Getting a few things out of the way before the meat.
The rule of three was an arbitrary rule applied to the old skill tree, it could have been changed at any time, even if the old tree had been kept. That said, it's a very nice change, though not specifically relevant to a discussion on the new tree, and I wish folks could separate the ideas.
Likewise, the engine decoupling is a separate matter.
The quirk changes could have been done independently, or not at all, but for various reasons, they chose to bundle them. (Had to be done, because the skill tree would have made mechs with the quirks too potent, you say? And yet, all those mechs that did not have the quirks just improved their potency by that same amount, by more as a relative amount, in fact, than the quirked mechs would have had they been untouched, so, hmm. We'll ignore that too)
So, the skill tree.
I do think the skill tree needed to be updated. The change was long overdue.
I absolutely love not needing to switch modules around. Yay for that.
The skill tree change was supposed to clean up some of the problems with the old tree ( forced useless &/or low value skills), and is supposed to create greater diversity on the battlefield. I believe a secondary goal was to discourage boating.
Secondary goal of less boating? Um, no, big time. Boating seems just as popular now as it ever was.
Useless or low value skills? Still present, in spades. Forced purchases still. I will grant that in some areas (firepower, mostly) an effort was made to allow you to get around some of the inherently useless nodes, but not every tree allows that. So, ooops?
Diversity. Hmm. Yes and no.
For reasons I'll go into in the wall of text below, I have to fall on the no side on this.
0 for 3. Hmm.
The cost.
Let's get past people thinking of a new player's first mech or two. We're ignoring the rule of three thing, for reasons above.
Now, instead, realize that every single mech you buy comes with an extra 4.1M Cbill price tag to get it working as well as you can.
Obviously, for experienced players, that's largely irrelevant. For new players, who often struggle to get enough cash to re-tool a mech (put DHS on, change the weapons, etc.), that is quite the extra burden. (I'm not talking about alt accounts, because experience, person at keyboard skill, really does make a big difference in earnings, especially in low tiers).
It does feel grindy to work with, and part of that is...
Small values.
Each individual node, with a few exceptions, gives only a small benefit. Individually and realistically, you won't even notice a change from a single node, wherever you put it. Cumulatively, yes, the change is noticeable, but still isn't great.(but while skilling up... ugh.)
I know many people seem overjoyed to be able to individualize their mechs, but a lot of that difference seems kind of illusory (details below).
I've also seen people comment that even with a non-skilled mech, or one only partially skilled, it feels almost as potent. Though anecdotal, I concur. Fully mastered (91/91) mechs do feel more potent (some more than others), but typically not significantly so.
The UI.
Pretty bad. I'll just keep it simple and just say this seriously lacks in the QoL department, and overall needs some serious improvements it is very unlikely to get.
Final bit (sorta)
I still don't care for the tree as a whole. I don't think it manages to do what it was intended to do. Though it utterly lacks QoL and often seems limited in effectiveness, I still feel compelled to use it, and that wears on me.
If my only two options are the old skill tree or the new one, I will grudgingly take the new one.
Really, I'd like the third option: the greatly new and improved, user friendly, high QoL, effective numbers, role promoting skill tree. Sigh.
Wall o' Text
Given that the exact same nodes are available to every single mech, I think we've started by leaning towards low variety. I will grant that the percentages vary a bit based on weight class (in survival), and between Clan and IS (don't even start that), but by and large, there aren't significant differences.
Mobility
Spoiler
The general sentiment is that you are going to need to invest fairly heavily in this tree regardless of the mech. Exactly how heavily will vary a bit, but ignoring it or putting only a few points in is considered a mistake.
The upshot is that means, overall, less diversity of build, because typically a significant fraction of your points will be here (for most of my mechs it comes out to 15-20 points, some a little more or less).
For some mechs, various of these nodes are useless or of minimal value. For urbies, for example, torso yaw is literally of zero value. For many lights, torso speed doesn't seem to matter all that much. So, the gating is light to moderate
Auxiliary
Spoiler
The Aux tree... has its adherents. It is possible to ignore it completely (though I always try to spend the first point to get my second consumable slot).
However, people typically want to max one type (often coolant), which takes 7+/- points, so the overall points spent here are likely to be low.
I don't think this tree will significantly contribute to diversity.
