The 1-3-5 Philosophy of Skill Break Points

[Fortescue]

HELLO! Let us talk about game design! Specifically, class design, skill design, and player choice!

We begin with assumptions:

1) We're only talking about Class categories for the purposes of this discussion

2) Class points exist because ideally each class would have multiple viable builds / play styles

3) Diminishing returns per effective skill rank is both necessary for game balance (monsters) and won't ever go away

4) In many cases, skills that scale only with diminishing returns on combat stats (hp, damage, etc...) are underwhelming by the time the player puts 5 points in them. For a recent example, the new skill Bloody Butcher in the Bloodthirst tree is very hard to justify putting more than 2 points in!

--------------------------------------------------------------------------------------

With that out of the way, let me present the most workable solution I have figured out to both keeping diminishing returns on numbers within a reasonable curve AND to allowing for class build options with meaningful variance!

The 1-3-5 Philosophy!
I believe that the best way to give players meaningful decisions to make about where to spend Class points is achieved by implementing the 1-3-5 system of skill break points to as many skills as we possibly can as designers. Addon makers, unless you've figured out a way to work around diminishing returns, this applies to you as well!

To illustrate my point, take the skill Death Dance. In its prior incarnation most players only ever put 1 point in it, because it did its job of hitting multiple enemies for a decent chunk of damage with that level of investment. More points made it hit harder, but the diminishing returns combined with a long cooldown meant that most players did not consider it a very competitive option for their Class points vs. something like Stunning Blow that offered increasing duration on a very powerful debuff as more points were added, in addition to its own rapidly diminishing damage increase.

In the most recent revision, game version 1.2.2, Death Dance now offers a new perk at rank 3, an additional Bleed effect applied to enemies. This is great, but only for Berserkers really out of all the classes that have access to 2H Weapon Assault. This is because of their skill Bloody Butcher, and Death Dance being the only way Berserkers can reliably make things Bleed. However, there is again very little incentive to go past 3 points other than to slightly improve your odds of one shot killing enemies you strike with the skill, because it does have a very high initial hit damage value to begin with. Because Death Dance has a long cooldown (10 turns) and the Bleed effect only lasts 5 turns, there is the additional problem introduced of Bloody Butcher's primary benefit only being applicable 50% of the time, lowering the value of Bloody Butcher further! Finally, there is a 3rd problem I noticed while actually playing a Berserker today. As you get near level 20 you begin to run into a lot of situations where you have plenty of Stamina to use a skill like Death Dance, but no option to do so because of said long cooldown. You are left bumping into monsters instead of being allowed to trade your Stamina for their quick deaths! This slows down gameplay and is generally annoying in my opinion.

What is to be done about this? Well, when I designed the new Death Dance / Bloody Butcher combo I had envisioned a new cooldown mechanic for the skill, which lowered said cooldown for every enemy you struck with Death Dance. The reason is that this incentivizes using the skill as intended, vs. large groups of monsters. This did not make it into the game, likely because the on hit damage value was kept very high. However, there are now the 3 problems I described with the skill and its relationship to Bloody Butcher as a result. What can we do next to incentivize players going past 3 points? What we can do to incentivize non-Berserkers to reach 3 points in the first place, without making the Bleed effect too powerful from a raw damage aspect? How do we lower the cooldown without making the skill itself too powerful? These are the kinds of questions we must ask as responsible game designers who care about crafting meaningful decision points for players.

An example solution, which I hope DarkGod implements in 1.2.3, follows:

-----------------------------------------------------------------------------

Death Dance - Active
Spin to win, doing (70, 90, 105, 115, 120, etc...) % weapon damage to all adjacent foes, and an additional (50, 70, 85, 95, 100, etc...) % weapon damage in Bleeding over 5 turns. For each opponent struck by Death Dance, the cooldown is lowered by 1 turn.

At skill rank 3, the Bleed damage is doubled, and spread out over 10 rounds instead of 5.

