Jump to content

Skill Tree: Easy Way To Eliminate Gating, Help Mixed Builds, And Apply Quirks


  • You cannot reply to this topic
41 replies to this topic

#21 TheArisen

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 6,040 posts
  • LocationCalifornia

Posted 29 April 2017 - 10:34 PM

I was thinking of a similar system, especially after Mischief's thread. +1 this idea

#22 vandalhooch

    Member

  • PipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 891 posts

Posted 30 April 2017 - 12:09 AM

View PostYeonne Greene, on 29 April 2017 - 10:23 PM, said:


This is a non-issue; the naturally short range on a small pulse means that going for extra range is a pointless exercise in the first place. As a brawler, you would be going for shorter cool-downs or heat reduction.


I always choose range over cool-downs on my short range brawlers. But in your system, you have made the decision for me. The choices you make matching specific nerfs with specific buffs is in essence the developer deciding how to build mechs instead of the players. You have reduced player agency and that is never a good thing for a game.

Quote

Ignoring for a second that I was just tossing out an example, there is, in fact, a way to decide what counters what. Easy. Take lasers, for example:

What's the immediate mechanical counter to more damage? Longer duration (keeps the damage over burn the same). More range? Longer cool-down (makes the weapon clumsy to use up close regardless of skill level). Now think about what the secondary counters are. Repeat this exercise for every weapon type.


You, the developer are removing choices from the players and making them for him/her. By trying to control everything you have reduced the chances for emergent play to create new, unforeseen combinations. You have ruined the experience for a subset of players who like to explore all options by blatantly discouraging some combinations. It's a terrible, terrible game design idea.

Quote

At any rate, unless we're bundling skills into role packages that neatly arrange the perks and drawbacks, it's up to the players to decide which set of boosts best benefit the payload they've chosen. Is that not the entire raison d'être for the skill tree in the first place?


Your system reduces player input in those decisions because your chosen nerfs are telling the players the correct choices to make.

Quote

Another way we can side-step the problem is by assigning every 'Mech a limit to govern attributes placed onto the 'Mech, where players can pick negative ones to give themselves room to install more positive ones.


The whiners will never accept such a system now because it would be seen as "loss of progress." I get what you are describing and if the system had been built around that mechanic from the start then we would be on the same page . . . Alas, it's too late for that. For a game developer (and engineers the world over), perception is reality for most people. Just take a look at the Close Door button on elevators.

Quote

If you want to make it even more interesting, certain attributes could synergize positively or negatively with others to create larger buffs or debuffs. Heavy Gear 2 had such a system and it worked brilliantly in its PvP mode.


I'm assuming that it was built into the game or the mode from the start? Too late here.

Quote

From a mechanical perspective, why do we care? If it is mechanically sound, opinion doesn't matter and the pieces will fall into place like they should through enlightenement via experience.


Player perception is game reality. That's just how human brains work. No amount of logic is likely to alter those perceptions in a mind that is seeking agency in their entertainment. Notice how hard I've tried to get some module swappers to understand the inherent unfairness of the compensation system for iteration two versus one? Do raw numbers alter their opinions that the second one is "fair to all" just because they aren't "losing progress?" Opinions do matter and trying to make changes to a game with out taking them into account is just dumb.

Quote

This is not an MMO.


What? Of course it is.

Quote

This game has more in common with Counter-Strike and even Quake than it does any MMO,


Those are MMO's too. Any game that rewards play time with resources that can be exchanged for in game resources that can be used for future play has an economy. That's just the facts.

Quote

and I don't recall those having any sort of in-game economy. Quake even went F2P.


It's possible for a MMO to not have an economy. This MMO has one and thus it must have currency sinks.

Quote

That would be a problem if PGI stopped releasing 'Mechs, but they won't. That's why nobody is spending on modules (and often engines) in the first place, there's always a new 'Mech to buy and master with C-bills.


1 - Of course they aren't going to stop releasing mechs. It's part of their revenue stream.

