The most powerful thief
Dukinson
Member Posts: 46
My friend and I (he's crazier than myself about optimization) are planning another BG2 run. I like trying to get away from the simplistic multi F/thf and trying something a bit more tricky like ken/thf dual. We disagree on which is better overall seeing as there is downtime for a 9ken/thf vs the 7ken/thf. He prefers the instant recovery of skills with 7ken/thf but even between that and the multi f/thf, he prefers the multiclass kosherness (even bg1 and on it's cleaner). You lose 12 hp, +1 dmg/+1 hit and a Kai by dualing at 7. It doesn't seem like that long to wait for 2 more fighter and 2 more thief levels, but he does have a point it's at semi important parts of the game were power is needed.
It might be important to add we play nightmare mode and the gibber tweekpak to max out enemy hp. Magic is dead in our campaigns because of enemy hp, more powerful grandmastery, the overpowered might of GWW, and he hates resting. I suppose he thinks resting over and over is unfair. When you have near or over 20 GWWs, he does have a point, but ya, magic is dead so no point in offering thief alternatives like M/thf or F/M/thf for purposes of this conversation.
Finally, before anyone begins typing in replys, I would make this a two parter because it's still on topic. To open up more thief combos, I made an argument that it doesn't make sense you can specialize and dual class but your first class must be the specialized one. Why not the other way around? If I were to go 9 levels in fighter, and 39 in thief, am I not truly a specialized thief? Infact, I still can't decide whether a multi F/thf, dual Ken/thf, dual F/SD, dual F/Sin, or perhaps even a F/Swash is the highest utility/strongest damage dealer overall throughout the game.
So two points. First, which thief is the strongest? Second, does anyone here say dualing into a specialized class with EEkeeper is unfair? If so, why? Keeping in mind we impose a one specialized class limit (no dualing from a special to a special) as well as the points that shortys/multiclass are quite strong.
---------------------------
Situation:
-4 player party (Bard + tank + ? + ?)
-NM mode
-around 8 rests the whole game
-tweakpack full enemy HP
---------------------------
It might be important to add we play nightmare mode and the gibber tweekpak to max out enemy hp. Magic is dead in our campaigns because of enemy hp, more powerful grandmastery, the overpowered might of GWW, and he hates resting. I suppose he thinks resting over and over is unfair. When you have near or over 20 GWWs, he does have a point, but ya, magic is dead so no point in offering thief alternatives like M/thf or F/M/thf for purposes of this conversation.
Finally, before anyone begins typing in replys, I would make this a two parter because it's still on topic. To open up more thief combos, I made an argument that it doesn't make sense you can specialize and dual class but your first class must be the specialized one. Why not the other way around? If I were to go 9 levels in fighter, and 39 in thief, am I not truly a specialized thief? Infact, I still can't decide whether a multi F/thf, dual Ken/thf, dual F/SD, dual F/Sin, or perhaps even a F/Swash is the highest utility/strongest damage dealer overall throughout the game.
So two points. First, which thief is the strongest? Second, does anyone here say dualing into a specialized class with EEkeeper is unfair? If so, why? Keeping in mind we impose a one specialized class limit (no dualing from a special to a special) as well as the points that shortys/multiclass are quite strong.
---------------------------
Situation:
-4 player party (Bard + tank + ? + ?)
-NM mode
-around 8 rests the whole game
-tweakpack full enemy HP
---------------------------
Post edited by Dukinson on
1
Comments
Just consider:
Fighter -> Swashbuckler
Thief -> Wild Mage
Ranger -> Priest of Lathander
Fighter -> Wild Mage
With that said for context - I don't have a lot of experience with "illegal" setups. Dualing into kits can be as wowo said - very powerful but I'm not entirely sure if any of them break the game. I'd be really curious if others with more experience with unusual combinations had something to say on the matter.
You mention fighter/shadowdancer and fighter/assassin and fighter/swashbuckler. My initial thought is that only the swashbucklers will get to whirlwind and I find that ability to be extremely overpowered and therefore necessary in powergaming compositions. With that aside - just evaluating each of these on their own, I would have to ask whether you want to backstab as a normal part of game play.
