This is an old revision of the document!


Haste Nerf

It has been pointed out on several occasions that various points in the late game are clearly balanced for hasted player vs hasted monster, and this is pretty seriously unbalancing considering how large of an effect haste currently is (twice the speed is quite significant). We would like to tone haste (and slow) down on both the player and monster part, in order to make haste not quite so mandatory. Nothing has really been decided on yet in terms of specifics, but here are some ideas that have been brought up (all numbers are subject to tweaking). — doy 2010-09-14 07:40

Monster speeds

With the haste nerf implemented, various monster speeds were reduced. For monsters such as orb of fire (speed 20 reduced to speed 15) and electric golem (speed 20 reduced to speed 16) this was definitely good, as part of Haste's problem was that it was almost a requirement to kill some late-game enemies. However, I think some monsters had their speed reduced unnecessarily, especially monsters whose speed was just reduced by 1, and even more especially monsters whose speed was reduced from 10 (normal player speed) to 9.

I propose reverting a number of the speed changes:

  • Golden Dragon, Greater Mummy, Fiend, Ice Fiend, Shadow Fiend: restore from speed 9 to 10
  • Pit Fiend: restore from speed 7 to 8
  • Storm Dragon: restore from speed 11 to 12
  • Orb Guardian: restore from speed 12 to 14
  • The Royal Jelly: partially restore from speed 12 to 14 (was speed 16 previously)
  • (Somewhat unrelated, but I'll put it here anyway) Agnes: reduce from speed 18 to speed 16 (mainly just for consistency with the other spriggans, her depth could be reduced in return? could just be left as-is, though)

This isn't intended to be a buff to lots of high end monsters, just to make some speeds a little more sensible - and I guess I'll note that I mostly agree with the list later on this page under Rebalancing - it's quite short, looks pretty sensible, and also suggests keeping the number of additional monster speed reductions low. — marvinpa 2011-03-04 18:44

Haste

Flat boost

Make all haste +50% speed rather than +100%.

Scaled by power

Initial proposal:

  • Haste becomes an effect that gives somewhere in the range of +25% - +75% increased speed
  • Spell bases this on spell power (+75% should be quite hard to reach, near the top of the power curve)
  • Potion at flat +50%
  • Wand at +25% - +75% based on Evocations
  • Okawaru Haste at +25% - +75% based on Invocations
  • All forms of berserk get fixed +50% speed (?)

Things to consider:

  • Should some of these ranges be different for different sources?
  • What to do about berserk? I don't think it makes sense for it to scale based on power, but I could be wrong.
  • Could make the potion special in that it alone gives +100% speed (potentially also making it a bit rarer).
  • Is +75% a good cap? Or do we really want the cap to be more along the lines of +50%?
  • Having a fixed speed based on power could encourage doing lots of calculations to figure out what your exact hasted speed is at a given time - do we want to randomize the speed a bit each turn to make this more difficult?
  • What about Slow? It is good that Haste and Slow effects cancel each other out right now.
Why is this a good thing? Dedicated casters are going to spend some more exp on enchantments, maybe, while light hybrids get a little slower and marginal casters get left behind. Limiting poor casters to consumables (and/or one god) further widens the casting/AC gap, and it's hard to see why conjurers and the like should be given a speed advantage anyway. Haste should be a flat boost.

And yes, spell power isn't directly affected by armor, but it is affected by losing stats to str and exp to armor skill. — OG17 2010-09-14 12:53

Hybrids and casters have the same choice of where to put their XP. Both can choose to train enchantment more, or to focus on their primary skills. There's no reason why dedicated casters would put more or less exp on enchantment than any other build appart from enchanters and pure melee. Right now, they are going train enchantment until they can cast haste reliably and then forget about it. Having the speed bonus scaled to power will give a strategic choice, and it will be the same for everyone. — galehar 2010-09-14 14:25
Maybe I was unclear, but dedicated casters have better casting abilities than light hybrids, who in turn have better than armored casters. Power-scaled haste isn't going to be the same for anyone. — OG17 2010-09-14 15:01
Heavy-armoured casters tend to have worse success and/or higher power than light-armoured casters. — rob 2010-09-14 15:12
This has been extensively discussed before, on IRC, and the consensus was against power-dependant effects.

* a fixed rate is simpler to understand and code

  • since Haste is supposed to be cancelled by Slow, they should give the same bonus/malus
  • consistency between players and monsters
  • you can't really rebalance monster speeds in a varied rate, using a fixed nerf would make those ancient liches/greater demons far lesser threats later on than when they appear for the first time
  • only high-level Charmers would enjoy the best bonus. Since we'd have to balance for the average case, they'd receive a significant boost rather than a nerf. In the Ench split, Hexes may have plenty of early and mid-game oomph, yet unless we implement new spells (like proposed Possess Demon or Illusory Decoy), it'd be Charms (or Self-ench if we go that way) who would be better later – and that's what we would want to avoid.
  • what to do if you extend Haste from a different source?

