Skip to content

Uh, What? (Firkraag)

13»

Comments

  • tbone1tbone1 Member Posts: 1,985
    edited March 2017
    Or have a cavalier with the dragon slayer sword and shield and helmet pound on that big lizard while Aerie and Jan provide great support. Admittedly, Aerie looks better in the cheerleader costume than Jan does as the mascot for the Fightng Root Vegetables, though his pompoms are more impressive.
    Post edited by tbone1 on
    DJKajuruArctodusSunderBelgarathMTH
  • BorekBorek Member Posts: 513
    About to hit Firki again in my latest game, but found a nice way to deal with the Shadow Dragon with my Solo FMT that should work just fine. Project Image then use the image to set all traps, cast a bunch of Skull Traps, then use Lower Resistance to start combat. Image then got Remove Magic+wing buffeted out of range and killed but tbh there wasn't long left on its timer and that bought Spawnette back into the fight with the Dragon at low health from all the traps/ST's. Since i cleaned out the Twisted Rune before getting here that meant Staff of the Magi, simply whacked it once to dispell then switched back to swords to administer the final few blows whilst it flailed away at my stoneskins since it wasted it's remove magic on the Image :)

    Firki can be nasty though as he will detect if you have fire immunity and if you do usually casts a spell that makes you more vulnerable before then breathing. Anything other than a pure warrior can be in serious trouble due to Health, especially if his remove magic got rid of stoneskins/ironskins.

    If you are lucky enough to have Mordy Swords you can cheese up 3 pre-fight, haste them and send them in, AFAIK nothing he has can damage them, but unless you pop in to Dispell/remove/breach his Stoneskins he'll outlast their duration and he has a bad habit of wing buffeting them (main reason you need haste so they get back on him faster) then having a cheeky breath on the party.

    As for pure cheese then you can stack cloudkills from the wand from outside his vision and as long as you use the staff of the magi to immediately go invisible he should just sit in the multiple clouds and slowly choke to death between magic resists, lol.

    Couple of games i did wait before taking him on until extremely late and used Fallen Planetars, but he's such a beast in melee it would take more than 1, which was a bit disappointing lol. Haven't tested yet if Hardiness stacks with Iron Golem form for a multi-class F/M+, but if it does then Iron Golem shapeshift should be immune to his breath and the physical resistance should mean you can basically walk up to him and beat his face with impunity. Drawback is you need to wait until you have Shapeshift which i think means holding back until after underdark, but one of these days i will have to try it out :smile:
  • SomeSortSomeSort Member Posts: 859
    Borek said:

    As for pure cheese then you can stack cloudkills from the wand from outside his vision and as long as you use the staff of the magi to immediately go invisible he should just sit in the multiple clouds and slowly choke to death between magic resists, lol.

    EE fixed this cheese. If a dragon sustains heavy injuries and doesn't see any enemies, he'll automatically cast heal on himself. This heal is scripted, so there are an unlimited number of them.

    You can always pay attention, use Cloudkill cheese to get him down to near death, and then rush into his visual range before the Heal script triggers. That'll trigger all of his protections firing, though. Or if you really time it well, you can wait until he starts casting his heal and then rush in and interrupt him. That'll probably take a couple of reloads to time just right, though.
  • BorekBorek Member Posts: 513
    SomeSort said:

    Borek said:

    As for pure cheese then you can stack cloudkills from the wand from outside his vision and as long as you use the staff of the magi to immediately go invisible he should just sit in the multiple clouds and slowly choke to death between magic resists, lol.

    EE fixed this cheese. If a dragon sustains heavy injuries and doesn't see any enemies, he'll automatically cast heal on himself. This heal is scripted, so there are an unlimited number of them.

    You can always pay attention, use Cloudkill cheese to get him down to near death, and then rush into his visual range before the Heal script triggers. That'll trigger all of his protections firing, though. Or if you really time it well, you can wait until he starts casting his heal and then rush in and interrupt him. That'll probably take a couple of reloads to time just right, though.
    Just tried a few different strategies on max difficulty (not LOB, not that much of a sadist) as a solo FMT, although i wanted to use just magic, and not Project image and 5x spike traps lol. So this was essentially as a 17th level Mage, just with additional HP's.

    Couple of things i noted, Mordenkainens Sword will hold his attention until it gets wing buffeted away, drawback is they get shredded by Skull Traps, more on that in a bit.

    I was not properly prepped with Sequencer and Spell Trigger, so ended up having to cast a fair bit more than optimal, but the pre-buffing is extremely important:

    Protection from Fire obviously a must.
    Resist Fear highly advised.
    Stoneskin for obvious reasons.
    Spell Immunity Abjuration to stop his remove magic.
    Globe (or minor) of invulnerability.

    After that you should probably have Amulet of power + Robe of Vecna, otherwise he may well knock you unconcious with a Wing Buffet at the wrong time, which he tends to start spamming once he realises that you are essentially immune to everything he is throwing, plus he tends to Greater Mallison so you fail saves more than is handy. Standing with the wall directly behind you so you can't be knocked back also helps.

    Ideal Spell Trigger would be 2x Lower Resistance + 1 Greater Mallison (mine was LR, Breach, GM, since i didn't melee breach was no benefit)

    Ideal Sequencer 3 Skull traps (mine was GM 2 ST's).

    Fight sequence:

    Spell Trigger
    Cast protection from Magical Weapons
    Sequencer
    Cast Skull traps until dead if he isn't already.

    Depending on level you may want to open with a manually cast Lower Resistance, i didn't really have any problems with him having 11% MR, but never hurts to be safe. You should be able to do it w/o Spell Trigger, but in order to cast 2 or 3 Lower Resistance and a Greater Mallison you'll probably want to put up a Mordy Sword to focus his wrath, ideally with a Haste on it.

    If you are prepared to save and reload there's tons of options, but this is almost 100% guaranteed to work, just make sure you have a contingency stoneskin and a spare Prot. Magic Weap and all your level 3's as Skull Trap except for the Prot Fire. I'd also say you should have either Magic Missile or Chromatic Orb as level 1's on the rare off-chance you are not using the wall to prevent blow-back and he ends up healing, with Greater Mallison Chromatic Orbs have a so-so chance to blap him :wink:

    Of course 3x Spike traps is a lot easier, but i wanted to use spells. This was solo, a group should make it a lot easier although it's significantly harder to buff to full immunity and you definitely don't want the party accidentally getting caught in the skull traps >:)
  • Necromanx2Necromanx2 Member Posts: 1,246
    I use the Cloud kill cheese. But to make it work best, you need 3 casters with one fully loaded wand each. You cast out of view range 3 clouds a round. By the time you have 30 active per round, his heal does not kick off fast enough to save him.
  • SomeSortSomeSort Member Posts: 859
    If I've got my math right, since you're a level 17 mage and therefore have access to level 8 spells, a Spell Sequencer with 3x Lower Resistance and a Spell Trigger with Malison and 2x Feeblemind should give you an 84% kill rate in two rounds. If you cast two more Feebleminds the old-fashioned way, that should increase your odds of success to ~97.5% in four rounds.

