15 replies to this topic

[XtremWarrior]

This idea has probably already poped-up somewhere in here, but here is my take on it.

One of the most controversial aspect of current skill tree iteration is the fact that you have to take un-wanted/un-needed Nodes to obtain the most usefull ones.
The obvious goal of this is to prevent players to specialise their Mech to the limit without any tradeoff.

So what if the Trees happen to be a bit more linear, without any placeholder Nodes between powerfull Ones, but also have an increased cost in SP for the deeper Nodes of the chain?
Simple example: all Weapons Cooldown Nodes are in a direct line, but the Skill Points cost increase when reaching the 4th one. It then becomes 2 SP to unlock every next Nodes in the chain.

Good or bad idea? Share your thoughts. Thanks!

Edited by XtremWarrior, 05 March 2017 - 02:13 AM.

[IanDresarie]

I like the idea, but would prefer increased CBill/XP costs as the SP are already rare as they are, even if you would safe some points on the useless stuff like arm quirks.

[Alistair Winter]

I like linear skill trees and I hate being forced to take pointless skills that I don't want. They also clutter up my list of quirks with tiny bonuses that don't matter.

But I don't like increasing cost of more valuable skills, or "diminishing returns". As I have said in other threads, you have to ask yourself what the point would be. Is it simply to copy other games? What purpose would it serve in MWO? Because the consequences are obvious. If a skill tree has diminishing returns, then it's likely that the majority of players would grab the "low hanging fruit", because those skills are cheaper and you get more of them. A minority of players will go all the way to the edge of the tree, sacrificing versatility for some special skill, like enhanced ECM or whatever.

Furthermore, I don't really see a big issue with having a linear skill tree, reduce the number of skill points, reduce the number of nodes, and no random node locations at all.

[XtremWarrior]

Alistair Winter, on 05 March 2017 - 02:31 AM, said:

If a skill tree has diminishing returns, then it's likely that the majority of players would grab the "low hanging fruit", because those skills are cheaper and you get more of them. A minority of players will go all the way to the edge of the tree, sacrificing versatility for some special skill, like enhanced ECM or whatever.

That's the point, tbh. Wouldn't that be a good thing? I mean if players are grabing many low skills, that could increase diversity in Mechs Builds, wouldn't it?

[Alistair Winter]

XtremWarrior, on 05 March 2017 - 07:17 AM, said:

That's the point, tbh. Wouldn't that be a good thing? I mean if players are grabing many low skills, that could increase diversity in Mechs Builds, wouldn't it?

It depends. Let's say there are 3 skill branches and the system rewards specialization. In that case, you only end up with 3 different types of specs. On the other hand, if the system rewards versatility, then you may see people combining either 2 or 3 skill trees. In that case you will suddenly see 7 different types of specs. In that sense, a system that rewards versatility will increase diversity.

However, if specialization becomes far too expensive, then no one will specialize. Maybe everyone either picks 2 or 3 different branches. Or even worse, maybe everyone picks 3 branches, because versatiltiy is so much better. In that case, you will only see 1 type of spec in the game, and you've killed diversity.

[Pyed]

XtremWarrior, on 05 March 2017 - 07:17 AM, said:

That's the point, tbh. Wouldn't that be a good thing? I mean if players are grabing many low skills, that could increase diversity in Mechs Builds, wouldn't it?

Not necessarily. To give a simplistic example, let's just talk about 6 nodes and 3 points to spend total:

Weapon Range
+5%, 1 point
+5% (10% total), 2 points

Weapon Cooldown
-5%, 1 point
-5% (-10% total), 2 points

Weapon Heat
-5%, 1 point
-10% (-10% total), 2 points


I can spend my 3 points to get:
+10% Weapon Range
or
-10% Weapon Cooldown
or
-10% Weapon Heat
or
+5% Weapon Range, -5% Weapon Cooldown, -5% Weapon Heat

--

I think most people would take the 4th option, and there's no diversity now.

[Gentleman Reaper]

Alistair Winter, on 05 March 2017 - 07:27 AM, said:

It depends. Let's say there are 3 skill branches and the system rewards specialization. In that case, you only end up with 3 different types of specs. On the other hand, if the system rewards versatility, then you may see people combining either 2 or 3 skill trees. In that case you will suddenly see 7 different types of specs. In that sense, a system that rewards versatility will increase diversity.

However, if specialization becomes far too expensive, then no one will specialize. Maybe everyone either picks 2 or 3 different branches. Or even worse, maybe everyone picks 3 branches, because versatiltiy is so much better. In that case, you will only see 1 type of spec in the game, and you've killed diversity.

We're talking about mech build diversity, not skill tree diversity. PPC/gauss builds can focus purely on velocity and heat-gen in the firepower tree, then go for armor and structure, torso twist, and dissipation skills. A mixed build would want to get different nodes just to get the best out of all their weapons, and would have less points to spend elsewhere.