KiloByte 2010-09-14 23:27

+1 to a fixed rate. Making it vary with power can still be done later and need not be mixed in here. — rob 2010-09-15 12:27
My thoughts exactly. — evktalo 2010-09-15 12:36

Monster Haste

  • Cut most monster haste to +50%
  • Might be fun to let certain interesting monsters keep +100% (Executioners maybe?)
  • Could split it into separate monster spells - Lesser Haste/Greater Haste. Would it be worth knocking some monsters down to +25% (Minor Haste)? Or would that be too weak?

Slow

The effects of Slow should probably be toned down as well to mirror the effects of Haste - right now, Slow and Haste exactly cancel each other out, and I think that's a reasonable thing to preserve (doesn't have to be exact, but in the same range is good). Giving Slow -30% (for the fixed proposal) or a range of -20% - -40% (for the scaled by power proposal) would be in the right ballpark. Something else we could do here (if we decide to scale things) is make Slow scale up a bit better than Haste - this way, there's more of an interesting choice about whether Slow or Haste is a better decision in a given situation (if a successful slowing would make the speed difference between you and the monster greater than hasting yourself would).

Rebalancing

Several late-game monsters are currently strong enough that haste can be required in encounters with them, and so they should probably be toned down to some extent. A list of the worst offenders:

  • Orbs of fire: speed 20 → speed 15
  • Ancient liches: speed 12 → speed 10
  • Electric golems: speed 20 → speed 18 (?)
  • Gloorx Vloq: speed 20 → speed 15
  • random Pan lords: speed cap of 24 → speed cap of 18

In addition, we might want to move a couple uniques deeper, as they might not be as depth-appropriate if Haste isn't as strong. I think leaving Executioners at speed 20 would be interesting, they would remain something of a unique threat that way. More monsters than just this list will probably need to be toned down, but I think that keeping the list of changes reasonably small is a good thing. We certainly don't need to slow down the majority of things in the (normal) endgame, for instance - just the ones that people regularly have to haste for.

Player movement delay

If monster haste gets cut to 50%, player spriggan would be able to run from hasted normal-speed monsters. Similarly, the change would be a strong buff to swiftness and swift+fly.

Possible solution:

  • Make quickness mutation give -1, -2, -3. (centaurs at 8, spriggan at 7)
  • Make swift stay at -2, swift+fly 3.

Rounding issues

The multiplication and division by 2 has relatively few rounding issues (since most speeds are multiples of two anyway). Moving to different fractions (even the flat 1.5) will make this more noticeable (e.g. is x * 1.5 / 1.5 == x ?). It would be worth considering a finer time granularity.

The easy option would be to round randomly (div_rand_round). That would introduce more randomness.

Making haste give the player extra energy instead of reducing the time their actions take would help: Speed System Revision.

Discussion

Why not just change Haste and Slow into Enchantment/Translocation spells? Flavorwise you're not actually moving through space faster, you're just “warping” space to travel through less of it. Having to master two schools instead of one to cast Haste effectively should nerf it without changing its effects. -Wes

In this same vein, perhaps the spell could be bumped up to level 8 or so, so that it takes a much more serious investment to get it castable, while nerfing some of the tougher endgame monsters so that it isn't almost required for endgame. The problem with haste is that not that's it's particularly overpowered, it's that for how powerful it is, it's pretty easy to learn. As an upside, it wouldn't have to be balanced against invisibility, which tends to be almost universally a worse choice anyways. - adulus

If the problem is 'Haste is used in all endgame fights', why not buff some other glow-inducing enchantments to where they're viable in the endgame? The player would then have some interesting choices to make when analyzing a given fight (as he can't use more than one powerful buff without potentially picking up some nasty mutations; the penalty for glowing while undead would also have to be increased), and might also have to more thoughtfully allocate his experience if all these endgame buffs are put into different schools. Some tentative ideas, possibly overpowerd as-is:

Invisibility: currently largely useless in the endgame; the only enemies it works on tend to be the ones you don't need to buff yourself for anyway (the biggest exception being, perhaps, Tormentors). In particular, every 1 can see invisible. If many endgame foes had their See Invisible downgraded to nothing or Sense Invisible, Invisibility might be a viable competitor to Haste. If, say, there's a type of fiend that can't see invisible, players might want to use Invisibility instead of Haste for a fight where said fiend is present to avoid getting tormented while fighting other stuff. It might also be nice if there were a Zot monster or two besides the puny draconian mobs that can't see invisible.

Phase Shift: it might be interesting to make this spell give EV boosts that start with +8 and increase according to spell power… but have it cause glow. This would make it a superb buff against pure melee enemies (particularly for heavy armor casters; doubly so because, as a level 5 spell, it's easier for them to reach), but incompatible with Haste.

It just seems to me that glow is an underused mechanic, as many (most?) players only trigger it frequently by using Haste and the occasional controlled teleport. And a mechanic that could be useful here, at that. — dtsund 2010-10-01 20:41

The problem isn't directly that “haste is used in almost all endgame fights”, that's just a symptom. The problem is that haste is inherently quite overpowered, and making other spells equally overpowered isn't going to fix that. — doy 2010-10-03 01:24
Logged in as: Anonymous (VIEWER)
dcss/brainstorm/effect/haste_nerf.1324361579.txt.gz · Last modified: 2011-12-20 07:12 by XuaXua
 
Recent changes RSS feed Donate Powered by PHP Valid XHTML 1.0 Valid CSS Driven by DokuWiki