    The thing about Save or Die spells is if you spam enough of them, it ceases to be a matter of "luck".
  • BorekBorek Member Posts: 513
    SomeSort said:

    If I've got my math right, since you're a level 17 mage and therefore have access to level 8 spells, a Spell Sequencer with 3x Lower Resistance and a Spell Trigger with Malison and 2x Feeblemind should give you an 84% kill rate in two rounds. If you cast two more Feebleminds the old-fashioned way, that should increase your odds of success to ~97.5% in four rounds.

    The thing about Save or Die spells is if you spam enough of them, it ceases to be a matter of "luck".

    Unfortunately the problem with statistics is they result in a Dragon Burping and picking his teeth with your Femur every so often, if you are going to save and reload then it's not a problem but i wanted to get as close as possible to a no-reload solo Insane play-through. As it happens i died once in Hell to an Elder Orb imprisonment, a Maze caught me a few times from Irenicus' Suldanessar 1st fight before i realised i could just scoot off screen with staff of magi invisibility up, then pump in 2 ADHW's from off screen and blap him lol, although felt somewhat underwhelming doing that...

    Had to reload a few times with Adalon because for some reason she wouldn't move and trigger my skull traps, but never actually died. Easiest kill was the Black Dragon and totally non-cheesed that one, pre-buffed and started with Imp Alacrity then manually cast 2x LR, 1x GM, 5x ST without any sequencers/triggers used and he got flat out ripped. Obviously i was significantly higher level by then though.

    I did Watchers Keep before Spellhold, but i admit i just spike trapped Demogorgan and the most difficult of the final seal keys did require a few reloads, mainly because i forgot there was a limit on traps and the Beholder didn't die to just 1, seemed the 2nd was removed when i attempted to place more than the limit and i had already saved my game w/o knowing lol...Ended up requiring a planetar, Skeleton warrior and a Mordy Sword to distract Captain Kai and the Str debuff archer from butchering me due to me carrying too many Healing potions to move after the archer called shot me lmao.
  • SomeSortSomeSort Member Posts: 859
    Borek said:

    Unfortunately the problem with statistics is they result in a Dragon Burping and picking his teeth with your Femur every so often, if you are going to save and reload then it's not a problem but i wanted to get as close as possible to a no-reload solo Insane play-through.

    Again, provided you have access to the relevant spells by then, (Spell Trigger, Spell Sequencer, Greater Malison, Lower Resistance, Feeblemind), the Sequencer/Feeblemind combo should be available at Mage Level 16 and have an 84% success rate within two rounds. Yes, this means 16% of the time it *won't* be successful within two rounds. But that doesn't translate to a 16% chance of death; it translates to a 16% chance you have to activate "Power Word: Run Away!!!" and try again later.

    If you have a Chain Contingency, (either through a scroll or because you're level 18 and have 9th level spells), you can boost your odds of success within two rounds to 99%. And, of course, if you already have Improved Alacrity you can spam Feebleminds just as easily as you spam Skull Traps, and to greater effect.
  • BorekBorek Member Posts: 513
    SomeSort said:

    Borek said:

    Unfortunately the problem with statistics is they result in a Dragon Burping and picking his teeth with your Femur every so often, if you are going to save and reload then it's not a problem but i wanted to get as close as possible to a no-reload solo Insane play-through.

    Again, provided you have access to the relevant spells by then, (Spell Trigger, Spell Sequencer, Greater Malison, Lower Resistance, Feeblemind), the Sequencer/Feeblemind combo should be available at Mage Level 16 and have an 84% success rate within two rounds. Yes, this means 16% of the time it *won't* be successful within two rounds. But that doesn't translate to a 16% chance of death; it translates to a 16% chance you have to activate "Power Word: Run Away!!!" and try again later.

    If you have a Chain Contingency, (either through a scroll or because you're level 18 and have 9th level spells), you can boost your odds of success within two rounds to 99%. And, of course, if you already have Improved Alacrity you can spam Feebleminds just as easily as you spam Skull Traps, and to greater effect.
    True, you can run away and try again. You have the same chance of success every time you cast the spell, the fact you failed the attempt before doesn't matter to the game engine. Not sure what it's save is, but if it has a 60% chance to save, it has a 60% chance to save every time, it's why i dislike using spells that do nothing if saved against and prefer Skull Trap for Dragons and most tough monsters, if i don't just beat them to death with a melee weapon anyway.

    I mean technically you can just use Chromatic Orbs, they fit in all the sequencers and you can learn loads, the +6 save will become statistically low if you cast enough, but i definitely wouldn't want to do it in a no reload game.
  • SomeSortSomeSort Member Posts: 859
    For starters, I'm glad I put "if I have my math right", because I didn't. Firkraag has a Save vs. Spells of 6, which I was treating as a 30% chance to fail (i.e. he fails on a 1-6). But of course, a 6 would not be a failure, it would be a success; he only fails on a 1-5.

    That drops the odds of taking him out in two rounds with a 3x LR sequencer and a GM/2x Feeblemind sequencer from 86% to 80%.
    Borek said:

    True, you can run away and try again. You have the same chance of success every time you cast the spell, the fact you failed the attempt before doesn't matter to the game engine. Not sure what it's save is, but if it has a 60% chance to save, it has a 60% chance to save every time, it's why i dislike using spells that do nothing if saved against and prefer Skull Trap for Dragons and most tough monsters, if i don't just beat them to death with a melee weapon anyway.

    I mean technically you can just use Chromatic Orbs, they fit in all the sequencers and you can learn loads, the +6 save will become statistically low if you cast enough, but i definitely wouldn't want to do it in a no reload game.

    I resisted the siren's call of SoE (Save or Else) spells for the longest time for exactly this reason. "They're too inconsistent", I told myself. "There's a great chance they'll do nothing at all", I told myself. And this is completely true... if you're only casting them once. The "lightbulb" moment for me is the realization that they're actually tremendously consistent... if it's important enough for you to cast multiple times.

    Let's take ol' Red Wings, here. He has a save vs. spells of 6. Feeblemind provides a -2 penalty, and Greater Malison adds another -4 malus on top of that, bringing his save vs. spells to 12. This means he fails on a 1-11, and succeeds on a 12-20, for a 55% chance of death, (or, alternately, a 45% chance of survival). No-reload runs, as you point out, are far too important to leave up to a coin flip like that.

    But what if you are prepared to cast Feeblemind twice? The odds of him surviving two Feebleminds are (the odds of him surviving the first) times (the odds of him surviving the second), which is 0.45 * 0.45, or 20.25%. Which means the odds of him dying are 79.75%.

    If you're able to cast it three times? That 20.25% chance of survival gets multiplied by another 45% chance of survival, and his odds of lasting through all three drop to 9.1%. If you cast it a fourth time, the odds of him surviving fall to 4.1%. If you cast it a fifth time, the odds of him surviving are just 1.8%, which means you've got a method that has a 98.1% chance of killing Firkraag outright. That's not a coinflip, that's a virtual lock.