What we have now is the best system for benefiting all build-types, the choice it has isn't about picking exactly what skill you want, but how far you want to spec into one tree over the others. There's nothing wrong with this, so it pisses me off immensely when people wanting linear trees call the current trees a bad system. Don't get me wrong, the trees need to be adjusted so that build-specific nodes are off to the side (spread, arm pitch), but having less optimal skills as filler prevents the trees from helping boats more than the module system already does.

Edited by Gentleman Reaper, 05 March 2017 - 09:43 AM.

[Skribs]

A general principle in gaming is that hybrids should be overall stronger, but weaker at each individual skill. For example, in an MMO, if a hybrid can do 75% of the healing or damage as a dedicated healer or damage dealer, then they're effective at performing both roles.

That's the principle here. On something like a CN9-A with a zombie hugger build, a Crab that's just got a bunch of CT/Head weapons, or a Banshee you want to run as a tank, you might want to take 15% survivability over 10% survivability and 10% damage. On something like an LRM boat that you don't expect to be in a direct confrontation, 15% firepower seems better than 10% firepower and 10% survivability. On a light, maybe you want to go max mobility and forego other things. On a medium, maybe mobility and firepower are important, but survivability isn't.

Going more granular, something like a 3xAC5 Jagermech would probably rather get CD boosts than heat boosts, because it's already running cold. A 6xLPL Stalker would probably go the other way, because heat is the limiting factor in dealing damage. An AC20 Hunchback may prefer ballistic velocity, while an AC10 hunchback may prefer cooldown. A SRM6 mech may prefer missile spread, but a SRM2 boat may prefer cooldown, heat, or range.

So there are choices here, and it gives hybrid builds the "best" mathematical numbers, but players who specialize the ability to get more effectiveness out of their given role.

[XtremWarrior]

Pyed, on 05 March 2017 - 08:23 AM, said:

Not necessarily. To give a simplistic example, let's just talk about 6 nodes and 3 points to spend total:

Weapon Range
+5%, 1 point
+5% (10% total), 2 points

Weapon Cooldown
-5%, 1 point
-5% (-10% total), 2 points

Weapon Heat
-5%, 1 point
-10% (-10% total), 2 points


I can spend my 3 points to get:
+10% Weapon Range
or
-10% Weapon Cooldown
or
-10% Weapon Heat
or
+5% Weapon Range, -5% Weapon Cooldown, -5% Weapon Heat

--

I think most people would take the 4th option, and there's no diversity now.

Well, i see your point but even in your example, i really think that ppl would split into the different choices, as -10% Cooldown or Heat only is better than +5 in both, simply because min/maxing is always the best (read: most competitive) way to go.

[MischiefSC]

XtremWarrior, on 05 March 2017 - 07:17 AM, said:

That's the point, tbh. Wouldn't that be a good thing? I mean if players are grabing many low skills, that could increase diversity in Mechs Builds, wouldn't it?

Nope. A common misconception though.

Having many low skills doesn't change PPFLD > DOT, hitscan > low velocity and 60 pts to 1 location > 80 pts spread over several. All the things that create boating and meta are completely unrelated to the skill tree. Nothing in the skill tree will seriously impact weapon balancing unless it's in the neighborhood of buffing under-performing stuff by 30-50% and leaving top performing stuff alone. As the 'lots of low stuff' approach just gives minimal bonuses over a wide range of things, both good and bad weapons and systems, it has no impact on what does and does not work.

It just makes the skill tree largely pointless. So everyone has 3-5% bonuses to all weapon stats. Why not just buff all weapon stats by 4% and eliminate the skill tree?

A linear tree with increasing bonuses however gives you the tools to make otherwise mediocre builds viable when combined with quirks. So suppose 'spread' was 6 nodes going +1, +1, +1, +2, +2, +3. So 6 pts gets you 10% spread reduction. HOWEVER, on a sub-par mech you might have a mech-specific quirk of '+1 free node on Spread' or even better '2x Spread reduction'. So it could get a 20% reduction on the spread of missiles or LBX.

A 10% spread reduction is worthless to a laserboat mech and a GHR for example should have no real quirks around weapons; he could spend his points to get 10% heat/cooldown/range/whatever bonuses to his lasers (about what the GHR has right now with quirks) and the player with the GHR will feel they're getting value out of the points they spend on the skill tree.

However the guy in the 2xLBX10, 4xSRM4 Mauler with 20% spread reduction/velocity *plus* cooldown, range, heatgen and structure/armor quirks? He's now in a range to play with/against the GHR.

However if that Mauler just has the same 10% quirks the GHR does or even worse 5% quirks then he's ****. The GHR will destroy him, every time with equal skilled pilots. Because hitscan > varied velocity weapons, concentrated damage > spread.

Equal quirks or low level quirks are irrelevant. They do nothing to close the gap between bad weapons/builds and good weapons/builds. Forcing everyone to buy useless stuff under the idea that the guy with the terribad bracket build is actually getting a little bit (few %) more value out of the points both are forced to spend to get the worthwhile quirks doesn't change that at all.