If the answer is yes, forget the swash, if no - then consider the grandmaster dagger throwing swashbuckler which seems entirely unfair.
Now if we are going to backstab then HIPS vs x7 multiplier is a big question. I'd probably default to the bigger numbers but without a second thief you'll have a hard time, especially considering when you dual.
I'd ask @Lord_Tansheron if he has any thoughts on the matter or any of the other "crazies" on the forum.
As a final thought if you are allowing dualing into kits, you might consider going into a fighter kit to get whirlwind HLAs - I guess kensai would make the most sense if as a 13 thief -> and would do the highest damage but kensai without UAI would be very vulnerable.
For reference my favorite party setup is currently a 4 man powergame.
1 TANK (ranger/cleric or fighter/cleric or dwarven defender)
1 THIEF (fighter/thief)
1 BARD (skald)
1 PURE DAMAGE (archer or dagger kensai or swash->fighter)
Personally, I don't much like Thieves. They're mostly convenience to me and not actual power, but that's largely because of my mods and/or scruples about certain exploits.
If you don't mind going all-out in what you can do, then Spike Traps and Mislead -> backstab are some of the most ridiculous sources of damage in the game. The latter of course requires you to be a partial Mage to use effectively.
The best "legit" Thieves (i.e. allowed under default rules) I found are probably the following, in no particular order:
Kensai 9 -> Thief dual: some damage bonuses and later on UAI to negate downsides, plus gets to use SNT
Swashbuckler 10 -> Fighter dual: essentially a fighter with 100 in pop locks/traps and some Detect Illusions
Assassin X -> Mage dual: dual point can vary; has lots of interesting applications for Poison Weapon, plus ML backstabs
Fighter/Mage/Thief multi: good choice for small parties, where you combine all the various strategies; not that great in large parties
If you do allow "creative" combinations with dualing into kits etc., then the best one is probably Fighter 9 -> Swashbuckler dual. Max APR under IH, UAI, and sick damage bonuses. Alternatively, you could also do Thief 10 -> Kensai for less flexibility (no UAI) but slightly more damage and Kai.
I haven't tested Shadowdancer extensively. HIPS could lead to some interesting setups, but if you want to play like that it's probably better to go with Mislead.
Side note: I would strongly recommend dualing Fighters at level 9 in BG2. XP before lvl 10 is trivial in BG2, and WELL worth it. I don't generally recommend dualing at level 13 because it's not worth the cost, and level 7 is probably only a consideration in BG1 (but I am not too well-versed in BG1). If you're running Nightmare Mode, then it's level 9, no questions asked. Kill a bunch of mephits or goblins, and get those levels back in no time.
Thieves mainly exist to improve your quality of life. Bonus XP from traps/locks, stealing, scouting out blast zones, detecting illusions, setting up traps, UAI, backstab, pickpocketing, etc. When it all comes down to it the road to power in BG2 boils down to accumulating XP and loot/gear, and the rogue-types know a lot of shortcuts.
The multiclass F/T is the most reliable thief type. You get just about all the utility you need from a thief AND the end-game reliability of a fighter. You also get access to thief HLAs as a freebie.
Dual classed F/Ts can work depending on party composition.
I don't think that the kensai thief si so powerful, some better THACO and damage is useful, and kai for max damage Backstabbing is even more useful. But a well played backstabber usually can manage to have more than one backstab, he manage to have the enemy following him, turns a corner, hide and backstab again.
And if he con not backstab, specially with the settings you are using, a lev 7 or 9 kensai have big problems to survive, the way to 3M XP + 9 fighter levels is just too long.
Imo a multi is better, no down time, effective from the beginning, will reach a good power even if he levels slow, is not a caster, as fighter don't need lev 20 to be effective and as a thief he have plenty of skill points, is just slowed in the bacstab multiplier. But from 3M XP he will have both the HLA pools. In late game he will be more than a plain fighter (assassination, UAI for scarlet and scrolls) and better than a plain thief ( real fighter THACO and CS when backstabbing).
