Hi guy !

Need your help ! I try to understand why a tower defense game may be considered as easy or challenging. One of the aspect of the question maybe is the fact we treat temporal problems such as if they were spatial problems.

On this article, I explain you that and try to give some examples

https://lovetowerdefense.wordpress.com/2015/07/13/tower-defence-games-temporal-or-spatial-problem/

Can you give me some examples of games or levels of TD games you've found very challenging ? What was the problem to solve ? Why was you not able to solve it ? How have you find the solution ?

This distinction between temporal and spatial problem, do you think it's pertinent ?

Thank you in advance for your help.

There are also upgrades and combos to consider, and enemies that are strong or weak against different attacks. Part of that is the spatial problem of getting different types of towers in range of the same spot, but it's also the towers and upgrades that you choose. Sometimes games like Kingdom Rush let you bring the next wave early, giving you extra cash, so there's the problem of accurately assessing when you can let the next wave in, to better prepare you for the later waves.

One of the aspect of the question maybe is the fact we treat temporal problems such as if they were spatial problems.

No, we use placement of towers to to increase their effectiveness, this includes trying to increase the time they are being effective.

Need your help ! I try to understand why a tower defense game may be considered as easy or challenging.

Umm, of the top of my hat i 'd say:

1) complication, aka different types of towers/enemies/maps/spells/etc.

2) speed, how much time does the player have to react(how many reactions does he have to make in a short interval, or can he just set up his defenses and then click "next wave")

3) allowed rate of error; aka if the player can potentially dish out 1.000 damage, do the monsters have combined 900 hp or 750 hp.

The core is, of course, a spatialtemporal problem. area covered x damage per second. If you can slow a unit or change their path, that's equivalent to increasing the area. At this level, it's a pretty straightforward optimization problem, multiplication of the values gives you the output of a tower, which you can balance against each other. The derivative tells you how much benefit you'd get from increasing the area of a weapon versus increasing the damage. Divide by cost and you get efficiency.

Things get interesting, because each wave and tower has different properties. Splash damage weaponry is (are covered x damager per second x unit density). Sniper weaponry is particularly subject to (area covered x min(damage per second, unit hp)). Some towers may have bonuses when used in conjunction. Units often have strengths/weaknesses as well: immune to slow, flying over some weapons, different path taken, etc.

Each wave, then, has a different set of constraints. Ultimately, you just want to hit a given damage output per wave (enough to stop it). In the early waves you need to hit the current output requirement while optimizing your build for the future waves. Perhaps it's more effective to leave the prime location open for the first few waves, until you can put an elite tower there. Perhaps you can easily solve wave 3 with tons of splash towers, but that leaves you too suboptimal in wave 8.

The first time through, you're trying to build a balanced defense that can react to each new wave. Subsequent runs (or when told in advance), you can use the knowledge of what's ahead to prepare specifically for it. Some games are challenging because many options are bad: I've played some where upgrading towers is basically a trick: it's more cost effective to always build more. Some are challenging because they demand very exact layouts. Perhaps the most difficult actively mislead you: The best build for the first 9 waves is catastrophically wrong for wave 10. Instead you need to just barely muddle through the first waves with inferior builds so you've got the right one for the last wave.

In general, I think this temporal problem, how to optimize across different waves, is the biggest challenge. The spatialtemporal problem of laying out towers can certainly be tricky, but it's still usually a pretty straightforward question of efficiency.

Hi guys !

Thank you a lot for your answers ! Very fascinating.

No, we use placement of towers to to increase their effectiveness, this includes trying to increase the time they are being effective.

Hum, curious. Can you tell me how you prove this affirmation ? I've found an article published in 2012 which explain players often try to optimize area cover zone whereas increase the time tower is effective is more efficient.

Baylor Wetzel, Effects of Representation on Solving Complex Spatial-Temporal Problems, 2012

<http://www.aaai.org/ocs/index.php/AAAI/AAAI12/paper/viewFile/4893/5294>

If you look at the post I've published on my blog, I've given several examples where I've lost several times. The problem was precisely I tried to optimised the area cover zone. When I've realized the problem was temporal, I have found the solution.

Sometimes games like Kingdom Rush let you bring the next wave early, giving you extra cash, so there's the problem of accurately assessing when you can let the next wave in, to better prepare you for the later waves.

Hum, curious. Can you tell me how you prove this affirmation ?

I played a few games in my life.

I suppose my "no" should 've been a "nah" though,

as in: it's not realy incorrect, but seems to overly focus on a small (sub-)problem.

Can you give me some examples of games or levels of TD games you've found very challenging ? What was the problem to solve ? Why was you not able to solve it ? How have you find the solution ?

This distinction between temporal and spatial problem, do you think it's pertinent ?

You know, sometimes I'm just not sure why I lost or what I'm supposed to do. I've been playing Kingdom Rush Origins a bit lately, and it's getting really tough compared to the original. I feel like I pick a good spot and decent towers, but it's just not the right combination or something? I think that can be an odd part of tower defense games, where if they get challenging enough, it's not always clear what you're supposed to do to get better. But maybe that's just a higher level of play where you're meant to try a wide variety of strategies to get the perfect one for that level.

Hello,

I played a few games in my life.

I suppose my "no" should 've been a "nah" though,

as in: it's not realy incorrect, but seems to overly focus on a small (sub-)problem.

Uh uh, I understand.

Yeah sometimes I feel in the same situation. But the fact is if you continue to retry these kind of levels, you may continue to lose. Several TD games have been designed as thinking games. So you need to stop and think to determine why you have lost. What's happened in the map ? And if you do that, you find the problem and now are able to solve the problem, to find a good strategy.

For example, I've played to a very challenging mobile game. I've reached one level in which I have lost a first time, then a second, a third. I try to understand the problem, find a solution, apply and continue to lose. And so on. And I was about to surrender when I stopped playing the game and begin to think why I've lost. I've found the real problem, a solution and try to apply this. Successful !