    Why do I mention five times? Because with Spell Sequencer, Spell Trigger, and Chain Contingency, you can cast 3x Lower Resist, 1x Greater Malison, and 5x Feeblemind in the span of two rounds without even having to abuse the broken / cheesy aspects of Chain Contingency, (like the ability to cast multiple per round or to cast it while paused). Time Stop and Improved Alacrity dramatically increase that total. (In fact, with a casting time of 5, Feeblemind should be instant cast with Robe of Vecna, Amulet of Power, and Improved Alacrity, so you can empty your entire spellbook at once.)

    It's true that you can do the same thing with Chromatic Orb, but it will take a substantially larger volume of casts to get the same degree of certainty. For starters, with a +6 penalty and a base Save vs. Spells of 6, you literally can't kill Firkraag with Chromatic Orb without first hitting him with a GM.

    After you cast GM, he fails vs. Orb on a 1-3 and succeeds on a 4-20, which means he has a 15% chance of dying and an 85% chance of living. At those odds, you'd need to cast 5 Chromatic Orbs to get an equivalent "kill chance" of one Feeblemind. (5 Chromatic Orbs gives a 55.6% chance of killing, one Feeblemind gives a 55.0% chance of killing.)

    So whatever your desired risk threshold is, you'll need to multiply the number of Feebleminds by five to see how many Chromatic Orbs would be required. If you want to be 80% sure you'll kill Firkraag, you'll need 10 Chromatic Orbs. If you want to be 90+% sure you're going to kill Firkraag, you'll need 14-15 Chromatic Orbs. If you want a 95% certain kill, you'll need 18-19 Chromatic Orbs. It's simply not a feasible strategy for solo / no-reload.

    But yeah, Feeblemind? Very quick, easy, safe, and most importantly, reliable kill method.

    (As an aside: this idea that stacking a Save or Else spell a bunch of times makes it super-consistent is why Slow is so amazing. Sure, each individual enemy gets a chance to save for no effect. But if you're casting it on five enemies, the odds that all five make their save is virtually nil, just like the odds of an individual enemy surviving five consecutive casts of Feeblemind.)

    The other nice thing about stacking SoEs is that it scales really well with multiple mages. Let's say you have a party with Edwin, Aerie, and Jan, and everyone has 1.5m experience. That makes Edwin level 14 with level 7 spells and Aerie/Jan level 12 with level 6 spells.

    Edwin packs a sequencer with 3x Lower Resistance. Round one, Edwin triggers his sequencer, Aerie casts Greater Malison, and Jan casts Feeblemind. (Feeblemind's casting time is one tick longer than Malison's, so GM should hit first.) There's a better than 50/50 chance that you just ended the fight outright, before Firkraag really had a chance to do anything.

    But say he rolls a 12 or better and survives. Fine, next round all three mages cast another Feeblemind. The odds of him surviving to round three at this point are about 4%, which isn't bad odds at all for a tactic you can pull off without any cheese at all before you even leave for Spellhold. (There are even three copies of the Feeblemind scroll for you: Galoomp in Waukeen's Promenade, Sister Garlena outside Watcher's Keep, and one during the DeArnise Keep.)
    grisbossk
  • QuickbladeQuickblade Member Posts: 957
    I'm not real up and up on my dragon fighting, but why the focus on killing within 2 rounds?
  • ThacoBellThacoBell Member Posts: 12,235
    Less time for the beast to kill you. Especially important on high difficulty no reload runs.
    SomeSort
  • SomeSortSomeSort Member Posts: 859
    ThacoBell said:

    Less time for the beast to kill you. Especially important on high difficulty no reload runs.

    Yup, exactly this. Surviving two rounds with Firkraag is cake. Toss up some halfhearted pre-buffs for him to concern himself with, toss up a few halfhearted summons for him to blow up, have a Tank or two stand next to him and beg for a wing buffet, and he'll be so busy setting up the battlefield over the first two rounds that he'll be dead before he even has a chance to unleash his offensive arsenal.

    If you have a strategy that requires 10 rounds to execute, though, your back line is all-but-guaranteed to be eating a couple Dragon Breaths, which is an instakill to low-HP mages without extensive buffing. If your tank needs to keep him occupied for a half-dozen rounds, (if your tank even can keep him occupied between all of the inevitable wing buffets), he or she will need to weather six+ rounds of 3+ attacks per round at -5 THACO with 25 strength, which is over 250 damage from his strength bonus alone. (I don't know what the base damage of his claw attacks is, but it's certainly non-zero.)

    In BG2, survival is just killing enemies before they can kill you. Oftentimes the best defense is a good offense. If you can get a 90% chance to kill in one round, that's even better, but you can't reliably get to that point until you've got 9th-level spells at your disposal. In the meantime, a 2-round kill is quick, consistent, and safe enough.
  • BorekBorek Member Posts: 513
    @SomeSort You are missing the point completely regarding save or die and a 100% strategy, lol. The whole point of a guaranteed win strategy is that it wins every time, if you have to slink off, rest and come back for another attempt because RNG screwed you then it's not doing what it's intended.

    On top of that the Feeblemind Strategy in particular is annoying because you need to use level 5 spell slots, which are also used for breach, spell immunity and lower resist, all spells i use a LOT for general play, just to kill a tough single target enemy. If it fails to work and you need to reload sequencers and trigger then you need to be extremely high level or rest 3 times, 2 for reloading, 1 to refresh, plus they are 5th level so you can't even abuse limited wish with Neera/Wild Mage Bhaalspawn.

    Chain contingency is not even worth mentioning because you need to have cleared underdark to even get it w/o consoling and it's 9th level.

    Skull Trap is by far the most effective lower level spell for clearing enemies as long as you watch out for the splash damage, either with Minor Globe or just careful positioning, plus you can pre-stack them. Feeblemind on the other hand doesn't work on undead and some demons, is wasted on trash and for general play there's at least 3 extremely useful spells competing for the same slots, more if you include Animate Dead.

    Will it work, sure, not every time, if you cast enough of them it will, but you need to remember 3 castings that each have a 33.3% chance of success does NOT mean 100% success, that is the trap of using probability in a scenario where you do not have unlimited resources and previous failure has zero impact on your chances of future success.
  • SomeSortSomeSort Member Posts: 859
    Borek said:

    @SomeSort You are missing the point completely regarding save or die and a 100% strategy, lol. The whole point of a guaranteed win strategy is that it wins every time, if you have to slink off, rest and come back for another attempt because RNG screwed you then it's not doing what it's intended.

    On top of that the Feeblemind Strategy in particular is annoying because you need to use level 5 spell slots, which are also used for breach, spell immunity and lower resist, all spells i use a LOT for general play, just to kill a tough single target enemy. If it fails to work and you need to reload sequencers and trigger then you need to be extremely high level or rest 3 times, 2 for reloading, 1 to refresh, plus they are 5th level so you can't even abuse limited wish with Neera/Wild Mage Bhaalspawn.

    Chain contingency is not even worth mentioning because you need to have cleared underdark to even get it w/o consoling and it's 9th level.

    Skull Trap is by far the most effective lower level spell for clearing enemies as long as you watch out for the splash damage, either with Minor Globe or just careful positioning, plus you can pre-stack them. Feeblemind on the other hand doesn't work on undead and some demons, is wasted on trash and for general play there's at least 3 extremely useful spells competing for the same slots, more if you include Animate Dead.

