Tree of Savior Forum

Reorganizing the Class Trees/Widening the Class Pool for Early Ranks

Gameplay Feedback General Feedback
#1

Back when ToS only had 7 ranks and 13 unique classes for each of the 4 trees (not counting hidden classes), having only two “new” classes to pick from at each advancement felt reasonable. There wasn’t really anything to build into for the long run, because the long run wasn’t really too far off. Rank 7 being the end goal made the selectiveness of our advancements feel more enjoyable, even though our early options were few. Plus, Rank 7, like the earlier ranks, only had two “new” picks. It was a bit of a tight squeeze, but it was uniform and still felt relatively open-ended, and it was always at least somewhat clear what you were going to build into.

Now that we have two more ranks and 14 more unique classes than before (again, excluding hidden classes), the narrow selection of “new” classes with each advancement has started to feel much more constrained. Even though the game has continued to grow and expand, we’re still locked into those same early build choices. We still only have the same two (“new”) options. Late-game classes allow for more variety in their supplementary early classes, but those options are still tightly restricted by the small pool we have in the early game. You REALLY start to feel how little wiggle room you have. Rank 8 gives you 4 new classes to pick from (at least for Swordsman and Wizard), along with whatever comes next, and because of that late-game variety, everything BEFORE then feels more restricted by comparison. And even though you have those new classes and that openness to look forward to at Rank 8, you still have the same number of possible combinations in Ranks 2-7. You are still locked into the same pattern of choices. You feel forced to know EXACTLY what you want by Rank 8/9 and so instead of building a unique class, you end up “making do” with classes or extra circles you might not want or need. And when you take into account that skill damage is all percentage based now, this regimented stratification as a method of power balancing is at least somewhat unnecessary.

To simplify it, you have all these new cherries-on-top, but nothing is different about the range of your choices in which sundae you want.

As such, I would like to suggest that (at least Rank 2) the earlier ranks be widened to include one more option for advancement. This would allow for early builds to have more variety than they already do, and make the juxtaposition of Rank 8’s choices and earlier class choices feel MUCH less rigid. And as new classes/ranks get released, you widen subsequent early ranks to further increase the variety.

How this might look (for the sake of demonstration):

1 - Classes that offer new weapon types/playstyles lower in rank. This way, players can experiment with a newer playstyle sooner. And because of rank rollback, building early for a weapon type they may end up not liking is less of a gamble, since you can play around with it and go back if you changed your mind. Ideally, the wider class pools mean that your early build is less likely to be catered specifically to that new weapon unless the player is certain of their build and what they want out of it.

2 - Progressive Condensing. As new classes get released, you widen subsequent early ranks to further increase variety in builds. This makes it so that the early freedom is still there, and you still have a comfortable amount of options late game. (Operating off of the assumption that all classes released from here on will be Rank 8, with Ranks 9 and 10 being used to give them their three circles)

Swordsman:

Rank2 - Peltasta (Tanking), Highlander (New Weapon Type/Slash DPS), Hoplite (New Weapon Type/Pierce DPS)
Rank3 - Barbarian (Slash DPS), Rodelero (Tanking/Strike DPS), Cataphract (New Weapon Type/Mount Mobility/Pierce DPS)
Rank4 - Corsair (Party Support/Dual-Wield DPS), Squire (Economy/Party Support)
Rank5 - Doppelsoeldner (Slash DPS), Fencer (New Weapon Type/Counters/Pierce Damage)
Rank6 - Dragoon (Pierce DPS), Templar (Guild Functions)
Rank7 - Murmillo (Strike DPS), Lancer (Mount Mobility/Pierce DPS)
Rank8 - Matador (Counters/Pierce DPS), Retiarius (Crowd Control/Dual-Wield DPS)
Rank9 & Rank10 - Circles 2 and 3 for Rank 8 classes.

