44 replies to this topic

[BustedHipGaming]

Can someone explain to me why they made the skill tree a clump of mazes instead of nice, straight lines?

Like why is it that you don't just get to pick skill 1 -> 2 -> 3 -> 4 -> 5 all the time so you can just level up the things you want without having to buy a bunch of dead cells in the tree that you don't want, that don't have any relevance to your mech, etc.?

If it's because some skills would be more valuable than others, doesn't it make sense to improve the value of the less-valuable skills to make them competitive?

I dunno, I just can't wrap my brain around this maze-layout. Anyone know why they did it that way?

[RestosIII]

PGI is going to PGI, and they think you getting the skills you want should be a case of sacrificing points for no good reason to reach them. That's... about it?

[Kin3ticX]

They are doing it so that you have to invest more points to get all of something. For example if you want ALL speed tweak you have to get ALL mobility instead of just 4 or 5 pts into speedtweak.

They dont want you to min max it without a tradeoff

A linear tree is not going to happen

Edited by Kin3ticX, 12 March 2017 - 10:47 AM.

[oldradagast]

Two reasons for the maze, as far as I can tell:

- Trying to substitute GRIND for CONTENT. The bigger they make the tree, the more pretend "content" they've added to the game. Look at all those skill nodes, each doing almost nothing! It's beautiful! Look at all the work put into this! Yeah, right. But the key part is grind, grind, GRIND! Would you like to buy some premium time, cheapskate, to regrind all your mechs?

- Lousy video game design skills and failure to understand what the game needs (ROLES, not GRIND). Hiding the same mandatory skills everyone needs behind a wall of trash skills that are not needed is not "choice" or good video game design. Even World of Warcraft gave up that failed idea years ago, but here we are in 2017, with PGI reinventing the wheel. All this silly setup does is tick people off and make certain that they spend more time lacking "the same good skills every mech needs" as they eagerly get a jumpjet boost for mechs without jumpjets or get a missile boost for their mech that has no missile hardpoints. Just mediocre game design, and nothing more.

Edited by oldradagast, 12 March 2017 - 10:49 AM.

[BustedHipGaming]

Kin3ticX, on 12 March 2017 - 10:46 AM, said:

They are doing it so that you have to invest more points to get all of something. For example if you want ALL speed tweak you have to get ALL mobility instead of just 4 or 5 pts into speedtweak.

They dont want you to min max it without a tradeoff

A linear tree is not going to happen

That seems silly to me. Again, if speed tweak is so useful that it's a "must have" the response should not be "come up with some arcane method to siphon people's skills points away to punish them for taking it." It should be "let's make all the skills useful so that people actually have a reason to want a variety of things."

oldradagast, on 12 March 2017 - 10:48 AM, said:

Two reasons for the maze, as far as I can tell:

- Trying to substitute GRIND for CONTENT. The bigger they make the tree, the more pretend "content" they've added to the game. Look at all those skill nodes, each doing almost nothing! It's beautiful! Look at all the work put into this! Yeah, right. But the key part is grind, grind, GRIND! Would you like to buy some premium time, cheapskate, to regrind all your mechs?

- Lousy video game design skills and failure to understand what the game needs (ROLES, not GRIND). Hiding the same mandatory skills everyone needs behind a wall of trash skills that are not needed is not "choice" or good video game design. Even World of Warcraft gave up that failed idea years ago, but here we are in 2017, with PGI reinventing the wheel. All this silly setup does is tick people off and make certain that they spend more time lacking "the same good skills every mech needs" as they eagerly get a jumpjet boost for mechs without jumpjets or get a missile boost for their mech that has no missile hardpoints. Just mediocre game design, and nothing more.

That was my point - it just seems like a system that is so...off the mark.

[Kin3ticX]

Drewbicus, on 12 March 2017 - 10:51 AM, said:

That seems silly to me. Again, if speed tweak is so useful that it's a "must have" the response should not be "come up with some arcane method to siphon people's skills points away to punish them for taking it." It should be "let's make all the skills useful so that people actually have a reason to want a variety of things."

I know, I know, it triggered the hell out of me at first. You have to listen to the podcast to hear how the dev explains it. The way I had envisioned it (more linear) simply will not work. If you can cherry pick to the max then you get have everything that matters and none of the stuff you dont care about.

[MrJeffers]

Kin3ticX, on 12 March 2017 - 10:46 AM, said:

They are doing it so that you have to invest more points to get all of something. For example if you want ALL speed tweak you have to get ALL mobility instead of just 4 or 5 pts into speedtweak.

