Contact:

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

Post by Stratadrake » 13 Sep 2012, 06:54

Reg312 wrote:role of sensor units can be changed, who setted them to that role?
Any unit can explore the map and discover whatever things (enemy patrols, enemy base, etc.) in the process, but only sensors/commanders can be the "eyes" for artillery.
i remember games in 2.3x where you could make many artillery emplacements in invisible mode
(they just did not attaced until you build first sensors)
Meaning that if you wanted to strike enemy fortifications with fixed artillery pits you actually had to build a sensor tower nearby? That is how it should be, but in 3.x anything other than a wall/tanktrap will serve the sensoring role. (Technically even walls and tank traps have ZNULLSENSOR and its 8-square sight range, it must be special coding that exempts them.)
also in 2.3x was common situation when artillery draw all fire on 1 unit
Which is exactly how mobile sensors/artillery works - sensor only targets 1 thing at a time, artillery can only shoot that sensor's target. Defenses generally take aim at closest target to them - with artillery this means that if enemies are approaching from only one direction and all in a cluster your artillery will generally aim at only one of them at a time until/unless they spread out. Spreading out your artillery pits helps with this (like with using Vindicators against enemy VTOLs, without prophetic target selection all those Vindicators would unload only on the very first thing they see and you could slip a whole fleet of VTOLs through the reloading time.)
Strata @dA, @FAC

Reg312
Regular
Regular
Posts: 681
Joined: 25 Mar 2011, 18:36

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

Post by Reg312 » 13 Sep 2012, 10:28

@Stratadrake:
Stratadrake wrote:but only sensors/commanders can be the "eyes" for artillery.
...
Which is exactly how mobile sensors/artillery works - sensor only targets 1 thing at a time
it was changed, i dont see reasons for rollbacking that
when artillery strikes singe units one by one... it looks just stupid
Stratadrake wrote: without prophetic target selection all those Vindicators would unload only on the very first thing they see and you could slip a whole fleet of VTOLs through the reloading time
IMO that was just wrong,
you researching all game to get most advanced AA, and you see this mindless attacks with all rockets to 1 first vtol


Artillery/sensor system should be reviewed/improved/..
it worked wrong in 2.3 and it works wrong in 3.1/ i prefer variant 3.1

User avatar
Stratadrake
Trained
Trained
Posts: 197
Joined: 07 Sep 2008, 09:43
Location: Pacific NW
Contact:

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

Post by Stratadrake » 13 Sep 2012, 17:49

Reg312 wrote:@Stratadrake:
Stratadrake wrote:but only sensors/commanders can be the "eyes" for artillery.
...
Which is exactly how mobile sensors/artillery works - sensor only targets 1 thing at a time
it was changed, i dont see reasons for rollbacking that
Really? Obviously, if you only select one target for your sensor unit (especially if it is otherwise ordered to "hold fire") then its attached artillery has only one thing to shoot at. Whereas if you issue an attack queue of several targets (Shift+target) to your sensor then artillery can pick whichever target in the queue it wants, but ultimately, a user picking a target means "shoot this thing only", or at least "shoot this thing first".
Stratadrake wrote:without prophetic target selection all those Vindicators would unload only on the very first thing they see and you could slip a whole fleet of VTOLs through the reloading time
IMO that was just wrong,
you researching all game to get most advanced AA, and you see this mindless attacks with all rockets to 1 first vtol
The same thing used to happen with Ripple Rockets and Archangels due to it being able to unload the full salvo before the first one even hits.
it worked wrong in 2.3 and it works wrong in 3.1/ i prefer variant 3.1
"Variant" ?
Strata @dA, @FAC

User avatar
Stratadrake
Trained
Trained
Posts: 197
Joined: 07 Sep 2008, 09:43
Location: Pacific NW
Contact:

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

Post by Stratadrake » 03 Oct 2012, 06:49

On the accuracy fix, I saw the suggested patch but I can agree that it's probably not the best way to go about it. And since it was on my mind this afternoon, I thought up something:

-- All shots get some degree of scatter, not just misses. Why make a distinction? Remember the actual hit/miss is determined by a hitbox check; weapons lead moving targets, and projectiles are generally fast enough that (minor exceptions aside) if it's meant to be a hit it will hit.
-- Let's say the default scatter (representing a weapon with 100% accuracy) is 32 distance (PIE) units at a range of 4 squares. This should be suitable to get, say, 100% accuracy against a Viper or cyborg sized target at that range.
-- Scatter should be treated like an angular deviation; the scatter distance increases with range in a linear manner.
-- How does accuracy factor in? The "worst-shot" scatter is calculated in the form of (base scatter) / (rated accuracy). If a weapon has 50% accuracy, base scatter is 32 distance but worst-shot is 64; but regardless of how, this means that about half of shots land inside the target hitbox and half land outside.
-- Target size also affects accuracy; smaller targets are harder to hit.

It's not perfect, but the idea is to keep it as conceptually simple as possible. No "forced" miss trajectories; the trajectory either hits or misses, but it is calculated in a way to replicate the original base accuracy.

Alternately, if we wanted to do a really quick-and-dirty fix for accuracy, just implement one simple rule:

- No "miss" projectile ever inflicts direct-hit damage. Period. Even if a "miss" strikes the target hitbox, the target can only receive splash/incendiary damage (if any) from the hit. (We could always handwave that as, sometimes a shot simply deflected off the armor instead of penetrating or damaging it.) Vice versa does not apply - a "hit" projectile must still strike the hitbox to inflict damage.
Strata @dA, @FAC

User avatar
Iluvalar
Regular
Regular
Posts: 1803
Joined: 02 Oct 2010, 18:44

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

Post by Iluvalar » 03 Oct 2012, 21:09

So a 100% accuracy weapon would miss 75% of it shot against a cyborg at a range of 8 ?
Heretic 2.3 improver and proud of it.

User avatar
Stratadrake
Trained
Trained
Posts: 197
Joined: 07 Sep 2008, 09:43
Location: Pacific NW
Contact:

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

Post by Stratadrake » 04 Oct 2012, 17:24

What's the diameter of a cyborg?
Strata @dA, @FAC

User avatar
Iluvalar
Regular
Regular
Posts: 1803
Joined: 02 Oct 2010, 18:44

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

Post by Iluvalar » 05 Oct 2012, 02:48

32
1/2 of a viper (68)
Heretic 2.3 improver and proud of it.

Post Reply