Wizard:

Rank2 - Pyromancer (Fire DPS), Cryomancer (Crowd Control/Ice DPS), Linker (Party Support)
Rank3 - Psychokino (Crowd Control/Magic DPS), Elementalist (Elemental DPS), Thaumaturge (Support)
Rank4 - Sorcerer (Summons DPS), Chronomancer (Support)
Rank5 - Necromancer (Summons DPS/Magic DPS), Alchemist (Economy)
Rank6 - Warlock (Dark DPS), Featherfoot (Poison DPS/Side-Healing)
Rank7 - Sage (Magic DPS/Side-Support/Economy), Enchanter (Support, Economy)
Rank8 - Shadowmancer (Dark DPS), Onmyoji (Elemental DPS)
Rank9 & Rank10 - Circles 2 and 3 for Rank 8 classes.

Archer:

Rank2 - Ranger (AoE Missile DPS), Quarrel Shooter (Traps/Missile DPS), Hunter (Companion Support)
Rank3 - Sapper (Traps/AoE DPS), Wugushi (DoT/Poison DPS), Scout (Support/Stealth)
Rank4 - Rogue (Stealth/Evasion/Close-Range DPS), Fletcher (Crafting/Missile DPS/DoT)
Rank5 - Falconer (Companion DPS), Schwarzer Reiter (New Weapon Type/Mount Mobility/DPS)
Rank6 - Cannoneer (New Weapon Type/AoE DPS), Musketeer (New Weapon Type/Missile DPS)
Rank7 - Hackapell (New Weapon Type/Mount Mobility/Close-Range DPS), Mergen (AoE DPS)
Rank8 - Bullet Marker (New Weapon Type/DPS), ??? (???)
Rank9 & Rank10 - Circles 2 and 3 for Rank 8 classes.

Cleric:

Rank2 - Krivis (Passive Support/Magic DPS), Priest (Healing/Active Support), Dievdirbys (Passive Support)
Rank3 - Bokor (Summon DPS), Sadhu (Magic DPS/Support), Paladin (New Weapon Type/Support/Strike DPS)
Rank4 - Monk (Strike DPS), Pardoner (Economy/Support)
Rank5 - Oracle (Active Support/Economy), Druid (Magic DPS/Support)
Rank6 - Plague Doctor (Active Support/Magic DPS), Kabbalist (Math/Support)
Rank7 - Inquisitor (Strike DPS/Anti-Mage), Taoist (Magic DPS/Passive Support)
Rank8 - Zealot (DPS/Self-Support), ??? (???)
Rank9 & Rank10 - Circles 2 and 3 for Rank 8 classes.

As you can probably tell, I’ve put a great deal of thought into this. The above examples are just how I visualize it. I understand that there might be parts of this that aren’t feasible, and I also understand that the game is far from finished. If nothing else, I just wanted to put the idea out there and see what the community thinks. So please, share your thoughts!

Thanks for reading. See you all in the game!

5 Likes
#2

While this looks good in theory, there is one problem:

The Rank 2 Classes are supposed to complement the initial idea of the Class.
The best example is the Cleric Class, that advances into either Priest or Krivis at Rank2.

To sum it up, all 3 Classes have ± the same idea, bringing recovery and support skills into the Cleric Class tree.
If you add Dievdirbys, you bring in specialized gameplay/support mechanics before offering an expansion on the initial concept of the Class tree [which is covering the basic concept of MMORPG-Classes, Swordsmen as twohanded dealers(Highlander) and/or tanks(Peltasta), Clerics as healers, buffers(Priest) and debuffers/damage dealers(Krivis), Wizards as magic damage dealers (Pyromancer) and CC support/damage (Cryomancer) and Archers as bow-wielding skillspammers/debuffers (Ranger) and autoattackers/tacticians(Quarrel Shooter)], which might make it harder for people to grow into the game.

The main problem I have is that the first 1-3 Ranks are more of a gaming tutorial, as most Rank 1-3 skills don’t really have an impact on lategame, even if you invest the few thousand attribute points to level the enhancement attribute to max.
Master Circles might solve this initial problem in the future, but for now, it’s still more of a problem that the lower Ranks have lesser impact on gameplay [except for a few skills like Heal,Guardian,Surespell,Pain Barrier and Swift Step] and the Rank limitation that makes it less rewarding not to choose a Rank 6+ Class on the higher Ranks (especially with Rank 8 Classes being really strong).

I honestly have no idea how IMC plans to address this problem, with the possible upcoming powercreep in the following Ranks[10,11,12…] where people can keep stacking Rank 8 Classes on top of each other.

