Tweaks to scaling of spells

[Anthony]

Somewhat against my better judgment, purely because I'm worried that these are going to break the game in some unforeseen way, I've taken the suggestions and requests of a number of you to heart and have done a little reworking of the spells, specifically allowing them to scale up a little better. Good discussion about spell damage was had here - viewtopic.php?f=4&t=55&hilit=lightning. We also needed to up Azra's max spell levels to 15 instead of 9 (since the level cap is now 60 - I don't think we'll be changing the defender skill point maxes, but Azra needed it). All Azra spells should also start targeting "dark" 100% of the time soon, though that's not implemented yet. Yes, I realize there are still some spells that are better than others, and I'm fine with that, but hopefully now there's a little more balance. So, here are the new details:

Lightning. Previously, lightning became weak pretty quickly, even before NG+, in which is became entirely useless. It just didn't do enough damage, and I was constrained and unable to make it do more damage since it would then have a >1 psi gained to psi spent ratio, thus breaking the game. Lars added a new tool though that lets me break that. Lightning will now inflict "psi burn", removing psi rewards proportional to the HP damage done. So, if a lightning strike removes 20% of an enemy's HP, it'll "burn off" 20% of their Psi reward as well. So, the old lightning spell varied from 10 damage at level 1 to 200 damage at level 9. Now, it varies from 20 dmg at level 1 to 730 dmg at level 15. Furthermore, bonus damage is applied based on Azra's attack stat ((level +1) * 5 * mult), where the mult varies from 0.5 to 4.7. Thus, at level 60 with 15 points in lightning, it will do 730 + ((60 + 1)*5*4.7) = 2164 damage. The level 9 version of the spell was previously 200 damage, now at Azra's level of 40 that same spell would do 300 + ((40+1)*5*2.9) = 895 damage. For the moment, I'm keeping the cost at the same 10 psi, and the cooldown at 0.25 seconds, though I'm debating increasing the cooldown a little (to 0.5 or 1.0 seconds).

Frenzy. I tweaked this one down ever so slightly. Previously ranged from speed bonus of 40% to 65% at level 9. It now will range from 40% to 75% at level 15 (so level 9 is now 60% instead of 65%). Still costs 35 Psi and has the same 90 second cooldown.

Healing. This one got a small boost. Previously cost 25 Psi with a 5 sec cooldown, and ranged from 30 HP to 270 HP at level 9. The new cost has dropped to 20 Psi, it still gains 30 HP per level (now maxing at 450 HP at level 15), but also gets a similar addition to lightning ranging from 30% to 75% of Azra's attack stat ((lvl+1)*5), so at full max level it'll heal for 450 + (61*5*0.75) = 679 HP. This will fully heal nearly every defender (outside of knights and dragons) for 20 Psi. If you don't care about healing then this won't change your mind, but I think it makes the heal spell legitimately useful with a few stat points put into it (drop 5 points into it at Azra level 40 and it'll heal for 150 + (41*5*0.43) = 238 HP, which even at 40 will heal 40% - 50% of most defender's health.

Crystal. Big boost for crystal - it has a long cooldown and small range, so it should do some pretty significant damage. I did add the same Psi Burn to it though, so you're gonna hurt for the lost psi, but at least you'll live. Old version varied from 150 dmg to 1050 dmg at level 9, costing 45 Psi and with a 75 second cooldown. It also had a splash range varying from 1.25 to 2.50 tiles. While the cost and cooldown haven't changed, everything else got pumped. The ranged now varies from 2.00 to 2.70 (admittedly at level 9 it's only 2.40 now, but whatever, haha). The base damage varies from 150 to 4000 at level 15 (with level 9 at 2000). Bonus damage ranges from 400% to 1800% of Azra's attack stat. At level 40 then, level 9 crystal would do 2000 + (41*5*12) = 4460 damage to everything in range, and truly maxed out it'll do 9490 damage.

Pushback. I think this one may stay almost entirely unchanged. The only difference is that instead of maxing at 4.0 tiles at level 9, now goes to 5.5 tiles at level 15, still at 0.25 more tiles per level. Currently leaving the cost at 60 and the cooldown at 40 sec.

