Part Stats Balance/Polish, Round 3

[Misery]

It is definitely based on playstyle to some degree;  I tend to play very carefully and am always considering position and distance, and I tend to pull small groups away from the main groups, and stuff like that.  So I'm usually in a position where I can make overly effective use of the current regen and such.  It indeed wouldnt work quite as well if you've got things coming at you from every direction all at once, but that's relatively rare, even in Lion's Den missions.  That playstyle seems rewarded well enough for being defensive as it is, as it tends to simply lead to not getting shot very often.

I forgot what else I was going to say.

[Pepisolo]

It is definitely based on playstyle to some degree;  I tend to play very carefully and am always considering position and distance, and I tend to pull small groups away from the main groups, and stuff like that.  So I'm usually in a position where I can make overly effective use of the current regen and such.  It indeed wouldnt work quite as well if you've got things coming at you from every direction all at once, but that's relatively rare, even in Lion's Den missions.  That playstyle seems rewarded well enough for being defensive as it is, as it tends to simply lead to not getting shot very often.

I forgot what else I was going to say.

Hmmm, I might have to switch to my permadeath playstyle (using house rules) and see what I think then. When I'm in permadeath mode I go excessively cautious, for obvious reasons. That's kind of why I like permadeath as it forces me to play uh.... properly.

[Tridus]

It sounds like the numbers probably shouldn't have been raised so much initially. What would dropping the damage reduction bonus by difficulty do, given the lower attack powers we have now?

[Pepisolo]

This is a strange and not very helpful way for me to provide feedback, but hey, brief comments for every stat in a stream of conc fashion.

Weapon-specific:
+%Attack -- No probs. Useful in customization.
+%Ammo -- Good for specializing in certain weps.
+Range -- Super useful. Probably the closest to being exploitable (if you outrange enemy there's nothing they can do). Still don't mind it personally.
+%Splash-radius -- Useful...maybe too useful. The rocket launcher is a BEAST stacked with radius. Actually, Rocket Launcher seems comparable to the Plasma Rifle, but with more ammo. Maybe a little OP?

Shield-related:
+Shields -- Don't pay much attention (surprisingly). Quite easy to make exos tanky if you choose to.
+Regen -- Use it. Exciting skill. No probs personally.
+Damage-reduction -- Boring. Don't pay much attention, but it seems to work anyway. Sometimes an enemy hits me and does minimal damage.

Propulsion-related:
+Stealth -- Fun. Tactical. Useful!
+Overload -- FUN! Not that useful, but I still love it. Especially the giant explosion on death you get.
+Trap-skill -- weird. Almost like a non-skill. I pay minimal no attention to it. The trap avoiding, anyway. Didn't know this affects trap placement also ...for some reason.

Computer-related:
+Sensors -- Yeah, useful.
+Hacking -- Useful, but boring. When is hacking ever 100%? Where's the "oh s**t they're onto me!!" (siren sounds) after failing a hack.
+Virus -- Useful. Entertaining results. Not much danger in using it. Could be chance of success based.
+Mines -- Use these when I'm in trouble to barricade myself away.
+Sentry-turrets -- Cool. Useful.

Second-order:
+%Power-generation (exo-wide) -- Seems OK.
+%Attack (exo-wide) -- Useful.
+%Shield-related-stats (exo-wide) -- Don't pay much attention, but sounds good, and it's clear what it does.
+%Propulsion-related-stats (exo-wide) -- Don't really know what this affects...stealth? (My own fault for not paying attention)
+%Computer-related-stats (exo-wide) -- Know exactly what it does. More loot hacks.

Other:
+Power-generation - Highly prized. Always look out for these in the shop.
+Power-consumption - Not highly prized. 
-Power Cost of other parts in same system -- Try to use it cleverly, but seems a little weak. Also, much better in some systems than others (more slots the better of course).

[Winge]

I like regen as it is (mechanically speaking), but I could see either Regen or Damage Reduction stats going down a little bit.  Increasing the cap on Damage Reduction could work as well; right now, it's minimized at 1, which is a joke if you have any regen whatsoever.

[keith.lamothe]

Maybe I just need to do the regen-only-works-on-wait change on Expert only

Though having mechanics change between difficulties (rather than just numbers) is probably a no-no.

Maybe Regen should only happen when you kill an enemy   Nah, that should just be a new separate mechanic in an expansion if we get to do one of those.

Anyway, I'm just gonna tweak numbers for now, and leave mechanics changes for later if there appears to be a need.  I certainly would like to avoid mechanics changes at this point if we can, and I think most of you feel that way too.

