Fractal Softworks Forum

Please login or register.

Login with username, password and session length
Advanced search  

News:

Starsector 0.97a is out! (02/02/24); New blog post: Simulator Enhancements (03/13/24)

Pages: [1] 2

Author Topic: Let's talk progression and skill tree  (Read 7113 times)

Cosmitz

  • Admiral
  • *****
  • Posts: 758
    • View Profile
Let's talk progression and skill tree
« on: April 30, 2017, 01:16:57 PM »

Starsector has always been a bit of 'get the bestest dude and wreck face', but as the skill trees are starting to get filled out and with level limitations in-place, we're starting to hit the real point where people have to consider builds and really crunch the skill tree and figure out a level progression. Now, i don't want to delve in the actual individual skills, since these will probably be in-shift for a long time, especially with new skills relating to future outpost management/construction take out slots on the ballot, but I want to talk about the overall arhitecture of the tree and methodology.

Before we go in, this post is part of a series. I also delve into UI/UX on 8.0 here: http://fractalsoftworks.com/forum/index.php?topic=12078.msg205547#msg205547 and on the campaign progression here: http://fractalsoftworks.com/forum/index.php?topic=12256.msg208065#msg208065

The current system, as felt on my own skin and read about in the forums/discord, has some problems:

- Putting points in aptitudes feels 'wasted', compared to the old system were we got 'something' for them, and almost feel like a 'barrier to entry' a new tree
- There is a huge barrier to entry leading to decision paralysis for newer players when judging all the skills at a time for an upgrade
- Starsector at its core is and is meant to be a combat game in which you fly a single ship within a larger fleet combat situation, as such players feel like they have to juggle all the skills against the combat benefits of the Combat tree (or Carrier skills), since it's never much fun to get your ass whooped.
- Some skills are considered necessary and nobrainer choices / "of course i'll get Navigation"

But it's not all doom and gloom. Let's see what works well:

- Skills are chunky and provide a healthy benefit "+50% damage to weapons and engines".
- Ranks build on that and give identity and sense of strenght/power, within an easily graspable time-frame
- Absolutely no grind, the observe-plan-achieve loop is short and sweet

So, what's to do here?

Let's tackle the simplest issue, choice paralysis and aptitude points.

-----------

Aptitude points as a mechanic is a delaying tactic to not allow the player to get too strong, too fast. However, it fails quite quickly at that after the initial three points are spent, unlocking the entire tree. It functions more as a blockage to be removed, than a door that you open to other posibilities, and it still keeps those rank 3 perks in easy access (damage control). Blockagers, if used, should unlock entire new options or playstyles, not just allow you to focus down harder. Even using the old system where we get rewarded with something for choosing them, it still felt like unplugging a hole, it was just sweeter a bit.

Second part is choice paralysis, giving players entirely too much to think and gulp down in one go, especially with a no-respec system. This leads to a few things. They can choose not to use any points, making their lives harder and experience worse, until they know they want to go down a particular path, forcing them to prespec a lot of their build. Or they can become disenchanted with the premise that their choices might be wrong and require a second or third playthrough until 'they get it'; what skills are 'traps' (Command and Control) and which are "needed".


To replace aptitudes, the rest of the skill three , there are two clear methodologies to limit initial choice and provide a limitation of power.

- Unlock specialisation by focus, also known as a buy-in system. One of the examples of this is Borderlands:



Starting out, you have access to all rank 1 perks in all trees. You can unlock placing points in the rank 2 perks only after placing a certain number of skills in that branch. Rank 3 perks unlock after even more points are placed within the branch. This will eliminate aptitude 'wasted' usage, allow players to spread out in any branches they want without rushing to that OP rank 3 perk, and offer incentives to specialize without punishing exploration of the trees. Also, given that there is no 'buy in' and players don't have to consider rank 2 or 3 skills right off the bat to be able to engage, this will lower the barrier to entry.

- Level based perk unlocks

Can't remember a strict exact example right now, but i made this handy graph.



While arbitrarily enforced, this guarantees a certain lack of minmaxing and allows the game to not get too easy too soon, keeping higher ranks out of reach. It also strives to enlarge the build-up phase by providing instant-reward with a delayed fuse for build consequence. Usually a mix of these two mentioned systems can be seen often used together, like in Fallout 4.