2 - The decision to make modules swappable was in service to that revenue stream as you describe. However, that choice came at the cost of reducing modules' effectiveness as currency sinks. It was one or the other and PGI went with revenue. Can't exactly blame a F2P developer for that call, though.

3 - With the new skill system the pendulum has swung back to currency sink over revenue. PGI is counting on the collectors continuing to buy mech packs for real money as their C-bill stocks will likely be drained by the skill point costs (at a reduced rate compared to the original modules) and the introduction of new tech that can be retrofitted on old mechs. People who used modules as currency sinks in the past are being forced to permanently lose 50% of that currency and the returned C-bills will go towards collecting more mechs that will need real dollar mech bays. Without having access to all the details of PGI's finances, there's no way for outsiders like us to tell if this switch will turn out better or worse for them in the long run.

Quote

Players know this, and that's why having a respec cost is seen as a burden; it's already taking them long enough to amass C-bills to buy 'Mechs and equipment, now they have to buy the nodes.


Unlocking additional nodes will have a C-bill cost. That's the module system returning to its originally intended role, currency sink. Note that reactivating a previous unlock doesn't carry a C-bill cost but does carry an XP cost. This is an attempt to rectify a second problem with the old system. The hyper-inflation of the XP economy(s). Players who used the modules as currency sinks found themselves with massive loads of completely worthless mech XP and GXP. Grinding out the necessary C-bills for our modules left us with a second currency that has increasingly less and less value. I currently have 1.3 million GXP and have several mechs with 1-4 million mech XP. What use is it? Event rewards and cache unlocks often give players additional GXP. What use is that to me now? It shifts my perception of event rewards and caches towards "worthless" as well. That's not good for the longevity of the game. The XP cost to reapply a skill point will serve as a continuous XP sink and will serve to reign in the inflation of the XP economy. It might be too little too late as players have had years to amass the currency without anywhere to spend it, unlike C-bills.

Quote

All of these things are absolutely essential to creating a level playing field for themselves and that is a major problem. In-game currency for well-designed PvP games is never used for buying a global advantage unless that's the point of the game (i.e. DotA 2, but I'd argue that DotA 2 gold isn't analogous to C-bills at all, it's more akin to a strategic resource in an RTS of your choice).


Your "arguing" sounds a lot like an admission that I'm right but you don't want to admit it.

Take a look at end game content for MMORPGs like WoW. You going to argue that permanent, small stat boosts at exorbitantly high in-game currency prices are not a currency sink?

Quote

I really don't buy this whole C-bill sinks business. Unless you play an ungodly amount, it's the players who were around before the income was dramatically nerfed that are sitting on vast troves. I've been playing since February 2014, with 8,436 matches. It's not ungodly, but it's pretty solid. Do you know how many C-bills I have in my account right now? A hair under 5 million, out of a cumulative earnings of almost 1.1 billion. I bet you I'm pretty close to where most players are sitting. PGI cannot and should not bother trying to reign in or keep up with the legacy holders and current busybodies. That's a fool's errand.


I started playing the game four months AFTER you. I have 14,763 matches recorded. My cumulative earnings are 2.219 billion C-bills and I have 306 million C-bills sitting unused in my coffers right now. Who are you to say that your experience is THE player experience? You don't know that and making design decisions around your own personal experience alone is a terrible, terrible idea. It's what made you think that laser range buffs should be partnered with mobility nerfs. That's your experience and how you play the game. There is nothing "natural" about what you claimed were the proper pairings. It was all your personal opinion colored by your personal experience.

You are not THE PLAYERBASE.

Quote

I think that has more to do with a desire to make the balance game easier and to drum up excitement than it does creating C-bill sinks.


It can be both. In fact, it is both.

Quote

That's fine, and thank you for sharing. I'm also more or less just venting my frustration at the entire ordeal because I think it's all one huge missed opportunity to do something more. I worry that the Civil War update will similarly miss the mark.


I get that. And in no way am I defending every design decision PGI has ever made. However, it helps to at least look at alternative explanations for why things were done in a certain way by taking the broader picture.