At skill rank 5, each opponent struck lowers the cooldown of Death Dance by 2 turns.

Cooldown 10, costs X Stamina (I'm too lazy to look up the base price)

-----------------------------------------------------------------------------

So, lets deconstruct these alterations. First up, the base damage was mauled! Down by 100% vs. current 5/5 points damage of 220%. This is necessary for balance, as it answers the question of "how do we lower the cooldown without making it overpowered?". Next big difference, Bleed works from the start! This is important because it lets Berserkers get the synergy with Bloody Butcher with a smaller investment, if they don't think they need to keep their targets bleeding until they die. Finally, before you even get past 1 skill point the cooldown can now be lowered by 50% just by attempting to hit 5 enemies with it before you press the button. This encourages a new play dynamic where the player can decide if they want to just dish out the damage or try to maximize the number of targets they hit, causing more in that round, and later when the skill comes back faster. All in all, a great investment for 1 point that creates fun gameplay scenarios without being too powerful!

So what do we get for 3 points now? Well, a hard to detect in practice, but significant, power boost! This favors Berserkers more than other classes because now they can keep a single target Bleeding forever as long as they don't miss, enabling Bloody Butcher to work full time. The Bleed damage is doubled, but spread out over twice the duration, so unless your enemy lives for more than 5 rounds you won't actually see the damage boost as a player. As well, Bleeding is a stacking debuff in ToME, so now everyone who uses Death Dance can stack this effect very easily, a nice benefit for Berserkers and non-Berserkers alike. Again, very good investment for 3 points!

Finally, the deal closer! What could possibly convince a savvy player that Death Dance is worth spending 5 entire Class points on? Well, as I mentioned above, around level 20 on my Berserker I kept running into scenarios where I had Stamina to burn and just wished I could use Death Dance more often, but was unable to. By lowering the CD by 2 per enemy hit, Death Dance is now able to be used repeatedly vs. a group of 5 or more enemies limited only by your Stamina. Does this break the game? I don't think so, I won those fights even without being able to end them more quickly. There will be scenarios where being limited only by Stamina and repeatedly stacking the Bleed effect will lead to winning tough group battles that you couldn't have otherwise, but also scenarios where running yourself out of Stamina then needing to use Unstoppable during that same fight will get a Berserker killed. Testing and feedback would be required to ensure this is the right decision, but remember, 5 points in a single skill is asking a LOT!

Rather, it SHOULD be asking a lot, but very often it is not as ToME currently stands. This is because not enough skills offer compelling rewards for investing 5 points in them. In fact I would go as far as to say very few actually do, including many that I have designed! During my design phase of new Berserker and Skirmisher I tried to buck the system, thinking there would be something akin to hard damage caps if I could just scale things how I wanted to. I understand now that this was the wrong approach, and both of these projects suffered from my unwillingness to accept the diminishing return system as it is and work around it. I could tell though, and you can see it in some skills that offer significant boosts at 3 and 5 points, that this was going to be the way to deal with the limitations introduced by the diminishing returns. I just did not embrace the 1-3-5 philosophy quite as hard as I should have at that time, and some skills have suffered for it as a result!

So what can I do about this problem? Let me provide another example:

-----------------------------------------------------------------------------

Bloody Butcher - Passive
You revel in the blood of your enemies, allowing you to ignore (10, 15, 20, 25, 30) % of their physical resistance while they are bleeding. This value scales with Strength. (at 100 Str, double the listed values, does not reduce resistance below 0%)

At skill rank 3 your lust for blood is such that the act of making an enemy bleed reduces all incoming damage by 30% for that turn. (raise your resist all cap / resists by 30% for that turn only)

At skill rank 5 you are driven into a frenzy at the sight of foe blood, gaining 15% global speed as long as an enemy is bleeding.

