Jump to content
EleTD.com
Sign in to follow this  
Guest Shadowplay

Possible abuse with instakill towers

Recommended Posts

Guest Shadowplay

While testing out the open 4.0 Beta it occured to me that the way you have chosen to make the difficulty levels opens up for a drastic but with towers based on % life for an autokill.

As you play on lower levels the creeps start out with lower % of hp and are therefore exponentially easier to kill instead of the otherwise Linear flow in the difficulties.

I suggest you change the difficulty modifier to an aura or an auto generated "upgrade" for the creeps based on the difficulty people choose.

As each level of Armor gives you 6% extra EHP (effective HitPoints) it will be a simple matter of giving them an aura/upgrade which gives them, say +10 Armor for a +60% effective HP increase, thus eliminating the abuse.

Otherwise good job with the new version (although I dislike the Divine Shield for Mechnical Units and the fact that the Undead units' Reincarnation screws over bad hosts badly)

Share this post


Link to post

yeah, drowning tower is >50% armor ignoring on ve and >66% armor ignoring on vh. if it's not allready affected by armor?

Share this post


Link to post

The key is that now you have to go for a triple to have instakill, the new gold is as well a triple and far less useful then in 3.0 where it was a dual.

Im going to test it.

Tested, a good rusher could rush between waves 22-42, Well+Drowning is a good rusher build and can rush on VE.

Replay: http://rapidshare.com/files/88302909/Po ... B.w3g.html

Sorry, you need to have the map in "...\Maps\Beta Testing" to be able to see it. I choosed the wrong copy of the map. :?

As you see you can rush badly on VE if you are good (better then me, oc). I think that this concearn is real.

Share this post


Link to post

thinking about it, i think the best way to rebalance this tower should be adding a compensator to the trigger.

Being h the handicap as ve=0,5 and vh=1,0 (50% life to 100% life)

%instakill = 3000 life/creep actual life

better would be:

%instakill = (3000/creep actual life) x h

this would at least remove the advantage from the skill itself. Then just apply the instakill damage:

water damage = (max creep life - actual life) x 1,34. This way a darkness creep would survive the instakill with near to 0 hp. Enough to finish him off with arrows. A composite creep wouldn't, leaving the skill very useful.

Share this post


Link to post

disagree. how about it combined with armor reducers?

why not just make it affected by armor? And armor type.

It don't need to be imbalanced to be unique

Share this post


Link to post
water damage = (max creep life - actual life) x 1,34

This would be armor affected.

The thing is that simply being the actual life water damage would leave almost all of the creeps alive. I ex:

1 creep at 2000 life, 10000 maxlife and 15 water armor.

Instantkill --> Waterdamage = 2000 water damage

2000 damage, armor reduction 47% (15 armor), water attack to water armor 100% damage; 2000 x (1-0,47) x 1 = 1060 real damage. Creep alive with 940 hp.

1 creep at 2000 life, 10000 maxlife and composite armor.

Instantkill --> Waterdamage = 2000 water damage

2000 damage, armor reduction 47% (15 armor), water attack to composite armor 75% damage; 2000 x (1-0,47) x 0,75 = 795 real damage. Creep alive with 1205 hp.

So, if the skill is unable to kill creeps under normal conditions, whats the point on having an instantkill skill?

What we are trying to achieve is that:

1) The instant kill chance is adequated to the difficulty, remaining constant for how hard it is to damage creeps due to difficulty. If the chance is multiplied by handicap it should do the trick.

2) That the damage bears in mind the armor type of the creep, not instant killing at all in some cases, i.ex. darkness.

water damage = (actual life) x 1,34 x (1/armor reduction% (0,47 on vh))

would kill all creeps except darkness, and

water damage = (actual life) x (1/armor reduction% (0,47 on vh))

would kill all creeps except composite and darkness, which would survive with low hp.

The question just is what exactly you want to achieve by making the instant kill damage water damage. Doing it right away would make instantkill a "nearly instantkill"-skill.

What we could do is use the (max hp-actual hp) as a water damage, like:

1 creep at 2000 life, 10000 maxlife and 15 water armor.

Instantkill --> Waterdamage = (10000-2000)= 8000 water damage

8000 damage, armor reduction 47% (15 armor), water attack to water armor 100% damage; 8000 x (1-0,47) x 1 = 4240 real damage. Creep dies.

1 creep at 3000 life, 9000 maxlife and darkness armor.

Instantkill --> Waterdamage = (9000-3000) = 6000 water damage

6000 damage, armor reduction 47% (15 armor), water attack to composite armor 75% damage; 2000 x (1-0,47) x 0,5 = 1590 real damage. Creep alive with 1410 hp.

