The TD Paradox

[Grundy]

Ok, so it's not a paradox but I couldn't think of any other word to use.

So here's the problem....
Tower Defense maps are always either too easy, or too hard.

I cannot think of 1 current TD map that doesn't have 1 set strategy that always wins every time. This is boring because when you find that 1 startegy you will win every game without any challenge.

There shouldn't be one set way that is always a garenteed victory. There should be multiple strategies that can. But then, if there are many ways to win, then the map becomes too easy.

So, how do you make a map that's not too incredibly easy and doesn't have 1 set strategy that owns all others?

 

[Cloak_Master]

Hmm... very true in fact. But, you could always use a unit balancer and make all the towers realtive, but "different" in a sense. This would basically make every race the same, but people would be hard-pressed to realize it because in reality they'd be balance. Good luck on that anyways... lol.

 

[UnknowVector]

I think it has things do due with, spalsh and cold towers. Basicly the first thing a TD player looks for is a cold tower and if theres not one they lose, make no cold towers and balence it from there, nerf spalsh towers so they can't hit everything. Use some special towers with cool abilities to mix it up. And if a wave needs a certain tower to kill it (Not talking about AA towers mabey you should try to remove AA towers too) then nerf the wave.
If a wave is to blasted easy, then give it an ability.

Blizzard's TD has no cold towers and not much in the way of splash towers and no AA towers and look at it, I only play alone because no one wants to host it, but I've never used the same startagy two times and its flipping hard.

Notice maze TD also have new startgies becuase of the maze part, Minotar's Libarinith was almost an expection (I happen to know one of the best ways to defend in it and it wins bascily everytime). It wouldn't have been so bad in that sense if it was team instead of solo, I think thats something else than impacts easiness/strt-overuse.

Don't attempt to spam races, if you can think of only 2 good ones the don't try push for more. Two is hard enough to balence, espicaly in a team game becuause of player postions. In that regrad Green TD was good becuase everyone had a spawn but pink/grey (blue/teal in later versions) and red had a greater importance without compromiseing game balence.

I hope you can manage to make a TD that breaks the treand, I'll help as much as I can if you get a start on it because I would realy like to see something like this. GL

 

[Slaughter]

I thought you were building up to this: what about a td that behaves somewhat randomly.

After 10 fixed levels, the game starts mixing things up. Maybe five waves of normal armor come. Maybe not.

 

[Bartuc08]

make a honey comb TD where it looks sorta like a honey comb duh, and then make units spawn at random points and move in random direction (but random directions towards the middle) that would make it alot harder, with no way to make a possible starategy to always win

 

[Pied]

Most TD's use different armor types on different waves. So every way has another tower that is good for it.

But after a while you just know when the different armor types will come.

And the think with random waves would be cool. So random armor types and random speed/life. (higher speed = lower life and visa versa)

I would like to help with the map tho
PS: i like triggering

 

[Rinpun]

Well you're right, Grundy, it is a sort of paradox

...But here are some ways to solve it

1) Create various races of towers that specialize in something that can smash various waves in the game. THEN, have every single player have a randomized pick on their race! They'll have to know how to play with every single race in order to have a chance at winning. Of course, make sure that you can win with any race first. In this change, it is quite obvious that applying the usual TD ways of thinking (splash and cold...always) will cost you the game and strategizing makes it possible (but very hard) to win.

2) Weaken the towers and strengthen the monsters. Then, add some sort of ability to each tower. Perhaps you could make a combo ability so that two special towers (usually of DIFFERENT races or something so you'd have to plan with a partner) would deal a lot more damage together or something to rebalance the sway of the game. In this change, being uncooperative would cost you the game.

3) There's another TD that does this (and it make things harder), but make every player have a leakage point. Too many times is it noticeable that if one just holes up in the gray gatekeeper spot that one can destroy all spawns and win the game. The winning trick for that would be in stretching from the green and pink (usually lower left and lower right; the last defenses before the gray gatekeeper) to the gatekeepers location in tower mazes. If you make every player have a leakage point, things become hectic as everyone has their own spawns to worry about.