For Starsector though, i think the simple focus specialisation system is enough, levels being somewhat meaningless in terms of focused power, and bringing with it versatility more than anything


-----

The final two issues were 'no brainer' skills and the combat vs universe skills.

One of the core issues i see is that the separation of skills is done on a flawed 'type'/flavor basis. Combat can roughly be separated in armor/shield/guns/maneuvrability, all solo skills with some logistics. Leadership has a mix of carrier solo combat skills and some fleet performance bonuses. Tech is a jumble of solo combat skills, overmap skills and logistics. Industry covers two completely non-combat, secondary profession almost skills mixed with some fleet management.

While i can understand the current setup, it roughly sums itself up in some clear builds: 'solo-focused gladiator' (cbt/tech), 'carrier & command'(ldr/tech), 'explorer/surveryor' (ind/tech) and 'scrapheap challenge admiral' (ldr/ind), each with maybe some dips in the other trees for specific skills. It's not bad in itself, and guiding players down 'classes' is not a bad mechanic, but that's not what's happening here.

The larger issue is that that separation doesn't account for the actual layers in which the player will think of the choices... allowing the player to feel that he's gimped, or worse, gimping himself.

Some skills will always end up being more useful than others, and it's often a waste of time to try to super-fine-tune that the first picks won't break the game. Given how SS has a 'free' system and a wide approach to what you can pick to start out, it'd be almost impossible to balance this. Pick out the outliers and rework them, sure (Navigation), but the general balance of them should come out of players picking 'build-like' abilities in a group, instead of cherrypicking. (I'll get back to this in a bit.)

However, the balance between areas of the game itself, as they are grouped up, is a large problem. Given that we will probably get a lot more skills to come that will deal with outposts, maybe faction warfare, maybe a reworking of reputation and generally more skills that deal with pre/post combat fleet logistics and overmap economy and progression, leaving them in the current 'tree'/separation i think is a mistake.

I propose a new grouping which takes into account layers of interaction.



Skills are grouped in relation to the core aspect of the game: Ship combat in the context of fleet operations that work in a simulated world. And it's this layering that we use here.

Combat is as we know it. All skills that directly affect moment-to-moment action within the arena go here. Carrier skills and 'piloted ship' tech skills included. They're part of the things you put points in to be better in the actual fighty-shooty bits.

Admiralty (tactics) deals with 'soft' combat helpers, both in combat, like ECM and Coordinated maneuvers, as well as the 5 minutes before and after combat, Field repairs/Safety Controls and some aspects of Damage Control. It also deals with fleet bonuses in the actual arena.

Logistics (strategy) deals with long-term matters, the greater aspects of building a fleet, the economics of it, maintaining it and all the bits that tie the experience together. We have officers here, navigation, fleet logistics, Recovery Operations/economic skills.

Now, this might not look like it helps, since this just accentuates the problem right? Not when you get a point to put in EACH skill line individually.

This would require lowering the max level and various math but as it will stand, allowing a single point in each branch will allow the skills of different usages to not be in direct competition amongst skills of a different layer.

Builds will create, in the combat tree, as they are now, based on what you will fly and how you fly it, allowing you to be efficient in combat and enjoy flying your ship. The admiralty or industry trees don't give as much 'build' potential, but they are what they should be, things that support combat. However they allow interesting things to be done if you decide not to pick up those 'must have' skill in each branch, allowing niche utility.

------

I might have made it worse than it is, but as the game gets more complex and more overmap skills get added, catalysts and levers to pull to put combat into context, i think a system similar to what i highlighted will do more good for Starsector than the current breakdown and progression.

PS: Given 'rare' hull mods, as well as an aspect of collectionism regarding them, i think it's wise to remove them from skills. Either players think those are all that exist, or they specifically go for a skill just for the hull mod... either way, while it made sense in 0.7.2, i think it's safe to remove and keep the tree cut and dry focused on the skills.
« Last Edit: May 05, 2017, 07:47:41 AM by Cosmitz »
Logged

Gothars

  • Global Moderator
  • Admiral
  • *****
  • Posts: 4403
  • Eschewing obfuscatory verbosity.
    • View Profile
Re: Let's talk progression and skill tree
« Reply #1 on: April 30, 2017, 01:48:31 PM »

Ho, you put some effort into this! Unfortunately I have to say that I mostly disagree.


