Optimizing Tower Defense for FOCUS and THINKING - Defender's Quest


The design philosophy for Defender's Quest boils down to two guiding principles that play to the Tower Defense (TD) genre's strengths:

  1. Let the player FOCUS
  2. Test the player's THINKING

Note the verbs. We don't want to TEST the player's focus (ie assault them with strobe lights and dancing kitens and

This is a companion discussion topic for the original entry at http://www.fortressofdoors.com/optimizing-tower-defense-for-focus-and-thinking-defenders-quest/


A great article, as usual.

A few comments:

  1. The biggest flaw of Cursed Treasure (and many other games too) isn’t actually in timing issues (made harder without slow/pause) - it’s the random and its power. Unless I was doing something wrong, succeeding against the few ninja bosses essentially required grinding and returning later (with cheaper and stronger magic) OR luck (getting them sent back by a crypt, so they spent most of their invisible period wandering near spawn).

You say yourself that it’s good when one only has his decisions to blame - and I would say that games that want to get close to this should do their best to avoid RNG usage whereever possible. Critical hits are so popular mechanic that you can hardly do without them, but other than that, the design should try to avoid “random” things and convert them to direct buffs/debuffs whereever possible. (I.e. my suggestion for dark enemies would be that unless they’re lit up, they just take significantly reduced damage, even if it’s not exactly logical).

  1. I agree about air units. Usually they’re done quite badly - but there are games (i.e. Bubble tanks series) where most towers can hit them, and so the decisions for the player are about “will 10 versatile towers be enough or should I instead get 9 versatiles and one AA”?

  2. Number of towers CAN be high and successful at the same time. Bloons TD, especially the latest (5) does quite a good job, and also has daily challenges where some towers are disabled/restricted in upgrades, thus effectively forcing users out of their comfort zone and teaching them that even without their favorite tower, they can manage.

I agree especially about 2D vs 3D. Eye candy in general doesn’t make a good game :slight_smile:


Wow oh wow. I agree with virtually everything you’ve mentioned here that as I was reading this I literally at times wanted to shout “OMG YES!” “EXACTLY!”
And I suppose this is why I really enjoyed DQ1 and why I continue to almost everyday keep checking your blog, forum and everything to see what progress is like as I so eagerly await DQ2. As you have demonstrated from this blog post of yours, you clearly (in my opinion anyway) know what it takes and can tick virtually all the correct boxes that make an excellent TD game.

As I was reading this, almost the whole way through it made me think of a TD game that I recently tried out and well admittedly continued to play for a few days until I realised that rather than liking it, I was TRYING to like it and as such stopped playing it soon after. And a lot of the things you mentioned here is where this particular game went wrong. This was a game on Kongregate by the name of “Prime World: Defenders” (PWD).

3D is totally unnecessary which is what PWD was. I’ve never understood the point of it in a TD game. I always thought that if devs put time NOT into all the 3D hassle and animations etc. but instead, put the time and money into making more levels, developing on the storyline, improving on the upgrade system and all that other stuff, that would be a far FAR better trade off and make the game much more enjoyable in the long run. 2D games can be made to look really good and aesthetic too so I don’t see why they need to strive for 3D. And quite often, it can actually have a reverse impact as they sometimes can look horrible and messy.

I also hate that system where you have to build things exclusive just to take out a certain type of enemy (or as you mentioned Lock and Key enemies.) Some games really take this system too far. Looking at PWD, you had to have annoying air towers, then there were invisible enemies, then you needed flame towers to take out the swarms. I much prefer as you mentioned, other ways to reliably take them down. And yeah, you did have invisible enemies in DQ1 but you didn’t take the whole concept of exclusive enemies and exclusive towers too far and so it was fine.

You also mentioned about having 20+ towers to choose from and most of them ending up obsolete. PWD was pretty much like that. And in the end you could only take in a handful with you in battle. But even if you could take all of them in, you could only realistically still only use a handful. Like you said, I would much prefer a small selection of towers but have the ability of making them versatile by way of upgrades.

I also agree entirely about providing information. Complete information on your tower’s capabilities, on your enemies capabilities, full effects of upgrades etc. unless and as you mentioned, information is deliberately left out for the sake of adding a bit of a challenge (as long as it’s not done in an annoying way.) TD games and other RPG games are basically a game of numbers. You need the info to do the maths to be able to figure out whether you can do something and whether or not you need more. And when this is lacking in TD and RPG games and I cant make that decision because the game refuses to give me the info, it frustrates me to hell. As with a lot of TD games, this was also the case with PWD.

I think you did a very good job in DQ1 with balancing the capabilities of melee units with that of rangers. One thing I liked very much about the rangers was how they could NOT attack close range (even though the range area was still pretty large).

Although I don’t feel so strongly about the mazing thing, I do also prefer TD games where the map is predetermined and doesn’t involve mazing. I do somewhat like what GemCraft did though. Maps were most often predetermined but there were still at times limited scope to alter the enemies path. I didn’t quite like Desktop Tower Defence and another one I tried out a long time ago which I think was called Bubble Tower Defence or something as those relied heavily on mazing.

One thing I would disagree with… well… perhaps not disagree but say that I don’t mind, is the thing about scrolling. I can appreciate why it’s better to have everything on one screen but let’s just imagine you wanted to make a grand level of some sort and as such, you needed more space to really create a huge impact and make the level a really memorable one, maybe for the perfect send-off for that boss who has been a big enemy in your path for freedom and peace and whatever, then I wouldn’t mind if I had to scroll for that kind of a map. I guess it would depend on if you had to scroll in multiple directions, but if you could do it in such that you only say needed to scroll in one direction, it could be good. Or perhaps, if it really bothered players, you could make an option to zoom out so that everything could be seen on one screen but that would obviously compromise on the aesthetics so that could be a no-no.