-----------------------------------------------------------------------------

Now we have a skill that both synergizes extremely well with the (newer) proposed Death Dance AND competes for points with other skills, because getting both of them to 5/5 would make you extremely good at fighting groups of monsters! Note that, because this skill primarily exists to help Berserkers (a pure physical damage class) overcome enemies with lots of physical resistance, it does not have diminishing returns. This is because the resist system operates largely with the 0-100 realm of numbers and no part of this skill deals damage, heals, or restores a resource (areas that all need diminishing returns explicitly). Like Death Dance, this is a rewarding talent to spend 1 point in for any Berserker since you will be able to ignore 20%~ of physical resist from most enemies you meet while they are bleeding once your Strength is high, and for some players that might be plenty. Going further though, at 3 points you get encouragement to take on more risky behavior in the form of strong protection when using Death Dance. Finally, the blood frenzy induced 15% global speed boost is a fine thing to have even if the player doesn't take the proposed Death Dance past 3 points!

The design challenge is far from over, however. There are plenty of other Berserker skills that lack strong justification to take to 5 points, or even 3 in some cases. To create a genuinely interesting class with multiple strong build options requires a lot of planning and careful decision making. In the examples provided, however, I have given an outline of how to make an group fight specialist Berserker by going 5/5 in both proposed skills. Other builds should be able to compete with this, builds that demand those points be placed elsewhere, to facilitate other interesting play scenarios I haven't even thought of yet These alternate build specific boons could include better tanking, better single target damage, better debuffs, etc... All at the cost of ravaging groups of enemies


Conclusion!

My call to action for all of you who are interested in designing skills and attempting to make older skills more appealing, is to utilize the 1-3-5 philosophy thoroughly in your designs, creating within individual classes a greater range of meaningful player choices and play styles! It will require creativity at times, but I believe we can make ToME even better through improving player choice, and making their characters feel more like THEIR characters who are good at things that player specifically wants to be good at!

[Davion Fuxa]

Interesting philosophy for Talent design to try and have breakpoints in some of the Talents that have since had problems with their scaling. I don't necessarily think all Talents would need it, just some Talents that are struggling a bit. For Talents that really struggling, it more has to do with their use in combat (or lack of use) that could use a redesign.

Shield Pummel is an example I'll throw in that once you get 4 Talent Points invested in will hit its maximum Stun Duration pump out - but a player might use Shield Pummel a lot so their is incentive to investing another talent point to bring it to 5/5 is fairly high. That's fairly good design for a skill

The problem with Death Dance right now is that regardless of the Talent Points used, you can't use it again for another 10 turns - by which a group of enemies may be dead. There was really no incentive for increasing Talent Level past 1/5 either since it's damage scaling was ridiculous for a Talent that saw such little use. Adding Bleed for synergy with Bloody Butcher simply just made it so that Berserkers needed to bump it up to 3/5 Talent Points, and maybe a consideration for other Non-Berserkers looking to perhaps take advantage of the Bleed effect.

The idea for Death Dance is a good direction to probably pursue with revising it further. However, it has to take into account player decision making, the inclusion of other classes (Arcane Blade and Wyrmics), and plain balance towards other Talents - specifically Fearless Cleave but also Sweep and Whirlwind from the Two Weapon Fighting category (since Arcane blades for example might opt for them over Two Handed Weapons).

Lowering cooldown based on a player hitting enemies is probably going to hit a big 'NOPE' for player decision making - specifically because it entails actually getting into melee range of multiple enemies. Effectively it looks like Death Dance would turn into a 5/5 or don't bother taking talent. Additionally, some thought process needs to go into how Fearless Cleave interacts with Death Dance here. Fearless Cleave is technically a substitute AoE - useful for finishing off a group of really close weak opponents or when maxed out at 5/5 usable as taking out a small cluster of enemies. If Death Dance becomes too usable then Fearless Cleave is effectively turned into a 1/5 Talent which is bad design, since having to make a decision to use Fearless Cleave is currently a meaningful decision atm.