Edited by vandalhooch, 30 April 2017 - 12:10 AM.


#23 Appogee

    Member

  • PipPipPipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 10,967 posts
  • LocationOn planet Tukayyid, celebrating victory

Posted 30 April 2017 - 12:15 AM

View Postvandalhooch, on 29 April 2017 - 10:31 AM, said:

[another intelligent idea which will neither be seen nor adopted by PGI because [a] they have their own ideas and don't want new ones, and [b] they don't read the forum.

Sure, sign me up.

Edited by Appogee, 30 April 2017 - 12:15 AM.


#24 vandalhooch

    Member

  • PipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 891 posts

Posted 30 April 2017 - 12:21 AM

View PostTheArisen, on 29 April 2017 - 10:34 PM, said:

I was thinking of a similar system, especially after Mischief's thread. +1 this idea

Yeah. There was no way I was alone in thinking about something like this. It's a fairly simple concept and as others have noted exists in some form in other games.

View PostAppogee, on 30 April 2017 - 12:15 AM, said:

Sure, sign me up.

Should we set up a monthly newsletter?Posted Image

#25 Appogee

    Member

  • PipPipPipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 10,967 posts
  • LocationOn planet Tukayyid, celebrating victory

Posted 30 April 2017 - 12:22 AM

View Postvandalhooch, on 30 April 2017 - 12:21 AM, said:

Should we set up a monthly newsletter?Posted Image

Monthly...? We already have a daily one, here ;)

#26 vandalhooch

    Member

  • PipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 891 posts

Posted 30 April 2017 - 12:25 AM

View PostAppogee, on 30 April 2017 - 12:22 AM, said:

Monthly...? We already have a daily one, here Posted Image


Reddit? {cold, cold shivers run up and down your spine}








{you were killed by a grue}

Edited by vandalhooch, 30 April 2017 - 12:25 AM.


#27 ocular tb

    Member

  • PipPipPipPipPipPipPip
  • The Seeker
  • The Seeker
  • 544 posts
  • LocationCaught Somewhere in Time

Posted 30 April 2017 - 02:47 AM

I very much like this idea. You pretty much laid it out how I envision the skill tree should be, even if the exact number of steps and allotted skill points need further refining (I tend to think only a few steps per skill are needed, but regardless I like the idea). By having the trees be linear and ordered neatly side-by-side allows a player to look at the tree and those next to it and see what their choices are instead of trying to navigate the web, which can be confusing. I think if done right a simpler solution like this should be able to do what the web intends to but in a much neater and easier to understand manner.

I'd also like for some sensor skills to be given more attention by maybe introducing more mechanics and HUD elements to make the benefits more "tangible" for users. They could also just scrap the rarely used skills such as shock absorbance, improved gyros, speed retention, etc or replace them with new ones that offer better benefits. I suppose for the most useful sensor modules such as seismic you could gate them behind something less useful or you could just make them cost more i.e. 2 points instead of 1. Either way it accomplishes the same thing I suppose.

PGI make this the new skill tree please Posted Image

Edited by ocular tb, 30 April 2017 - 02:48 AM.


#28 vandalhooch

    Member

  • PipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 891 posts

Posted 30 April 2017 - 07:21 AM

View Postocular tb, on 30 April 2017 - 02:47 AM, said:

I very much like this idea. You pretty much laid it out how I envision the skill tree should be, even if the exact number of steps and allotted skill points need further refining (I tend to think only a few steps per skill are needed, but regardless I like the idea). By having the trees be linear and ordered neatly side-by-side allows a player to look at the tree and those next to it and see what their choices are instead of trying to navigate the web, which can be confusing. I think if done right a simpler solution like this should be able to do what the web intends to but in a much neater and easier to understand manner.

I'd also like for some sensor skills to be given more attention by maybe introducing more mechanics and HUD elements to make the benefits more "tangible" for users. They could also just scrap the rarely used skills such as shock absorbance, improved gyros, speed retention, etc or replace them with new ones that offer better benefits. I suppose for the most useful sensor modules such as seismic you could gate them behind something less useful or you could just make them cost more i.e. 2 points instead of 1. Either way it accomplishes the same thing I suppose.