 

[Slaughter]

Idea for random waves. Once you want it to be random, have a triangle:

     a
    b c
   d e f
  g h i j

Wave x is determined: it uses creep a. Wave x+1 uses creep b or creep c, and wave x+2 uses creep d or e if wave x+1 used creep b. And so on.

 

[Yoshii]

well ARKGUILD TD(NOT THE NOOB 100 tower version imtalking original or 15.34?) is pretty hard if do one wrong thing and still be kinda hard if u do strat due to noob hah
reason people dosame strat over and over in a td to win is because of efficiency, if you found a way to win ladder game 100% woulnt you use same stratover and over?

 

[Grundy]

to try to break the mold i came up with a new TD idea.

every wave will be random. (but they will have health/armor set appropriately)

and to mix it up even more each round has a "mutation" and a "deviation" choosen at random.

the "mutations" are special abilities, like increased movement speed, bonus HP, evasion, phase shift, etc...

the "deviations" are more like special events, extra spawns, 1 extra hard mob that spawns with the rest, a 20% reincarnation, things like that.

the mutations will be ordered 1 to n in an array, and each ability will have as many chances as it's index in the array.
example:
mutation[1] = phase shift
mutation[2] = bonus health
mutation[3] = speed
mutation[4] = evasion
add the indexes, 1+2+3+4 = 10
mutation[1] has 1 in 10 chance = 10%
mutation[2] has 2 in 10 chance = 20%
mutation[3] has 3 in 10 chance = 30%
mutation[4] has 4 in 10 chance = 40%
so some mutations are more likely than others, and they will be loaded into the array randomly so they wont be the same chances every time you play.

the deviations will be choosen at random every round.
towers will have different abilities, like critical strike, fast attack speed low damage, high damage slow attack speed, long range, blah blah

i dont know exactly how the different towers will play out vs the different mutation abilities but maybe critical strike towers arent good vs evasion towers because they rely on crits and if they dodge a crit you lose a lot more damage, against phase shift units maybe a fast attack speed tower is good so you can get in a lot of hits between uses, or maybe not i dont know, i'd have to test it out first.

anyway, point is, the mutations will have different chances every time so your strategy and what kind of towers you build will change depending on what mutations are most likey to happen and what are least likely. and the deviations might force you to use at least a little bit of micro if you either want to focus on a boss first or take out the weak ones and save the boss for last, or if you want to kill reincarnated units first or whatever i dont know

 

[mixmax2]

just put in a bunch of RANDOM triggers , like spawn random units, spawn at random points, spawn at random times, spawn with different stats, hmmmmm......*runs off to start a td.

 

[AgentPaper]

Thats the type of "good vrs certain types" you should have, where its not just "attack this type and it does more damage, period" type thing, but it is just the way it works out. and for phase shift, more damage slower would do more overall, since its not wasting shots while it's in the phase shift, for example...Lets say we have 2 towers, each does 100 dmg a sec. the first one does 300 damage every 3 seconds, the second does 10 damage every .1 seconds. versus a normal unit, they will do the same damage, but versus a phase-shifting one, the slower does more damage, lets say they shoot for a total of 10 seconds before the monster passes by:
second:
1 300 dmg
2 units phase shifts
3 finish phase shift
4 300 dmg
5 unit phase shifts
6 finish phase shift
7 300 dmg
8 unit phase shifts
9 finish phase shift
10 300 dmg

Thats a total of 1200 dmg, for the second tower, it goes like this:

1 10 dmg
2 unit phase shifts
3 finish phase shift
4 100 dmg over 1 sec, until coldown is over
5 unit phase shifts
6 finish phase shift
7 100 dmg over 1 sec, until coldown is over
8 unit phase shifts
9 finish phase shift
10 100 dmg over 1 sec

Thats a total of 310 dmg, much much less. of course if the cooldown was longer than 1 second, the difference would be less, but you get the picture. This is the type of thing that should make one thing better against another, air units should be attackable by (if not all) most towers, but they go over the maze, so if you have more mazing towers and shorter range, you do less damage to them. for a evasion unit, nothing would really do more or less damage, as the chance of miss takes te same amount from all types of shooters. same thing for speed and hp, unless you have something that does a % of total hp in damage (bad idea). These are still good to ix things up though.