The Two Handed Assault category has to somewhat fit in to the grand scheme of things - specifically in relation to Dual Techniques and Shield Offense. In generally there needs to be some sort of thought process for designing the specific weapon trees so that a player has some meaningful distinctions to note between the different trees. Right now Shield Offense caters well to Sword and Board players by allowing for strategic blocking and powerful counterattacks, Dual Techniques features some of the most powerful but simple techniques in game that cover a wide variety of uses. The old Two Handed Maiming was weak in comparison because it didn't offer any real advantage while featuring a lot of subbar talents (which literally included everything but Warshout). The Two Handed Assault is a fairly good improvement because it sort of holds its own niche by offering 'advanced utility' that doesn't make the category necessarily more powerful or weaker then the talents offered in the other categories. This balance needs to be kept.

Changing the talent so that it more closely aligns with Berserkers, and notably makes it an 1/5 'Talent Filler' or 5/5 'Or Nothing) talent is probably not a good idea unless you plan to to throw Death Dance into a Berserker only category, which is a consideration - could swap it with Shattering Blow, which would turn the Two Handed Assault into really strong utility category for dealing with enemies. If left in Two Handed Assault though, then the talent should be made as useful as possible to other classes that can use it as well.

Really, I think that there are a few more lessons to throw in here - including the famous KISS rule (Keep it simple stupid) because your also making the talent really more complicated then it needs to be.

To throw in a concrete idea for Death Dance if it remains in Two Handed Assault -

-Lower down the cooldown so that use of Death Dance is encouraged a bit more, but not enough to orphan Fearless Cleave: I suggested a value of 5 here. Just lowering the value will make the skill more useful without discouraging players from using it even without hitting multiple opponents.

-Knockdown the base damage to where it would be very useful with only 1/5 talent points invested but not necessarily really powerful: I'll suggest starting at the simple 120% of weapon damage here. Essentially with just a single talent point spent, you hit multiple opponents around you for a little bit more then full weapn value at a cost of stamina. This is a good cost ratio if you need to hit multiple opponents and haven't 5/5 Fearless Cleave; somewhat comparable if you did.

-Make the damage scaling per increasing Talent Levels meaningful, so that it isn't an overpowered 1/5 wonder while increasing it to 2/5 through 5/5 a player gets a notable amount of value back: Sweep and Whirlwind both scale fairly well with a 70% value, Whirlwind in particular which also starts at 120% goes up to 190%. Notably though, Dual Techniques have awesome talents - Dual Strike merrily laughs at Stunning Blow, Flurry is friggin Flurry. Let's makes Death Dance match Whirlwind with it's 120% to 190% value, but with the reduced cooldown of 5 on Death Dance compared to Whirlwinds 8

-Think about the bleed effect. Death Dance can do without it, it's a powerful Talent enough with just base weapon damage on multiple opponents on all your flanks. Maybe ditch it and think about another way to make Berserker make use of a Bleed effect - notably I think making one of the Bloodthirst Passives, since that is where the Bleeding matters anyways.

[Fortescue]

You're missing the point that number tweaks are not going to ever work, because diminishing returns must apply on all number tweaks. Either the skill is good at 1 point and scales poorly (don't go past 3!!!), or is crap at 1 point and never becomes good, because the returns must diminish. I also don't get why you think any of these changes would be bad for the other classes, or what this has to do with Shields or Dual Wield stuff.

Also the "too complicated" bit doesn't jive. None of the breakpoint rewards I listed are hard to understand. Ugh. Also what you said about Fearless Cleave / this overlapping, because they don't. None of this jives at all and I don't get where you are pulling this from. Are we playing the same game?

[edge2054]

Those diminishing returns are there to prevent exactly what you're trying to encourage. 5/5ing talents isn't supposed to be a 'good' thing until late game when you're running out of places to put talents that synergize with your build and when multiplicative damage increases are at their peak.