For myself, I've only got a couple light mechs with more than that spent (strikes, cap assist, and UAV's).
Jump Jets
Spoiler
Either you jump or don't. If you do, you might want to put points in the tree.
Except as an experiment, I doubt most folks will invest heavily here. Going for lift speed and vent calibration is all that's really needed for a pop tart, so that's 8 points, without going through nodes you might not want. Less than 10% of your total points will account for the vast majority of builds, and many, even jump capable, will probably not spend even a single point.
Not a big source of build variance, in my opinion.
I have a few mechs invested here, but only a few (one poptart, one spotter, one light sniper).
Sensors
Spoiler
Go big or don't bother, mostly.
The way the sensor tree is laid out, in order to get anything needful, you have to invest heavily. (exception, advanced zoom can be picked up for only 4 nodes, so not too bad if you're into it)
Along the way to the good stuff, you pick up target info gathering, sensor range, and maybe some target decay. Oh, and possibly some target retention. Given the incredible dearth of those modules being used previously, many folks consider this tree to have a very high forced gating. They are handy little bonuses, but the general feeling on all those things is who cares?
Radar Deprivation used to be considered a must have module, but picking up full RD will cost you 17 nodes (for 5 nodes of RD).
Seismic also was very popular, and isn't much cheaper (12 nodes for 2 seis).
In either case, you've delved so deep for the one, imo, you might seriously consider taking the other (5 points more from seismic to RD, only 3 more from RD to get full seis)
ECM is the one piece of equipment that essentially requires you to spend skill points to make it effective. It has a 13 node skill tax for the 2 ECM. You do end up with 2 nodes of RD, and 1 more node to get 3/5 (which, given that your ECM'd, is probably enough?).
Sensors is one area I suspect will see a lot of variance of builds. Many people simply won't want to bother because of the gating. Some might just go for zoom. LRM builds will likely want a bunch of target decay. ECM, as mentioned, desperately needs investment here.
I would also argue that the nodes here will actually make the mech feel/play differently, so they really do impact the game.
I have mechs that follow each of these patterns of use, from nothing at all, to the great majority of the tree, so although I may be an atypical example, my builds definitely vary here.
Operations
Spoiler
Going by limited value nodes, I think this tree has some of the worst gating.
Again, we see 3 types of nodes that represent modules that were basically never used in place as gating mechanics: speed retention, hill climb, and improved gyros.
It also contains nodes for a skill that was considered of low value, i.e. quick ignition.
That said, this tree also contains another 2 node types that are considered absolute must-haves: cool run and heat containment.
It will cost you 20 points to get all five cool run nodes (out of 25 in the tree). In doing so, you will get at least 2 Ignition nodes and a varying number of hill climb, gyro, and speed retain nodes (depending on the path you chose). You will also automatically get 2 or 3 of the heat containment nodes.
Heat containment is cheaper, at 13 nodes, again with the tax to get there, but hey, you can totally avoid buying any cool run nodes if you want to. I'm not sure who would choose that, but it is possible.
Ultimately, like sensors, the tax on this tree is so high, and the value of the junk nodes so extremely low, that you really need to ask yourself how much heat problems you have with the mech in question. The more heat issues, the more you'll want to spend, especially if you have a large number of heat sinks on the mech (because both cool run and heat containment are % benefits, so high HS mechs get more value).
So, variety, yes or no? Hmm. Most mechs do have some heat issues, so will benefit from investment here. Full investment (ie. 20 points)? Possibly not.
Yeah, I guess there will be some variety here, I'm just not sure how much that will show.
I have minimal variety here. I think I've got a handful of mechs that have the full 20. Most of my mechs have 9 or 11 points spent (which gets me 2 containment and 2-3 cool run nodes). A handful of my builds might differ from those. I used to invest more heavily, but many of my mechs are medium or light, and it was pointed out to me that mechs with few HS (less than 17 or so, iirc) probably benefit more from the heat gen nodes in firepower, and the nodes you pick up to get those are of better value than the junk in operations.
Side question: why is hill climb in operations rather than mobility?