    Will it work, sure, not every time, if you cast enough of them it will, but you need to remember 3 castings that each have a 33.3% chance of success does NOT mean 100% success, that is the trap of using probability in a scenario where you do not have unlimited resources and previous failure has zero impact on your chances of future success.

    For starters, I don't know what you're talking about in the last paragraph. Yes, I know that three spells with a 33.3% chance of success don't add up to a 100% chance of success. I already walked through the math in this thread. I said that each feeblemind gives you a 55% chance of success, so five feebleminds gives you a 98.1% chance of success. Not a 275% chance of success, which is what the number would be if I just added 55% five times. Probability of survival = (1 - P)^n, where P = probability of a single cast succeeding and n = number of casts. (Probability of death = 1 - probability of survival.)

    There's no such thing as a "100% strategy". If there was, no-reload runs would have a much higher success rate. Every encounter carries risk. The longer the encounter drags on, the greater that risk is, so safety favors strategies that end encounters quickly and decisively.

    (Speaking of 100% strategies: "[abusing] limited wish with Neera/Wild Mage Bhaalspawn" is far more likely to force a reload during the course of a run than facing off against Firkraag with Feeblemind.)

    I don't really have a rebuttal to your claim that Feeblemind is less useful than Skull Trap for general purpose play. That is a true statement. Defeating Firkraag requires different tactics than defeating a nest of spiders or a Sahuagin patrol. Similarly, in 99% of encounters, I don't bother casting Spell Immunity: Abjuration, but see if you ever catch me squaring off against a demilich without it active. In unmodded BG2, there's really no reason to ever cast Spell Shield... unless you're doing the Unseeing Eye quest without the Shield of Balduran, in which case it's the best spell in the game.

    There is not a single encounter in the entire game that warrants preparation more than Firkraag and the Shadow Dragon. They are dragons that are just sitting around waiting for you to come kill them. You know they're there. You don't stumble into that battle by accident. You go up against them only when you want to, so it makes sense to resort to excessive preparation even in "no meta-knowledge runs" in a way that you wouldn't against, say, the Twisted Rune or Kangaax.

    But back on the topic of "100% strategy". As I said, every encounter involves risk, and the best chance of survival lies in ending dangerous encounters as quickly as possible. The more rounds you give Firkraag, the more chances he gets to dispel your protections and drop a Dragon's Breath on you. Let's also assume that we want to compare "legit" strategies rather than exploits (fake talking), and additionally, we want to avoid any sort of trivializing cheese (stacking traps or skull traps before engaging so they all blow up and kill him immediately after he goes hostile).

    Since time can be measured in terms of spells cast, (one per round), the strategy that kills Firkraag with fewer casts is the "better" strategy. So let's compare Feeblemind to Skull Trap. Both will want to open with 3x Lower Resists and a Greater Malison. (Technically the Malison isn't mandatory, but it'll speed up Feeblemind and reduce the average number of casts for Skull Trap by one, which is net neutral.)

    Also, let's assume you're not level 18 as a mage, because that renders this whole discussion moot. Vecna + Amulet of Power + Improved Alacrity means you can empty all of your Skull Traps *AND* all of your Feebleminds in an instant. (This is why I brought up Chain Contingency, by the way; you brought up Improved Alacrity earlier in the thread, and I was saying if you already have access to any of the big three level 9 spells-- Contingency, Alacrity, or Time Stop-- the Firkraag fight was trivially easy anyway.)

    So, with a level 16 mage, a skull trap deals 56 damage, 28 if he saves, which he has a 55% chance of doing after Malison. This gives you an average expected damage per cast of 40.6, and an average casts to kill of five. But that's an average; if he saves five times in a row, (something he'll do 5% of the time), odds are he's still alive after five casts and it takes a sixth, or if he saves against that, a seventh. And this is also assuming average rolls; if you get unlucky with your damage rolls, he could even take an eighth or a ninth. (Hell, if we're talking about "100% strategies", if you want to be 100% positive that you've killed Firkraag no matter how badly you rolled for damage, you'd need to be prepared to cast Skull Trap *23 times*.)

    Now, calculating exact kill probabilities is pretty wicked complicated with Skull Trap, since each cast involves the outcome of seventeen different independent dice rolls, (16 for damage, one for saves). But I can say for sure your probability of victory after one skull trap is 0%, and your probability after two casts is essentially zero. (Technically, if you roll sixes on all 32 dice and he fails both saves, you'll kill him with eight points to spare. The odds of one perfect skull trap are about one in three trillion, the odds of him failing his save to boot are half that, and then you'd need to square that for two of them.)

    After three casts, you will have dealt on average 122 damage, but if he fails all three saves (9% chance) that'll be 168 damage, and if you get lucky on your rolls you could conceivably take him down. But I'd guess your odds are maybe around 1% at best.

    After four casts, you're starting to get decent odds that you've killed Firkraag. It's still unlikely, but fairly plausible. There's about a 24% chance that he's failed either three or four saves, and the expected damage after three failed saves and one successful save is 196. With two or more successful saves, expected damage drops to 168 or less, but let's say you have a 1-in-4 chance of having killed off old Red and Scaly at this point.

    With five casts, you only need two failed saves to have an expected damage output over his max HP, which is going to be in the rough neighborhood of a 40% proposition. (At this point, it's not worth running the exact numbers). So let's say five casts gets you 50/50 to kill Firkraag.

    With six casts, the balance really starts to tip in Skull Trap's favor. Now, instead of the pressure being on Firkraag failing saves to die, the pressure is on Firkraag making saves to survive. Firkraag needs to make all six saves in order to keep the expected damage from Skull Trap below his 184 total HP, and the odds of him making all six saves are about 2.8%. Even if he makes them all, some really lucky damage rolls could still kill him. (On the other hand, even if he fails some, some unlucky damage rolls could spare him.) Overall, let's assume that six skull traps gives you about a 98% chance of victory.

    By seven skull traps, it would really take something spectacular for Firkraag to survive. Basically, he needs to make all seven saves (1.5% chance), *AND* he needs to get pretty lucky with the damage rolls. Say he's got only about a 0.5% chance of survival here.

    By eight skull traps, Firkraag is pretty much dead. We're looking at a fraction of a fraction of a percent chance of survival. (Although remember, in terms of a 100% guaranteed kill, it's technically possible for Firkraag to survive up to 22 consecutive casts.)

    Now, let's compare those percentages from Skull Trap to the ones from Feeblemind. Again, we're assuming that the mage led with 3x Lower Resistance and a Greater Malison.

    After one cast, Feeblemind has a 55% kill rate and Skull Trap has a 0% kill rate.
    After two casts, Feeblemind has a 79.8% kill rate and Skull Trap has a 0% kill rate.
    After three casts, Feeblemind has a 90.9% kill rate and Skull Trap has a ~1% kill rate.
    After four casts, Feeblemind has a 95.9% kill rate and Skull Trap has a ~25% kill rate.
    After five casts, Feeblemind has a 98.2% kill rate and Skull Trap has a ~50% kill rate.
    After six casts, Feeblemind has a 99.2% kill rate and Skull Trap has a ~98% kill rate.
    After seven casts, Feeblemind has a 99.6% kill rate and Skull Trap has a ~99.5% kill rate.
    After eight casts, Feeblemind has a 99.8% kill rate and Skull Trap has a ~99.99% kill rate.