[Davion Fuxa]

You have to design the class so that everything fits in - the other talents that the character with the talent has access to, as well as when the player has to make a decision between another category with the talent over another category that would do without it (Arcane Blades and Wyrmics have the option of choosing to use Two Handed Weapons, but they can also ignore them in favor of going with Shield and Board or Two Weapon Fighting (Dual Techniques or Mindstars) if they wish.

In looking at other talents in the Two Handed Assault category, the thing about Fearless Cleave is that you can use it as a fill-in AOE while Death Dance is on cooldown to finish off a group of enemies - that's a good thing that helps make Fearless Cleave have a lot of utility both for advanced up a tile toward an enemy but also if you are already close to a bunch of wounded ones. Overpowering Death Dance nerfs Fearless Cleave because it renders Fearless Cleave only useful for advancing on enemies because you can just use Death Dance quickly soon again if you want.

In comparison to Dual Techniques and Shield Offense versus Two Handed Assault, an Arcane Blade will be making a choice on whether he wants to use Two Handed Weapons over Sword and Board or Two Weapon Fighting. Arguably right now, the buff to Fearless Cleave and Execution have balanced out Two Handed Assault enough that Arcane Blade players MIGHT consider using Two Handed Weapons over the other weapon styles. Changes to the Two Handed Assault have to take into account that Arcane Blades will be benefiting already quite a bit with the current setup - adding more benefits is going to start stacking Two Handed Assault too much compared to Dual Techniques or Shield Offense when it comes to decision making on which category to pursue. Wyrmics are a similar yet slightly different story in that they have Two Handed Assault and Shield Offense to choose from, and snazzy players can go pick up Mindstar Mastery if they want to use those.

*****

In regards to the Death Dance changes itself, it is perfectly plausible that the talent can be made so that it works well at any Talent Level. Whirlwind which I practically just re-suggested for Death Dance, is a good example of a Talent that was designed well enough that it works well at all Talent Levels. A Player can get 1/5 Whirlwind (Starts at 120%) for a bit so he has a little AoE capability early on. He can leave it as a One Point Wonder talent and never touch the talent again, but should he boost it up to 5/5 he gets a Talent that nearly does double his weapon damage and hits all the enemies around him (final value is 190%) and in combination with prodigies like Windblade and Giant Leap or just from use with a simple Sweep he can obliterate a group of enemies.

Whirlwind is strictly very Vanilla, but it works extremely well at what it does and is strong on all points. Additionally it is in a category with 2 very powerful single target abilities - Dual Strike and Flurry - which is an impressive feat to pull off. It's a good example where very fine tuning of damage values can produce and impressive hotbar of buttons.

If Death Dance had a sort of damage scheme similar to these talents, we wouldn't need to think about any 1, 3, 5 planning. Simply put, having a good base damage for use at 1/5 Talent Investment and a good damage scaling mechanism per talent level would go well in improving the talent. For example

Take the old Death Dance formula of 140% - 210%. The 70% difference, or 50% increase in weapon damage multiplier was too low when combined with the 10 turn cooldown. Because you didn't really have lot of ability in a follow up AoE with Two Handed, these two factors led to Death Dance being a One Point Wonder. If however, the old Death Dance Formula included something simple like a small reduction in cooldowns (say you get a cooldown of 10 at 1/5, cooldown of 9 at 2/5, 8 at 3/5, 7 at 4/5, and 6 at 5/5) then it could heavily improve the Talent gradually to the point where a player might be somewhat interested in perhaps making greater use of Death Dance at full Talent levels.

The following example above is a good idea of how old Death Dance could use a very simple setup to improving its potential use. It could also be applied to the current revised Death Dance setup - maybe altering the Bleed Effect slightly (maybe have Bleed have 3 + 2*Talent Level/3 Duration so that you get a 6 Turn Cooldown/6 Turn Bleed effect. Ironically that also hits a 1, 3, 5 setup where the Bleed Effect starts off at 3 Turns at 1/5 Death Dance, goes up to 5 Turns at 3/5 Death Dance, and hits 6 Turns at 5/5 Death Dance; however I got to this point by simply using a very simple scaling mechanism with a focus on making the talent as useful as possible at all Talent Investments.

