What would you remove from Defender's Quest?

[sir.pinski]

As I'm working on thoughts for a potential sequel, there are some elements of DQ that I think I want to remove or substantially overhaul. We've talked a lot in another thread about what new features we want to add, but sometimes trimming the fat is even more effective.

- Enemy Armor. I think we'll keep some sort of damage-reduction, but armor in its current form just devolved into having to chuck down knights to hack off the armor first. So perhaps more accurately, I want to get rid of "armor break".
- Tower damage. This was highly debated early on for me - I don't like tower defense where the towers get damaged or destroyed, but as an RPG it seemed like having armor and hit points was going to just make more sense. Having spent time with it I'm leaving heavily towards dropping any direct damage against towers. If we still have armor it will likely just affect other stats. Enemies may still have the ability to stun or slow a tower, but probably not direct damage or destruction.
- Skill tree tedium. We've discussed this a little before, but I'm definitely aiming to streamline the skill tree substantially to avoid all of the single point tick-ups that you have to apply each time you level up each hero.

I'm sure this list will keep growing for me as I analyze the current game, but is there anything else in the current game that you find tedious, annoying or absurd that you would pull out for a sequel?

[Polaris2013]

-Enemy Armor I have mixed feelings, but I think I agree with you about the "Armor Break" skill. At least it should not be so effective that you can just put guys with armor break in the front and not worry about armor after that. (perhaps make armor break a Boost 5 skill so it is not so accessible?)

-Tower Health I agree it is generally annoying to have to worry about health in a Tower Defense game. However, there is no mazing in this one so I feel like it needs something to replace that and tower health is not such a bad substitute. Maybe instead of "dying" the defender just gets knocked unconscious for a few seconds and they can't fight during that time. The difference from having debuffs do the same effect is that health has a more controllable "threshold" where you can get overwhelmed before something bad happens, whereas debuffs tend to just get applied as the corresponding enemy type walks past. In other words, with health you can "level-up" past the point where the debuffs hurt you, but with standard debuffs that typically does not occur. Also, concerning armor I 100% agree it should have some other effect than defense. Just adding armor is such a boring thing to manipulate, that should definitely be changed.

-The skill tree I like because it gave a lot of personalization / strategy that you don't get in other TD games. But I admit there was a problem with it. Namely the number of level ups you get in the game. Multiplying the level up pace of ~1 level up per character per map * 36 characters it gets tedious. But that is not a problem with the skill system, it is a problem with the level up pace. This can be attacked in many ways, such as changing the number of characters, and changing the expected number of level ups per map.

The skill tree gives the ability to make each tower fairly unique, but when you have so many it is too difficult to remember what each one's role is so I start making them all the same so I can remember what they do (most of the classes I have 2 archetypes, one for status debuffs the other for damage).

What I would like to see is you get your plot characters and that takes up 6 slots, and then you get your retinue, which is 12 slots to fill with any characters you like. No class restrictions on your retinue, so it could be 12 berserkers, or 2 of each class, etc. This gives more personalization / strategy on how you construct your team as well as mitigating the level up complexity. It also adds identity to characters without eliminating your ability to reuse assets.

Another possibility would be to have a skill tree for the plot character only, and all the others follow that path, but the downside to that is you can't have different characters of the same class do different things, for instance, one set of knights that do armor break and another set that do knockback/stun. So I don't really like that idea, but I'll throw it out there for discussion.

[sir.pinski]

[Reply to Whisperling...Polaris ninja'd me by about 5 seconds]

Haha, interesting. Well, that's part of the reason that I started this thread. I find the skill trees to be tedious and boring, and I do know I've seen complaints that adding skill points to an already-obtained skill only minorly affects it. Yes, I could up the difference per skill point, but that could do some weird things to balance - I'd need to try it out and see. Part of this is that I'm a gamer who does not like micromanagement, as I think I mention in nearly every thread. I prefer strategy to tactics and want to try to feature interesting, important decisions while optimizing and automating as much of the minutia as possible. This is why I think we'll move to a streamlined "skill tree" setup.

I'm not dead set on removing damage to heroes. It could be an alternate approach - maybe a "slain" hero is just knocked unconscious for a recovery time (which a healer may have a "revive" spell to reduce). You're right that a lot of options close if we don't allow enemies to attack, though many of the status effects could still apply even if we don't involve direct damage.