Anyway, to finish up, I really like how you ended your blog post, “Now go forth and make an awesome tower defence game”. Although I cant make games and am on a career path that takes me no where near making games, soooo many times I’ve played TD games and other RPGs too and thought gosh I would’ve done this differently, and this too, and I would’ve tweaked this stat like this and come up with all sorts of criticisms in my head. The thing about DQ1 is that I didn’t have many of those thoughts. It had characteristics in the game which when I played it, I felt like, if I made a game, I would’ve done this and this too.

And also one more thing, (sorry I promised the point before would be the last), and this is quite off point, but thank you so much for not going down that annoying Free2Play model when deciding you were going to make games. I cant begin to say how many potentially good games have been ruined with this annoying rip-off model. PWD again had this annoying model. It only truly satisfies a very small minority (ie the rich who can flash out money at will) but mostly serves to irritate the majority, even those who are willing to pay. I don’t know why developers continue to insist on this rubbish nuisance. Do they really earn a lot out of it? I wish there were some statistic or something which showed the earnings of games that went down this route as opposed to a conventional pay-a-one-off-price-to-play. I’ve only ever paid for games that are a one-off buy to play but not once have I paid for some special currency in an F2P. I really liked how you got a good demo up on Kongregate and used that to grab attention of the wider audience. Yeah there were some rants, but they were mostly by kids who couldn’t afford or aren’t allowed to buy it. I will never understand why developers go down that F2P route but I’m glad that someone of your talent didn’t waste it by making a game that fell in to that category. Although I do think that there are perhaps ways to make F2P so that it isn’t annoying and yet make money for the devs, I think it’s extremely delicate to work with to ensure both parties are satisfied.

Anyway, I think this comment of mine is a bit too long. Got kinda carried away. Thanks for taking the time to post these stuff even whilst you’re busy making DQ2. Really enjoy reading all your posts even though I barely comment.
I just felt like I had to comment on this one though. :stuck_out_tongue:


With respect to panning, zooming and 3D; there is one game that manages to weave these elements into tower defense and still play well. That game being Defense Grid. It makes smart use of the camera where panning is ‘locked’ to the floor plane by translating the camera along two axes, leaving the distance along the 3rd axis (representing height from the floor plane) constant. The camera essentially has a fixed orientation, but the geometry can be rotated about a fixed up-axis, plus the angle of the camera w.r.t. to the floor plane is not strictly top-down, its somewhat closer to eye level. The trick to making 3D work well within a Tower defense context is to impose reasonable constraints that limit the degree of rotational freedom so that it mimics the typical visual range of a person viewing the world from a flat surface as opposed to say a fighter pilot (with full six degrees of freedom). This is where Siegecraft loses points…too much freedom can disorient a player causing loss of focus.

One more thing. The Cursed Treasure level you referenced in your article…you can get a perfect rating without using meteors…you have to cut out the trees in the middle and build more crypts. This will take care of the ninjas with ease. When you hit the last wave, make sure you have a den ready to trigger invisibility on the boss as soon as possible. The second its visible use the frenzy skill to amp up your towers (some of which may have the fear buff assuming you ramped them up accordingly)…takes that boss out in no time.


Hi !

Thank you for this very interesting post ! Fascinating and the comments are too !
I personnally agree with several things. Just disagree about the scrollable maps. I think the principle goal of a TD game is to understand how the waves work. You have to understand which kind of wave, of monster, the way he will travel. And then you begin to think about the good strategy to use.

In such games as Monster and Dwarves TD, you have to scroll to see the entire maps. But once you have understood from which waves come from, where they go, you apply a strategy and scroll the maps is just a supplementary way to challenge player. Challenge because you have to scroll to look at the wave’s way, to scroll to build towers at the good place at the good moment.
But I understand your point. I’ve tested a chinese tower defense game on Android. A bad mazing game. At one moment, you just see a strong wave coming and destroy all of your tower. So you lose and you have to reset the game. At each time you play this level, you have to scroll, look at the wave, trying to understand what they do and how destroy it and then trying a strategy. It’s awful !


I think you make a very good point about Bloons TD 5. I adore the daily challenges in that game, because it’s very easy to get complacent with the same tower set-up on every level. So being forced into using a small selection of uncommon towers is always a very interesting challenge.
Defender’s Quest has similar aspects with the layout of some levels, with limited defense points or the paths simply laid out in a way that challenges you to build an unusual defense.
I hope to see more such creative constraints in the sequel!


September 2014? I’ve been ignoring my RSS feeds for a while!

I just wanted to add a quick note of thanks for this insightful post - I’ve been wanting to write a TD game for years now and still haven’t gotten around to it. But your post (and the comments too!) are very useful. (I wholeheartedly agree about the Prime World comment, although I uninstalled it in the end not because of this, but because it kept loosing my save game and resetting the campaign.) The whole thing with specialized tower types is something I agree with. Actually I agree with all of what you said, it nicely firmed up vague impressions of my own and “one day” when I write a TD game I’ll have to keep all this in mind. I loved DG, DG2 and DQ1. I’ve been playing iBomber too but that seems to suffer from the opposite - not enough tower choice (and you still need dedicated AA) and no story or interaction (aside from the obvious), things which DG and DG did well I thought.


This is why I love DQ: the devs don’t add in frustrating and useless mechanics. Flying and lock-and-key mobs are my biggest gripe with other tower defense games. I can’t wait to play the new game!