    It takes seven casts of Skull Trap for it to finally catch up to Feeblemind, and by that point whichever route you went will have given you a near-certain victory, anyway.

    And these assumptions are about as biased in favor of Skull Trap as we can get. Again, by level 18 it's a moot point, so level 16/17 is the best showing Skull Trap is going to give. If you're trying with a level 12 mage, you're dealing 75% as much damage per cast, so it'll take more casts to net a kill.

    One advantage of Skull Trap is its damage will stack with any other sources of damage on Firkraag, perhaps from party members. Of course, one disadvantage is that it's not exactly party-friendly, so this is less of a concern for this comparison. If you're bringing friends to the battle, getting loose with Skull Traps is as likely to force a reload as anything else, so "Edwin + Jan + Aerie spamming Feeblemind" scales a lot better than "Edwin + Jan + Aerie spamming skull traps".

    This post is kind of going off the rails, but the larger point is this: if you want the best way to kill Firkraag, use exploits. If you want the best way to kill Firkraag without exploits, use cheese. If you want the best way to kill Firkraag without exploits and without cheese, use Feeblemind. It's the best, safest, most consistent, and easiest-to-use strategy for taking him down quickly before he can wreck your day.
    SkatanThacoBellgrisbossk
  • BorekBorek Member Posts: 513
    Like i said, 3 Skull Traps in your sequencer, Prot Magic weapons, then keep casting skull traps, he's chewing 7 before he can even hit a stoneskin. At that point you'll probably be out of 3rd level slots on a mage unless specialist (assuming you cast Prot.Fire) and can merely switch to unloading Magic Missile since if he's not dead he'll be nearly dead.

    Now same scenario with Feeblemind, you need Spell Immunity Abjuration and if you want to be sure with his Magic resistance that means either 3 LR's in a spell trigger and a manually cast Greater Mallison or manually casting a Lower Resistance, which is a 2nd 5th level slot. You have 5 slots, 6 if specialist, but question is would you risk having zero Breach spells, for this strategy lets say you are solo with no intention to melee, which considering none of your spells do any damage is wise. This means you have your Sequencer to unload 3, and at best 4, 5 if specialist. However, you have zero wiggle room, if you hit RNG bad luck which does often happen he's sitting at full health, lol, you've not even tickled him.

    You're betting everything on him failing 1 out of 7 or 8 saves with feeblemind, of which you freely admit the spell has pretty much zero use for most other aspects of the game, thus is going to require a special set of resting just to prepare.

    Skull Traps however are guaranteed, you *may* not kill him with 7, but if not you have more than enough 1st level slots for magic missile or Chromatic orb, 2nd level slots for Acid Arrow, etc etc because you have been damaging him all along, even the worst RNG will not prevent you from winning, if it goes horribly wrong you whack up a 2nd Pro.Magic Weapons and switch to 1st and 2nd level damaging spells.

    And that my friend is the point of the 100% strategy, it always works, there is no need for probability, he WILL die every time, without cheese, using this strategy, feeblemind, no, sometimes he won't.

    Added bonus is everything i'm casting is useful for generic play, so high chance i don't even need to rest to prepare, just need to save my sequencers and trigger for the big finale. On top of that you can *probably* do this strategy solo at level 13 (12 for specialist) which gives you 2 Prot magic weapons, but you will need to manually Lower Resistance as no spell trigger/Sequencer. Most definitely need spell trigger for the Feeblemind strategy though, unless you want to reload a lot. Obviously it's a serious risk w/o sequencer as you have less spells, doing less damage, but 2 prot magic weapons, plenty of stone skins, 4 Skull traps, 5 MM's, 5 Acid Arrows, a decent chance at least.

    Another good point for my strategy is you have spare 5th level slots for a breach or 2, which means the option to step in and melee him when he's low on health is on the table. I haven't tested on the ToB Dragons yet, but this worked flawlessly on all the Dragons in SoA including Adalon, although i was higher level by then so capping damage on the Skull Traps. Yet another nice feature is since they are not directly targeted spells they cannot be turned or deflected. In favor of Feeblemind however is that it cannot be blocked by either of the Globe of Invulnerability spells.
  • SomeSortSomeSort Member Posts: 859
    edited March 2017
    Borek said:

    Like i said, 3 Skull Traps in your sequencer, Prot Magic weapons, then keep casting skull traps, he's chewing 7 before he can even hit a stoneskin. At that point you'll probably be out of 3rd level slots on a mage unless specialist (assuming you cast Prot.Fire) and can merely switch to unloading Magic Missile since if he's not dead he'll be nearly dead.

    Now same scenario with Feeblemind, you need Spell Immunity Abjuration and if you want to be sure with his Magic resistance that means either 3 LR's in a spell trigger and a manually cast Greater Mallison or manually casting a Lower Resistance, which is a 2nd 5th level slot. You have 5 slots, 6 if specialist, but question is would you risk having zero Breach spells, for this strategy lets say you are solo with no intention to melee, which considering none of your spells do any damage is wise. This means you have your Sequencer to unload 3, and at best 4, 5 if specialist. However, you have zero wiggle room, if you hit RNG bad luck which does often happen he's sitting at full health, lol, you've not even tickled him.

    You're betting everything on him failing 1 out of 7 or 8 saves with feeblemind, of which you freely admit the spell has pretty much zero use for most other aspects of the game, thus is going to require a special set of resting just to prepare.

    Skull Traps however are guaranteed, you *may* not kill him with 7, but if not you have more than enough 1st level slots for magic missile or Chromatic orb, 2nd level slots for Acid Arrow, etc etc because you have been damaging him all along, even the worst RNG will not prevent you from winning, if it goes horribly wrong you whack up a 2nd Pro.Magic Weapons and switch to 1st and 2nd level damaging spells.

    And that my friend is the point of the 100% strategy, it always works, there is no need for probability, he WILL die every time, without cheese, using this strategy, feeblemind, no, sometimes he won't.

    Added bonus is everything i'm casting is useful for generic play, so high chance i don't even need to rest to prepare, just need to save my sequencers and trigger for the big finale. On top of that you can *probably* do this strategy solo at level 13 (12 for specialist) which gives you 2 Prot magic weapons, but you will need to manually Lower Resistance as no spell trigger/Sequencer. Most definitely need spell trigger for the Feeblemind strategy though, unless you want to reload a lot. Obviously it's a serious risk w/o sequencer as you have less spells, doing less damage, but 2 prot magic weapons, plenty of stone skins, 4 Skull traps, 5 MM's, 5 Acid Arrows, a decent chance at least.

    Another good point for my strategy is you have spare 5th level slots for a breach or 2, which means the option to step in and melee him when he's low on health is on the table. I haven't tested on the ToB Dragons yet, but this worked flawlessly on all the Dragons in SoA including Adalon, although i was higher level by then so capping damage on the Skull Traps. Yet another nice feature is since they are not directly targeted spells they cannot be turned or deflected. In favor of Feeblemind however is that it cannot be blocked by either of the Globe of Invulnerability spells.

