Disable dynamic EMP time

[Tobi]

Changed it.

Now unit gets stunned if it paralyzeDamage > maxHealth, and paralyzeDamage shouldn't get canceled by low-paralyzeTime weapons anymore.

[Argh]

Ok, so now weapon damage table --> armorType --> damage --> paralyzeDamage + damage? Just flat out?

I wrote an instant paralyzer variant in Lua:

Code: Select all

	if paralyzer then
		 local h,mH,pD,cP,bP = Spring.GetUnitHealth(unitID)
		Spring.Echo("h = "..h,"mH = "..mH,"pD = "..pD,"cP = "..cP,"bP = "..bP)
		if pD < mH then
			return damage + mH
		else
			return damage + pD
		end
	end

First hit == paralysis. Second hit == paralysis + current paralyzeDamage.

Only problem is that the treatment of the paralyze time is funky. On the first hit, a weapon doing 1000 damage seems to do roughly 1 second of paralyze.

On the second hit, it's nearly 7 seconds, due to

Each slow update (every 16 frames) paralyzeDamage is decreased with maxHealth * (16/30/40) = maxHealth*0.025*16/30.

This isn't going to produce very logical behavior- things with lots of hitpoints, even if set up to be extra-vulnerable to paralysis, become free of paralysis faster, arbitrarily.

It should just ramp down 1000 units / second, so that building timing and balance is straightforward. 1000 paralyze damage == 1 second game-time. If people want to make Units recover faster / slower than that, they can use Lua to easily manipulate the logic then.

Then you can get rid of all the extra cruft, like having a timing variable in the WeaponDef (or just treat that as an override- if defined, ignore damage, multiply that number by 1000).

Or... even better yet... fix it however you want to, but let us turn that whole system off with a mod option. Now that I know how it works, I'd rather do it in Lua.

[TheFatController]

So just to get my head around all that, if a BLADEWING_LYZER has "paralyzetime=7;", will there still be circumstances where a unit can be paralyzed for longer than 7 seconds?

[Argh]

Yes. Because the proposed change means that it's the step-down relationship between maxHealth and time that now governs how fast that paralyzer damage drops back down below maxHealth.

[Tobi]

So just to get my head around all that, if a BLADEWING_LYZER has "paralyzetime=7;", will there still be circumstances where a unit can be paralyzed for longer than 7 seconds?

With only my change the only case in which this can happen is if it is being / has just been paralyzed by another weapon, which has paralyzeTime > 7 seconds.

[Tobi]

Only problem is that the treatment of the paralyze time is funky. On the first hit, a weapon doing 1000 damage seems to do roughly 1 second of paralyze.

On the second hit, it's nearly 7 seconds, due to

Each slow update (every 16 frames) paralyzeDamage is decreased with maxHealth * (16/30/40) = maxHealth*0.025*16/30.

This isn't going to produce very logical behavior- things with lots of hitpoints, even if set up to be extra-vulnerable to paralysis, become free of paralysis faster, arbitrarily.

It should just ramp down 1000 units / second, so that building timing and balance is straightforward. 1000 paralyze damage == 1 second game-time. If people want to make Units recover faster / slower than that, they can use Lua to easily manipulate the logic then.

Then you can get rid of all the extra cruft, like having a timing variable in the WeaponDef (or just treat that as an override- if defined, ignore damage, multiply that number by 1000).

Or... even better yet... fix it however you want to, but let us turn that whole system off with a mod option. Now that I know how it works, I'd rather do it in Lua.

The way you suggest it units with more health would stay paralysed much longer. Currently, independent of hitpoints, effective remaining paralysis time is reduced with 1 second every second.

The unit does 1 second paralyze damage on first shot because it first has to waste damage on getting the unit paralyzed in the first place. This is the first maxHealth damage. In your example the remaining damage happens to be good for 1 second of paralyzation. The second shot can be fully used to add time, so this happens to add 7 seconds. This goes up to the time specified by paralyzeTime in the weapon.

So if you want weapons that pretty much instantly do the full paralyzeTime of paralyzation, just make damage for paralyzation weapons a few orders of magnitude higher then damage for normal weapons.

[TheFatController]

So just to get my head around all that, if a BLADEWING_LYZER has "paralyzetime=7;", will there still be circumstances where a unit can be paralyzed for longer than 7 seconds?

With only my change the only case in which this can happen is if it is being / has just been paralyzed by another weapon, which has paralyzeTime > 7 seconds.

Excellent.

[Pxtl]

Personally, I liked the "damaged units are more easy to stun" behavior, but I agree that "thus, dealing damage effectively also deals stun-damage" side-effect was undesirable.

Either way, since my modding projects are in torpor, I'm hardly a stake-holder.

Anybody know the OTA behavior?

[Gota]

Was there emp in OTA?
In any case IMO hurt units should not become paralyzed faster.
I think it makes more sense for the paralyze to use the units full hp as a base.

[Pxtl]

Was there emp in OTA?
In any case IMO hurt units should not become paralyzed faster.
I think it makes more sense for the paralyze to use the units full hp as a base.

I don't know whether to laugh or cry at reading this.

Yeah, there were paralyze units in OTA. The Arm spider and tactical missile launcher building were stun units.

[Google_Frog]

Please make this optional. CA emp is balanced around damaged units being easier to paralyze and the ability to keep units paralyzed by damaging them.

[Gota]

Was there emp in OTA?
In any case IMO hurt units should not become paralyzed faster.
I think it makes more sense for the paralyze to use the units full hp as a base.

I don't know whether to laugh or cry at reading this.

Yeah, there were paralyze units in OTA. The Arm spider and tactical missile launcher building were stun units.

I support laughing if you must pick one :)

[TheFatController]

Please make this optional. CA emp is balanced around damaged units being easier to paralyze and the ability to keep units paralyzed by damaging them.

I'm sure you'll have no problem recreating the old functionality with lua as BA had to do with the reclaim change.

[Google_Frog]

Please make this optional. CA emp is balanced around damaged units being easier to paralyze and the ability to keep units paralyzed by damaging them.

I'm sure you'll have no problem recreating the old functionality with lua as BA had to do with the reclaim change.

Which change? There's lots of options about reclaim in modrules so you should be able to find it there.

[TheFatController]

Where they replaced the crazy reclaim formula with the sensible one, but as BA was balanced on the crazy formula it had to be reinstated with a Gadget, no modrule was provided to access the old formula.

[Otherside]

Please make this optional. CA emp is balanced around damaged units being easier to paralyze and the ability to keep units paralyzed by damaging them.

Do not want :/

EMP is overpowered because of the stupid perma stun

[Neddie]

If it is overpowered, it is because it wasn't balanced correctly given the design. This mechanic can be balanced as well.

[Google_Frog]

Where they replaced the crazy reclaim formula with the sensible one, but as BA was balanced on the crazy formula it had to be reinstated with a Gadget, no modrule was provided to access the old formula.

There's a difference. The old crazy reclaim formula was just a holdover from TA.

The different EMP behaviors on the other hand are based on simple formulas and each have their merits for different game designs.

[imbaczek]

you can increase/decrease the amount of paralyze damage in UnitPreDamaged.

[Gota]

Maybe CA should think about rebalancing emp so they are balanced with this new mechanic since the one that is currently implemented is far less intuitive.