I'm a little torn on the smoke monsters. The whole point was that you'd have to do unusual things with your healers, but they were a single-solution enemy - if you didn't use healers you had no other option. We did pump up the healers' powers to hit 100% later on (I don't know when you first played the game), but even then it was often complained about. If we did something like that again I'd want to make sure there was at least an alternate solution to dealing with them. Armor is in a similar situation - it wasn't interesting before, you just had to toss in knights first. It was one-dimensional and thus didn't make for particularly dynamic or creative gameplay.

Moving units is something I've considered, and there might be a way to make it work. I don't think you'd like a 10 - 30 second move time in practice, but there may be other ways to make something like that make sense. At the same time, it does go somewhat against my philosophy about tower defense: that it should be more hands-off than most games. I'm specifically avoiding juggling-style mechanics (like what Desktop Tower Defense would involve) because I found that type of interaction to detract from the main gameplay and to be annoying. Certainly there are people who enjoy it, but it's not the type of game I want to make or play myself. :)

[sir.pinski]

Polaris:

- Yeah, I think we're generally in agreement on the armor for enemies. "Armor Break" is an odd thing numerically that just doesn't work the way I have it set up in the current game, which is why I have enemies with 1000 armor while heroes are doing only 100 damage. In the next one, either armor break with do a % armor damage, or armor will just decrease by a % rather than a fixed amount, or armor will stay low and maybe "armor break" will be more of a temporary status effect. Or we just won't have it at all and will instead concentrate on high-damage and piercing attacks.

- I proposed a similar solution in the post I put up at the same time as you. I can see a "knock out" option making some sense, and you're right that having some sense of danger does keep things more interesting. I'll want to play around with the numbers a bit though - right now, you're generally either in no danger or immediately dead. There isn't much in-between play and I think that shows a weakness in the system that needs to be addressed.

- I totally understand that the skill tree gives personality, but I don't think my proposal really takes much of that away. What I'm proposing is that you only make major decisions about which skills to select, rather than appropriating points each time you level up. I like that level ups occur a lot, and I like the large size of your army. What I'm thinking at the moment is more of a branching class system. Perhaps the "Squire" class at level 10 can choose between "Swordsman" and "Fencer", and then the "Swordsman" at level 20 can choose between "Knight" and "Berserker", or something along those lines. You'll branch out, have lots of choices, and be able to know what each one is good for. It'll almost play out as more of a job system.

I do like the suggestion of more flexibility to the non-hero unit make-up - I can see something like that working. Though the risk of exploits and balance problems seems much higher when you can create an entire army of the same class, which may be a deal-breaker.

[Polaris2013]

edit: and then I read your reply... oh my =(

Oh I just had another idea about skill tree!

Azra is the only one who gets skill points, but her skill points can be assigned to any character. Leveling up would automatically unlock new skills for use on each character so you don't have to put the first point in a skill just to be able to use it. The skill points you choose are only to upgrade skills, not unlock them.

This actually works quite well with the discussion point I posted above where only the main hero gets a skill tree. Instead of Azra being the only one to get skill points, only the plot hero(s) get skill points. All of his retinue shares from his skill point pool. So when Slak gets a skill point, he chooses which berserker gets an upgrade and which skill they upgrade. Eg. he could upgrade his own Double Strike skill, or another Berserker's Flurry skill, etc. Possibly he gets skill points faster when there are more characters of his type (or possibly not, as that could regenerate the original problem). Skill points are gained on a separate pace from level ups. Maybe every 10 levels a character gets their own skill point just for themself that can't be shared, so newbies don't fall too far behind.

I really like this idea more and more as I think about variations on it.

[sir.pinski]

I think what you're suggesting is actually very close to what I'm suggesting for the skill tree. Essentially I'm trying to whittle down the number of decisions so that they are interesting and important. An alternate way to the branching classes would be to only have 3 skill points per skill in the tree and dole out points much rarer. Either way, I want to reduce the amount of minor tweaking done and keep it more strategic. Your suggestion is a good one, and may in fact be easier than trying to work out a class tree. :)

For the teleporting, my response is that your suggestion doesn't seem all that much less convenient than what you can currently do (sell and rebuild). I'll have to keep pondering to see if it would make sense to build out something for this or not.

