Jump to content

"skill-Salad" Format Of Tree Provides Little Choice, Differentiation - 3 Options For Improvement Via Trade-Offs


2 replies to this topic

Poll: Skill Tree - Meaningful Choices (8 member(s) have cast votes)

Which option do you prefer to make the skill tree choices meaningful?

  1. I like the current Skill Tree (3 votes [37.50%])

    Percentage of vote: 37.50%

  2. Option 1 - Skill Groups with Escalating Costs (3 votes [37.50%])

    Percentage of vote: 37.50%

  3. Option 2 - Skills with Direct Trade-Offs (2 votes [25.00%])

    Percentage of vote: 25.00%

  4. Option 3 - Selectable Trade-Offs (0 votes [0.00%])

    Percentage of vote: 0.00%

  5. None of These (Other) (0 votes [0.00%])

    Percentage of vote: 0.00%

Vote Guests cannot vote

#1 Leroy the Walrus

    Rookie

  • Ace Of Spades
  • 6 posts

Posted 04 March 2017 - 12:43 PM

TL;DR: The current skill tree is big and complex, but is still very "generic" and bland, and doesn't provide meaningful choices. The skill-tree should allow you to pick skills in a way that enhances your performance in some roles, but high levels of performance in one area should mean correspondingly reduced performance in others.

==========================================

One of the central promises of the new skill system was supposed to be diversification, allowing customization and more choice in roles/specialization.

It does not succeed at this.

One of the biggest reasons is the giant, interleaved, "skill-salad" structure of the trees. There are repeating segments of incremental, generic buffs.

Take the "Lasers" sub-section of the "Firepower" tree, for example. If I wanted to improve my heat-generation and cool-down rates (for a laser-brawler build), I also end up getting the range and duration skills along the way. If I wanted the range skills (for a laser-skirmisher/sniper), I also end up getting the heat-gen and cool-down skills. Two different builds, but they both just end up devoting points into a generic set of "laser better" skills. There's no real "choice" or differentiation between the two.

The same can be said for most other segments of the tree, and the the skill tree as a whole. The end result is a 91-ingredient, ~6-Million C-Bill recipe for... vanilla cupcakes. Sure, you can change what weapons you put on it, and you get 70% of the same recipe, just now it's a vanilla cupcake with chocolate frosting. The practical number of actual skill-tree options is still pretty low.

Proposed Solutions -
The goal here is to provide meaningful, impactful choices. For something to be a real choice, not just an obligate upgrade investment, there need to be pros/cons to the choice. The choice should involve trade-offs that provide an advantage in some situations, but a weakness in others. I've outlined three possible approaches below.

Option 1 - Sorted Skill "Groups" with Escalating Costs within Group

The idea here is to let people pick which aspects of skill group they want, but have skill point costs increase as overall investment in the group increases, requiring more sacrifices in other groups if you intend to max all aspects of one skill group.

Example - "Lasers" Skill Group
4 sub-categories in the group, say 5 levels each. The first 5 points allocated anywhere in the group cost 1 point each, the next 5 in the group cost 2 points each, then 3, etc.

Range: OOOOO
Duration: OOOOO
Cool-Down: OOOOO
Heat-Gen: OOOOO

So, for a very efficient cost of 5 skill points, I could max-out my range for sniping/skirmishing. Or I could max my cool-down for quick-hit brawling. I could also use the same 5 points to get 2 levels of heat-gen and 3 levels of cool-down, for a little more sustained brawling mix. But, if I wanted to max 2 skill sub-groups, it would cost 5 for the first 5, plus 10 for the second 5, for a total of 15 points invested to get 10 skill levels. You could use that to get range and heat-gen, for sustained skirmishing. Or range and duration, for more focused pokes. Or a more generalist mix. But, if I wanted to go all-in, and get all laser skills, the costs would keep going up. The idea is I can take the aspects I want to enhance my build under certain circumstances, but if I try to maximize laser functions under all circumstances, I start making very significant trade-offs in the points leftover for other trees, such as mobility or survivability.

Option 2 - Direct Trade-Offs in Each Skill
The idea here is that every skill has an inherent bonus, as well as an inherent cost. Each skill becomes much more impactful, because it enhances performance in some aspects, but directly inhibits it in others.

Example - More Speed, Less Heat Efficiency
Investing points increases your maximum speed, but reduces your mech's heat-dissipation. Increased speed gives you flexibility to determine when/where to engage, and gives you an edge during the engagement, but lower heat efficiency means you suffer in engagement duration and frequency.

Example - Faster Weapon Cool-Down, Shorter Range
More points increase your weapon cool-down and DPS, but reductions in weapon range mean you have to get closer or be more exposed to fully utilize those benefits.

While this setup has the potential for very interesting choices, it also presents a balance/implementation challenge for the dev team. Specifically, pairing the right bonus with the right drawback could be difficult, or providing the right selection of skills so you can select the bonuses you want for your playstyle without having an overwhelming number of bonus/trade-off options.

Option 3 - Selectable Trade-Offs
A bit of a hybrid between Option 1 and Option 2. Like Option 1, you have skill "Groups" where you can select from sub-groups of performance. But, to unlock a bonus to one sub-group, you have to select a trade-off drawback from another subgroup to offset it. So, if I want to increase my laser cool-downs, I'd have to selectively take a loss in either heat-gen, range, or duration. In this way, mech customization would be more in the form of deviations from a "nominal" performance, rather than straight upgrades. One plus to this system, aside from putting role-related choices in the hands of the player, is that it cuts-back on the "newbie penalty" - i.e. a mech without skill points allocated simply becomes a generalist by default, with less specializations, but isn't inherently inferior to a leveled-up mech. Seeing as PGI seems hell-bent on the grind though, I don't expect they'd ever go for this.

#2 Pz_DC

    Member

  • PipPipPipPipPipPipPipPip
  • Private
  • 1,108 posts

Posted 04 March 2017 - 12:53 PM

You've read my mind man. Was aiming to post same thoughts but You've don't it for me :P

#3 kazlaton

    Member

  • PipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 170 posts
  • LocationNew Jersey

Posted 04 March 2017 - 01:51 PM

Option 1. I think option 2 would be to hard for PGI to get right.





1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users