    You can't just stuff Skull Traps into your sequencer. Whether you're using Skull Trap or Feeblemind, you need to lower his resistances first, which means those are going to be stuffed into your Spell Sequencer. Which means no sequencing your Skull Traps unless you also have Spell Trigger at your disposal, same as for the Feeblemind strategy.

    If you leave his resistances untouched, it will take you 3x as many Feeblemind spells... but it will take you 3x as many Skull Traps, too. (Actually, since we're concerning ourselves with tail probabilities and "100% strategies", leaving his resistances in place will increase the required casts far more than threefold.)

    And as I said, you'll also optimally want to drop a Greater Malison whether you're Feebleminding or Skull Trapping, since it will speed up both approaches. So that's not a difference between one and the other. And you don't need Spell Immunity Abjuration for the Feeblemind approach any more than for the Skull Trap approach. And PfMW protects you just as much while Feebleminding as it does while Skull Trapping. And so on.

    The optimal buffs, the debuffs, the setup, absolutely everything will be exactly the same for both approaches. Basically, your Firkraag strategy is going to be "Cast ________ spells to set up, then start spamming X", with the only difference being whether X is Feeblemind and Skull Trap.

    EDIT TO ADD: I confused Malison (lvl4) with Lower Resistance (lvl5). You won't be able to sequence LR in a Spell Sequencer, so you'll need Spell Trigger for that either way, or else you'll be casting it manually. This does mean that Skull Trap has the advantage of being Sequencable (1x Malison + 2x Skull Trap), whereas the optimal Feeblemind approach would require stuffing that sequencer with a lesser SoD spell in Polymorph Other, which lacks the -2 save penalty, which will slightly lower the kill chance. Here's the updated chart if you assume that the first two casts of Feeblemind are actually Polymorph Other, instead.

    And then after spamming X, here are the odds he'll be dead after each cast:
    After one cast, "SoD Spells" has a 45% kill rate and Skull Trap has a 0% kill rate.
    After two casts, "SoD Spells" has a 69.8% kill rate and Skull Trap has a 0% kill rate.
    After three casts, "SoD Spells" has a 86.4% kill rate and Skull Trap has a ~1% kill rate.
    After four casts, "SoD Spells" has a 93.9% kill rate and Skull Trap has a ~25% kill rate.
    After five casts, "SoD Spells" has a 97.2% kill rate and Skull Trap has a ~50% kill rate.
    After six casts, "SoD Spells" has a 98.8% kill rate and Skull Trap has a ~98% kill rate.
    After seven casts, "SoD Spells" has a 99.4% kill rate and Skull Trap has a ~99.5% kill rate.
    After eight casts, "SoD Spells" has a 99.7% kill rate and Skull Trap has a ~99.99% kill rate.

    So yes, I'm betting on him failing one out of seven or eight saves. But the odds of him failing one out of seven or eight saves are 99.7%. If you fight Firkraag 300 times, he will successfully survive your Feeblemind onslaught once. One time. Out of three hundred. I can live with that.

    You're right that I can never get my odds to 100.000% with Feeblemind, no matter how many times I cast it... but I've probably played through a lot. Maybe 20 times. Let's say there's someone who is far more into this game than I am, and she's going to complete it 50 times. And let's say that person is so impressed with my Feeblemind strategy that she uses it on Firkraag every single time through the game. If she casts eight Feebleminds at Firkraag every time, and she plays through the game 50 times, then not only was she incredibly likely to kill Firkraag successfully each time, she was also incredibly likely to kill Firkraag successfully every time. There's a better than 90% chance that she killed Firkraag without a hitch in all 50 of her playthroughs.

    If you try the "spam Save or Die spells" strategy 20 times, there's a 96% chance that you got 20 kills without any problems. Again, not only is each individual kill highly likely, it's *so* highly likely that using this strategy over and over and over and over again is almost certain to result in success every time.

    Meanwhile, Skull Trap can eventually get to a 100% success rate, but this does not mean it's a guaranteed kill. It's only a guaranteed kill if you can also prevent him from dealing lethal damage to you in that time. Remember, he's an angry red dragon who is actively trying to kill you and who has a fire breath attack that deals more damage than most mages have maximum hitpoints.

    The great advantage of Feeblemind is that after 7 or 8 casts, it's about as likely to have killed Firkraag as Skull Trap is... but after 1, 2, 3, 4, or 5 casts, it's orders of magnitude more likely to have killed Firkraag. If you fight Firkraag 100 times, he'll be dead after four casts of Feeblemind in 94 of them. He'll be dead after four casts of Skull Trap in about 25 of them. (Assuming you're level 16 or 17, the most optimal levels for the Skull Trap approach; if you're lower level, he's much more likely to still be alive.)

    The longer you stay in a fight with Firkraag, the more chances he's going to have to kill you. If there was a spell that dealt 1 magic damage that bypassed all resistances, and you could cast it unlimited times without resting, then 184 castings of that spell would result in a dead Firkraag with 100% certainty. But who on earth would advocate a strategy that required you surviving for 18 turns against an angry red dragon? Time to kill matters a great deal when punching above your weight class.
    Post edited by SomeSort on
    SkatanThacoBell
  • SomeSortSomeSort Member Posts: 859
    Borek said:

    You're betting everything on him failing 1 out of 7 or 8 saves with feeblemind, of which you freely admit the spell has pretty much zero use for most other aspects of the game, thus is going to require a special set of resting just to prepare.

    I also wanted to address this separately. I did not admit anything close to that. I said this: "I don't really have a rebuttal to your claim that Feeblemind is less useful than Skull Trap for general purpose play. That is a true statement. Defeating Firkraag requires different tactics than defeating a nest of spiders or a Sahuagin patrol."

    The statement "Feeblemind is less useful than Skull Trap for general purpose play" is not anywhere close to the statement "Feeblemind has pretty much zero use for most other aspects of the game".

    The vast majority of enemies you meet in the game are low-level mooks. They have low HP, low resistances, and come in large groups. The best spells for dealing with these most common encounters therefore deal solid area damage.

    But these encounters are also the least challenging part of the game, so a spell that excels in this particular department isn't exactly dramatically increasing your overall power level. It's mostly a convenience to be able to wipe out a pack of Hobgoblins all at once instead of one at a time.

    The most dangerous encounters typically involve a small number of extremely dangerous foes, instead-- perhaps, as with Firkraag, even just one. For encounters like this, area of effect is meaningless, (or worse than meaningless, detrimental to the health of your partymates). Instead, you want single-target firepower. And Feeblemind is fantastic single-target firepower, as evidenced by the fact that it's the most consistent and reliable tool you're going to have to take out some of the most dangerous enemies in Shadows of Amn without 9th level spells.

    Basically, Feeblemind is Finger of Death. Or rather, Finger of Death is basically a special sequencer that casts both Feeblemind and Magic Missile. Both carry the same -2 penalty. Except Feeblemind is a level 5 spell instead of a level 7 spell, meaning you get it five levels earlier, can cast 3 or 4 more per day for most of Shadows of Amn, and it competes for less valuable spell slots, (Breach and Spell Immunity are great, but every Finger of Death means one fewer Mordenk's Sword or Project Image). That's a great spell!