Putting points in aptitudes feels 'wasted', compared to the old system were we got 'something' for them, and almost feel like a 'barrier to entry' a new tree

It feels that way because that is precisely what it's meant to be! And you provide the perfect example why. You said Navigation is a OP skill that you'd want every time. That would be the case, if it were freely accessible. But since it is locked behind the tech aptitude, you only want it if you plan to get other technology skills, too (as long as you don't think it's worth 6 CP, then it means it is really very OP). Only one of my three characters picked it, actually.


Skills are grouped in relation to the core aspect of the game
Now, this might not look like it helps, since this just accentuates the problem right? Not when you get a point to put in EACH skill line individually.

That seems like it would gut the core aspect of character building, i.e. deciding what your character is really good in. Instead of a fighter, a salvager, a scientist or an admiral, now there would only be fighter/admiral/salvager/scientist hybrids with different sub-specialization in all areas. I can understand wanting to play such a character, but it's certainly not good game design.
Logged
The game was completed 8 years ago and we get a free expansion every year.

Arranging holidays in an embrace with the Starsector is priceless.

Cosmitz

  • Admiral
  • *****
  • Posts: 758
    • View Profile
Re: Let's talk progression and skill tree
« Reply #2 on: April 30, 2017, 02:00:17 PM »

If Navigation would be a '6 CP' skill, then it should say so on the tin, and we'd rework the entire skill tree on a point-based system. But right now, it isn't. It's put on the same pedestal as Electronic Warfare. and Gunnery Implants.

As for the matter of character building, right now you can reach level 40 and lock yourself out of half of the experience (salvaging remnant and surveying) because you don't want to put 9 points, roughly a quarter of your total CP, in them. But you can 'do' combat without any skills in anything. But that's not the singular drawn-out issue here.

It's part of how the game is consistently acting on different layers, while it clearly has a singular focus layer, that of the arena combat that's been in the game since it first demoed out. And then it allows the player to invest into things which don't even related to that core aspect of the game.

There's a reason magic find and gold find mechanics were removed from RPG loot games, as they drawn peolpe into making 'long term strategy' builds, at the expense of a shittier moment to moment experience.
Logged

Megas

  • Admiral
  • *****
  • Posts: 12118
    • View Profile
Re: Let's talk progression and skill tree
« Reply #3 on: April 30, 2017, 02:05:31 PM »

Quote
- There is a huge barrier to entry leading to decision paralysis for newer players when judging all the skills at a time for an upgrade
For now, I will just say the following:

I just hit level 40, and not one skill point was spent.  I wanted to try various things in the simulator.  Decision paralysis can be a problem.

If I want to build a combat-focused character aside from carriers, there are not enough skill points.  You need to spend everything on combat for a marginal increase.  Meanwhile, I can play carriers and probably do better without any skill investment.  I can spend some points to hire more officers to do more fighter spam, then spend leftover points on fun non-combat stuff like Surveying or Navigation.  Far more efficient than trying to build a Combat nut who cannot solo fleets.

I agree the aptitudes are no fun.  That was the reason why I posted that one topic asking how much players spend at aptitudes.  While many seem to be around 9 or 10, there are a few extremes at 6 or 12.  Personally, I would love to see the aptitudes gone in part because there are not enough skill points to do very much.  And this is without more skills coming like outpost support and possibly others.  But, if there are exceptions that only need 6 or all 12 instead of the popular 9 or 10, fine, I guess aptitude serve some purpose.
« Last Edit: April 30, 2017, 02:13:17 PM by Megas »
Logged

BillyRueben

  • Admiral
  • *****
  • Posts: 1406
    • View Profile
Re: Let's talk progression and skill tree
« Reply #4 on: April 30, 2017, 02:46:47 PM »

Can't remember a strict exact example right now...

Diablo II or Torchlight had that kind of system if I understand your example correctly.