Dragon Fire. As this one isn't a desperation move, nor as well-targeted, as lightning and crystal are, I'm going to try not to put psi burn on it. It does need a substantial boost though. The old version cost 100 psi and had a 120 sec cooldown. Base damage previously ranged from 150 - 500, and the fire damage was 10% of that damage ranging from 10 to 25 seconds. Thus, total damage ranged from 150 + (150*0.1*10) = 300 damage, to 500 + (500*0.1*25) = 1750 damage. Not bad when you consider it applies to everyone on the screen, but the cost was high and in NG+ it was just too little damage still. First off, I've dropped the cost to 75 psi. Next, the base damage varies from 150 to 900 at level 15 (with 500 still at level 9). The times have been stretched a bit so they now range from 11 to 25 seconds across all 15 levels. However, significant damage boosts based on Azra's attack stat have been added, from 200% to 900%. As a result, at spell level 9 and Azra level 40 Dragon Fire will do 500 + (41*5*6) = 1730 immediate damage. It would then do an additional 1730*0.1 damage for 19 seconds, yielding a total damage of 5017 over 19 seconds. Completely maxed, it will do perhaps way too much damage, totaling 12,758 damage over 25 seconds. However, with the 120 second cooldown, I don't *think* this will break anything, plus it would take 15 of your 60 total skill points to get to that level.

So, that's the new plan for the spells. I'm going to do a little similar tweaking for a few of the passive stat boosts (like HP and armor) for defender skills, but probably nothing too drastic. Thoughts are of course encouraged and appreciated. I haven't play-tested these yet, this is just on paper (and won't be in the test build until Lars' next deploy).

[coyot]

Good news . Seeing the numbers on Dragon Fire, I think it should apply Psi Burn - at least for the immediate damage done (I guess the burn effect would mean a lot of extra work for Lars, not worth the effort).
(Quick check - does the PSI burn take into account remaning HP of creeps? That is, if a creep is at 5% HP and the lightning has power of 20% of creep's HP, will the PSI-burn be 5%?)

[Anthony]

Good question on the burn of remaining HP. It does not, but I'll fix it now so that it does. I had meant to check that myself, thanks for the reminder.

For Dragon Fire, the up-front damage is much less than the overall damage, so adding Psi Burn to just that part wouldn't make a huge difference, and you're right that doing it that way would be super difficult too. We could probably have it burn off 25% of the relative damage or something like that, but doing full Psi burn I worry would hamstring Dragon Fire too much. I'd rather just tone down the damage if needed. Though it'd take a lot of skill points to do that much damage, so hopefully there's now some decisions to be made on where you want to focus your skill points. Or I suppose you can just max out 4 of them and ignore Healing and...Crystal or something. Well, whatevs - I think this'll work to provide some longer term value for the spells.

[coyot]

Good question on the burn of remaining HP. It does not, but I'll fix it now so that it does. I had meant to check that myself, thanks for the reminder.

For Dragon Fire, the up-front damage is much less than the overall damage, so adding Psi Burn to just that part wouldn't make a huge difference, and you're right that doing it that way would be super difficult too.

You misunderstood me - I thought it would be actually hard to add Psi Burn to FULL effect of dragon fire - and easy to add it to the direct damage one.
(Full effect would be especially hard to measure if a burning (spell-induced) creep were hit by Niru's fire, thus combining burns).

So, instead of making it psi-burn-free spell, just implement PSI-burn based on direct damage only - that should be simple...

[Anthony]

Oh, gotcha - yeah, that would be pretty easy. I just worry that the penalty would be too big. People were complaining about how expensive the 100 Psi was before. If a Psi Burn is added, the spell effectively gets much more expensive.

[Yinan]

Actually, i think the Psi Burn is a bad idea. Even if you could get more PSI out of a creep by killing it with Lightning... so what? The Same PSI spend in a Tower would be much better, so already using PSI for a spell is a waste of PSI, and now we're getting punished even more for doing it.

In my Opinion, Psi Burn is even more of a reason to not use any direct damage spell. Add it to Dragonfire and its the best dump skill because it will be a complete waste of PSI.

[coyot]

Oh, gotcha - yeah, that would be pretty easy. I just worry that the penalty would be too big. People were complaining about how expensive the 100 Psi was before. If a Psi Burn is added, the spell effectively gets much more expensive.