[Lin]

The "job" system makes sense, especially if it'd somewhat minimized the tedium of having to remember what skills you've maxed on which guy. I'm hoping there'd be some way to visually differentiate the specializations too, or at least make them easily visible once you click the character (as opposed to having to specifically go into each one's skills to remind yourself what that one can do, then figure out how many you have who can do that already, where to place them, etc etc). With so many around I rarely make the best use of them, and yes, I too tend to just do pretty much the same on all of them just so I can remember what they're good for.

Part of the reason I hardly went back to the normal game after I started playing Hero Mode was because I actually knew exactly what everyone does/needs to do and was forced to plan better. Don't get me wrong, I love micromanaging in the right game, but yes, this specific game, on later levels with long battles and so many guys, you do loose track of who what where. And pausing every three seconds to upgrade each and every one of them (including going into skills and trying to remember what the initial plan was) can make a long battle really really long without adding much to gameplay and strategical thinking.

The other question would be, would it make sense if specialization could/should alter what that class can do on initial placement, ie before upgrades.

The one thing I wouldn't touch is the heroes. They should be able to freely upgrade. You don't want a hero tied down to a class that forces only specific upgrades. The boss should be able to dab in all the jobs, right?

And yes, part of me wants to touch nothing because this game is awesome and works. But then again, a sequel should be a new game, at least in part. Also, happy new year. ;)

[sir.pinski]

Lin:

Yeah, remembering who is who is definitely a challenge right now. One of the reasons we let you customize the colors of the characters is to give you some control over what they look like to give some visual feedback, but it's really too minor and free form to work well. I'll need to make sure it's doable with our artists, since hand-animating a bunch of classes is very time consuming, but hopefully we can work out something that would make sense (perhaps using some overlays to minimize animation). For the heroes we'll just have to see what makes sense - at some level it depends on the story too. I hesitate a little to separate out the heroes too much without going full on "hero mode" on them (i.e. where they are substantially bigger and stronger than other characters). Though I suppose that could be interesting too. Hmmm... (Also, thanks for the compliments and happy new year to you as well!)

Whisperling:

Thanks for the suggestions. :)

[lars.doucet]

I'm very interested in the jobs system myself - the one major downside to it though is that it often induces you to level up each character multiple times.

So, one thing I thought was interesting was how in the Penny-Arcade Rainslick 3 game, was that character classes are something you EQUIP, and that those classes level up independently of the character they are attached to, so you can swap them around. I thought that was pretty neat.

One idea I've been toying with is a merger of this and the FF Tactics jobs system. The "job" levels up so any specializations you put into it stay there, but you can equip it to a different character. So, class-specific skills and bonuses stick with the class, whereas character-specific stats and bonuses stick with the character. It might take some streamlining to keep it from getting overwhelming, but I think it could work.

One challenge that arises is explaining this system in UI and story terms. One way to do it is to tie the "job" with a special piece of equipment, so, say, you make someone a knight by giving them a knight sword. Perhaps the sword is sentient or made of shiny super tech or something, and that justifies how the sword itself "learns" over time, and also justifies why it can be swapped around. You might customize one character with character-specific stats that work really well with that "job," so that's the natural place to put that sword/job, but you still have the flexibility to move it to another character without them having to be a level-1 terrible knight.

Anyways, just thinking...

[Lin]

@Phoenix: Good point about the story. It makes absolute sense to allow room for the story to dictate in which direction heroes go. So refreshing to hear an answer like that. As for customization, really, the simplest overlays would be fine as long as they do the job and allow you to recognize/remember easier. (I'm seeminly one of the rare few that didn't mind the art at all, perhaps because story and gameplay trump all else in my book. I thought it was cute and whimsical.)

In the same line, @lars my initial reaction to the "equipping a job" idea is, again, story. Feels nice to think of the little doodles as folks with more or less experience in battle. I felt quite attached to the late comers to my party, poor things needed protecting or occasionally watched a battle from a safe position just to see how the job's done. Useless little buggers. But on second thought I realized I don't really know any sentient swords personally, so I shouldn't trash them too early. What I'm saying is... I hope he or she gets lines. ;)

Really, all you need to excuse any developing decision is a good writer with enough say in the planning process, and you have one, so I'm not too worried.