PGI make this the new skill tree please Posted Image


Nearly everyone is always going to go for Seismic Sensor and to a lesser extent Radar Deprivation. Putting those behind gates just builds resentment in players and reduces the opportunity to build a trade off system with the rest of the Sensor nodes. The sensor node webs are going to look nearly identical from player to player in the current iteration. There won't be any sort of trade off at all in that regard.

#29 MischiefSC

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPip
  • The Benefactor
  • The Benefactor
  • 16,697 posts

Posted 30 April 2017 - 07:24 AM

It breaks my heart that this idea is better but won't get used.

#30 TheArisen

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 6,040 posts
  • LocationCalifornia

Posted 30 April 2017 - 08:59 AM

View PostMischiefSC, on 30 April 2017 - 07:24 AM, said:

It breaks my heart that this idea is better but won't get used.


Yeah it seems that way. Maybe this thread might get acknowledged at least. Because Pgi does read the forums but generally doesn't post.

#31 vandalhooch

    Member

  • PipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 891 posts

Posted 30 April 2017 - 03:24 PM

For fun, I went through and created my vision of alternate skill trees for each category.

Firepower - Energy

Firepower - Ballistic

Firepower - Missiles

Survival

Mobility

Operations

Jump Jets

Sensors

Auxiliary

Some numbers:

Minimum number of active nodes for "mastery" (boating one weapon category and no jump jets) = 72
Maximum number of active nodes (all three weapon types and jump jets) = 112
Total number of nodes that could be unlocked per mech = 287

#32 MischiefSC

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPip
  • The Benefactor
  • The Benefactor
  • 16,697 posts

Posted 30 April 2017 - 05:27 PM

Yes, but that looks like it would be clean and efficient and create a variety of builds that are all viable.

How would that fit into MWO? Where's the needless complexity? Confusing caveats that are not obvious without searching? How does this brutally punish new players vs experienced players? Where's the clear additional motivation to boat a single weapon type?

This would never work.

#33 Y E O N N E

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPip
  • The Nimble
  • The Nimble
  • 16,810 posts

Posted 30 April 2017 - 09:35 PM

Poof!

Spoiler


#34 Shifty McSwift

    Member

  • PipPipPipPipPipPipPipPipPip
  • 2,889 posts

Posted 30 April 2017 - 09:46 PM

View Postvandalhooch, on 30 April 2017 - 03:24 PM, said:

For fun, I went through and created my vision of alternate skill trees for each category.

Firepower - Energy

Firepower - Ballistic

Firepower - Missiles

Survival

Mobility

Operations

Jump Jets

Sensors

Auxiliary

Some numbers:

Minimum number of active nodes for "mastery" (boating one weapon category and no jump jets) = 72
Maximum number of active nodes (all three weapon types and jump jets) = 112
Total number of nodes that could be unlocked per mech = 287


1. Still too many clicks!!!221`!!! (joke)
2. It leaves a lot of choices that will never get used at all and might as well not even be there, or be included as bonuses to other buffs, who is going to spend points on hill climb or capture accelerate against choices like top speed or heat containment etc.
3. You made it way too easy to get all the good upgrades barring radar dep, at least that was scaled out well comparatively.
4. It is too specialised, which minimises actual choice/sacrifice, i.e. it is a min/max lovers paradise.

#35 MischiefSC

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPip
  • The Benefactor
  • The Benefactor
  • 16,697 posts

Posted 01 May 2017 - 08:18 AM

The idea is that you should be able to spend points across segments. Currently I have to get hill climb to get speed tweak - which means I've wasted points I would have used to actually get something interesting.

If I can give up .75% of Speed Tweak to get 7.5% Hill Climb, I might do that. It's not that hill climb is worthless - it's that it's not worth using in a gated system where I have to give up a ton of anything to get it. Same with Seismic, Radar Derp or 99% of stuff on the tree.