Well, that's the point - big damage spells are there really to pull you out from very tight situations, typically an unfortunate combination of waves/specials. Not a big deal if it hurts financially.

[Anthony]

Actually, i think the Psi Burn is a bad idea. Even if you could get more PSI out of a creep by killing it with Lightning... so what? The Same PSI spend in a Tower would be much better, so already using PSI for a spell is a waste of PSI, and now we're getting punished even more for doing it.

In my Opinion, Psi Burn is even more of a reason to not use any direct damage spell. Add it to Dragonfire and its the best dump skill because it will be a complete waste of PSI.

Unfortunately the math doesn't work out that way. If Lightning does 1000 damage and costs 10 Psi, it has a 100 : 1 dmg to Psi ratio. Meanwhile, most enemies have somewhere in the 20 - 60 HP : Psi ratio. As a result, you make a profit by using lightning on enemies. Let's say an enemy has 50 HP per Psi, and has 5000 HP overall. It will then yield 100 Psi. Now, you can kill it with 5 lightning bolts, costing 50 Psi, and you'll get 100 Psi back. As a result, you could play the entire game just by casting lightning over and over again, making a profit, and continuing. The only exception is if an enemy has more HP than you can fully kill with lightning, but that's rare outside of a few bosses. So...Psi Burn is the answer I'm afraid.

[larsiusprime]

I will weigh in here on the psi burn issue - the big difference between a spell and a tower is that a tower has a specific range, whereas a spell can target anything (or EVERYthing). So, even though a defender will make back more PSI than it cost to summon/boost over time, it's unable to target any arbitrary enemy on the screen, and it's also vulnerable to attack.

Therefore, if a spell can get you back more PSI than it costs to cast, especially if it has a low cooldown (lightning has basically none), you can just sit back and spell-spam every enemy without ever having to place a tower. It well and truly breaks the game - waaaaaay back in early development we had this exact imbalance and testers could beat every single level without ever summoning a character.

If you place defenders perfectly, spells are going to generally be a worse PSI investment than defenders - that's intentional. Whenever you screw up, get in a tight spot, or encounter something unexpected, though, a spell can bail you out. The additional boosts, especially the ones linking them to azra's attack level, should put them in a place where they're going to actually be useful again, (ie, lightning can actually take out strong enemies instead of just poking them with a toothpick).

[coyot]

Actually, i think the Psi Burn is a bad idea. Even if you could get more PSI out of a creep by killing it with Lightning... so what? The Same PSI spend in a Tower would be much better, so already using PSI for a spell is a waste of PSI, and now we're getting punished even more for doing it.

In my Opinion, Psi Burn is even more of a reason to not use any direct damage spell. Add it to Dragonfire and its the best dump skill because it will be a complete waste of PSI.

But spells are nit intended for you to make profit...

The goal of spells is to enable you yo beat tough levels where your group's firepower is just about on par with the average HP /second strength of incoming waves - and therefore any irregularuty in how the incoming HPs are spaced can be a problem.
Spells are a safety measure and the game would be playable just fine even if spell-killed enemies yielded no PSI at all...

[Marak]

This goes way beyond the "tack something into place to make Direct Damage spells useful" that I was not-really expecting to happen.

I'm thrilled. It makes me question my ignore Lightning/Pump Frenzy/ignore Healing/Pump Crystal build that I've been using since my 2nd play-through - that's great! Lightning that does more than tickle? Excellent. A bigger, stronger Crystal? Fantastic; Crystal was already my favorite "oh crap!" spell and now it's going to be better.

The changes to Dragon's Breath make it worthwhile to cast for reasons other than "I'm swimming in Psi and roflstomping this battle and it's off cooldown and it looks pretty, so hey". Now, you could use it to seriously soften up waves of monsters that are overlapping (say, some snails that spawn after a slow wave of Zombies) and the new 75-Psi cost makes it a little more likely that you'll be able to whip it out when needed instead of only when you're already dominating a map, as is the case currently.

Also, the buff to Healing opens up a lot of options for quite a few maps, including the hardest modes of all the maps pre-Healers-joining-the-Party. Faster, better, cheaper heals sounds like a great way to challenge yourself on those early maps with Attack Mobs without having to go back later with Healers in tow and a few more Levels under your belt. Again, great!