Minor addition: I'll throw in one more vote for removing that darn lightning. Or at least really REALLY minimize the use of it somehow. Until hero mode++ I hardly ever touched it, and when I was forced to start using it it really brought the game down. If it's not a boss battle I want to sit back and sip on my coffee after I'm done planning. This is a TD at heart, so it's the strategy that should make or break a level, not clicking.

[sir.pinski]

Agreed on the lightning.

To Lars and Lin regarding the equippable classes, my gut reaction is against it since I feel like it minimizes the characters and the attachment to them. The dress system in FFX-2 was interesting but the result is that no one *really* had a particular class. You could kind of customize them to be what you wanted, but to me it still all felt very...squishy? I dunno - maybe there's a way to make it work, but personally I prefer more permanent growth and identity.

To clear up what I was talking about with my branching class idea, it wouldn't be like FFT jobs. You wouldn't level up multiple jobs for a single character. Instead, as a character levels up, they are able to select a path down a branching job tree. It's a single path that you take, and in practice is more like a skill tree than a job system, but it is streamlined and framed as a sort of "job" to help make the selection and visual feedback clearer. I'll have to draw up a chart or something. Also Lars - we should talk over some of these ideas now that we're getting more serious about a sequel. :)

[The Rainmaker]

the skill tree was too simple... if anything you need to make it so there are more branches and perhaps different ways to spec classes entirely. If you want to streamline it simply make the skill ups every other level and make each point count way more than it does now. Heck I dont care if you make each skill a 1 point investment but as if it had maxxed out impact and skillups every 5 levels if you really want to make them have impact.

The only thing I would remove from defenders quest is A) Archers insane range, if you need to give them insane range make it directional. Like make them face a direction and give them a range that way. B) Smoke enemies. They are super annoying especially on the hero mode challenge as you only get one healer.

I dont think removing armor from enemies really works so much as you could 1) make knights effective but far less so against them or 2) make all classes have the capability of being able to be effective against them.

I think another thing is boost is a bit too important. Maxing out boost on one character was far better than throwing 2 boost 3 characters out there. I think boost needs to be removed and some other formula needs to be reworked. Perhaps either a specific game mode if not a straight out implementation where boost is a global powerup and is less powerful based on the amount of characters you are using.

IE rough example Boost 2 with only one character on the screen doubles his damage and health defense and speed but if you have two characters on the screen it only does .5x more of those stats and so on. Could add even more levels to it than just boost 5 but with great diminishing returns the more levels you put into it to the point its not worth the psy cuz the cost is so high for the next level.

Then you could do something interesting like make someone be able to recall a character and not get any psy back at all then if they drop them somewhere else it will cost the initial psy cost to gain that character (you could also make this scale upwards the more characters you drop) and then still be affected by the global boost.

Something along those lines maybe? There would have to be some balancing work done but I think a global boost would be more interesting at least and you could make it cost more psy and could balance the games difficulty better due to it.

[Polaris2013]

This topic should have been called "what to do with the skill system," haha!

While there is a lot to like about having a small number of major choices for each character build, this is my concern. In a normal TD game you get a set of towers to choose from and you can upgrade them to make them stronger and/or change their function. This is not very different from an RPG where you get a set of characters to choose from and you can upgrade them to make them stronger and/or change their function. In fact, I fear the RPG version may be worse because you have to pick your "towers" before the map starts and can't change them to suit the needs of the map. You are essentially getting nothing from all the complexity of the RPG aspect of the game if you strip it down and make it too simple. You are just getting the complexity of an RPG and the gameplay of a standard TD, but with less freedom.

I think the unique thing about Defense Quest was the depth of the characters and the ability to add subtle nuance to them. I agree it was tough to keep track of them, but I think that illustrates the point that you can't have both depth (complexity and nuance) and breadth (lots of options). You have to choose one or the other. It's why I previously suggested reducing the character count- so you can focus more on depth.

If you instead want to keep options broad, it seems to me that you should seriously evaluate whether the RPG format is the best genre for the game you are trying to make. It is my contention that such a game will work better without RPG mechanics.

[丸々]

I found myself skipping every cutscene, so i would not complain if that were removed.

[Whimper]

I loved the humour in the writing, so please don't get rid of the cutscenes! It's easy enough to just skip past them already. Here are my thoughts on the other systems in question:

1. Skill trees. I like being able to customize the towers' abilities a lot. That's what makes this feel like an RPG to me. But since it's repetition and tedium you want to avoid (or reduce), what I'd suggest is to find a way keep the custom abilities but ditch the point-by-point economics.

