10 replies to this topic

[Skribs]

Forward

This is an untested design that is made to show the big picture of what I am talking about. It is an untested build that is probably horribly unbalanced. If your comments are about the details of my post (i.e. specific costs, individual skill values, skills I should add/change/remove) then that’s interesting, but it’s not about the meat of the post, which is the big idea around this type of a skill tree I am suggesting.

Big Picture

• Linear skill trees that give us a choice in how we upgrade
  
• Increasing cost in those trees in terms of skill points
  
• Items that are binary yes/no (such as whether or not you have Advanced Zoom) are single-skill nodes with a higher cost associated with them
  
• Quirk system can mostly be replaced with additional skills

Basic Tree Design


In the build, you’ll see a black box next to each skill. That isn’t the rank of the skill (well, in a lot of cases they match up), but rather the skill point cost associated with it. There would generally be no prerequisite to grab any skill, except within each line.
In the example, I assumed a Jagermech (or similar) with no Lower Arm Actuators and no Jump Jets, so you will see that Arm X and Jump Jet skills are grayed out. Keep in mind that the overall design here is just a way to visually show what a lot of people had suggested, and different visualizations of this design are perfectly okay.
As a quick example, let’s assume you have 30 skill points to spend, and you want to spend them in Defensive. You could get 2 of the 5-tier skills maxed (1 + 2 + 3 + 4 + 5 = 15) and end up with 15% armor and 25% structure. Or, you could get 3 skills up to Tier 4, and get 12% armor, 20% structure, and -4% incoming crit chance. Or you could get Tier 3 Armor Plating, Skeletal Density, and Reinforced Casing, Tier 2 of Quick Ignition, and Fire Control or Improved Gyros. A balanced build could get Tier 2 or 3 in a lot of things, while a focused build could get Tier 4 or 5 in a few things.
Now, before you say “mathematically everyone will spend 1-2 points in everything so they can maximize the maths”, remember that not every skill will be super useful to everyone. Lights may not need torso upgrades. Energy boats may not want cooldown. Dakka boats may not want heat efficiencies. If you would rather sneak up from behind enemies or use LRMs, you might not value defensives as much as firepower and mobility. So different builds would still want to focus on different things. What this does is make it so that if you value defensives, you have to choose how much you value them – do you want to put in 3 points, 4 points, or 5 points into the defensive abilities?

Applying Skills:

In the upper left corner (which is currently blank) there could be a readout with your currently used skill points, total skill points purchased, skill points available to purchase through XP, and an XP -> SP conversion button. You would purchase SP from XP up to (or more than) the amount needed for the skill you want, and then you could acquire it. Respeccing would mean taking a skill point off, which would refund the skill points, and let you reapply them elsewhere. (Notice how I didn’t say a C-Bill cost or an XP cost on re-applied skill points).
Binary Yes/No Skills

Binary yes/no skills (like Advanced Zoom, Seismic Sensor, etc) have a higher point cost associated with them. Currently I have them between the cost of Rank 3 and Rank 5 of any of the other skills. This is to reflect the varying values to which the community usually places on these particular traits. Maybe some of them should be more expensive, but like I said – this is an untested, unbalanced version.
Jump Jets, AMS, spotter weapons, and Capture Accelerator I put as 3-point skills as a kind of bridge-the-gap between the binary skills and the full-tier skills.

Elimination of Quirks

I think the quirk system solves the problem of underperforming Mechs, but if bonus points were given to those Mechs, we could do away with most quirks, especially modest weapon quirks. Now, mobility quirks I think give Mechs character, and possibly some armor/structure quirks as well to help reinforce specific components, but outside of Mechs that have absolutely horrible hardpoints (like the Spider 5V, 5K or the Locust equivalent) where you really need the help, quirks should mostly find their way into bonus skill points.
Bonus skill points would simply be free additional points that you could use on those Mechs when they are first purchased and/or periodically as they are leveled up, and the amount of bonus points received would be based on how badly the Mech underperforms, and if there are any chassis or variants that can do their job better (for example, a Stalker 4N would receive a bigger bonus than a 3F). Something in the Dragon chassis would have a significantly higher amount of bonus points than a Stalker (i.e. the 3F may get 3 bonus points, the 4N may get 8 bonus points, and a Dragon may receive 20 bonus points).

Skills Not Present

Some skills I have not included in this build. Those skills include (and reasons why they aren’t in there):

• Consumable Skills, because they feel like the game is going towards Gold Ammo (from World of Tanks)
  
• Crit Skills, because I thought they would be overpowered if universal, but too specific a boost if I left them under missiles
  
• Weapon Heat Generation, because they are redundant with Cool Run and Heat Containment
  
• Gauss Charge, because it doesn’t seem to be a very popular thing anyway (longer time before you can reset the charge on a miss)
  
• Magazine Capacity, because I forgot to put it in
  
• Arm Pitch, because arms tend to pitch off-screen already for me
  
• ECM Boosts, because I don’t think the effectiveness of ECM should be put behind a skill wall

Some of these (crit skills, weapon heat, gauss charge, magazine capacity, and arm pitch) I could see the case for putting in, especially Magazine Capacity. The ECM boosts and consumables I think should stay put off the tree.