I'm not an expert here by ANY stretch of the imagination, but the numbers look good to me "on paper", as you've said. And since this system is basically nothing but adding Moar Maths to the spell damage equations, they should be fairly simple to tweak if something somehow becomes Too Good.

Final Thought: now that Azra's attack power actually does something, does that mean that some pure Attack Power Books are in the works? That would be an interesting alternative to Books That Do Unique Things; you would have to choose between Unique Effect Books and "Make All Your Spells a Bit Better" Books. Or even throw in a third tier, where the effect isn't as good/as unique but also has some Attack Power "baked in". Of couse, you guys have probably thought of all this already and have all the Book ideas more or less laid out on the table, as it were, but I thought I'd throw it out there nonetheless.

At any rate, I gotta say I really appreciate the detailed post and the heads-up on this topic, so thanks! Looking forward to Gold even more now, if that's possible.

[Anthony]

Thanks everyone - good to get some positive feedback for a change, haha. I'm sure we'll do some tweaking once more people get their hands on it, but as you surmised it's just some simple numbers to change to get the balance closer. Now I just have to get those new unique items a little more under control, haha!

As for book ideas, I'm honestly trying not to come up with any new ideas. Lars and James had a bunch, but I wanted to reign them in a bit so we could focus on finishing. This whole process has been a continuous stream of "ooh, what about this?" and "wouldn't it be cool if?" At this point, we're trying to do nothing new and have pretty much a code freeze. The more ideas you give us, the longer it takes for the Gold version to come out - so stop sending awesome ideas! Or at least we should start up a "Defender's Quest 2" thread to collect ideas. Of course, we don't know if we're doing a sequel or anything (we each have new game ideas that are quite different too!), but I think I can say that no new features or items are going to be added to Gold that aren't already in there.

(Also, I jest on the "positive feedback" joke - you guys are critical, but because you care about the game, and that means a lot!)

[Yinan]

Still don't buy it (the reason for PSI Burn).

Following your logic, i could apply the same thing to towers:
I'm pumbing in like 50 PSI for Berserker first level, and after some time he has done like several thousand damage. Following your arguments, that would mean that psi-burn has to be added to the Berserker as well, because you get more PSI out of it then you paid...

I don't see a difference. Saying "towers have a shorter range" is not a real reason for me. So what? They still do a lot of damage and give you back more PSI then they costed, so i don't see the reason y spells can't. I mean spells are ALREADY a waste of PSI. Not only because they don't do much damage (as it is now), but they are also a one time thing.

But well, it's your game. But it will be all the more reason for me to simply not use damage spells but put all my points in Push Back, Frenzy and Heal, ignoring the other threee because they're a waste of PSI.

[coyot]

Still don't buy it (the reason for PSI Burn).

This game is supposed to be played mainly with the defenders, the role of spells being support (pushback, frenzy heal) and firefighting (damage spells).
In order to make the damage spells worth casting at all in later NG levels where HPs of common creeps are really huge and lightning 200 simply doesn't make difference, spells have been pumped up as we've asked.

Lars and Anthony don't want the game to be a "defend your castle by shooting spells" one, they want you to rely on defenders. And in order to make us happy about damage spells AND not allow us to play the game in a way they don't want, they've introduced PSI burn.

Following your logic, i could apply the same thing to towers:

Sure you could, if you had a "castle defense shooter game" with added fighters and changes to their mechanics would mean that people could play the game as a TD instead of a shooter.

But well, it's your game. But it will be all the more reason for me to simply not use damage spells but put all my points in Push Back, Frenzy and Heal, ignoring the other threee because they're a waste of PSI.

Well, since heal is not really needed as it is now, you'll find by level 30 that you have nowhere else to put spellpoints into . And just a reminder - the goal of the game isn't to collect PSI, it's to defeat all the creeps.
Sure, if you play the game at a leisurely pace, you will probably never NEED damage spells, but once you go for any of the number of possible self-imposed challenges, you'll soon realize that to defeat this or that particular wave combination, upgrading a defender with 100 PSI simply doesn't help, but a large crystal explosion does

[larsiusprime]

I will point out, that through using mod functionality, you can remove psi burn from any spell you want, as well as tweak the numbers to whatever you think they should be

We still think this is the right decision for the core game, but I understand everyone won't agree, and that's what mods are for - it's impossible to please everyone with one fixed design.