[Skribs]

I think the idea is variety in the skills to tailor your Mech's builds to a desired role, and not in making a Mech better. For example, right now there is basically a cookie cutter because if 1 point in defensives gives you 3% survivability and 1 point in firepower gives you 1% firepower, it's obvious that defense is the best way to go.

However, if it was 1 point in either gives you 3%, 2 or 3 points gets you an additional 2% each, and 4 or 5 points an additional 1% each, then you can choose which is more important.

I will agree with you, though, that the skill tree itself will largely be irrelevant next to MechLab build and pilot skill. As such, the costs associated with it and the time invested seem like terrible gameplay mechanics.

[MischiefSC]

Skribs, on 05 March 2017 - 03:29 PM, said:

I think the idea is variety in the skills to tailor your Mech's builds to a desired role, and not in making a Mech better. For example, right now there is basically a cookie cutter because if 1 point in defensives gives you 3% survivability and 1 point in firepower gives you 1% firepower, it's obvious that defense is the best way to go.

However, if it was 1 point in either gives you 3%, 2 or 3 points gets you an additional 2% each, and 4 or 5 points an additional 1% each, then you can choose which is more important.

I will agree with you, though, that the skill tree itself will largely be irrelevant next to MechLab build and pilot skill. As such, the costs associated with it and the time invested seem like terrible gameplay mechanics.

Except you want 3 pts invested to be worth more than 2 pts by a growing amount to promote focusing a mech in 1 direction, otherwise you'll get more value out of a moderate bump in 10 things instead of a above moderate bump in 5 things.

[JC Daxion]

What is the difference if you take a few skills you don't need, or have to pay double? this makes no sense.

Typically the only nodes that add zero help are AMS with no AMS, a weapon slot with out that weapon, and Arm nodes on mechs that have all torso based weapons.

Pretty much everthing else is a buff, even if it is a tiny one, of thought of as not needed.. (fall damage nodes are under this, they help if you want um or not.)


IMO the only thing need changing are to move some nodes up to very early in the trees, so a decent boost can happen for your first 20 points spent.

Kinetic burst, torso twist, anchor turn, cool run, and heat contain.. You know all the skills that start in basics

[MischiefSC]

JC Daxion, on 05 March 2017 - 03:57 PM, said:

What is the difference if you take a few skills you don't need, or have to pay double? this makes no sense.

Typically the only nodes that add zero help are AMS with no AMS, a weapon slot with out that weapon, and Arm nodes on mechs that have all torso based weapons.

Pretty much everthing else is a buff, even if it is a tiny one, of thought of as not needed.. (fall damage nodes are under this, they help if you want um or not.)


IMO the only thing need changing are to move some nodes up to very early in the trees, so a decent boost can happen for your first 20 points spent.

Kinetic burst, torso twist, anchor turn, cool run, and heat contain.. You know all the skills that start in basics

Except they help eliminate the cost/reward factor and create the illusion of value. Points spent on weapons you're not going to use are just wasted points. They serve no purpose. You may as well buy AMS buffs with no AMS. Yet they create the illusion of value to justify why when you have to spend 3 points to get a 1% improvement in cooldown on your lasers it's 'but you got a buff your ballistics!'. Which is irrelevant. It also means that mechs with one hardpoint type (like the Catapult or Black Knight) have to pay more to get their weapon bonuses than mechs who are at least getting some tiny benefit.

It forces people into a mediocre mish-mash of perks, which eliminates the point of a 'trade-off' system of skilling up.

The only decisions to be made in this skill tree is 'buy the narrow selection of worthwhile stuff which requires you to buy stupid/pointless/irrelevant stuff' or 'waste your points on on things that are not as good a deal for the points spent and be at a disadvantage to those who picked the good stuff'.

There's no tradeoff. JJ perks are never going to be as useful as armor/structure. With all the nerfs the mobility quirks are a 100% requirement. Slight variety in weapon/sensor quirks based on mech but because the values are so low it's literally less meaningful to your mechs performance than modules currently are.

[XtremWarrior]

JC Daxion, on 05 March 2017 - 03:57 PM, said:

What is the difference if you take a few skills you don't need, or have to pay double? this makes no sense.

Typically the only nodes that add zero help are AMS with no AMS, a weapon slot with out that weapon, and Arm nodes on mechs that have all torso based weapons.

Pretty much everthing else is a buff, even if it is a tiny one, of thought of as not needed.. (fall damage nodes are under this, they help if you want um or not.)


IMO the only thing need changing are to move some nodes up to very early in the trees, so a decent boost can happen for your first 20 points spent.

Kinetic burst, torso twist, anchor turn, cool run, and heat contain.. You know all the skills that start in basics

Take any Energy HP only Mechs like Black-Knights. You want to get all the Heatgeneration and Cooldown Nodes, then be prepared to have increased velocity on your Lasers, and + ammo for your Lasers...
Those are not Nodes i don't want, those are Nodes i can't use.
Paying more is actually the exact same thing as buying useless Nodes, but buying utterly useless things seems like a really bad game mechanic.