Still, this wouldhave the inconvenience to leave alive creeps with a low max hp value. Which brings me to the point, what sense does it make to spare the life of weak creeps?

The best solution is that instant kill does a fix amount of water damage, like 15000 water damage on instantkill, which is affected by armor type and difficulty.

A weak creep at full health, element armor; 1 creep at 10000 life, 10000 maxlife and 15 water armor.

Instantkill --> Waterdamage = 15000 water damage

15000 damage, armor reduction 47% (15 armor), water attack to water armor 100% damage; 15000 x (1-0,47) x 1 = 7950 real damage. Creep alive with 2050 hp.

Same creep at full health with weak element armor; 1 creep at 10000 life, 10000 maxlife and 15 fire armor.

Instantkill --> Waterdamage = 15000 water damage

15000 damage, armor reduction 47% (15 armor), water attack to fire armor 200% damage; 15000 x (1-0,47) x 2 = 15900 real damage. Creep dies.

A stronger creep almost dead, element armor; 1 creep at 5k life, 100k maxlife and 15 water armor.

Instantkill --> Waterdamage = 15000 water damage

15000 damage, armor reduction 47% (15 armor), water attack to water armor 100% damage; 15000 x (1-0,47) x 1 = 7950 real damage. Creep dies.

Same creep, with oposite element armor; 1 creep at 5k life, 100k maxlife and 15 darkness armor.

Instantkill --> Waterdamage = 15000 water damage

15000 damage, armor reduction 47% (15 armor), water attack to dark armor 50% damage; 15000 x (1-0,47) x 0,5 = 3975 real damage. Creep alive with 1025 hp.

This way the skill would be:

-Longer useful on easy games or on harder ones combined with armor reducers, would after a certain point not instant kill on harder difficulties

-Be longer useful against fire waves, and start sooner being useless against darkness

-Some kind of critical strike skill with chance depending on current health more that just instant killing regardless conditions

-Still be usefull and not ignoring armor

So my suggestion is:

Multiply chance% x h (handicap)

Make the tower deal 15000 water damage instead of simply killing the creep.

Share this post


Link to post
Guest Shadowplay

I fail to see why it is so hard to just give the creeps less armor on lower difficulties? It would definately be easier, look better (as all creeps would start out with 100% life) and he ALOT faster to fix with a simple dummy unit in the middle of each player lane with a corresponding armor aura - the buff could be Very Easy, Easy, Medium, Hard, Very Hard - that would also make it easier to see what difficulty a player is playing on

Share this post


Link to post

The only advantage I see is merely cosmetic. You hava a strong visual feedback of your difficulty (or the difficulty of other players), that could oc be replaced by something else.

But I disagree that it would be easy to implement as values, denying handicap and extreming the armor differences between difficulties would mean to rebalance many things again to keep the whole thing as it is, not to mention that it would mess up the way armor reductors work right now.

Share this post


Link to post
Guest Sancdar
I fail to see why it is so hard to just give the creeps less armor on lower difficulties? It would definately be easier, look better (as all creeps would start out with 100% life) and he ALOT faster to fix with a simple dummy unit in the middle of each player lane with a corresponding armor aura - the buff could be Very Easy, Easy, Medium, Hard, Very Hard - that would also make it easier to see what difficulty a player is playing on

You do know that armor is different between difficulties, right? I'm a proponent of modifying max life as well, but I don't believe that a system based purely on adjusted armor would work perfectly well.

Share this post


Link to post
Guest Shadowplay

Yeah you are probably right - come to think of it, it would be a massive hazzle to rebalance the values.

But then I dont get why the fix isnt as easy as a trigger giving a specific upgrade to the creeps. This could be based on the Grunts hp upgrade and with some relatively simple JASS code it should be possible to make the HP increase in percentages rather than a fixed number.

It would (with some extreme simplifications) not include more than:

IF difficulty = very hard THEN nothing

IF difficulty = hard THEN upgrade HARD tech (*which would give -10% max hp*)

IF difficulty = medium THEN upgrade MEDIUM tech (*which would give -20% max hp*)

IF difficulty = easy THEN upgrade EASY tech (*which would give -30% max hp*)

IF difficulty = very easy THEN upgrade VERY EASY tech (*which would give -40% max hp*)

or whatever the percentages are. The point is that upgrades are vastly overlooked in coding in my oppinion and they would be more than adequete to fix this in an easy way - as long as one remembers to start with very hard as a starting point and then give a negative "upgrade" to all creeps spawned at a players location to have lower max hp

Share this post


Link to post

@Shadowplay: Your method does not work in ETD because players can have varying difficulties amongst each other.

Share this post


Link to post
Guest Sancdar

Good point, that would only work in games with 1-4 players.

Share this post


Link to post
Sign in to follow this  

×
×
  • Create New...