[coyot]

May I suggest toning down the base power of the Crystal spell? (Not in-map crystals).

I'm currently past the first Zelemir - and in Eztli-Tenoch's forces, Azra level 12 and no books, level 1 crystal is still strong enough to wipe out revenants, worms and swimming revenants in any wave. And with just one point invested, it'd wipe out even the shadow worms easily.

The reduction needn't be big, but I think that without skill point investments, the crystal spell should not be able to wipe out any creeps that are rouhgly 2.5-3* the level of Azra. (Azra level 12 does 410 dmg with Crystal 1, worm level 30 is 334 HP, worm level 36 would be near 400 - and THAT should not be 1-hit kill for the crystal spell unless the player decides to put at least one or two skillpoints into it.

[MTaur]

It seems to me like psi spent on defenders should be better even without psi burn, or else spells just aren't balanced too well. It should take, maybe, ten seconds for a lvl 1 defender to do its psi value's worth of spell damage, as measured against lightning. If the psi payout per creep health is so high, then either the level is too easy, or defenders are not strong enough.

But maybe that's how we got the worthless/OP dichotomy in the past. A sort of OP magic system with psi burn and cooldowns is one way to make it rewarding but not broken.

It just feels like the solution should be rebalanced psi payout, not an ad hoc cap on the power of lightning. Though I guess the plus side to an OP magic system with this safety measure is that it opens up the luxury to blast lots of weakened stragglers, maybe allowing the maps themselves to get harder overall to balance everything out. It feels like an odd approach to me, still.

[MTaur]

I don't remember what has and hasn't been said so far, but a more natural-feeling fix, imo, would be a sort of spell fatigue system.

The cost of casting Lightning increases per cast, gradually returning to base cost over time. It could be a rather dramatic curve, so that a player feels little guilt for casting once or twice per wave, but still absolutely can't afford to spam it. There could even be a hard cap rather than endless cost scaling, sort of like a cooldown timer, but a more continuous one where you only "heat up" a little each cast, and you can keep casting until you heat up all the way, at which point there's a few seconds of full heat before the bar starts falling and you can cast again; maybe a yellow state for "the next cast will put you into overheated state". A hard cap like this could either take the place of scaling costs, or could be used in conjunction with a more forgiving version of scaling costs. Or it could take the place of all psi costs entirely.

Maybe psi burn works fine in practice. It just sounds fishy to me. I'm sure just going into the files and modding out psi burn would upset the balance and open exploits, and I'm sure it's a valid solution. It just feels fake and it would be a bit hard as a fan to mod something else in its place

(I also think a new resource for spells, like mana instead of psi, would also solve the problem. Even when it works, the looming specter of Psi guilt just makes me paranoid that I've already lost, taking any spell use as a sign of a possible beginning of the end. btw, I haven't played since the last full release, so I can't speak to what the new pre-release feels like at all.)

[Anthony]

coyot - yeah, the crystal spell may be a bit too powerful, but the long cooldown helps balance away from that. It gets me out of some rough spots pretty effectively, which is getting necessary in NG+ (but perhaps at the detriment of making the standard game a bit too easy).

MTaur - the balancing point for spells traditionally has been cooldown rather than cost. Many spells will generate a profit if you're in a tight spot, but the long cooldown means you have to be judicious with the timing. I like your idea of a progressive cost/cooldown, but a) it's a bit late for that now and b) I don't think it fundamentally changes most of the spells in practice. The one exception is lightning, which is a quickfire spell with minimal cooldown (0.25 seconds). In this case, we have to be careful that you can't turn a profit or the game can be won just by hitting the lightning spell enough (assuming the 0.25 second cooldown doesn't get prohibitive, which it would eventually, but not for a while). Psi burn is perhaps a bit of a hack, but I think it patches up the problem pretty cleanly without having to do some pretty major overhauls.

[coyot]

I think crystal would work fine if the base damage got nerfed by some 30-50% and compensated with higher increase per level... that is, I don't mind for the spell to have a big kick, but it should require some point investment.
It's OK if it catches up just within a few/several levels...

killing a bunch of worms spawned from montrosity is the typical use case in normal game - and IMO it should not be possible with level 1 crystal... at least in non grind modes, playing the levels as soon as possible...