With him I will avoid thief kits, assassin is slowed in skill points and as multi is double slowed, bounty hunter also but get 20 x level so maybe, shadowdancer I don't know, I play not EE.
If you don't like backstabbing the fighter/swash multi can be crazy.
It's not that multiclass is bad per se, it's just that I personally don't think the trade-offs are worth it. Also if you're going multiclass already, there's little reason not to go F/M/T instead, especially on NM.
That's for (mostly) unmodded games, naturally. Mods can change things considerably. If you're running with SR/IR, for example, it may very well be worth it to not run any duals at all, just multis/hybrids.
However your best bet is playing a F/T multi simply because of Hardiness, especially in Nightmare Mode.
About dual he will end with the rogue THACO, the down time to dual at 13 is more then 2M XP, imo a lot of game before having the toon ready, level 9 is better. But this shorten the down time, not much reaching UAI, is still 3.25M XP (at 13 is 4.25M). And before UAI the K/T is a glass cannon.
Imo the K/T is the classic end game monster that suffer his big problems for the hardest half of the game (since the power curve of the party is steaper of the enemies one) and in the second part have (maybe) a little advantage. the K/M is different as he can use his magic to offset the lack of AC.
I agree in dualing from Berseker, it can be done early and after the down time the toon can have a good AC.
I still prefer the multi, useful in every stage and after 3M the HLA offset the missed GM and the rest.
Sure, everyone has 200+ HP at least but my character already had -20 AC in chapter 2 so he was avoiding every physical attacks and had Mirror Image / Stoneskin to block the criticals. I have 4.000.000 XP and I'm still in chapter 2 without leaving Athkatla. So far I think you can still destroy anything with a Fighter/Mage hybrid sort of build (which includes F/M/T, F/MC, Fighter => Mages etc).
For a character without spellcasting like a F/T, I would definitely bring Hardiness to the table or at least dual later on. I agree that the early game is the hardest part of the run, but still a level 7 or 9 Kensai will have a really poor THAC0 anyway later on and doesn't quite cut through the early game neither (like any builds with Poison Weapon for example). Plus with the huge experience increase, dualing at a higher level shouldn't be too difficult.
I think ToB is going to be really really painful to play without Hardiness, it increases your EHP by at least 64% without any source of regeneration, healing or any other sources of damage resistance.
From my play so far, I've found damage to be quite relevant still. Leaving dangerous enemies alive too long can cause cumulative problems. While you can't steamroll-brute-force your way through fights the way you can outside of NM, a good offense remains the best defense.
In the end, it's rarely ever been about IF you can beat the game, but HOW. Of course you can do it in many, many, MANY different ways, that's not in question. What bears most discussion is how well those many ways compare to each other.
Same for this discussion. There's no doubt that Thieves can be made powerful, but that isn't the question - the question is relative power levels, pros and cons, stuff like that.
I admit I haven't tested Thieves down to the last details. I never really liked them, because I can't find their niche. Perhaps once Sneak Attack is properly implemented, it'll warrant another look.
So I'll better let talk who uses it.
I would add this is a 4 person party. A bard and a tank and a damage dealer are the other 3 slots. Ofcourse, the thief I play could take the place of the damage dealer as a thief, like a F/sin, but then another thief would be needed to deal with traps until the sin walks the long road on his thieving abilities. So this is complicated for sure and we've spent hours talking about it.
However, I'm still not convinced I can justify dualing into kits quite yet as I do try to keep the game as fair as possible. The reason I make the attempt, is that there really are few kits you can make any use of. Multi-shorties are so powerful in the difficult NM mode. A few kits like archer, bard, and kensai are great, but I can think of few others from the large potential list. Even dualing isn't all the strong. You must play human, a weak race as well as enduring down time before recovery of starter class abilities.