    (Aside: if you're concerned about having enough Feebleminds to last the encounter, this also means you can freely substitute in Finger of Death as much as you'd like without changing the odds.)

    You're not going to want to use it all the time, no. It's a waste against 95% of the mobs you run across. But 95% of the mobs you run across are a joke, not a threat. Feeblemind is a spell that increases in usefulness the more dangerous the situation becomes. It's at its best when the game is at its hardest.

    And it has a ton of general utility beyond just dragonslaying. For instance, if you cast Feeblemind at a character with a blue circle and it fails, the character doesn't go hostile. If it succeeds, no nearby NPCs realize what you've done. Imagine the potential here for, say, Mencar Pebblecrusher. Reduce his crew to a bunch of drooling idiots and then chunk them at your leisure. Or Tarnor the Hatchetman. He demands your money, you insult him, he orders his friends to attack you, only to realize that his friends are just staring at him slack-jawed.

    My personal favorite: Haer'Dalis has been reduced to a mindless tool by Mekrath. What better way to free him than by likewise reducing his captor to a mindless tool? One doesn't have to be a poet to apply poetic justice.

    Feeblemind is also one of the greatest tools for a thief who cares about his reputation and wants to minimize reloads. A Feebleminded NPC will not turn hostile when you pick his pockets, even if you fail your attempt. Feeblemind a bunch of noblemen around Athkatla and then relieve them of their possessions as you want with no risk. I often like to use Nalia for this purpose; given her tough talk about nobles in general, I pretend she's delighted to play Robin Hood on some stupid, rich fools.

    (Just don't feeblemind anyone plot-critical.)
    SkatanThacoBell
  • BorekBorek Member Posts: 513
    Dude really, spell trigger for LR, sequencer for Skull traps.

    There's no point saying anything else, i posted a 100% strategy, you posted a 99.8 (at best) strategy claiming it's as good, i explained in detail many times why it isn't, you say it is.

    Feel free to run a solo no reload game with feebleminds, it *might* work, but chances are you'll be resting a heck of a lot more than i do with my strategy.

    Anyone that wants to kill a Dragon every time, in 1 sitting, with no reloads or cheese, use my strategy.

    Let me repeat so it's clear, using multiple damaging spells to whittle him down means even if you need to switch to lower level spells after exhausting Skull Traps, you can do so, if you fail with all your Feebleminds you retreat, rest 3x and try again, or get eaten.
  • tbone1tbone1 Member Posts: 1,985
    Cavalier with sword & shield & helmet. Aerie & Jan waving pompons.

    That is it.
  • ThacoBellThacoBell Member Posts: 12,235
    Borek said:

    Dude really, spell trigger for LR, sequencer for Skull traps.

    There's no point saying anything else, i posted a 100% strategy, you posted a 99.8 (at best) strategy claiming it's as good, i explained in detail many times why it isn't, you say it is.

    Feel free to run a solo no reload game with feebleminds, it *might* work, but chances are you'll be resting a heck of a lot more than i do with my strategy.

    Anyone that wants to kill a Dragon every time, in 1 sitting, with no reloads or cheese, use my strategy.

    Let me repeat so it's clear, using multiple damaging spells to whittle him down means even if you need to switch to lower level spells after exhausting Skull Traps, you can do so, if you fail with all your Feebleminds you retreat, rest 3x and try again, or get eaten.

    If you are a high enough level to cast spell trigger, then you are high enough level to facetank him. Its like saying all you need to beat Firekraag 100% of the time is a gate spell or summon planetar. At those levels, it doesn't matter.
  • BorekBorek Member Posts: 513
    Spell trigger is level 8, meaning level 16 mage, you need at least level 18 for Planetar, Gate you can't even get until Suldanessar. A mage isn't going to face-tank anything, but yes the idea is you pre-buff so that he cannot strip your buffs or damage you whilst you cast. Not dying is not the issue, it's how to most effectively kill him with spells. I mean sure as an extremely high level FMT i can just place 3 spike traps and explode him, but the object is not to cheese him, it's to drop him with spells, not force-talking him or abusing the game mechanics.

    Both of us arguing agree that once you get Mage HLA's/9th level spells it's a done deal, but the problem with waiting that long is as a rule the sword is not so useful since you've probably blown away everything else except for a few challenging encounters.

    So the question is what is going to work 100% of the time BEFORE you attain demigod status (AKA 9th level spells)?
  • SomeSortSomeSort Member Posts: 859
    edited March 2017
    Borek said:

    Dude really, spell trigger for LR, sequencer for Skull traps.

    There's no point saying anything else, i posted a 100% strategy, you posted a 99.8 (at best) strategy claiming it's as good, i explained in detail many times why it isn't, you say it is.

    Feel free to run a solo no reload game with feebleminds, it *might* work, but chances are you'll be resting a heck of a lot more than i do with my strategy.

    Anyone that wants to kill a Dragon every time, in 1 sitting, with no reloads or cheese, use my strategy.

    Let me repeat so it's clear, using multiple damaging spells to whittle him down means even if you need to switch to lower level spells after exhausting Skull Traps, you can do so, if you fail with all your Feebleminds you retreat, rest 3x and try again, or get eaten.

    I did not claim that my "Save or Die" spells strategy was as good as your Skull Trap strategy. I claimed it was better.

    After eight casts of our respective spells, your strategy is 1.002 times more likely to have killed Firkraag than my strategy. After seven casts, it's 1.001 times more likely. That is the kind of edge that Skull Trap provides. That kind of edge is completely inconsequential from a real-world standpoint.

    Meanwhile, after six casts, my strategy is 1.008 times as likely to have killed Firkraag. After five casts, it's nearly twice as likely. After four casts, it's almost four times as likely to have gotten a kill. My strategy is EIGHTY SIX times as likely to give you one dead dragon after three casts, and it's infinitely more likely to have killed Firkraag after one or two casts, since it's essentially impossible for Skull Trap to kill Firkraag in two casts.

    This is at level 16. If we go down in levels to 12 or 13, the relative edge my strategy provides only grows.

    You are obsessed with 100%-- or, more accurately, you are obsessed with 100.0000000%. As I said, the minimum possible damage from a Skull Trap at level 16 is just 8 points, (all dice roll one, Firkraag makes his save). With 184 hit points, in order to get a truly 100% guaranteed kill of Firkraag, you'd need to be able to cast 23 skull traps. If you give a mage 23 casts of Feeblemind, the odds of killing Firkraag are 99.999999%, which you apparently find unacceptably low, so I can only conclude that you need to take it out to at least seven decimal places before you'll be truly comfortable with your odds.

    This is known as Zero-Risk Bias, or the human tendency to perceive inconsequential reductions of risk as truly consequential when they occur at the very tail edge of the distribution. If I told you a strategy improved your odds of success from 50% to 50.1%, you'd rightly discard that improvement as completely meaningless. But when a strategy improves your odds of success from 99.9% to 100%, suddenly that's the entire ballgame. Both improvements are still just 0.1%, though. You're chasing 0.1% gains at the end and in the process completely ignoring 50, 70, 80% gains in the middle.