3 Likes
#3

I still use the starting skills and they are still useful even in late ranks…

Sleep is useful on all levelsand ranks, so are safety zone,full draw, pain barrier…

what matter is the mindset of the player and sure dmg might go down, but they have an utility…

#4

Regarding your first point:

I understand the necessity of keeping it simple in the early ranks. I tried to take this into account as I went through and moved stuff around. (Though, moving Hunter instead of Sapper was probably not the best move in retrospect, as Hunter is [arguably] a more complex class. The same could be said for moving Linker instead of Psychokino.) In terms of Diev v. Bokor, I feel Diev is incredibly simple in comparison to summons management (Step 1: Drop statues. Step 2: Ok, go back to fighting. Step 3: Profit???). And take into account that the addition of Hoplite is just swapping out the 1H sword for a different weapon/method of DPS (1H spear), in the same way Highlander does (2H sword). “Wanna keep your shield but ditch the sword?” and “Wanna ditch the shield and make your sword bigger?” are more-or-less the same thing. In this specific case, I feel that there’s no real issue moving Hoplite down a notch, as the first circle doesn’t really introduce any greater degree of unfamiliarity than Highlander does.

As this is laid out now, I recognize that Archer and Wizard trees might need a bit more thought, but I feel that the Cleric and Swordsman trees are reorganized in such a way that there wouldn’t be that steep of a learning curve.

But like I said in the op, this is all for the sake of illustrating my point. I’m not devoted to the idea of Ranks 2 and 3 being widened. Keep the simpler stuff in the early game, then once players have acclimated to it, widen the middlemost classes (Ranks 4-6[7?]). That would also be fine. My overarching point was that our options need to be widened SOMEHOW, because if classes just keep piling into Rank 8 and everything before then stays the same, nothing is really changing about the number of possible combos available to us before then. The choices you end up making might change to suit the newer classes, yes, but you’re still operating within the exact same constraints. You’re doing different things, but nothing is ACTUALLY different.

The proposed change would allow for a more diverse community of players and more unique builds, which is something I would really like to see. (Wiz3ele3lock2/3 has been the single most popular build in the entire game since day one. Most of the runner-ups haven’t changed all that much either.) Playing the meta should feel more like a choice and less like a necessity. I feel more variety in individual ranks’ choices would help do so.

Regarding your second point:

I feel that the issue of balancing the early skills, while valid and definitely crucial, is something that can only be solved when/if this change ever happens. Things are already in pretty hefty need of rebalancing (PvP is basically only for magic DPS now, and clerics have almost no weaknesses or shortcomings). But if rebalances keep happening in the context of bottlenecked class variety and then (hypothetically) this change comes, everything will be back to square one. Everything will need to be rebalanced all over again (which is part of why I don’t really expect this change to happen).

By gradually slotting down higher-ranked classes as new ones come out, this helps to buffer the powercreep, and incentivizes proper nerfs and buffs where they are needed, which would ideally gradually even out the playing field. If early-rank classes are supposed to be weaker, that’s fine. In that case, move higher ranked classes lower down the list and balance them out accordingly so that growth maintains a steady pace and Rank 8 doesn’t become stockpiled with the “only classes that actually matter.”

Like you said, how this problem gets addressed and where IMC goes in the future is up to them. But offering suggestions and at least putting ideas out there, which can simultaneously give them a pulse on how the community feels about the issue, is highly productive for both the Devs and the community.

#5

This is a cool idea, this gives us more options on builds, but some filler classes may be discarded (not a big deal to most I believe). When you said widening, I was hoping for a new class for each rank XD obviously too good to be true though.

1 Like
#6

Haha, for sure. I think it’s pretty safe to say that anything new is gonna be for Rank 8 and up from here on out.

#7

I’ve seen several talks about that, because it’s indeed an issue the more you think about the future.
You still have the normal restrictions at lower ranks, then once you’re rank 8 you’ll have to choose between at least 4 classes.
I think this is also why IMC is working on “Master” classes.

