14 replies to this topic

[RudWolf]

So after having look at new update to Skill tree in latest PST, I have to say that I am quite happy with changes but there is still way to go.

Main issue of current tree system is apparent from the Weapon branch. Previous installment encouraged weapon boating, which was somewhat fixed with merge. Now the unified skill tree benefits all weapons in low SP investments and encourage specialization with increased SPs unlocked - so far so good.

Problem:

I do understand that the position of nodes is arranged in the way that getting to juicy stuff requires spending SP on less appealing improvements and I am OK with that. However this arrangement encourages investing in the weapon upgrades which I am not currently sporting. For example I was trying to skill up my Marauder to be more effective with PPCs, but I came to the point where I would have to take Laser nodes to be able to get to other beneficial nodes (esp. heat is crucial). But I don't want to use lasers!

It would be possible to indefinitely rearrange the weapon tree, until it will encompass all the possibilities, but thats is going to be long and painful proces. And what are going to do when a major overhaul will be needed? Refund all nodes every time?

Second thing - because of merging, the three itself become tremendously huge leading to lesser readability.

Solution:

I know it is radical and not a new proposition, but I think it would be better to choose different system and instead of a tree create system of tiers. Each tier will require certain investment of SP before allowing advancement to next one. By clever distribution of upgrades among the tiers it would be possible to create similar system to current one, but without the restriction to take unwanted upgrades.

The idea to prevent boating is to fill lower tiers with mostly generic upgrades and start adding weapon-specific improvements progressively with tier increase. Parallel system like this will yield same results like current implementation but without the shortcomings of tree linearity.

Bonus: this system can allow invest in one upgrade several times and thus get rid of filler items (like 10 levels of same upgrade), however in this case, the number of points per tier would need to be limited, to prevent over maxing one atribute.

Example:

So with tier system, considering 1SP/Upgrade and 5SP needed for tier advacement the weapon upgrade branch could look something like this:

Tier 1:
Cooldwn 1, Range 1, Velocity 1
Cooldwn 2, Range 2, Velocity 2

Tier 2:
Cooldwn 3, Range 3, Velocity 3, Heat 1, L-Duration 1, M-Spread 1, LBX-Spread 1, G-Charge 1
Cooldwn 4, Range 4, Velocity 4, Heat 2, L-Duration 2, M-Spread 2, LBX-Spread 2, G-Charge 2

Tier 3:
Cooldwn 5, Range 5, Velocity 5, Heat 3, L-Duration 3, M-Spread 3, LBX-Spread 3, G-Charge 3
Cooldwn 6, Range 6, Velocity 6, Heat 4, L-Duration 4, M-Spread 4, LBX-Spread 4, G-Charge 4

Tier 4:
Velocity 7, Heat 5, L-Duration 5, M-Spread 5, M-Rack 1, LBX-Spread 5, UAC-Jam 1, G-Charge 5
Velocity 8, Heat 6, L-Duration 6, M-Spread 6, M-Rack 2, LBX-Spread 6, UAC-Jam 2, G-Charge 6

etc.

Probably would make sense to make the SP tier unlock requirement progressive to tier levels, buts thats more up to the fine tuning of the system. But the main point is - this will prevent investment to unwanted nodes while still keep need to invest to generic nodes first. I am sure that similar progression could be think up for each mech skill-branch

Other improvement for consideration:

Also, as I have stated during PTS1, I still think that reasoning behind mech specialization would be more straightforward. This could be supported first by lowering overall SP number and by correctly adjusting number of maximum SP vs. available upgrades. This ration should be about 1/3, so there would be more initiative for more conscious choices in a way: do I want to improve offensive, defensive, mobility, support or be jack-of-all trades?

Lowering the actual amount of maximum SP and number of choices would make whole skilling proces much more clearer and synoptic.

Edited by RudWolf, 08 March 2017 - 09:15 AM.

[MechaBattler]

It's designed to favor multi weapon builds. Let's face it, boating one weapon has been dominant for the longest time, and will continue to be so. But at least this skill system attempts to encourage multi-weapon builds by squirreling away extra cooldown and other skills behind the specialty skills.

I too initially tried to come up with a more linear tree. But the more I used it, the more I liked how it benefited my non-boating builds.

[R Valentine]

MechaBattler, on 08 March 2017 - 09:22 AM, said:

It's designed to favor multi weapon builds. Let's face it, boating one weapon has been dominant for the longest time, and will continue to be so. But at least this skill system attempts to encourage multi-weapon builds by squirreling away extra cooldown and other skills behind the specialty skills.