[keith.lamothe]

Ok, I've put in the changes for 0.926 that came from this thread.  Might be some more, dunno, we'll see.  Anyway, for the moment this is the devestation y'all have wrought upon the game this time:

=== Parts/Stats Polish, Round 4 ===

* We're getting into the last stages of polish here, so trying to address pretty much any stat issue that's sticking out to players.  It will seem like a lot of change at once (which isn't great for isolating variables for observation)  and we generally take the "do a big change, then pick a middle ground if it was too much" approach rather than the "do a small change, then do another change in the same direction if it was too little" approach, at least at this stage in a game's life.  But please bear with us, we're making progress and we're still listening
** As usual, most of these changes have no impact on parts already generated in old saves.  But things like bot stats and inherent exo stats will generally change immediately, even in old saves.
** Thanks to Tridus, nas1m, Misery, Pepsisolo, Teal_Blue, Winge, Histidine, windgen, and others for inspiring these changes.

* The Nano-Rifled Barrel part can now go in any weapon, not just projectile weapons, as it is the only part with +attack as a primary stat.

* The robot stat growth exponents are now separate for shields and health, as shield levels were feeling fine on Hard/Expert after the last change to bring the exponents down, but the attack levels were feeling low.  So the attack exponents are now:
** Normal/Boss on Hard from 1.12/1.09 => 1.15/1.10 (numbers before last change were 1.16/1.10).
** Normal/Boss on Expert from 1.13/1.10 => 1.18/1.13 (numbers before last change were 1.19/1.13).

* Previously damage reduction had felt very underwhelming on Hard/Expert, so last time we added a 2x/3x to that stat on Hard/Expert.  Then it started feeling OP.  That's probably because of the attack exponents going down significantly at the same time, so since those are going back up it's probably not good to entirely remove the buff to damage reduction on those difficulties.
** So it's going from 2x/3x => 1.5x/2x.
** Also, damage-reduction can now only negate up to 90% of incoming damage; you can't use it to prevent the last 10% (that's still subject to regen, but you'll need more than a token regen to heal it, probably).

* Previously Regen felt underwhelming (under the recently introduced new model, at least) so last time we buffed its range of values from 5-to-110 all the way to 15-to-330, but that's seemed pretty OP.
** So now it's going from 15-to-330 => 10-220.

* Apparently at some time after the trap-skill values were last set the tile levels came way down (or maybe the trap skill values were just set wrong, dunno).  Either way, it's become trivially easy to get trap skill to 17, which is enough to dodge all traps on day 50.  So:
** Trap-Skill value from 3+level => 1+(level/3).  So the highest trap skill you can get on an item is now 9.
** Trap Skill now also occurs as a variant effect on fewer parts (all primarily-propulsion parts kept it, most others did not).

* +%Propulsion-related-stats has been feeling underwhelming since the three related stats (stealth, overload, and trap-skill) are relatively secondary compared to what +%Attack and +%Shield-related-stats can get you.  The same is true to a lesser extent for +%Computer-related-stats.  So:
** +%Propulsion-related-stats value range from 6%-to-30% => 12%-to-60%.
** +%Computer-related-stats value range from 6%-to-30% => 7%-to-35%.

* "-power cost of all parts in this system" has felt useful but ultimately underpowered for many for a while, so:
** Value from level => level*2.

* Power reactors continue to be harder to come by than is really quite fun, so their occurrence rates have been boosted 25%.

* High durability was too easy to get for pretty much all exos in the last version, but it was particularly egregious for the Assault Exo, so we're trying:
** Inherent boost to shield-related stats from +50% => +30%.

* +%AOE-radius has been pretty OP for a while.  It's also one of those "really hilarious when it's OP" things so we're not really itching to nerfhammer it.  Still, some adjustment should be made.  So:
** +%AOE-radius value range from 25%-to-100% => 25%-to-75%.

If you have any pre-emptive feedback or otherwise wish to chime in, please do so

[Misery]

Lessee....

Trap-skill:  Wont this impact mines and turrets too? That could be a bit of an issue.   On that note, turrets might do with a bit of a buff, as it's hard to get really any damage out of them, particularly when bots with somewhat decent attack values come by.  If those bots just look at the turrets in a funny way the damn things shatter in fear.   Even with high trap skill, which I usually have on my Science guy, these can be a little underwhelming.