The problem is that while you’ll have a good and big choice of classes at R8, that’s not the case for earlier ranks. You have to fit 13 different classes and 3 circles for each of them (so a total of 39 classes) into 7 ranks… or spread them into 9 ranks and not take any R8 classes.
The tree is basically too thin at the beginning and widens suddenly and A LOT at R8. I also feel it’d be smoother and interesting to spread the classes a bit more like you proposed.
(Some people also proposed to switch R8 as the rank of choice to R5, R6 or R7)

Also - it’s my own perception - but it makes it feel like the “real deal” starts at R8 and before that the classes are more the fillers/utilities for your build, rather than classes as important as your later ones. They did solve this issue a bit by making most skills have percent-based damages rather than flat ones though.

1 Like
#8

I was thinking about this in the past weeks and even came up with a idea of rank reorganization and I was going to make a suggestion thread here but my english isn’t good so I was struggling a lot when typing and was delaying my post.
Here is some things I was thinking: Building Rank6 classes like Fencer, Druid till C3 now is the same thing as building Corsair back when we only had 8 ranks. You don’t have many good options for the last rank and you feel that the build is incomplete because you can only take 1 rank of some other class, and when rank 10 arrive it will be even worse, imagine in the case of Fencer C3>Matador1, when they implement the rank 10 it will be possible for Matadores to get C3 and I would imagine that they will implement some skills or attributes for C3 that will make the Rank1~2 almost not worth(or incomplete), so builds including rank7 C3 classes will be highly underpowered compared to builds with rank8 classes.
What I was trying to suggest is separating the existing classes in distinct groups, giving us more options to early ranks classes and also freedom to them to add classes not only to Rank8 group, but to the early ranks groups too in the future(like they do with Hidden Classes now).
Of course with the change it will be needed some adjustments and balance, but I think it will be easier for them to balance classes according to certain group than balancing 8 different ranks.
You can see in the image below my idea: Base Rank 1 will remain the same, Class Rank 2 will unlock all Rank C classes, Rank 5 will unlock all Rank B classes and Rank 8 will remain the same, unlock all Rank A classes.

image
image.jpg1223×604 338 KB

3 Likes
#9

funny thing is if they implement that… it will be another major overhaul update and people will be rushing to try again and the servers will crash again… and pooofff… everything is silent. :joy:

#10

People logging in and trying things is good for the game, the game not handling this is… IMC fault, but I think they improved this in the past months.

2 Likes
#11

I actually really like this idea, A LOT. Even better than mine, tbh. It also circumvents the issue of “progressive condensing” and the constant need for rebalancing as new classes come out I mentioned before. Widening the Ranks by changing how they’re formatted (Ranks 2-4 have Pool C, Ranks 5-7 have Pool B, Rank 8+ has Pool A), while keeping them in relatively the same place… really smart :astonished:

Still, I think @Asriel has a fair point regarding higher-complexity classes being introduced too early. Giving players more variety is absolutely necessary, but throwing too much at them right away might be overwhelming. But to be fair, that really depends on the person, plus there are tons of veteran players who can offer up their own experiences and feedback on the classes for new players to read through.

The issue of how skills get balanced at each advancement concerns me a bit, though. It doesn’t change the fact that higher-level content comes at later ranks, so keeping them in three separate pools might result in a few dead classes, if they aren’t balanced out properly. The only fix I can think of off the top of my head is by making stat growth-per-point more impactful (gain more stats per point invested). Or even giving each base tree their own stat growth factors. By this, I mean how your stats grow and change with each point is decided by the classes you pick (Classes with Phys DPS receive bigger gains from investing STR/DEX than others, Tanks get bigger gains from CON than others, Magic DPS gains more from INT, and Healer/Support get more benefit from SPR). The growth scaling depends on which pool you choose from, and will stack with subsequent pools (So Pool C will give lower stat gains for a c3 Hoplite than, say, a c3 Dragoon. Each stat would also have a scaling cap, so going strictly into DPS doesn’t turn you into an unstoppable powerhouse). This might encourage people to spread out their stats more and play with different class combos that they wouldn’t otherwise consider.

But then comes the issue of how the PvE/mobs get balanced. If they’re too weak, the game’s too easy. If they’re too strong, the game’s too hard. That might be something that IMC would need to tackle, cuz on that front, I got nothin’.

#12

I really like yours distribution, and I want to add that each “pool of ranks” must have a hidden class, meaning four hidden in Pool C (one of each class tree) four Pool B and four Pool A.

1 Like