I too initially tried to come up with a more linear tree. But the more I used it, the more I liked how it benefited my non-boating builds.

It's been said over and over again, the current design does not encourage multi weapon builds at all. Spreading out things and gating them behind useless things actually reinforces the boating meta. Just because I had to take 2 points in lasers that I can't even fit(because I have no laser hardpoints) doesn't mean I'm suddenly going to fit more than just gauss/ppc. It just means I had step over a few speed bumps. With all the nodes spread so far out, you can't diversify effectively. It's still far more effective to concentrate on one weapon type while grabbing all necessary general nodes. Nothing has changed.

[Prof RJ Gumby]

If you need to take nodes that are NOT for your weapons to get a universal upgrade you want, this means that this universal upgrade was not intended to be taken by a mech with your weapons. It's that simple. Yes, you can DECIDE that it is worth it and spend 2,3 or 4 nodes just to get that one upgrade, but it's YOUR CHOICE. This is how you pay for upgrading your weapons beyond the intended range.

People should stop thinking they NEED to take ALL the weapon nodes that benefit their weapons in a way. No you don't. You just can, if you're willing to bear the costs.

If you think there's not enough heat reduction skill for PPCs, just say that there's not enough heat reduction skill for PPCs.

EDIT:

Kiran Yagami, on 08 March 2017 - 09:44 AM, said:

It's been said over and over again, the current design does not encourage multi weapon builds at all (...)

This system was proposed by the players to make the new skills LESS beneficial to boats than the module system was. And it is. Partly because you may want (WANT, not NEED) to waste multiple points for additional laser bonus, while somebody with PPCs/Gauss and lasers can manouver his node path so that he won't even need to have any "empty" nodes to get more bonuses.

Previous system penalised non-boating to a degree with weapon modules, this system penalises boating to a small degree with node placement. This is not made to make boating bad. It's all made to even things a bit. That's all. And it's goddamn beneficial to balance, even if only a bit, so stop whining people about how your feelings are hurt with that nodes you feel you need to take even if they're not put there for your builds.

Edited by Prof RJ Gumby, 08 March 2017 - 09:54 AM.

[MechaBattler]

Kiran Yagami, on 08 March 2017 - 09:44 AM, said:

It's been said over and over again, the current design does not encourage multi weapon builds at all. Spreading out things and gating them behind useless things actually reinforces the boating meta. Just because I had to take 2 points in lasers that I can't even fit(because I have no laser hardpoints) doesn't mean I'm suddenly going to fit more than just gauss/ppc. It just means I had step over a few speed bumps. With all the nodes spread so far out, you can't diversify effectively. It's still far more effective to concentrate on one weapon type while grabbing all necessary general nodes. Nothing has changed.

Not being able to grab every general skill without spending on skills you don't need isn't a change? It's a targeted nerf against boating. Although at the percentages we're working at, it's not a huge nerf.

[RudWolf]

All the arguments either supporting current skill tree or bashing it are valid. However the benefit of proposed Tier system is that it is powerful tool how to find good compromise between both - It is scalable and resistant to massive re-branching if PGI will find out a flaw in a setup. What will happen if one of the path will prove to be more imba than other and whole branch would need redesign, would there be a complete refund on all accounts and mechs again?

I believe that most of current/potential issues can be adjusted by clever positioning and spacing of upgrades among the tiers. Its is also possible to adjust costs per Tier or play with unlock/maximum SP available per tier. The issue of hiding some generic upgrades behind specific ones could be adjusted by dividing those generic upgrades in more levels, making them as expensive as being obstructed by "unneeded" upgrade. Or they can have diminishing returns in higher tiers, making them less obvious choice.

There are plenty ways how system like this could be further balanced. Tree system is too rigid and quality of nodes is too dependent on how they connect with others.

EDIT:

Also lot of problems of current system are also coming simply from inflation of available SP. In my example spending 20 points on maxing the potential of my PPCs (speed and heat, cooldown is questionable due high heat) would be significant choice if I could spend only 50SPs total, but becomes obvious choice which needs to be hampered if I have 91 to spend.

Edited by RudWolf, 08 March 2017 - 10:55 AM.

[R Valentine]

MechaBattler, on 08 March 2017 - 09:58 AM, said:

Not being able to grab every general skill without spending on skills you don't need isn't a change? It's a targeted nerf against boating. Although at the percentages we're working at, it's not a huge nerf.