They dont want you to min max it without a tradeoff

Except a trade off implies that you give something else.
Min/Maxer invests fully into mobility to get speed tweak, and then uses less SP to invest only in specific weapon tree to support 1/2 weapon types, use extra points for operations/sensors
Jack of all trades builder invests into sensors tree to get seismic, radar dep, lightly into mobility tree to get twist,arm ranges,lightly into a few different weapon type, lightly into survival, etc

What did the min/maxer give up? Nothing! They have the exact same boosts on the shared tree as the jack of all trades, and they have deeper depth on things that give them advantages. They are more mobile and their weapons hit harder, more often from futher range, with lower heat. The jack of all trades has a bunch of varied 1-3% bonuses, in a bunch of different things.
The specialized build still wins. This doesn't shake up anything or prevent min/maxing. Specialized builds are still superior.

[Kin3ticX]

MrJeffers, on 12 March 2017 - 10:57 AM, said:

Except a trade off implies that you give something else.
Min/Maxer invests fully into mobility to get speed tweak, and then uses less SP to invest only in specific weapon tree to support 1/2 weapon types, use extra points for operations/sensors
Jack of all trades builder invests into sensors tree to get seismic, radar dep, lightly into mobility tree to get twist,arm ranges,lightly into a few different weapon type, lightly into survival, etc

What did the min/maxer give up? Nothing! They have the exact same boosts on the shared tree as the jack of all trades, and they have deeper depth on things that give them advantages. They are more mobile and their weapons hit harder, more often from futher range, with lower heat. The jack of all trades has a bunch of varied 1-3% bonuses, in a bunch of different things.
The specialized build still wins. This doesn't shake up anything or prevent min/maxing. Specialized builds are still superior.

It prevents min maxing. There are many skills you can shave in a linear tree and then go in and get tanky and all the weapon quirks you desire(assuming you still had 91 points). Thats the crux of it. You cant cherry pick just the select skills, you have to invest more points to stack.

Edited by Kin3ticX, 12 March 2017 - 11:01 AM.

[RestosIII]

Kin3ticX, on 12 March 2017 - 10:56 AM, said:

I know, I know, it triggered the hell out of me at first. You have to listen to the podcast to hear how the dev explains it. The way I had envisioned it (more linear) simply will not work. If you can cherry pick to the max then you get have everything that matters and none of the stuff you dont care about.

Maybe there's a reason we don't care about that other stuff. Maybe because it's,.. worthless? For instance, arm skills on a Jagermech. There's no point, but if you want speed tweak, you need to grab them. It's things like that which could get changed and make the system so much better.

[BustedHipGaming]

Kin3ticX, on 12 March 2017 - 10:56 AM, said:

I know, I know, it triggered the hell out of me at first. You have to listen to the podcast to hear how the dev explains it. The way I had envisioned it (more linear) simply will not work. If you can cherry pick to the max then you get have everything that matters and none of the stuff you dont care about.

You *should* be able to have just what you care about. the point is that the stuff you didn't pick should actually be valuable enough that you have to carefully consider what you buy, not just go "speed tweak, heat management, etc." off the standard checklist of important stuff. If all the skills were actually useful it would not matter if you got just what you wanted and nothing else, right?

Kin3ticX, on 12 March 2017 - 11:00 AM, said:

It prevents min maxing. There are many skills you can shave in a linear tree and then go in and get tanky and all the weapon quirks you desire(assuming you still had 91 points). Thats the crux of it. You cant cherry pick just the select skills, you have to invest more points to stack.

Again, if all the skills are actually useful, there is no such thing as "min maxxing". The problem is that some skills are just really useful and others really need a boost. If the less-useful skills were given enough OOMPH to make them competitive then it would not be an issue. the problem isn't min-maxxing; it's the lack of skills that are actually useful.

RestosIII, on 12 March 2017 - 11:01 AM, said:

Maybe there's a reason we don't care about that other stuff. Maybe because it's,.. worthless? For instance, arm skills on a Jagermech. There's no point, but if you want speed tweak, you need to grab them. It's things like that which could get changed and make the system so much better.

This, exactly!

[Spheroid]

Somebody must have preferred Civilization Beyond Earth to Civ 5.

[RestosIII]

Spheroid, on 12 March 2017 - 11:06 AM, said:

Somebody must have preferred Civilization Beyond Earth to Civ 5.

Speaking of Civ 5, damn I forgot how good that game is. Modded the crap out of it recently, and I'm playing it while I wait for new tech news. So fun.