I'd like to see things like surveying and salvaging removed from the skill tree. I would rather them be equipment based rather than skill based. You want to salvage that research center? I hope you have plenty of salvage rigs and related supplies (maybe even a specific crew type). They break the feeling of skills to me. Everything else in this game revolves around combat, and now suddenly you have two skills that have almost nothing to do with combat whatsoever.
Logged

Cosmitz

  • Admiral
  • *****
  • Posts: 758
    • View Profile
Re: Let's talk progression and skill tree
« Reply #5 on: April 30, 2017, 02:48:30 PM »

Wanted to use a classic RPG but didn't want to confuse with 'skilling up' a skill.
Logged

Megas

  • Admiral
  • *****
  • Posts: 12118
    • View Profile
Re: Let's talk progression and skill tree
« Reply #6 on: April 30, 2017, 02:51:40 PM »

Surveying helps combat indirectly by giving the player tons of cash early in the game to spend on superior hardware.  It is a bit like those gold-find or magic-find skills in the game, although you cannot mule your finds to another character to reap the rewards.  Surveying is sacrifice long-term power for short-term gain.
Logged

BillyRueben

  • Admiral
  • *****
  • Posts: 1406
    • View Profile
Re: Let's talk progression and skill tree
« Reply #7 on: April 30, 2017, 02:55:41 PM »

Yes, but they remain the only two skills in the entire game that I feel are All or Nothing. If you are going to put in points in to these, you are almost certainly going to level them up to max. And they gate off a lot of the game if you don't invest.
Logged

Megas

  • Admiral
  • *****
  • Posts: 12118
    • View Profile
Re: Let's talk progression and skill tree
« Reply #8 on: April 30, 2017, 03:00:21 PM »

Agreed; no argument from me.
Logged

Gothars

  • Global Moderator
  • Admiral
  • *****
  • Posts: 4403
  • Eschewing obfuscatory verbosity.
    • View Profile
Re: Let's talk progression and skill tree
« Reply #9 on: April 30, 2017, 03:09:24 PM »

I'd like to see things like surveying and salvaging removed from the skill tree. I would rather them be equipment based rather than skill based. You want to salvage that research center? I hope you have plenty of salvage rigs and related supplies (maybe even a specific crew type). They break the feeling of skills to me. Everything else in this game revolves around combat, and now suddenly you have two skills that have almost nothing to do with combat whatsoever.

As things are now, I'd tend to agree. But I believe this is just the start of bigger changes, there will be many more skills that are not directly related to combat in the future. It's just weird right now because combat is already so fleshed out in the rest is still bare bones. I see it as a transitory problem, not a fundamental one.
Logged
The game was completed 8 years ago and we get a free expansion every year.

Arranging holidays in an embrace with the Starsector is priceless.

Alex

  • Administrator
  • Admiral
  • *****
  • Posts: 23987
    • View Profile
Re: Let's talk progression and skill tree
« Reply #10 on: April 30, 2017, 03:31:55 PM »

One quick note: I keep seeing the sentiment pop up here and there that surveying and salvaging skills "gate off" content. I'm not sure I really get that. Salvaging is eminently possible without it (blow it up, then use the scavenge ability), you just get less of it - but still a lot.

Surveying... its hard for me to see "being able to survey every planet in the game" as "content". I mean, it's a new thing to do and a bit fun just because of it, but it's very much a stub at the moment, what with the reward being primarily monetary. As it stands, it wouldn't be good design, but for this release it was a choice between not having surveying at all or having it be a money-generator. Ultimately, it's supposed to determine which planets you have access to using for <whatever>.

In fact, both of those are "you can do the thing better" skills; just in the case of planets, "better" means "more selection" as opposed to "better results from same selection", but that's not a fundamental difference.

I'd like to see things like surveying and salvaging removed from the skill tree. I would rather them be equipment based rather than skill based. You want to salvage that research center? I hope you have plenty of salvage rigs and related supplies (maybe even a specific crew type).

I don't think that'd be a good idea - it would just force players to get all that stuff and always carry it around, and likely engage in surveying if it wasn't really their cup of tea in the first place.

I'd like to see things like surveying and salvaging removed from the skill tree. I would rather them be equipment based rather than skill based. You want to salvage that research center? I hope you have plenty of salvage rigs and related supplies (maybe even a specific crew type). They break the feeling of skills to me. Everything else in this game revolves around combat, and now suddenly you have two skills that have almost nothing to do with combat whatsoever.

As things are now, I'd tend to agree. But I believe this is just the start of bigger changes, there will be many more skills that are not directly related to combat in the future. It's just weird right now because combat is already so fleshed out in the rest is still bare bones. I see it as a transitory problem, not a fundamental one.