No, it isn't a change at all, because I still end up with all the necessary general nodes and all the weapon specific nodes I need. I have a few nodes here and there that do nothing, but that doesn't change my build at all. The exact same thing could be accomplished if they took all the useless crap out of the way from the general nodes and just reduced the total number of skill points I get. In fact, the two are one and the same. That's all useless nodes are. A deduction from the total number of skill points. It does not benefit me to fit multiple weapon types. Not even close. I'd have to reach out and stretch even further, picking up more less than desirable nodes to attempt that and I'd just end up with a build that is sub-optimal. The convoluted tree works against multi-weapon builds more than it works against specialized builds, so specialized still come out on top.

[MechaBattler]

Well I guess you'll just have to keep plugging in useless nodes then. I would just put the points elsewhere.

[R Valentine]

MechaBattler, on 08 March 2017 - 11:33 AM, said:

Well I guess you'll just have to keep plugging in useless nodes then. I would just put the points elsewhere.

Why? It's sub-optimal to do so. That's the whole point. The new system does not discourage boating at all. It is more taxing on mixed builds than specialized builds. Our boating meta hasn't gone anywhere. Mission not accomplished. This is what you keep missing. Boating is not going to go away with this tree. If anything, it'll get worse.

[RudWolf]

Well boating will not get away with any form of skill tree by any means. Boating is beneficial in other ways than just spatial +bonus - convergence, cooldowns and range are important factors adding to the benefit of boating. It would probably have to be mangled by other mechanism, probably in a way of improved ED - one giving penalization for firing same weapon types (recoil for ballistic, overcharge for energy, overheat for missile).

Edited by RudWolf, 08 March 2017 - 11:47 AM.

[MechaBattler]

You would rather spend 2 points on say duration to get 0.8% more cooldown? 3 points for that isn't even remotely worth it.

Better off putting those points to mobility, mech operation, or even improving your coolshot if you're running lasers.

[RudWolf]

With so many points to spend atm I would, but that just adds to obfuscation of whole system, inflates the skill list in mech details making it harder to read -> bad design

My aim is to simplify things and find cleaner and elegant solution, but still keep the initial intention (although I don't like it)

[R Valentine]

MechaBattler, on 08 March 2017 - 11:47 AM, said:

You would rather spend 2 points on say duration to get 0.8% more cooldown? 3 points for that isn't even remotely worth it.

Better off putting those points to mobility, mech operation, or even improving your coolshot if you're running lasers.

The mech operation tree is just as convoluted, so I'd just be traded one set of wasted points for another. That's the thing, saving 2 useless points here just ends up being 2 useless points elsewhere. EVERYTHING is gated behind stupid sh*t. That's why you can't really mix your loadouts. You end up with more wasted points, not less. At least specializing hits the fewest dumb nodes possible. It's still the optimal form of play.

[MechaBattler]

There's useless and less useful. Duration skills when you have no lasers, that's useless. Hill climb, it can still be of use on certain maps. Missile spread with no missiles, that is useless. Quick Ignition can be useful when you're running a hot laser build and someone brings a flamer to ruin your day. LBX spread when you're running regular ACs, useless. Having 2 coolshoot when you're on the verge over heating and you need that last alpha to finish your enemy? Priceless. (80,000 C-bills actually.)

If .8% cooldown is more useful to you and worth 2 points in something truly useless. Then more power to you. I won't stop you.

[Prof RJ Gumby]

Kiran Yagami, on 08 March 2017 - 11:54 AM, said:

The mech operation tree is just as convoluted, so I'd just be traded one set of wasted points for another. That's the thing, saving 2 useless points here just ends up being 2 useless points elsewhere. EVERYTHING is gated behind stupid sh*t. That's why you can't really mix your loadouts. You end up with more wasted points, not less. At least specializing hits the fewest dumb nodes possible. It's still the optimal form of play.

... in your opinion. If you claim that wasting nodes to go optimal, feel free to do so. What's the diferrence if it's optimal.

One thing I can agree with is that the trees are a bit too bottom heavy, especially the mobility one. Speed tweaks (all of them!) are pushed as far down and to the sides as possible, actually making it a "go full or go home" setup. This and a few other arrangements like that punish people who actually want to tweak their mechs carefully, balancing between the trees. These will most often end up with very bad skills.

It's not an unfixable flaw, but quite a painful one. All the tree needs is to have some of the most popular (often for a reason) nodes to get relocated to upper parts of the tree. Like 1 or 2 speed tweaks early in mobility, 1 or 2 seismic and radar derp early in the sensors, etc. This would make investing heavily in some trees more optional, not an actual necessity. Balancing usage of nodes between trees without always going heavy in 2-3 best ones should be a viable option. It isn't now, sadly

Edited by Prof RJ Gumby, 08 March 2017 - 12:57 PM.