[Kin3ticX]

some skills are just flat out superior, like cooldown, or speed, or accelleration, or health while target info gather and such will never be as desired

[RestosIII]

Kin3ticX, on 12 March 2017 - 11:11 AM, said:

while target info gather and such will never be as desired

Wait, what? That was one of the only modules in the game I liked.

[MrJeffers]

Kin3ticX, on 12 March 2017 - 11:00 AM, said:

It prevents min maxing. There are many skills you can shave in a linear tree and then go in and get tanky and all the weapon quirks you desire(assuming you still had 91 points). Thats the crux of it. You cant cherry pick just the select skills, you have to invest more points to stack.

No it doesn't because the trees are shared, a non-min/max player who wants to buy speed tweak and a min/maxer who buys speed tweak still pay the same costs.
The "opportunity" cost of burning a minimally useful point or two don't offset the gains by specializing, because you are either getting the same skill as everyone else, or burning one or two SP to get greater gains than people who don't.
A linear tree actually does.
Eg. Weapon Range - Skill 1-3. Each level costs a point. So level 1 - 1 SP, lvl 2 = 2 SP, 3 = 3 SP, so total for the tree is 6SP. Taking lvl 2 means actually giving up a lvl 1 somewhere else. You actually have *less* SP to spend. Likewise on LVL 3, means giving up 3 lvl 1 skills, or a lvl 1 and a lvl 2 somewhere else.

In that system the jack of all trades fairs better because they actually have more skills - they have more overall boosts in many categories rather than a few high boosts in a few categories. Rather than the current design where 3-5 of the specialized skills are truly useless (as in weapon/mobility items that don't have any application to your build) but you still have a better mech skill wise than the person who spends lightly across all the fields.

[FupDup]

Kin3ticX, on 12 March 2017 - 11:11 AM, said:

some skills are just flat out superior, like cooldown, or speed, or accelleration, or health while target info gather and such will never be as desired

PGI can fix this by adjusting the values of the skills and the number of nodes you need to max them out. Something like Target Info for example should be able to be maxed out with only 3 or so nodes.

Also, cooldown is pretty much useless with its current values (7.15% for Clan and I forgot what value for IS).

Some skills like Hill Climb just need to be redesigned completely.

[CMDR Sunset Shimmer]

The current skill tree in MWO has nothing on Final Fantasy X "Sphere Grid"

and ya'll call yourselves gamers.

[Bishop Steiner]

Kin3ticX, on 12 March 2017 - 11:00 AM, said:

It prevents min maxing. There are many skills you can shave in a linear tree and then go in and get tanky and all the weapon quirks you desire(assuming you still had 91 points). Thats the crux of it. You cant cherry pick just the select skills, you have to invest more points to stack.

I disagree , mostly semantically.

There are two basic schools, or definitions of "MinMaxing".

The one we see used here, currently, in MWO, is to invest the MINimum amount, for MAXimum game and benefit. The way MinMaxing builds in MWO works, one really is not sacrificing or trading off anything when building to the Meta. Because everything that matters, can be MAXimized with no loss.

Traditional "MinMax" did very much imply a tradeoff. Your warrior skimped point on Charisma, Wisdom, IQ, etc to maximize pts on Strength, Stamina and Agility. Because those were the primary needs of his class. But now if said warrior needs to barter or negotiate, or solve a puzzle? He's screwed. Likewise the person who flipped the script to make a Sage or Wizard.. but now needs to climb a cliff, walk a tightrope or win an arm wrestling contest.

And then you had the more generalized characters or builds who didn't do any one thing to the insane levels... but also didn't have egregious weaknesses.

Sou you MINimized something important to MAXimize something you perceived to be more valuable, but in so doing, left yourself open to being exploited.

Again, there are no such tradeoffs in MWO currently. Whereas now with the Node Web design... if you want that Speed Tweak, or Radar Derp or Seismic... you have to decide... do i REALLY need it? Because you are certainly going have to give up something else to get there. And as such... you likely will see a lot of Gunboats...no longer using Speed Tweak. You will see less overall saturation of maxed out ECM, let alone Radar Derp or Wallhack, unless you are legit investing in Sensors (which makes sense.. for certain mechs, limiting those more to Scout, Spotter and E-War builds).

It's funny the community complained about ECM, Radar Derp and Wallhack more years... and now that they can't just free lunch them on everything... they complain about that.

The very things you started to notice and figure, that Russ tried to explain between all the spew in the chatbox, were exactly what I figured out, on my own, from hours and hours of testing, and seeing how everything interacted. And then Chris and Russ literally confirmed my every finding on function and intent.