From a pure legit standpoint:
I like 9 ken/thief dual class. Backstab and boots of speed and make for some ridiculous damage with staff of the ram. Many fights I was able to zip to a hiding spot and explode enemies, every single round. Even w/o staff of the ram, I was able to do this technique and give our pure damage dealer a run for his money. Impressive considering I was originally there to steal/disarm/lockpick. This last run was with the f/thf multi halfling. I could have done it even better with the same exact setup (more scrolls and potions of invis for more backstabs). My goal is to beat that damage and beat the damage dealer this time. To find the most powerful thief. I don't believe a 10swash/thf can beat the 7-9ken/thf or a multi f/thf without backstab, but I could be wrong and am impressed with the great feedback in this thread so far. It's giving me pause for consideration. With only 8 sleeps, could a thief/mage of some variety still be viable? I'm currently still thinking no... my brain! My crazy BG2 buddy and I will be on skype for hours about the feedback here.
From a justify-into-dualing-kit standpoint:
If I can justify it, I still feel backstab must be worked in. 13-thief/swash is interesting but requires another thief. This means I would need to surpass the damage loss from the pure damage dealer AND I will create item competition by with the requirement of this second thief. Another loss. Same goes for f/Sin. Another thief would be needed. I really like the potential of the f/Sin with mega backstabs and a +1/+1 dmg/hit to boot.
So to recap, the ideal thief I want to discover will do the most damage for being present, kill the most enemies by the end of the game. If he cannot take care of stealing/disarming/lockpicking, he must TRULY be a monster to justify his position in the party for requiring another thief in the crew to pick up the thieving slack.
---------------------------
Situation:
-4 player party (Bard + tank + ? + ?)
-NM mode
-around 8 rests the whole game
-tweakpack full enemy HP
---------------------------
The solution would seem someone with high sustained damage that does not depend on external sources. That basically gives you two options: backstab, or high-damage (melee) attacks.
For backstab, you'd need some way to repeat. Mislead is out since you can't rest, so that leaves running behind a corner (pedestrian) or Hide in Plain Sight (Shadowdancer).
For melee, you can increase your per-hit damage or your APR, and conveniently enough a fighter dual will do both (Berserker->Thief, Kensai->Thief, or Fighter->Swashbuckler all do it in some way).
Multiclass actually works less well with limited rests, since you can't make full use out of the Fighter HLAs (all of which are activated limited daily use) and lose Grand Mastery. You do however get better THAC0, which can help offset the exhaustion penalties.
I'd personally probably go with Kensai->Thief. The reason is twofold: it doesn't rely on daily use for its damage bonuses (unlike the Berserker) and it offers both good sustained melee damage *and* the ability to backstab (unlike the Swashbuckler). It will however require more babysitting to avoid becoming the punching bag of some random enemy.
You can either go for the legit Shadowdancer => Fighter dual, or the ridiculously OP Fighter => Shadowdancer one and slaughter everything with ease.
Or pick the Kensai => Thief road as @Lord_Tansheron said but I would really advice you to dual at least at level 9, you will really need the additional HPs and kit bonuses if you are fighting through Nightmare Mode with something like 8 rests.
You get the Fighter's strength, APR and GMM + Assassins poisons (very strong), +1hit/+1dmg innate bonus, thieving skills (TRAPS!!) and backstab.
I started a Fighter, got to level 7 in BGEE and put 4 points into Daggers. In BGEE, you can get with that to lvl 8 Thief and GM in Daggers. (Used Daggers because of flavour and the fact you can fight with them and throw them and there are some really good throwing daggers in BG2EE.
If you are a powergamer - a Shortbow and Scimitars/Longswords is probably a better option.)
Assassin starts as a lousy Thief, but in time he will grow and lack of Thieving points will not be an issue anymore.
I've played this character in my multiplayer game with a friend Fighter/Cleric multiclass. Yes. Without magic. No, we did fine. Yes, we killed, back-stabbed, mangled and otherwise processed everything in sight.
Or, in other words, I don't think that in a multiplayer game a competition between players is productive. In BG like in basketball is often detrimental, a playmaker that has a good percentage of success in throwing the ball at the basket * is always welcome, a playmaker whose goal is to have himself the best score in the game (so throws that ball at the basket way too often) is very seldom a good player and very seldom is appreciated by the coach and the other players.