[Fortescue]

The following example above is a good idea of how old Death Dance could use a very simple setup to improving its potential use. It could also be applied to the current revised Death Dance setup - maybe altering the Bleed Effect slightly (maybe have Bleed have 3 + 2*Talent Level/3 Duration so that you get a 6 Turn Cooldown/6 Turn Bleed effect. Ironically that also hits a 1, 3, 5 setup where the Bleed Effect starts off at 3 Turns at 1/5 Death Dance, goes up to 5 Turns at 3/5 Death Dance, and hits 6 Turns at 5/5 Death Dance; however I got to this point by simply using a very simple scaling mechanism with a focus on making the talent as useful as possible at all Talent Investments.

It doesn't matter how you arrive at a 1-3-5 solution, my examples were just giving big, very obvious benefits that makes character planning easier without having a chart in a wiki showing you at which break points you get which durations. That is the other thing 1-3-5 is really good at, making it easier to communicate to players why they would want x number of points in a skill.

I know that other classes have access to 2H Weapon Assault, I'm still not sure how my proposed change above would negatively impact them. Up to now 2H was the least popular option on any class that could choose to dual wield instead afaik, so being more competitive with DW / S&B is fine. If S&B needs tweaks to stay competitive with DW and 2H so be it, give it tweaks. I agree that the DW attack category is really good, but I don't agree that getting to 5 in any of the skills in it is necessarily a great idea EXCEPTING that you wouldn't have anything else worth putting your points in. THAT is a problem in and of itself that would be remedied by every skill offering compelling reasons to get to 5 points. If they all have substantial benefits, where you spend your final points actually matters.

[Davion Fuxa]

It doesn't matter how you arrive at a 1-3-5 solution, my examples were just giving big, very obvious benefits that makes character planning easier without having a chart in a wiki showing you at which break points you get which durations. That is the other thing 1-3-5 is really good at, making it easier to communicate to players why they would want x number of points in a skill.

I know that other classes have access to 2H Weapon Assault, I'm still not sure how my proposed change above would negatively impact them. Up to now 2H was the least popular option on any class that could choose to dual wield instead afaik, so being more competitive with DW / S&B is fine. If S&B needs tweaks to stay competitive with DW and 2H so be it, give it tweaks. I agree that the DW attack category is really good, but I don't agree that getting to 5 in any of the skills in it is necessarily a great idea EXCEPTING that you wouldn't have anything else worth putting your points in. THAT is a problem in and of itself that would be remedied by every skill offering compelling reasons to get to 5 points. If they all have substantial benefits, where you spend your final points actually matters.

Thing is, many talents in the Technique/Dual Technique category ARE really good at 5/5. If you maximize Flurry at 5/5 it becomes one of the top tier single target attack abilities in game. Dual Strike is a consideration to max because I'm pretty sure all classes that have access Dual Strike have access to Cunning/Dirty Fighting and a player could continue a small Stun chain on a tough opponent to attempt to bring him down. Whirlwind is probably the best melee ranged AoE in play since it has the low cooldown at duration 8 and the highest upfront damage potential at that cooldown. Sweep is a weaker form of Whirlwind pretty much with the same cooldown of 8 but it's damage scales at the same rate as Whirlwind which means talent points invested here are worth more if a player wants additional focus on AoE ability; plus it gives Bleed which Marauder's and Rogue's have associated talents (Bloody Butcher and Scoundrel Strategies) and in the case of Cunning/Scoundrel it is accessible to anyone who picks up Tricks of the Trade.

