Venny Posted August 6, 2012 Element Tower Defense possesses a gross imbalance on the strategic level, especially among high skill players. Namely, 4-element builds are supreme. If you attempt anything else, someone rolling a 4-element build will utterly obliterate you. Here's how the 4-element build functions: -By wave 30, have level 2 in three elements, allowing access to a level 2, 3-element damage tower. -By wave 40, have level 3 in either Earth+Fire, Water+Nature, or Light+Dark. All damage towers can achieve this because all 3-element damage towers share one of these combinations. -Proceed to completely annihilate waves 45 without building anymore towers, and maybe 1 or 2 at the most for waves 46-50. In the process you have accumulated a ton of net-worth from interest. -Grab level 2 in one more element, unlocking 2 support triple towers (some combination of slows/damage amp towers). -Roll into fruit round with significantly more net-worth than all opponents who rolled something else. -Winning! Top tier builds are either 3-3-3-2 OR 3-3-2-2 + an extra pure tower. 3-3-3 + 2 extra pure towers can also work, simply because this build functions the same as 4-element builds through wave 55, though they typically lose to 4-element builds due to lack of synergy bonuses. Now, I would like to clarify that I have absolutely no problems with the usual surge of net-worth that 4-element builds create from waves 41-50. End-game defenses are more interesting when they can grow large and complicated. So what am I suggesting? Buffing the only other build set-up that exists: 5-element defenses. 5-elements roll either 2-2-2-2-2-1 for periodic towers or 2-2-2-2-3 for a specific elemental pure tower of your choice. 5-element builds offer a lot of synergy that can make up for the lack of a level 3 forge/well/trickery tower. HOWEVER, they do NOT have the ability to compete with 4-element builds during waves 41-55, and during that time 4-element builds develop such an enormous net-worth advantage that 5-element defenses crumble instantly during the fruit round. The key difference between 5-element and 4-element builds are the forms of support by wave 41. By wave 41, 4-element builds are rocking their level 3 wellspring/forge/trickery towers and are receiving 90% increased damage output from their towers (100% from trickery). 5-element builds can drop 2 different level 2, 3-element towers that increase your defense's effectiveness by 30% either through slowing or damage amplification. Together you have 69% increased output, which still falls short of a single level 3 well/forge/trickery tower. Now there are non-calculable synergy affects between your towers that make up for this, but here's where the problem originates: You spent WAY more generating that 69%. Going into wave 41 a 4-element build will have 2 level 2, 3-element damage towers and one level 3 wellspring/forge/trickery tower. All in all you've committed 13150 minerals to active defenses. Going into wave 41 a 5-element build can have 2 level 2, 3-element damage towers, but they'll have to put down TWO MORE level 2, 3-element towers to get both of their supports. That's 20000 minerals to active defenses! 6850 more minerals than a wave 41, 3-3-2 build! The saddest part is that you're only about tying them in damage, despite spending over 50% more than a 3-3-2 opening. So what's the true problem? Both builds have 2 damage towers. The 4-element damage towers have +90% damage, while the 5-element damage towers have an effective +69% output. What's pathetic is that the 3-element support towers are barely covering up that 21% difference, which brings me to the heart of this topic. 3-element support towers are horribly, horribly underpowered and need massive damage increases. Specifically, ALL 3-element support towers should deal damage about 90% of an actual 3-element damage tower. With the 10% support effect, a level 1 3-element support tower would have 99% output compared to a 3-element damage tower. In practical terms, an individual support triple would BE EQUAL to an individual damage triple. To show just how grossly underpowered support towers currently are, I am going to show some damage-per-second-per-mineral (DPSPM) figures. This number is imperative to tower balance, because ultimately the DPSPM of your defense determines how efficient your defense is, and thanks to the interest mechanic, how large your defense spans by the end game. For my baseline I am going to use the Flood Tower for many reasons. The flood tower is widely accepted to be balanced, a viewpoint I agree with wholeheartedly. It also possesses no special abilities of any kind (from a damage perspective). Most importantly, the flood tower's ability to change its splash radius allows it to be compared to almost every projectile tower in the game, including Incantation Towers, Muck Towers, Polar Towers, and Jinx Towers. Here's the data for Flood vs Muck and Polar towers: (DPS/DPSPM versus a single target) Tower:________ Dmg Spd DPS Cost DPSPM Flood Tower___ 300_ 1 300_ 1500 0.2 Muck Tower____ 200_ 1 200_ 1500 0.133333333 Polar Tower___ 200_ 1 200_ 1500 0.133333333 Drowning Tower 1500 1 1500 5000 0.3 Mire Tower____ 1000 1 1000 5000 0.2 Glacial Tower_ 1000 1 1000 5000 0.2 That's right. Mire and Glacial towers, LEVEL TWO triples, are matched by a LEVEL ONE Flood Tower. You are paying 3500 extra minerals for a 30% buff. Here's the data for Flood vs Incantation towers: (DPS/DPSPM versus a single target) Tower:___________ Dmg_ Spd_ DPS________ Cost DPSPM Flood Tower______ 300_ 0.67 447.761194_ 1500 0.298507463 Incantation Tower 80__ 0.31 258.0645161 1500 0.172043011 Drowning Tower___ 1500 0.67 2238.80597_ 5000 0.447761194 Enchantment Tower 400_ 0.31 1290.322581 5000 0.258064516 An Enchantment Tower is WORSE than a Flood Tower. Once you have one Enchantment Tower down you are better off dropping level one flood towers even with no other buffs present. Here's the data for Flood vs Jinx towers: Tower:________ Dmg_ Spd_ DPS________ Cost DPSPM Flood Tower___ 300_ 0.31 967.7419355 1500 0.64516129 Jinx Tower____ 600_ 1___ 600________ 1500 0.4 Drowning Tower 1500 0.31 4838.709677 5000 0.967741935 Hex Tower_____ 3000 1___ 3000_______ 5000 0.6 Just like the others, Level 2 Jinx performs like a level 1 flood or slightly worse. In summary, not counting buff effects, Flood is 61.3% more effective than Jinx, 73.5% more effective than Incantation, and 50% more effective than muck/polar. There are other disparaging factors. Incantation has 12 range, and thus shouldn't equal 90% of Flood Tower's DPS, but the difference should only be minor. Nova Towers, Erosion Towers, Ion Towers, and Barb Towers cannot be compared to a Flood Tower, but Nova Towers and Ion Towers CAN be compared to Quake Towers (Just the Radial DPS). Here is the data for Quake vs Nova and Ion towers: Tower:_______ Dmg Spd_ DPS Cost DPSPM Quake (avg)__ 80_ 0.67 120 1500 0.08 Nova_________ 75_ 1.5_ 50_ 1500 0.033333333 Ion__________ 50_ 1___ 50_ 1500 0.033333333 Seismic (avg) 400 0.67 600 5000 0.12 Supernova____ 375 1.5_ 250 5000 0.05 Plasma_______ 250 1___ 250 5000 0.05 Nova has larger radius than Quake's radial damage, and Ion can move, but Quake also possesses extra single target damage. Nova and Ion are the worst so far. Unfortunately Erosion Towers and Barb Towers have no comparisons, but from personal experience I can say that they join Ion as the worst triple towers in the game. Support towers need massive damage buffs. They should compare to a damage tower of similar handling/function. Some are not far off (Muck/Polar), others a bit worse (Incantation/Jinx) and some need crazy damage buffs (the rest). On a one-for-one basis any single triple tower should match up against any other single triple tower. Here are the new damage values I am suggesting for each support tower. Tower______ Old_ New_ %Increase Muck_______ 200_ 270_ 35.00% Mire_______ 1000 1350 35.00% Polar______ 200_ 260_ 30.00% Glacial____ 1000 1300 30.00% Incantation 80__ 120_ 50.00% Enchantment 400_ 600_ 50.00% Jinx_______ 600_ 850_ 41.67% Hex________ 3000 4250 41.67% Nova_______ 75__ 160_ 113.33% Supernova__ 375_ 800_ 113.33% Ion________ 50__ 110_ 120.00% Plasma_____ 250_ 550_ 120.00% Erosion____ 50__ 125_ 150.00% Corrosion__ 250_ 625_ 150.00% Barb_______ 50__ 120_ 140.00% Spike______ 250_ 600_ 140.00% I gave Polar/Spike towers less of a buff to account for their superior range compared to Muck/Erosion towers. These exact numbers don't have to be used, but they do give an idea of about how much each tower needs buffing. Nova, Ion, Erosion, and Barb in particular need more than double their current damage to catch up to par value. They are the worst towers in the game currently. The numbers will need tweaking in the end. What I'm suggesting is simply a new baseline for the EleTD community to try and move on from there. Going by my suggested numbers, a few towers might need a slight damage drop after testing, and a few might need just a little more damage after testing, but they'd place each 3-element support tower into their expected range of performance. COUNTERING COUNTER-ARGUMENTS No doubt some are wondering, "If support towers deals 90% of a damage tower, effectively matching them on a one-versus-one basis, why build damage towers at all?" Simple. Wellspring, forge, and trickery towers. These provide a quintessential damage buff at a bargain price. 4-element builds especially rely on damage towers because they can receive the level 3 buffs from their maxed out wellspring, forge, and trickery towers. Even for a 5-element build, you will always have 2 level 2 buff towers, and you're GOING to use them as the game progresses. As far as the end game is concerned, damage towers would provide AT LEAST 87% more base damage than a support triple. ([100%/90%] * 1.3 * 1.3) = 187% damage relative to support tower. If rolling a 4-element build, then that number becomes 211% damage relative to a support tower, or 112% more base damage contributed to your defense. As far as the end game is concerned, support towers are still very much a support tower. Another valid question, "If support towers deal 90% of a damage tower, effectively matching them on a one-for-one basis, wouldn't a level 2 support tower have a slight edge on a level 2 damage tower?" Yes, but ONLY during a brief phase of the game (wave 31-40), and only SLIGHTLY. Once you get to level 2, your support tower will have their 90% effectiveness increased 30%, resulting in 111% output versus a pure damage tower. However there are major limitations. You can only maintain this rate of efficiency so long as you unlock new support towers, and from waves 31-40 you can only have ONE level 2, 3-element support tower, so at best you can only have ONE tower outperform your opponent on an efficacy scale. You could spread out, but a 5-element player doesn't want to drop redundant towers. They