Rogues, both bard and thief, are utility classes, that can deal a huge damage. They can be the best damage dealers in the party, but imho only in a single player game, because there they are charname and the goal of the player is to have charname shine. In a multiplayer, if there is a designed damage dealer is often wise to let him cover the role and help him, with good backstabs, but also helping in tanking when needed, helping in plain mlee when backstab is not possible or somehow less productive for the party and so on.
As I told this is only my humble opinion, feel free to build your toon as use it the way you like, as long as you, and the other multiplayers are happy everybody is happy.
*I am not a native english speaker, I hope that everybody understand what I mean for throwing the ball at the basket, that maybe is not the usual way to express that concept.
@gorgonzola I definitely understand your point about "taking kills" but I think that as long as all the players are making the correct strategic choice then you can get some good data. For example if the thief backstabs an almost dead enemy - this COULD be a good choice, if that enemy is very dangerous or on a squishy, but it also could be a terrible choice if it is weaker and dying soon to a summon or arrows that are mid flight. Obviously trying to last hit will skew the data in an undesirable way. In my campaigns my multiplayer friends all compete a bit on kills but we understand the results can vary considerably and many other factors of contribution must be considered.
@dukinson I think your best composition will be.
fighter/cleric - multi (dwarf)
fighter/thief - multi (halfling)
archer - (half elf)
skald - (half elf)
However kensai9->thief has some damage advantages over the fighter/thief except for lack of whirlwinds. There are 9 mislead scrolls that you can get (that i know of) so you'll be able to utilize those heavily in watcher's keep and TOB. I think the fighter/mage/thief doesn't work out as well with such limited sleeping arrangements.
If you are really convinced you want to try one of the illegal setups - you'll be happiest with fighter7->assassin. You can dual early and still function as the thief through the first dungeon - you'll want to use master thievery potions for pickpocketing and you'll want to ignore detect illusion until you can get 200-300 into move silently and hide in shadows. With assassin's 15 points per level it will take you quite a long time to get everything you need but you will get barely enough at the start to function properly.
Obviously you can use ranger/cleric multi but you'll need to be half elf and lose the +5 saves from shorty and you'll lose 9 hit points by end of game and you'll get thaco / saves slower - but you gain............iron skins - which is arguably all worth it (assuming you don't consider it cheating).
My personal record is 28 days for beating the entire game with a similar party. Only one sleep in the expansion. The largest difference is that my fighter/thief didn't backstab and instead used throwing daggers exclusively (because i'm lazy). I believe a properly micro'd backstabber can easily exceed this and so a game time of 27 days should be possible.
On a final note if i ran my similar party again I'd probably try out the fighter7->swash dagger thrower as it seems to just get EVERYTHING. I do consider this cheating though so I wouldn't consider the end result official (by my own crazy standards that needn't apply to anyone else - LET ME HAVE MY THING).
Fighter 13 -> Swashbuckler 39. 2 APR, full Grand Mastery, 8 THAC0, +7 to hit, +8 AC, +7 damage per hit, no equipment restrictions, no need no backstab or restrict weapon choice, may equip Gauntlets of Extraordinary Specialization for +1/2 APR, +1 to hit, +2 Damage, which a kensai can't. You'll run slow early SoA, but just get better and better by the time you hit 2.1m and beyond, which shouldn't take long with a limited party size.
+9 damage per, +8 to hit, and 2.5 APR pre-GM and 5/4/6/5/5 saves.
No need for Belm, so offhand Crom Faeyr or something and get all five awesome hits per round instead of 4, or dominate harder than anyone except Archer kitted rangers with a 5 APR Short Bow of Gesen or 4 APR Sling of Everard, while busting out Carsomyr for pesky mage types.
And the tank? Once they get their class abilities back, that's going to be them. -9 AC in Armour of the Hart, -13 with 18 Dex, -14 Cloak of Displacement, -15 Helm of X, -17 Gaxx, -18 Ring of Earth Control, -19 for Axe of the Unyielding, -24 with a shield, and lower still against anything except Crushing.
Haha, I kinda sigh as well at the rest rule. It does further diminish what few classes are truly powerful. However, my crazy CRAZY cohort is insistent on low rests.
