Willothius, on 15 February 2017 - 11:20 PM, said:
The reason is that this way you really have to choose to 'specialize' (deep into a tree of expensive nodes), or to be an all rounder (unlock more and cheaper nodes).
Example: get your SRM cooldown aaaall the way to -15% but no laser buffs cause you're out of points, or settle for only -5% and plenty of points left for some laser and ballistic stuff. Right now, that's also possible, but increasing the cost strongly promotes this.
It's called Diminishing Returns, it actually is a proven principle. Science!
You do realize that having to invest farther into the current system is a matter of diminishing returns, right? It may not be as overt as your example, but the more you have to invest to reach advanced levels of the tree, the more you have spent to get there. Sure, trees could be made linear to remove the thinking part when you are upgrading, and we could lose those upgrades along the way that would still require nodes along the way, but does removing what some view as a "useless" node change anything if you are looking to change node costs instead? Current system forces you to buy things you may not want as added cost to the more desired nodes. To maximize some of those abilities, you are buying many nodes along the way which accomplishes the same thing (look at mobility, Speed Tweak requires a big investment just to get started, and even more investment to max out since you have to get the vast majority of that tree, which addresses high node cost).
Here's a case study from post #729 that actually relates to the weapon trees: Once I gave though to the cost benefit of those nodes, boating didn't hold much appeal to me because the return I saw from it didn't feel like enough to go all out a in a single tree unless I was dealing with a mech that already runs a single type of weapon (Warhawk and Nova Prime). It allows the mechs that are built as boats to keep their edge as a boat, but on a more mixed build I'd rather mildly buff a few different trees than go all out on a single one. My mad cat prime for example, has 2 large pulse lasers 3 medium lasers, and 2 LRM 5s (sometimes you gotta shoot over enemies rather than through allies). If I look at the weapon nodes critically and apply some of my experience from the current module system, making only upgrading one of my laser types doesn't make sense. My goal would be to have my large pulse lasers be ready to fire as quickly as the medium lasers and for my medium lasers to extend their range a little bit more.
5% 4%heat bonuses on them would be nice, but its not worth shortening the medium laser beams too much because of extra heat and trying to sync timing with my Large pulse lasers. Looking at the trees as I type this out, I would be most interested in getting cooldown for my pulse lasers, maybe a little extra beam duration reduction to better sync it to the medium lasers. Spending an extra 7 nodes to get a
5% 4% heat reduction would be nice, but that would be only be a be addressed only if I had enough nodes at the end. For my medium lasers I'd be most interested in their range. Lets say that i did have a few extra nodes left over (the LRM 5s wouldn't get quirked as they are only for harassing or avoiding team damage). I would have easy access to an extra 3 nodes for heat reduction on both of those mechs. Instead of spending 7 nodes to get 4.8%heat reduction on my pulse lasers, I would only need to spend 6 nodes to get 4.8%heat reduction, 3 from the lasers tree, 3 from the pulse laser tree. Using current smurphy's values, that means I'd get a 2.4%reduction from the 20 heat of the pulse and 2.4% heat reduction from the 18 heat of the medium lasers, that is 0.48 and 0.432 heat respectively adding up to 0.912 heat. That would even provide a slight edge over the point heat reduction I would receive if I maxed out only the pulse tree which would provide a reduction of 0.8 heat.
If you look at what the weapons trees offer, you can see that total percentages are minuscule and that highly desired nodes like Decreased Heat are farther down the tree. That implies that there is an higher investment necessary to gain as much as a grand total of 5%. In order to get the full 5%, you have to spend no less than 14 nodes. To get a portion of that, say 3 %, you only have to buy 9 nodes. The current system lets you get other benefits along the way to acquiring those heat reduction upgrades whereas a multipoint node system lets us know that we'd get nothing for paying more. Even if a person doesn't care about the nodes on the way to what he is after, there is an increased cost built in to the current system already, especially for more valuable upgrades. We could then argue about the value provided by each of those nodes, but we return to a discussion of small percentages that already required making an initial investment to max the upgrade out. If we reduce the value of each node as you progress in a tree, you end up punishing mechs that were built as boats (Nova, Warhawk, Black Knight, Catapult) while disproportionately incentivizing mixed builds. People currently boat and will continue to boat if that's what makes them happy, it's nothing game or groundbreaking. But I just provided an example above in which a build with 2 kinds of lasers and lrms were used. Using a mixed laser format and not maxing either tree ended up giving me more heat efficiency than boating a single type of weapon or weapon group. Further more, understanding that the benefit of the weapons trees are limited to compensating for a weakness or specializing a particular mech talent (the upgrade percentages aren't huge numbers) allowed me to explore a mixed build that didn't just produce less heat than maxing a single tree, but catered to the specific way I want to play a mech.
If you could explain why something like maxing out the ER Laser tree yields an overwhelming advantage, I'd like to hear it. A case study for this is the Nova. We'll examine the 10-12 medium laser build. When we want to upgrade its lasers, we are obviously looking to maximize heat reduction to help it cope with its payload. As the single type of weapon the mech and its soft nature, we would like to maximize its range as well. Now this is where the tree alone doesn't tel the whole story; a nova runs really hot, and requires you to split your weapons into groups. I run two groups of 5 lasers (because 6 lasers at a time is just too dam hot). Unless we are on a frozen map, I can only pull of 2 shots, maybe 3 shots before I would shut down. Buffing laser duration to be shorter would be good in the sense that more damage is delivered in less time, BUT what good does laser cool down do for the nova? If I consider that I'd spend 5 nodes to get a 5% laser cool down, just to allow my self to risk over heating (the normal cool down is a great, built-in means of slowing your fire down), those are points wasted. The stats look better on paper, but realistically it substantially increases overheating risk enough to be undesirable. Further, everyone who has played the PTS, has noticed a substantial increase in heat as well as decreased dissipation and it is clear that it is becoming a prominent factor in the game. So the Nova now, a dedicated laser boat which already struggles to manage its heat load, is put in an interesting situation. It can deliver a lot of damage, doesn't have to worry about ammo, but it being a laser boat means that it is going to have to constantly battle with heat. It will be forced to wait much longer between firing full weapons groups and is at a big disadvantage if someone pushes in on it because it will struggle to keep up fire without overheating significantly. I will still happily play the Nova because it is one of my favorite mechs, but the even with the laser tree, my play style will have to be greatly adjusted to the new heat scales which will affect how much fire that mech can put down as well as make me question the value of maxing out weapons trees.
There's a lot of subtlety in these trees when looked at in the PTS context that seem absurd when look at in the context of the live server. I encourage you to try the effectiveness of a few laser boats and a few mixed builds and see how the heat management and plays style are influenced, even if its for the sake of just proving me wrong. This was a round about way to suggest the current tree layout is better than a linear tree and that and that boating isn't necessarily as good as mixed building. I could give examples with mechs like the Orion IIC, Hellbringer, and Black Knight as well, but that's a whole lot more typing.