The good upgrades are not 'easy'. They're a given. They should be; they're good. The question is, do you create a system that rewards giving up a *bit* of the good upgrades to get *a bit more* of the mediocre stuff.

It's about good tradeoffs.

None of that in the current system. Smart people will ignore the sensor tree, max all but 2 nodes in mobility and ops, pick up 19 pts in weapons. Depending on the mech maybe slightly less nodes in weapons and some extra module stuff. Everything that isn't that is making bad tradeoffs, like someone taking AMS overload module instead of a weapon cooldown module, only worse.

#36 Shifty McSwift

    Member

  • PipPipPipPipPipPipPipPipPip
  • 2,889 posts

Posted 01 May 2017 - 08:32 AM

You say that as if any system won't have min maxed paths and "meta" builds, in fact you guys are advocating for even more min maxing than the web provides, which doesn't just mean less real choice, but less usage of the built systems considered useless or bottom of the meta.

Either that or you would be advocating an even bigger web, with somewhat easier connections allowing you to take lesser percentages of any given buff.

I mean you say that mobility is a be all end all tree, and that structure in particular is useless, but to a mech with good speed and mobility already may actually benefit more from other trees, and straight survivability buffs are still just that, a light mech may get 5 or 10 armor points on a particular structure point, but a good light pilot can stretch that out pretty damn far vs most weapons.

It all seems pretty relative to circumstance, which is better than the system we have by a decent measure.

View PostMischiefSC, on 01 May 2017 - 08:18 AM, said:

AMS overload module


AMS overload is uber meta for crushing the dreams and spirits of any LRM boats, so it is a vital part of QP Posted Image

#37 Shifty McSwift

    Member

  • PipPipPipPipPipPipPipPipPip
  • 2,889 posts

Posted 01 May 2017 - 08:39 AM

And no I am not saying I wouldn't take mobility in 90% of the cases that come up, particularly because I tend to favor bulky mechs who would need it. And I also say that without fully comprehending the engine changes, so there may indeed be no actual cases where a mech is in a place where it has no mobility issues, or need for the mobility tree, but assuming there's at least a few who have tolerable natural mobility rates, the weapon tree in particular would definitely be optimal after the usual operations etc.

#38 vandalhooch

    Member

  • PipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 891 posts

Posted 01 May 2017 - 05:57 PM

View PostShifty McSwift, on 30 April 2017 - 09:46 PM, said:


1. Still too many clicks!!!221`!!! (joke)
2. It leaves a lot of choices that will never get used at all and might as well not even be there, or be included as bonuses to other buffs, who is going to spend points on hill climb or capture accelerate against choices like top speed or heat containment etc.


I used the max values PGI had created in their current Skill Tree iteration. I adjusted the number of nodes for each skill in order to try and create at least some minimally tempting trade offs. I was trying my best to simply rearrange the system they already have. Given more free reign, there definitely additional value changes I would make to further enhance the trade off nature within each category.

Quote

3. You made it way too easy to get all the good upgrades barring radar dep, at least that was scaled out well comparatively.


All the good upgrades? Some of them, like Seismic, are so considered must haves within their category that no level of trade offs would ever work. In those cases there's no real reason to try and create one. Just make their use cost enough nodes to create tough choices about how to spend the rest of the nodes.

I also constrained myself to not having more than 10 nodes in any column, total max could not exceed PGI's current system, and total nodes needed to be somewhere near current system total. Given more time and data, I'm sure we could all tweak the nodes and values to get a more refined and 'balanced' system.

Quote

4. It is too specialised, which minimises actual choice/sacrifice, i.e. it is a min/max lovers paradise.


Every system of customization in every game ever created gets min/maxed. It's what some players always do. You can't fight it. In fact, fighting it, can create resentment in a large portion of your players. Why would you do that?

Make sure that the min/max still have choices to make beyond their "must haves" and you might have something for everyone.

Edit:My first goal was to eliminate gating of high value nodes. Gating isn't necessarily a bad mechanic as mechanics go, but its perceived as "wasting resources" by many players.