Yep. Also: Sensors, Navigation.
Logged

BillyRueben

  • Admiral
  • *****
  • Posts: 1406
    • View Profile
Re: Let's talk progression and skill tree
« Reply #11 on: April 30, 2017, 03:59:33 PM »

One quick note: I keep seeing the sentiment pop up here and there that surveying and salvaging skills "gate off" content. I'm not sure I really get that. Salvaging is eminently possible without it (blow it up, then use the scavenge ability), you just get less of it - but still a lot.

It doesn't really feel that way though. I might still get 50% of the potential salvage after blowing up something then salvaging the field, but it feels like I am getting 10%. I'm not sure what the percentages are (and it really isn't that important), but it feels like there are parts of the game I am not seeing simply because I didn't invest 3 or 6 points in a skill. It's a complicated thing and I'm not sure there is a correct answer.
Logged

ChaseBears

  • Commander
  • ***
  • Posts: 224
    • View Profile
Re: Let's talk progression and skill tree
« Reply #12 on: April 30, 2017, 04:04:23 PM »

the more i play the less i like aptitudes in 0.8.  they feel like wasted skill points, and the worst part is its the lowest levels where you have the most skill points being nonbeneficial.  You might has as much as 1/3-1/2 your skill points put into aptitudes so you can get at the actual extremely useful aspects of a skill, since the really good stuff requires going all in

On the fip side you have that you're encouraged to spam into all aspects of a tree rather than making choices to keep your efficiency up.  I hesitated and ended up just maxing out the entire industrial tree because it was still useful and i wasn't 'wasting' skill points on aptitudes.  then you have the weird thing where i put 1 point into leadership and just dumped 1 point into like 5 different leadership skills.

tl;dr- remove aptitudes, slightly reduce skill point gain.

One quick note: I keep seeing the sentiment pop up here and there that surveying and salvaging skills "gate off" content. I'm not sure I really get that. Salvaging is eminently possible without it (blow it up, then use the scavenge ability), you just get less of it - but still a lot.

It doesn't really feel that way though. I might still get 50% of the potential salvage after blowing up something then salvaging the field, but it feels like I am getting 10%. I'm not sure what the percentages are (and it really isn't that important), but it feels like there are parts of the game I am not seeing simply because I didn't invest 3 or 6 points in a skill. It's a complicated thing and I'm not sure there is a correct answer.

this is a good point. perhaps it would help if it was more explicit feedback rather than the player just giving up and taking the 'loser option'.
« Last Edit: April 30, 2017, 04:07:33 PM by ChaseBears »
Logged
If I were creating the world I wouldn’t mess about with butterflies and daffodils. I would have started with lasers, eight o’clock, Day One!

Cosmitz

  • Admiral
  • *****
  • Posts: 758
    • View Profile
Re: Let's talk progression and skill tree
« Reply #13 on: April 30, 2017, 04:40:10 PM »

One quick note: I keep seeing the sentiment pop up here and there that surveying and salvaging skills "gate off" content. I'm not sure I really get that. Salvaging is eminently possible without it (blow it up, then use the scavenge ability), you just get less of it - but still a lot.

I know, that's why i didn't touch on them specifically. But we can be sure to get more of these overmap/world skills, and it seems healthy to put them as separate things with their own resources to judge what you want to engage in.
Logged

FooF

  • Admiral
  • *****
  • Posts: 1378
    • View Profile
Re: Let's talk progression and skill tree
« Reply #14 on: May 01, 2017, 08:17:25 AM »

I've said it many times but spending skill points on aptitudes would feel a whole lot less "wasted" if there were token bonuses that went along with them. By design, they should not be as powerful as investing directly into a skill but they should not make the player regret placing a point into them (and not having another point to follow-up with it). Instant-gratification is definitely A Thing™ in video games and making us wait is, unfortunately, a hard-sell.

A "buy-in" system would work (I've thought about proposing something similar), but all it really does is give us more points to play with. It would be far simpler to increase the level cap or give "bonus" points at logical points of leveling (i.e. level 5, 10, 15, etc.) if that was the ultimate end desired. What I can get behind is the argument that in the early game, a disproportionate number of skill points might be in no-reward aptitude points versus skills. In fact, I think that ratio is still too high in the end-game.

Personally, I think the aptitude system is fine as a mechanic to slow down progression and to make meaningful decisions. That said, it could still be improved and small aptitude bonuses and ~10 more skill points along the progression path would go a long way to making it more palatable.
Logged
Pages: [1] 2