Dualing at 13 will mean a long wait and another thief needed for the party and create more item competition. This is why it's so hard to get away from the archer as the pure damage dealer. They need so little desired gear. Same deal with the bard. They let the tank and the thief horde all the best items. Tank gets all defense gear, the thief most of the offense gear.
If I were to go "dual into kits", backstab is so powerful that the assassin seems like the clearest choice. Think of all the invis potions and scrolls in the game, each a mighty backstab. Swash is strong, but no backstabs vs +8 dmg seems like a tough sell. The tanking would be nice but no cleric spells and (even in NM mode) a long wait to get the primary class back would be a rough situation if he were to be the primary tank. AC isn't everything, especially in the expansion. He would be a nice offtank though.
I think this campaign will be done by the books. I don't even allow myself to grandmaster in the beginning. It's odd, you can't high master a weapon as a level 1 fighter but you can in bg2 with an autoleveled fighter. This is essentially ingame allowed cheating.
@gorgonzola
I would respond but Tredvolt basically covered it. It's all about teamwork and the overall function of the group.
@Tredvolt
I'm liking
1 - fighter/cleric (tank)
2 - 9kensai/thief (offense thief)
3 - Archer (offense)
4 - Skald (Bard)
I believe the halfing multi F/thf is so reliable and a strong contender for the best thief but perhaps not the most powerful. Also, recently playing a campaign with one, I feel I know what this character can do right now. Another playthrough with him with what I know now would still be fun though.
A 9k/thf will make great use of the shield amulet and has -2 ac to help out. It won't be such a terrible "glass cannon" early on. There are other items that give ac but the thief shouldn't have to worry about this to much as he will be using two handed swords and staves to stay at a greater distance while summons and the tank soak most of the hits. Should I work in throwing daggers somehow? I haven't decided.
I kinda consider the ranger cleric cheating, he's getting a spell(s) he shouldn't. He isn't a shorty, so I'm torn on how unfair he really is though.
The next time I play, I have to try a dual 7fighter/assassin. If it turns out to be absurd, I won't play dual into kits. Multi kits is clearly cheating so I'll never do that. Even if I dual into kits, the primary class must be a standard class. At first glance it seems closer to fair as you must be human and most kits are just weaker if you dual out of them or singleclass them.
If we allow dualing into kits, it is quickly clear that the vast majority of kits benefit from being dualed /into/ rather than /out of/. It is no mistake that the game does not work this way by default.
That said, for folks who like customizing the game to their hearts content, and keepering characters around annoying and arbitrary game restrictions, knock yourself out! The nice thing about the whole BG cycle is that we get to choose our definition of what a good/fair game is
If you allow it in your game are you really increasing the amount of possible builds? I mean, if you can play a fighter -> assassin why would you ever play an assassin -> fighter? (Which is otherwise a decent build).
This said, if I might break down the maths of that "tough sell" for a moment.
Assassin Stealths, Approaches, Attacks. Assassin THAC0 is generally going to be worse than Swashbuckler THAC0, but assume it hits and is wielding Staff of the Ram: 108 average damage, and let's say... 2 more non-backstabs if you're not mislead cheesing, for 139 average damage before strength. Assuming 22 Strength, that's 169 damage on an average round of attacks if the assassin gets in all three attacks, 354 if you're using assassination that round.
Impressive, certainly, particularly as a first strike.
Swashbuckler meanwhile approaches and whirlwinds with a single handed Crom Faeyr (to keep it simple, . Ten attacks = 115 damage average, +70 swashbuckler bonus (assuming the swashie is level 35+), 7 extra attacks worth of bonus Mastery damage (+5 for improved GM?) is +35.
That's around 220 damage before strength, which at +14 per hit from Crom Faeyr's 25 strength is 360 damage in a round.
And if you start mainhanding the Flail of Ages at 8 APR + 2 Crom Faeyr APR then you can increase that by around 60, meaning if you're running 5 APR when dual wielding (2 @ level 13 + GM + 0.5 Gauntlets + Dual Wielding + Haste) then you're looking at around 204 damage per regular round of attacks.