Firepower
Spoiler
Here's the thing about the firepower tree. It seems like you can get a lot out of it, and in some ways you can, but the individual nodes are of such low value that you have to spend a huge amount of skill points to get anywhere. It is the largest skill tree by far (68 nodes). On the plus side, I suspect most folks will not be maxing the tree, and the layout is such that you can usually pick a path that gets you more of the node types you personally value most (for that mech). On the downside, all the nodes you want are scattered all over, so without spending massive amounts, you are unlikely to get all of them
Range nodes 15% max @ 1% per node. Maximized cost: 25 points (how many points in the tree to get all nodes of that type, shortest paths, I think)
If you are a short range &/or a missile build, these don't do much for you. A 15% boost to a c-SPL gives you 25m more optimal range. That's still under 200m optimal. Missiles... well, missiles of any kind should not be fired at their max range (hello spread and travel time), so range is kind of worthless for any mech that has missiles as a main weapon. Oh, and regardless of what you want, your very first 3 nodes in firepower will be range nodes. Once you get past the first 3, range nodes are peppered throughout the tree, but you can often go around them.
Heat gen, 10.5% max @ .75% per node. Max cost: 31 points.
Rare indeed is the mech that doesn't want these. They start a bit deeper, but hey, you were going there anyhow, right?
Cooldown nodes, 12% max IS, 9.6% Clan @ .75% or .6% resp. Max cost: 34 nodes.
Again, who doesn't want cooldown?
So, to get the maximum for any of these node types, there is some crossover between the other 2 types, so you get some of each regardless.
What about the weapon specific nodes? Basically, velocity (5 nodes), ammo (rack or capacity, 2 each), laser duration (4 nodes), gauss charge (2), missile spread (2), high explosive (2), and UAC jam chance (2) are all peppered here and there.
If you carry a decent amount of ammo, the ammo nodes are very nice, since they apply per ton. Laser duration, with a value of 10% (Clan) or 15% (IS) is also fairly nice. The rest, except UAC Jam, are perhaps nice, at best, but I'd only buy them if I didn't want the points in another tree. The reason is, the difference the special nodes make are just too low to put any sort of priority on them. 5% missile spread? 15% missile crit chance? 10% Velocity? An additional second (1s) of holding a gauss cannon at max charge? Meh. UAC Jam chance is total garbage, btw, because it lowers the base 17% jam chance by 5%, i.e. turns that 17% into 16.15%.
Anyone hoping they could make a specific mech a master at one or two weapons... just can't. The specific nodes aren't there. Anybody who wants to make their mech into a firepower machine of doom... probably can't afford to spend that many points (40+?) doing it.
So, while there is likely to be a great deal of variety in the nodes chosen here, and in the number spent, the values are low enough that I suspect there it won't make a huge difference on the battlefield. People are likely to have spent some points in the tree, so the difference between mechs is probably smaller in the firepower tree than you might think.
Verdict: not a whole lot of difference, for a whole lot of points.
For myself, I've got a few mechs with basically nothing spent in this tree, and at most I think i don't quite hit 30 spent. Lots of my mechs are in the teens somewhere to get a couple weapon specific things and either heat or cooldown (range I tend to be meh on). So, as I said, some variety in points and path, but not a great deal of difference in the feel.
Survival
Spoiler
I've saved the best, in some ways, for last.
This tree can make the biggest difference on the battlefield, and show the most variety. Yet, paradoxically, it also does not.
What?
The big plus to this tree is that the low value node gating is minimal. The most common "junk" node in the tree is reinforced casing, which lowers the chance your mech will be critted by 1%. Assuming that is a straight reduction, and I believe it is, that's actually nice to have. Not great, but nice. (it is not some lame tiny % of a small % like UAC jam)
The other junk node here and there is shock absorbance, which lowers fall damage. Less nifty, but there aren't too many of these (5/35), unless you invest heavily in the tree. You can also avoid 2 of these, if you have AMS and want to enhance that instead.
So why the paradox in the tree? The trick here is that you are getting a % increase to your armor/structure (decreasing as tonnage goes up). The armor % increase runs from 10% (@100 tons) to 26% (@ 20 tons), structure values are approximately double that (a little more @ 100 tons, a bit less @ 20) That's nice and all, but since those values get spread across your entire mech, they don't actually amount to much in any one location. For example, an armor maxed 70 tonner gets 7 points more in each side torso, which is not even a single LL shot. The maxed structure bonus would be about the same.