Here's an example of how I might do it:

Bob the Berserker (and every character type) has Attack Value A, Body Value B, Critical Value C, Dodge Value D, and Elacrity Value E. Every level achieved, Bob gets 15 points distributed among this array automatically. The default priority for Berserkers is A 4, B 3, C 1, D 2, E 5 - fast and hard-hitting, but not particularly precise and not that durable. Players never manually put points into these, but they may choose from a few ways to prioritize this array. Three types of berserkers, say. Default, one that ups Critical but lowers speed, and one that ups Body and Ddoge at the expense of attacks.

Players can happily leave a character following this scheme all career long, or can switch between levels.

Next, every few levels, a character gets a choice between one of a few unique bonus powers. These are one-time purchases that provide a very tangible, visible improvement or a new ability. The damage, effects, and other features of these abilities will depend on the character's stat array, not on putting points into the ability over and over.

Sticking with our example above, after a few levels Bob gets a choice between Flurry or Power Strike, or whatever. "On a Critical Hit, Bob attacks all enemies within an X-tile radius" or "On a Critical Hit, Bob does triple damage." A player never gets to have both, so this one-time choice is an important moment for Bob the Berserker. For a default berserker, either choice is good, since they attack very frequently with their high speed. For a slower, crit-heavy berserker, the Flurry option lets them hit more enemies than usual while the power strike lets them take advantage of their frequent crits to deal heavy damage. You get the idea.

With a skill system like this, the player doesn't have to micro-manage every tower's level-by-level point adjustments, but the chance to make important decisions about meaningful abilities remains. Design the abilities well enough, and the choosing will be fun instead of boring.


2. Armored enemies. Like swimming enemies or shooting enemies or enemy-spawning enemies, these represented a moment where I had to adjust my strategy from what was previously an always-win situation. Finally, something that didn't just die to my rangers' barrage of plinky little arrows! I liked the challenge, and I liked that Knights provided one way of meeting it. Smashing the armor right off the bad guys was very rewarding at first, especially after having had a few missions without that possibility. The problem is that Knights are the best and most obvious way to beat armored enemies, all the time.

The way to make it fun is to provide multiple solutions. What about a spell that makes armor brittle to attacks over x amount of damage? But the spell has a range of 1, so you have to risk your wizard to cast it. What about an arrow-type that makes armored opponents move at half-speed? Of course, to buy that arrow option, the ranger would have to give up a juicy alternate choice during level-up...

[aerouge]

Hmmm I would not remove anything from the game. But maybe rework the points you mentioned.

Skilltrees. Yes it´s tedious to click 60 times per character to distribute skillpoints. Especially when reskilling for something like the endbattle ;-) But I see this more as an point for UI-Improvement than as something to leave out. I actually like the skilltrees ;-)
TLDR: Keep Skilltrees and just improve the UI


Enemy Armor. Hmmm I have mixed feelings about it. I think it adds a nice variety to the game and opens up different builds. Do I spend psi and skillpoints for armor breaking knights, or do I try to force my way with armor piercing rangers? The point is that in some maps some creeps have such a ridicoulus high armor stat that they are untouchable for anybody until the knights broke off the armor. Still I think enemy armor is a nice gameplay element that enriches the game and opens roads to different classes and builds in the future. Like Thieves/Assasins with 80% Armor Ignore Skill "Weakspot", Rangers with 20% Armor Ignore "Piercing Arrows" and Knights having a 100% armor pierce "Heavy Weapon" or smth like that.
TLDR: I think that the current armor-break mechanic can be thrown out, yet that armored enemies should remain.

Tower Damage: Mixed feelings again. In the current version my characters take damage forcing me to think about where to place the healers, when to turn off their attacks (Berserker vs Porcupine Cultist-Priests), casting heal spells etc. It does get heavy on micromanaging but I think that this game benefits a lot from this. It is basically something that sets it appart from other Towerdefense games. In other games you can find a perfect solution for every map on youtube while in DQ there sure is an optimal placement / upgrade routine still most maps ... especially on Game + ... require a lot of interaction due to Tower Damage. And this interaction is one part the sets this game appart from other tower defense games I played.
TLDR: Tower Damage makes the game pretty hectic, yet is a very fun mechanic in from my point of view as it adds replay-value to the same maps :)