Second goal, do not penalize mixed builds. With a max node per mech system, weapon boating is incentivized. Every node you don't spend in weapons is one you can spend in another category. Or, given a limited number of nodes left over from the other categories, maxing the buffs for one weapon category is more bang for you buck. Making trade offs occur within categories and having category limits reduces the advantages of boats over mixed builds. What about jump jets? In the current system who is ever going to use any of those nodes when other categories are so much more valuable? You talk about my system leaving nodes not used at all? The current web system leaves entire categories unused. That's a waste of resources.

Third goal, create a simpler system for quirking mechs. I'll admit the current node values don't leave much room to significantly improve a poor performing chassis by giving them extra nodes within certain categories. But, if we were allowed to alter the values somewhat from PGI's theoretical max, we could create meaningful quirking opportunity. Imagine doubling the energy range percentage per node but reducing the total nodes in the category a little. Sure, someone could use the standard node total to exceed the +10% range bonus but they would hardly have anything left for other perks. But, a quirked mech that got +3 energy nodes could really get some value out of the higher value nodes.

Edited by vandalhooch, 01 May 2017 - 06:16 PM.


#39 MischiefSC

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPip
  • The Benefactor
  • The Benefactor
  • 16,697 posts

Posted 01 May 2017 - 06:05 PM

View PostShifty McSwift, on 01 May 2017 - 08:32 AM, said:

You say that as if any system won't have min maxed paths and "meta" builds, in fact you guys are advocating for even more min maxing than the web provides, which doesn't just mean less real choice, but less usage of the built systems considered useless or bottom of the meta.

Either that or you would be advocating an even bigger web, with somewhat easier connections allowing you to take lesser percentages of any given buff.

I mean you say that mobility is a be all end all tree, and that structure in particular is useless, but to a mech with good speed and mobility already may actually benefit more from other trees, and straight survivability buffs are still just that, a light mech may get 5 or 10 armor points on a particular structure point, but a good light pilot can stretch that out pretty damn far vs most weapons.

It all seems pretty relative to circumstance, which is better than the system we have by a decent measure.



AMS overload is uber meta for crushing the dreams and spirits of any LRM boats, so it is a vital part of QP Posted Image


Again dude, no. The point is this -

Just like with the current model, the weapon perks and such are capped. So for example cooldown is capped at 7.5%. Like fastfire for example in the current model.

The difference is that you could dial that back to 7% and spend 5% in cooldown to get 5% hill climb or whatever.

Make sense?

There is minmax in the existing one. It just has no tradeoff. The one on the PTS is the same - there's very little, if any, real tradeoff and everyone (who doesn't make terrible choices) is going to have hillclimb too.

There's always min/maxing. That's the only reason to have a skill tree; you're giving up one thing to increase another.

What the new PTS model does NOT have is any real equality of value for tradeoffs. There's just the clear good choices and clear bad choices. A real system for tradeoffs would let you give up a *bit* of the good stuff for *a slightly bigger bit* of the mediocre stuff. That would drive tradeoffs.

The new skill tree doesn't. Not to any real degree we don't already do with modules, in fact I have more real choices to make with modules than I do with the new skill tree.

#40 sycocys

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Moderate Giver
  • Moderate Giver
  • 7,697 posts

Posted 01 May 2017 - 06:16 PM

Since the game is set up that essentially all players are in the same pool for matches, anything other than a zero sum or at the most slight upgrade is just a waste of their time even putting on paper much less continuing to run tests on.

The entire problem with the original skill system is that it gave vets/spending players a huge mechanical advantage over new or less funded players. So if the new system isn't designed with drawbacks to counter specializations all that happens is a ton of time spent creating a more complex version of the same exact crappy skill system that the game launched with.

The last proposal I looked into didn't do anything to make the game more interesting, it wouldn't do anything to offer variation anywhere but the lower tiers of play - and only until those players realize their poor choices and respec or leave.

Without a zero sum balance mechanic all you get is more of the same instead of encouraging players to develop their own skill tree that fits their playstyle or need of a role.





1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users