    Also, your assumption that survival is a given, (to quote from your last post: "Not dying is not the issue, it's how to most effectively kill him with spells"), strikes me as an example of the Pseudocertainty Effect. Not dying is definitely *not* not the issue. In no-reload, not dying is literally the only issue! Taking survival as a given definitely changes the calculus, but in my experience, when facing dragons one should never take survival as a given. If survival is really a given, then who cares what strategy you use? Plink him to death with unenchanted darts for all it matters.

    In fact, that's what a successful Feeblemind does. It doesn't actually kill Firkraag. It just turns survival into a given, and from there you're free to plink him to death with unenchanted darts.

    EDIT TO ADD: To some extent, this whole idea that one must choose a Save or Die approach or a direct damage approach is a false dichotomy, anyway. I keep talking about Feeblemind, but not because Feeblemind itself is the key spell, but because Feeblemind is the lowest-level Save or Die spell that carries with it a save penalty, putting it into a cost/benefit sweet spot. Really, a mage could just as easily use Polymorph Other (level 4), Disintigrate (level 6, and with the Gore turned off it won't destroy loot), or Finger of Death (level 7). (Even something like Blindness, while not a true fight-ender, would be a pretty close approximation.)

    Finger of Death is the real interesting option. At mage level 16, you'll have two level 7 slots, or three if you're a non-Illusionist specialist. So for your first three castings, use Finger of Death instead of Feeblemind. You've still got exactly the same chance of landing an insta-kill, but if Firkraag makes all three saves, you'll still deal an average of 75 damage to him, which is right in line with what you'd have expected from Skull Trap if you'd cast that three times and he'd saved each time, instead, (84 damage, and those Skull Trap saves will be easier for him to make). From there, you can transition to either the Feeblemind strategy or the Skull Trap strategy to finish him off.

    This hybrid approach gives you the advantage of the pseudo-certainty at the tail end, (as you say, magic damage spells can reach 100% chance at a kill, while Save or Death spells can only approach it), while also giving you the early advantage of a pure "Save or Die" approach of having a good chance of ending the fight outright in just a couple rounds.

    Of course, this approach relies on being a high enough level to get multiple castings of Finger of Death, but not a high enough level to cast Improved Alacrity. But then again, if you're much below level 16, the magic damage approach becomes less and less appealing, anyway.
    Post edited by SomeSort on
    ThacoBell
  • BorekBorek Member Posts: 513
    You should read what is written not what you want to see, there's a fancy word for that as well btw, google it if you like.

    The point about not dying was that neither of our posted strategies have ANY realistic chance of the character actually dying since we both agreed that pre-buffing as described makes one invincible for 8 rounds (2x Protection from Magic) plus at least 1 round more (stoneskin), plus additional castings of stoneskin plus a potential contingency of Stoneskin or Pro.Magic Weapons. With Resist fear, spell immunity Adjuration, Protection from Fire, Stoneskin and in my case Minor Globe of invulnerability (not needed but avoids potential splash damage) he simply cannot touch the Character, the only danger is wing buffet during casting, which with Robe of Vecna and Amulet of Power is not going to happen, and if you stand with the wall directly behind you isn't an issue regarding being blown backwards.

    But then again i went over this before, so PLEASE READ in future and stop making pointless arguments about non-issues. The ONLY issue is certainty of killing, since worst case scenario is spells failing to kill him and having to flee when out of slots.

    Which, as i have repeatedly pointed out, is ONLY an issue with YOUR strategy since it relies on Saving throws being failed and not mine which relies on damage dealing.

    Yes, 7x Skull Traps at level 16, all saved, all for minimum damage (feel free to work out THOSE Odds btw) will do 56 damage, but then you have 5 Magic Missiles for minimum of 50 damage plus 6 Acid Arrows (1 slot on resist fear, 2 from minor sequencer) for 72 damage over 5 rounds minimum, plus you have 4 level 5 slots, 1 level 6, 2 level 7 and a level 8. So you've got 178 guaranteed damage plus plenty of slots to deal the other 6 damage. And this is worst case scenario in so far as the mage is not a specialist, doesn't have any items to increase spell slots, simply being a specialist will add 30 more damage, as would using Dak'kon's Zerth Katana, there's also 2 different rings of wizardry.

    So yes, it's guaranteed even if you manage to hit the depths of bad luck and bottom out on absolutely everything. Unlike Feebleminds, hence my point about you being wrong, probability does not equate to a guarantee, not when you are dealing with a small sample size of spells such as this scenario, something you should, again, look up, since it is extremely important in statistics and probability calculations.

    I agree, you are more likely to kill the Dragon with less casts, that is not my concern when devising a guaranteed kill strategy, i mean just look up the word Guarantee, lol. If i wanted a "high probability" strategy then i may consider using Save or incapacitated/death spells, but at no point have i used anything other than the words Guaranteed and 100%, when i start using High Probability and 95%+ feel free to chime in with your well known fact that casting lots of Save or die/incapacitate spells are highly likely, but not GUARANTEED, to result in victory.
  • SomeSortSomeSort Member Posts: 859
    Borek said:

    Yes, 7x Skull Traps at level 16, all saved, all for minimum damage (feel free to work out THOSE Odds btw) will do 56 damage...

    ~1 in 1.82 * 10^89, which is somewhere around 100,000,000 times greater than the number of atoms in the observable universe. (But is not, crucially, zero.)

    As for the rest of it, I fear there's been a fundamental misunderstanding; you seem to be under the impression that I'm putting forward the "Save or Die" strategy as a 100% guaranteed kill. I'm not. In fact, I will go so far as to guarantee that this is the very first time I have issued any guarantee of any kind in this thread.

    From the beginning I have talked about acceptable levels of risk. To me, a 99.8% success rate is an acceptable level of risk, especially for a strategy that is cheap to implement, simple to execute, non-exploitative, and broadly generalizable, (e.g. it works with a solo level 16 caster, but it also works with a full party of level 11 characters, and with anything on the spectrum in between). You disagree. To you, 1 in 1.82 * 10^89 is an acceptable level of risk. I happen to agree. Your computer has a better chance of spontaneously combusting and killing you in the process than it does of rolling that many 1s in a row. But at the end of the day, it's all just a question of risk mitigation.
  • BorekBorek Member Posts: 513
    Indeed, risk mitigation being my aim, mitigating to absolute zero when possible. Anyone who has played this or other games knows all too well that sooner or later the Random Number Generator God will bend you over and ruin you.

    I am actually playing a solo no reload game atm, not going to hit Big Red for a while though. I did steal the Feeblemind scroll, just made sure i was spot on 250 pickpocket skill although i believe it's still 1% failure chance on some NPC's.

    Already got the Katana and the game engine automatically awards any Kitted mage multi class as a specialist, i went for the obligatory Kensai/Mage/Thief, so i will have a genuine 100% kill chance by the time i hit level 16 mage and have all the required spells, just missing Protection from Fire and Spell Trigger for now.

    Probably use a Project image though, after all why risk the real me and waste all my spells when i can use 1 7th slot >:)
Sign In or Register to comment.