Admittedly, it ain't perfect, it will have issues, and they will need to be on the spot (for a change) with fixes. But overall, the Web design makes a lot more sense, and is a lot less exploitable than people seem to realize yet. And of course, some of the complaints still are literally because some of the ezmode exploits are not quite so straightforward and easy, any more.

At this point though, I reckon everyone mostly has their mind made up, so only real thing left is to wait for the 21st. In the meantime I do have a few mechs I need to add a couple thousand xp to, to get the best results from the switch, so I reckon I'll be doing that, more than worrying too much about what the forum mob is up to.

CMDR Sunset Shimmer, on 12 March 2017 - 11:22 AM, said:

The current skill tree in MWO has nothing on Final Fantasy X "Sphere Grid"

and ya'll call yourselves gamers.

oooooh... SKITTLES!!! I LOVE SKITTLES!!!!!

[Queen of England]

If I understood Chris in the NGNG podcast correctly, the goal of the maze is to produce declining marginal costs for people who invest heavily into a tree. That is, getting the first valuable node is quite expensive as it is gated behind a bunch of filler nodes, getting the second valuable node is less expensive since you've already purchased several of the prerequisite filler nodes, and so on.

Essentially, they want to reward old-school min/maxing (as Bishop defines it above). The plan is to have 'mechs that take nothing out of several trees so they can take everything in the trees they want.

I'm not convinced the trees actually do this, but my understanding is that is what they're intended to do.

As an example, take the Sensors tree. I assume the only nodes in here people will actually want are Seismic Sensors and Radar Deprivation. (I think Target Info Gathering is important, but the values assigned to nodes in this tree are far too small for them to be worthwhile.)

I went through the tree and determined how many points you needed to spend to get the skill nodes you want. Your first Radar Derp is quite expensive - 5 points! Then you can pick up 3 additional nodes for 5 points total (2, 2, and 1 point each). However, if you want to continue investing, you need to pick up a bunch of additional filler nodes on the right side, so your next node is as expensive as your first was. This node unlocks two more cheap 2 point nodes and you're done with the tree.



This tree is basically set up to reward two different levels of investment - you can either spend 10 points to unlock 4 nodes, or you can push through and spend 19 points to unlock 7 nodes.

Edited by Queen of England, 12 March 2017 - 12:04 PM.

[Bishop Steiner]

Queen of England, on 12 March 2017 - 11:56 AM, said:

If I understood Chris in the NGNG podcast correctly, the goal of the maze is to produce declining marginal costs for people who invest heavily into a tree. That is, getting the first valuable node is quite expensive as it is gated behind a bunch of filler nodes, getting the second valuable node is less expensive since you've already purchased several of the prerequisite filler nodes, and so on.

Essentially, they want to reward old-school min/maxing (as Bishop defines it above). The plan is to have 'mechs that take nothing out of several trees so they can take everything in the trees they want.

I'm not convinced the trees actually do this, but my understanding is that is what they're intended to do.

As an example, take the Sensors tree. I assume the only nodes in here people will actually want are Seismic Sensors and Radar Deprivation. (I think Target Info Gathering is important, but the values assigned to nodes in this tree are far too small for them to be worthwhile.)

I went through the tree and determined how many points you needed to spend to get the skill nodes you want. Your first Radar Derp is quite expensive - 5 points! Then you can pick up 3 additional nodes for 5 points total (2, 2, and 1 point each). However, if you want to continue investing, you need to pick up a bunch of additional filler nodes on the right side, so your next node is as expensive as your first was. This node unlocks two more cheap 2 point nodes and you're done with the tree.



This tree is basically set up to reward two different levels of investment - you can either spend 10 points to unlock 4 nodes, or you can push through and spend 19 points to unlock 7 nodes.

Yep.

And even I admit, I don't know how successful it will be across the board. People will still find ways to optimize, but if those "meta" builds are even marginally less efficient than before, while allowing people with less "optimized" builds (aka non boats) to close the gap and or find other useful functionality (like my CPLT-C1 using a full Sensor investment to boost it's LRM capability, while also giving it well, better sensors), then it is overall, a win.

But...

I do expect tweaks and value changes to be needed, and somethings may prove outright broken, initially.

You are trying to juggle 160 individual Skill Nodes among several 100 individual chassis with their own quirks, base stats, etc, with dozens of weapons, and the variables of equipment, maps, etc. Things are going to have issues.

Edited by Bishop Steiner, 12 March 2017 - 12:27 PM.