Dual Techniques lacks the 1, 3, 5 formula yet it simply produces the best Talents in this game. It's a category that should give pause for thought on whether your 1, 3, 5 formula is needed - personally I don't think your suggesting a bad formula, but it is just one strategy as means to designing talents.

I'll note as well that I also could have suggested that Death Dance's Bleed duration simple increase at a 2+Talent Level Duration as well. I only suggested starting at 3 and ending up at 6 because the Bleed effect is important to the Berserker Class and a 3 Turn Bleed duration should be enough for minimalist talent point spenders (2 Turn Bleed Duration I kind of consider as being too low for synergy with Bloody Butcher).

[Sradac]

I say put in more work and do a 1-2-3-4-5 breakpoint system, to make every level of a talent feel interesting, instead of going from 1-2 or 3-4 and going "yay. I do 3% more damage. Goody."

[Fortescue]

I say put in more work and do a 1-2-3-4-5 breakpoint system, to make every level of a talent feel interesting, instead of going from 1-2 or 3-4 and going "yay. I do 3% more damage. Goody."

Would be great, but unfortunately that would double the amount of work required and this is already increasing the amount of work by quite a bit. Usually 2 points gets you a good return under the current system, 4 is a lot less great, but with a solid reward at 5 it will make up for that 4th point and the terrible scaling return on the 5th point.

[HousePet]

While adding breakpoints can solve this sort of issue, you seem to have missed the underlying issue.

Firstly, the logic of 'numbers tweaks are never going to work' is faulty. Look at Stunning Blow, its numbers only, yet people love it.

What you really need to consider, is that you actually get something decent for spending a talent point.
If a talent only has damage scaling, then the diminishing returns will hurt it.
Stunning Blow however, has damage scaling and duration scaling. So spending a point does more than just a small amount of damage. Its basically has double the scaling.
Another example is Fireflash: Its damage and radius scale. If you consider the potential total damage, its damage x pi x radius squared. That makes the radius scaling count twice, so its three times scaling.

So yes, the diminishing returns numbers can be uneconomical is some cases.
Yes, breakpoints can work for making improving the reward for spending a point.
But it isn't the only solution.

[Mankeli]

Thing is, many talents in the Technique/Dual Technique category ARE really good at 5/5. If you maximize Flurry at 5/5 it becomes one of the top tier single target attack abilities in game.

Eh, flurry is probably one of the best examples of a talent that is already extremely great at just one point since you get the best thing at level 1: more hits on enemy which means more procs (and more damage, naturally).

It doesn't matter how you arrive at a 1-3-5 solution, my examples were just giving big, very obvious benefits that makes character planning easier without having a chart in a wiki showing you at which break points you get which durations. That is the other thing 1-3-5 is really good at, making it easier to communicate to players why they would want x number of points in a skill.

If skill levels 1,3,5 were all that mattered in this new system, why would we still have levels 2 and 4? If DarkGod was to go with your suggestion, let's make it 1-3 point system then and balance things (skill points etc) around that. I don't like having useless skill levels as filler.

I agree with you completely that communicating the effects of a given talent at given skill levels should be done more elegantly in TOME. I remember it being really hard to assess the usefulness of new talent trees especially, because I had no way of knowing how those talents would scale with levels (except the part that they have diminishing returns of some sort).

That's why I've proposed a system where you could add points to any skill you possess at any time without actually being able to use that skill: The player would have an unlimited amount of "virtual talent points" so he/she could decide if that new talent tree is worth unlocking for that last talent in said tree, for example.

So yes, the diminishing returns numbers can be uneconomical is some cases.
Yes, breakpoints can work for making improving the reward for spending a point.
But it isn't the only solution.

Yeah, 1-3-5, or whatever, doesn't have to, and most likely shouldn't, be an universal solution. Adding breakpoints to skill levels that give diminishing returns on investment is basically about countering the effects of those diminishing returns. While I think this works on some cases, I think balancing the rate of diminishing returns themselves can work in others.