And for the record, even with +4 from hiding vs -4 from Whirlwind Attack, the Swashbuckler with Crom Faeyr is still more likely to land those hits in the first place.
If I'm being unreasonable with the assassin, by all means propose alternatives, but hopefully this better demonstrates the benefits of the Swashbuckler a bit more clearly than just a lone "+8". The assassin may possibly outdamage them on a single round, but the Swashbuckler benefits on every attack, every weapon, every round. and against every enemy - nobody is immune to Swashbuckling. Better saves, more proficiencies, higher late game HP, better hit chance, and Fighter HLAs. The thief class doesn't actually have much to offer past assassination and UAI.
F/swash is an example, F(7)/Assassin is another, high APR and GM for a fully operational Assassin, +7 backstab and lots of poison werapon, with a very short down time.
Imo kitting the multiclasses is no more OP, The multi slower leveling balance well the GM and faster progression of the second class of some duals, a multi Kensai/M is different but not overpowered compared to a dual one.
I am not referring to the minimal reload of this topic, just comparing illegal kits with vanilla ones in general.
Fighter/Swashbuckler multiclass gets full backstab progression, -5 base THAC0 (only Kensai does it better), +5 Damage, and +6 AC, Assassination and Greater Weapon Whirlwind HLAs... And Timestop traps. And yes, Backstab applies to the extra Swashbuckle damage.
Fighter/Shadowdancers get full backstab progression and full skill points (still no set snares skill though, but for HLA traps I believe they use the bardic auto-success model).
Worst(best) of all are Mage/Thieves, since they don't dual as well.
Swashbuckler/Mage loses weapon specialisation for Swashbuckling, but again, full backstab and +1 Mage spell slot per level, because mages with kits get +1 spell slots - any kit. Whichever way you run the dual, they're missing out on a lot.
So let's say that both the dualed into kit and multi can create, but don't always do, characters that are OP in comparison of the best duals from a kit. And the developers was wise in allowing only the last.
If someone chose to dual into a kit or to give a kit to a multi he have to judge by himself as the result sometimes is just an other flavor, different from legittimate duals but with similar power, and sometimes is at a completely different level of power.
As a personal opinion I feel that using kits in multi or at the end of dual classes clearly opens up some overpowered combinations but I never feel good using these. Maybe it is that the existing combinations have so much depth and potential that I don't feel the need to go outside these boundaries, or maybe its because it somehow disrupts the sanctity of the experience.
Obviously opinions differ, and I will always disclaim that i fully believe each of us can find how they most like to play the game.....
But for myself, I think I have recently decided to draw a line and opt out of pursuing this gray area of character creation. I've even stopped using illegal race/class combinations and started focusing on just the core rules. As @Dukinson mentioned I recently didn't even realize that the proficiencies at the start of bg2 are bugged - and so I've manually been restricting myself. (you shouldn't be able to grandmaster that soon)
Lastly I'll just mention mine and some others apparently preference toward less sleeping. While it obviously does mess with balance - I feel that sleeping can be heavily abused and take away from game. I find it much more exciting to work with limited resources and carefully plan how many of those resources I use for each battle instead of blowing all my whirlwinds - spells and abilities on each fight.
To me (limited resting)
- increase the difficulty of the game by limiting your resources
- puts more strategic depth into the encounters (no whirlwinding kobolds)
- places a sense of urgency on our heroes
- creates more decision making as far as what encounters are worth pursuing
- from an RP stance, our heroes are battle beaten and worn, low on resources and hit points, after a long period of battles they emerged victorious but the foes were worthy and took their toll on the adventurers.
The down side to low rests
- some classes and combinations become obsolete. For example Mages want to click that rest button after every little fight because why not? Their spell pool is so small that they could reasonably exhaust it nearly every fight. Truly this is an unfortunate aspect, but it doesn't seem to hit the variety of compositions that badly.
For those that haven't tried the low rest environment, and want the added challenge - you might actually like it. Fatigue is a real issue and you'll start to look at your abilities in a whole new way as you carefully portion them out. The goal becomes defeating the battles with expending the least effort and taking the least damage, rather than just defeating the battle and resting back up to max power.