So, like the firepower tree, a large (35 point) expenditure of points, gets you some improvement, yes, but perhaps less valuable than it might seem.
There are exceptions. Any mech that has armor &/or structure quirks, especially big ones (like BSW's armor, or the structure on a GAR-A), gets a huge benefit from high investment in this tree. That's a big deal, yes? Sort of. Basically, it means that any mech that is already tanky from quirks can be made REALLY tanky by using the survival tree. Just play to your quirks (if you have armor quirks, focus on the armor nodes, same for structure). That means the tree doesn't really increase diversity of build, however, since you are just further enhancing already present characteristics (and in doing so, you have fewer points to spend elsewhere that might differentiate the chassis, <shrug>).
Another benefit of the way this tree was built (and I compliment PGI on that) is that because the %'s are higher at lower tons, the tree remains useful at any weight, though to what degree is debatable.
So, while I see that there will be a great variance in how many points people are putting here, I'm not sure that will be felt on the battlefield, except where the quirks were already being felt.
I admit that I've spent a variety of points here, some I've kept, some I've respec'd. The ones I've kept heavily invested, no surprise, tend to be the ones with armor/structure quirks.
And all that boils down to not as much diversity as one would hope.
If you a, don't have a habit of running off cliffs or high places and b, don't jump in such a way that you harm yourself on landing, then yeah, shock absorbance is junk.
Valueless, no. Low or limited value except, perhaps, to only a small amount of players, yes. Which makes it junk. just like speed retention is junk (very low value to most, if not all, players).
Mate. Rule of three could have been changed anytime, but so could a lot of things. It wasn't though, it was changed with the skill tree, so people talk about it along with the skill tree.
Just because a large portion of the playerbase, is too scared to move a small laser somewhere else on their mech unless a web or youtube guide says so, does not mean there is an issue with the skill tree.
Its this simple. People can specialise their mechs how they like, and base it more on their specific weight class.
For example.
I have a viper, which is a very fast light medium. Do i put skill points into structure and armour? No, of course i don't. What is the point? Most of the skill tree is based on percentages of base values.
So a % of a small number is a very small number. To clarify, lets say i increase structure points by 30%. If i have 12 leg armour on a light mech, 30% structure would be roughly 2 structure points extra. This is a pointless gain.
In comparison, if i had 30% extra structure on an assault mech with 90 leg armour, that would be roughly another 15 points of structure, which is a much greater benefit.
So play to your strengths. If your mech is fast, skill up mobility and sensors. If your mech is tough, skill up survival. If your mech has a lot of ballistics, skill up velocity and some of the firepower tree. Etc etc.
Alternatively, you can skill up for weaknesses, if the benefit is worthwhile.
There are many options for specializing mechs now. The fact that people are too scared to stray from the perceived status quo, or what meta mechs or wherever the sheeple get their build ideas from these days, is not a reflection of the skill tree.
Its a reflection of the lack of drive from gamers in general these days, including this community. All people seem to do these days is ask for guides on how to do things. No one experiments, no one tries crap out for themself. They chase the meta, without even understanding half of why it is a meta, or what the implications of the meta are or aren't on their gameplay choices.
Then because of their limited understanding, before diving into an arguably convoluted system that actually expects some understanding of the game to make best use of, people instead complain.
Step up or step off. The skill tree is a good feature, the old system was a load of bonk.
No one experiments, no one tries crap out for themself.
I used to, but now I can't unless I want to waste SP or a combo of GXP/XP and CBills to lock skills so I can take it into Testing Grounds to see if it works. If it doesn't, if I want to tweak and adjust, or if I want to take another path on the Skill Tree, well...time to pony up some more of my earnings.
Does this mean you are for a Testing Grounds that allows us to test mechs before committing earnings to the tree? If not, then you cannot use that statement.
Fact is, my handful of mechs that I did Master with the new Skill Tree are builds you will not find on any of the "meta build" sites. So I'm one you can count that does not follow the lemmings.
I could also write a dissertation on why the Viper is THE SINGULAR PERFECT reason why the current Skill Tree is flawed. But, I'm done arguing with those that like to just take one point to beat on instead of looking at the entire argument and seeing how the point fits into the bigger picture. It's evident that PGI only listens to them.
7