want to mix and maximize synergy. Dropping an extra copy of your level 2 support would come back to hurt you from wave 41-50. Also, even if a fast level 2 support tower has a slight edge from waves 31-40, your opponent will get back at you during waves 41-50 with their level wellspring/forge/trickery towers. Yes, even with the suggested buffs to support towers, 4-element builds still have a slight edge during 41-50, but that's a GOOD thing. Under the changes, support has a slight edge during waves 31-40, and then damage press has a slight edge during waves 41-50. From wave 51 onward it is anyone's game. Without the changes, damage press beats support from waves 21-60 (when you actually begin to differentiate). There are no options if you are playing competitively. Damage press or fail. One ADDITIONAL suggestion I'd like to make here, because it's on topic: buff wellspring/forge/trickery damage by 50-60%, bringing it in line with the new Jinx tower. That should compensate for the slight edge that a wave 31 support build would experience, and would also eliminate the annoyance of receiving wellspring/forge/trickery from your first two elements in random, because they'd actually be worth building. SELLING POINTS Buffing support towers to this extent will improve the game in many ways. New Players When new players start playing, they are going to experiment with elements because they have no idea what each combination yields. Some of these people are going to grab support towers first, and they are going to expect it to match the damage towers that their opponents picked first. Quite frankly, they're justified in thinking so. When their support tower gets overrun while a single damage tower cleanly demolishes each wave, they are going to leave with a negative impression. Also, tower defense veterans that start playing EleTD for the first time are going to identify the support towers just by reading the effects, and will likely aim to create a strong synergy-oriented defense (5-element build). When they realize heavy-synergy builds can't work in EleTD, they'll misinterpret that as a sign that EleTD doesn't promote tower defensing strategy and look for a new tower defense. To be honest, they'd also be justified in thinking this, because right now you are funneled into damage press builds. Improving support towers will eliminate the strategy-funneling effect and should lure in more TD-Veterans. Given the complexity of EleTD, I'd say TD-Veterans are the #1 targets for the player base. Tower Wars Correct me if I'm wrong, but I believe the reason support towers don't spawn creeps was due to their outright inability to kill what they spawn. Lining up support effectiveness with damage towers should resolve this issue. 5-Element Endgame vs 4-Element Endgame Imbalance Even if you ignore the net-worth disparity between the two, 4-element still trumps 5-element. 5-element spends far more minerals on actual support towers, and so they have less damage towers to operate from, equating to less base damage to receive the benefits of all these support towers. It's not much, but the result makes 5-element a little weaker than 4-element, even when forcing equal net-worths. As stated before, support towers sure as hell don't touch a damage tower in the end-game scenario, but the little extra base damage to your net defense should fix up 5-element against 4-element. Diversity Increased diversity in PLAYABLE strategies will improve EleTD as a spectator game, drawing more interest for new and intermediate players to stay and watch the resolution of the match. It's always fun to see a new concept unfold before you. Buffing support towers will DRASTICALLY increase the number of viable mid-game and end-game builds, allowing for more diversification of strategy. All Random I think all EleTD veterans can agree that random (especially All-Random, no Same-Random), is the most fun way to play EleTD. The only problem is that Random has always been plagued by imbalance. You had to deal with the "Bad Towers" like the old oblivion, the old laser, and of course single element towers and all the 3-element supports. The game right now is approaching an incredible apex. The damage towers are remarkably balanced and the single element towers work great. Supports are the only "Bad Towers" left. Right now if you get a level 2 damage against your opponent's level 2 support then you've probably already won because you'll slaughter waves at twice the efficiency if not more than that. If support towers get buffed, we'll finally have the dream scenario we've all wanted: Perfectly Balanced Random No more "God damn it I got XX combination". EVERYTHING is viable. Some you are more comfortable with, some maybe a little tweaked compared to the rest, but pretty much you can roll with whatever you're given. I know that I'm not asking for a small change. In fact I'm against a small change. Half-way in-between won't cut it. Support towers HAVE to hold their own against damage towers or nothing will change. Balancing triples 1-for-1 against each other will revision EleTD strategy on all levels, but the improvements will speak for themselves. Newbie-friendlier, more viable strategies, balanced random matches... The changes are at least worth testing. Go to top Share this post Link to post
Karawasa Posted August 7, 2012 Excellent read. I will include some variant of this in the next update. Go to top Share this post Link to post