And the AoE stuff.... one thing you could do is increase the rate at which damage drops as it gets further from the center of the blast.  I dont really think this would lower their usefulness at all, because one way or another you're getting free damage on potentially lots of foes, and probably outright killing at least a couple of them.  And if there's an ammo-refil station around you can usually feel free to fire more than one explody thing into the area.   The changes to that item type though sound good.

[keith.lamothe]

@Misery: Fair enough on the turrets.

FWIW, what you're seeing in game right now is that the turret gets a boost to health and attack equal to 5+1.13^((trapSkill-3)/2) ... oh, wait, yea, you and math.  You've told me about that.

To put it more concretely, in 0.925 you get this much turret health at these levels of trap skill:

(one spreadsheet later)

Ew.  The numbers are actually laughably low.  You basically have to stack like 53 trap skill to cap out, and that's only a 26% boost to health.  Which is 78 extra points of health.  Whooptedo

Similar on the attack side.


Hmm. ok.

So I'm thinking of instead having trap skill give a flat +% boost per point.  Given that 9 on a single part will be endgame levels, how about +10% per point?  So a turret with one endgame trap-skill part backing it up is almost twice as durable and does almost twice as much damage.

Or is that still pitifully low?  Even with 3 end-game trap-skill parts that's only a quadrupling of power of something that isn't all that strong to start with.  Then again you get a ton of turrets by endgame so we don't want to go too nuts with the stats on the individual ones.

Put another way: how much more durable do you want endgame sentries to be than no-bonus sentries? And how much more offensively powerful?

[Misery]

I'm thinking it's gonna take a bit of experimentation to get the numbers on those right.  With bot attack levels changing, and their health being different from how it used to be.  Obviously the turrets dont need to be doing the kind of damage a heavy weapon does, but nor should they be on the level of a light machine gun which takes many, many hits to kill anything. 

Hmm, what are the turrets' max stats for health and attack as they are now?  I dont think I used them all that much near the very end of the game, since they were scaling up really badly as it went on.

[keith.lamothe]

mm, what are the turrets' max stats for health and attack as they are now?

Base stats for a sentry right now are 300 health and 120 attack.  If my math is correct in that the max boost you can get from trap skill is between 26% and 27%, then max health is 381 and max attack is 153.

Is that what you're seeing in your game?  Seems like if the numbers were bad someone would have said something.  Because that's worse than mines, probably.  Mines at least do some damage while they're walked over.


And yea, it will take experimentation, I'm just looking for first-guess numbers.

[Histidine]

Bots also gain range over time while sentries don't (soon outranging sentries quite handily unless you place them just around a corner), so letting sentries gain range with trap skill would be nice as well.

Considering the endgame bots will often have at least 5x their starting health and 10x their starting damage on Hard, even quadruple strength sentries doesn't sound particularly excessive.

[Misery]

mm, what are the turrets' max stats for health and attack as they are now?

Base stats for a sentry right now are 300 health and 120 attack.  If my math is correct in that the max boost you can get from trap skill is between 26% and 27%, then max health is 381 and max attack is 153.

Is that what you're seeing in your game?  Seems like if the numbers were bad someone would have said something.  Because that's worse than mines, probably.  Mines at least do some damage while they're walked over.


And yea, it will take experimentation, I'm just looking for first-guess numbers.

Wow, that's the max stats?   Haha, and here I'd thought I just somehow wasnt managing to get enough trap-skill stats equipped.  I didn't think that was actually the max.  But yeah, that's..... way too low.  I can do plenty of damage with mines, particularly since you get like a zillion of them, but those turrets..... that wouldnt even scratch much of anything. 

Histidine is right, multiplying it by 4 or 5 wouldnt really be excessive.   Though, I'm thinking of the higher difficulties when I say that;  on lower ones, it's very different.  Might be a bit much on those since bots have much lower attack/shields?

[Tridus]

Given the way that the trap skill numbers work now, here's the health/attack numbers on sentries with 18 trap skill (two capped items) at various bonuses per point. Math might be off, it's early morning.

Bonus per trap skill	Health at 18TS	Attack at 18TS
10	840	336
15	1110	444
20	1380	552
25	1650	660
35	2190	876

So I guess the question after that is just how beefy should turrets be? Their attack triples between the start and end of a game (nearly) at 10% per point, nearly quadruples at 15, and gets pretty nice on its own at 20. After that they're looking rather strong.

[Misery]

The ones at 15 or 20 look about right to me, given the damage level and health of enemies around that time.  It's not too weak while not being too damaging either;  using multiple turrets at once would be the thing for stronger foes.