One idea to add some variety is player-trained creeps, that move automatically and fight the incomeing creeps. the player would buy a unit, and hen it would move and fight the creeps along the path. (you might have some trouble making sure it meets them where the towers are, though) there could be different types better for various tasks, like melee units that do very little damage, but can hold a unit and take hits well, and ranged units that have little hp but do much damage, and flying spellcasters that cant attack, be attacked, or block pathing, but can cast spells to weaken the enemy or bolster your troops/towers. This would add a whole new level of depth to the game.

 

[Slaughter]

Interesting analysis, paper. Don't forget that the fast-attacking tower causes the creep to phase shift more often, thus limiting their forward progress more.

TD makers: don't forget that the creep makes no progress through the maze when phase shifting, so it isn't necessarily an advantage. Normal creeps are vulnerable to attacks when walking...and so are phase shift creeps. Now blink...that is a whole different story...blinking creeps a la Blizzard TD are such a pain.

 

[Shadowy Fear]

Making 'lane abilities' or area abilities could be a cool feature too. I'm not talking about giving the grey gatekepper more gold; making unique, reactable objects, like rocks you can push to create an avalanche in certain areas, would be an interesting way to play. For another example, a player's main builder might stand on a circle, remain frozen for 20 seconds, and then unfreeze as a neutral unit on the side of the lane fires missles at the creeps. Don't make it just massing towers - make it so that there are multiple ways to win but none of them are easier than the rest.

Another thing to do is to make the creeps stronger or weaker depending on how well you destroy them.

 

[AgentPaper]

Actually, both the fast shooting and slow shooting towers will stop it just as much, as long as the cooldown+cast duration are larger than the cooldown of the slow tower.

 

[Grundy]

and for phase shift, more damage slower would do more overall, since its not wasting shots while it's in the phase shift, for example...Lets say we have 2 towers, each does 100 dmg a sec. the first one does 300 damage every 3 seconds, the second does 10 damage every .1 seconds. versus a normal unit, they will do the same damage, but versus a phase-shifting one, the slower does more damage, lets say they shoot for a total of 10 seconds before the monster passes by:
second:
1 300 dmg
2 units phase shifts
3 finish phase shift
4 300 dmg
5 unit phase shifts
6 finish phase shift
7 300 dmg
8 unit phase shifts
9 finish phase shift
10 300 dmg

Thats a total of 1200 dmg, for the second tower, it goes like this:

1 10 dmg
2 unit phase shifts
3 finish phase shift
4 100 dmg over 1 sec, until coldown is over
5 unit phase shifts
6 finish phase shift
7 100 dmg over 1 sec, until coldown is over
8 unit phase shifts
9 finish phase shift
10 100 dmg over 1 sec

Thats a total of 310 dmg, much much less. of course if the cooldown was longer than 1 second, the difference would be less, but you get the picture.

no, they would still both do the same total damage it would just be spread out more because towers dont stop doing damage just because 1 target is phase shifted. as soon as it casts phase shift the tower picks a new target to attack.
but since they keep phase shifting they keep picking new targets and sometimes they both attack the same target at the same time (or close enough so that both have a projectile in the air at the same time), 1 will hit then it will phase shift then both towers pick new targets, but the 2nd projectile is still aiming for the same unit thats now phase shfited, it does no damage. that's 300 damage you just lost right there.
if the same thing happens for the fast attacking towers then you lose 10 damage. and then they are both attacking new targets so no more damage is lost