Thoughts

• I haven’t actually figured out what a good skill point cap would be, although 60 sounds like a good number to me to start with.
  
• Increasing the starting cost (and total point cap, obviously) so the cost is 4-5-6-7-8 (for example), then getting a single 5-tier here would be 30 points, which would be enough to get 7.5 skills to Tier 1 or 3.33 skills to Tier 2 (as opposed to the 1-2-3-4-5 system in which a 5-tier skill is equal to 15 tier 1 skills or 7.5 tier 2 skills).
  
• As I mentioned above, this is just a way to visualize the basic structure. Which skills are included, what value each adds, the cost of each in terms of skill points, the total number of skill points you can get, how many tiers each has, how many bonus points each chassis get, the visual style of the chart, etc. etc. are all up for grabs.
  
• Simply removing the Rule of 3 and changing Pinpoint to a 10% armor buff or -5% crit buff (or similar), among some of the other changes this patch (i.e. mobility decoupling, crit changes) is still a viable option that will have a lot less testing and learning curve for players.
  
• This is my second version of an alternate skill tree that I have posted. The first was a big-decision tree based on WoW. I feel that was sleeker and easier to understand than this one, but this one offers more choice to the player than the current tree on the PTS, while still possibly being easier to figure out.
  
• MWO would be fine without a skill tree. The MechLab offers enough customization, and the C-Bill grind to get new Mechs, engines, and upgrades is enough to give people something to farm. Heck, I'd play it if we just had anything unlocked, because Mechs are fun to play.

I'm curious to hear your thoughts on this. I'd appreciate feedback, although like I said in the first paragraph: if your comment is "you forgot magazine capacity" or "seismic should be 14 points instead of 13", that's detail stuff, and my goal was to show the big picture and the overall functionality of this type of system, and not the specific costs or returns of each skill node or tier.

[Kuaron]

Wow, that’s very very close to what I’d imagine to be the ideal solution.

So very much “yes, this, almost”.

It has free choice, progressing cost, consolidation of weapon-specific effects, balancing of individual nodes (to some degree)*, …

*) This means combining unimportant abilities to less nodes to make them cheaper. This would affect things like arm twist and quick ignition, where 5 nodes are too much to be worthwhile (better condense them to 3), as well as armour plating and cooling and that like, where you could go up to 7 (with less bonus per node).

Edit: Unwanted modules like Hill Climb would have to be much cheaper than popular ones, like seismic, too, not just double.

Edited by Kuaron, 07 March 2017 - 01:30 AM.

[Hestan]

That tree would be about ideal for a skills tree. Simple and straight forward.

The end of the quirkening would also be nice and your solution is intriguing. The really bad performers would end up with a lot of extra points to spend. The one thing your system doesn't address is the differences in IS and Clan tech and attempts to balance that. I think a battle value system would be better for something like that.

[l33tworks]

Kuaron, on 07 March 2017 - 01:23 AM, said:

Wow, that’s very very close to what I’d imagine to be the ideal solution.

So very much “yes, this, almost”.

It has free choice, progressing cost, consolidation of weapon-specific effects, balancing of individual nodes (to some degree)*, …

Dude you just oxy moroned. How can you say feee choice and consildation in the same sentance?

I like the idea of linear skill trees however the weapon specific nodes should INDIVIDUAL not consilidated. This is even worse than pgis implementation because at least with pgis method you can choose some weapon attributes individually.

What are you guys smoking asking for a linear Tree with free choice then having a fully consolidated Weapon skills? The whole point is to be able to individually select cooldown duration velocity range etc to taylor your mech. Otherwisw we may as well stick with the current skill setup bevause consolodating things gives you no choice at all.

[Kuaron]

So what exactly is your problem with free choice of nodes and not free boating-empowerment at the same time?

[ArmageddonKnight]

Ah yes thats much better.

With the excepttion of some missing skills like ammo capacity which u mentioned need to be added.

The number of skill points u should be able to use in a skill tree pike this should be highly restricted.

What we dont want is the abiliy to get firepower, survivabilty , and maneuverability all at the same time.
Idealy you want to be limited to 2 fully decked out areas and half another. So full firePower, full Survivabilty and the rest spread amongst 1 shot skills like ammo capcity,and few points in maneuverability maybe.

[Skribs]

Kuaron, all good points.

Quote

The end of the quirkening would also be nice and your solution is intriguing. The really bad performers would end up with a lot of extra points to spend. The one thing your system doesn't address is the differences in IS and Clan tech and attempts to balance that. I think a battle value system would be better for something like that.

I would say that something like a Spider 5K could still have a heavy energy quirk, for example, but the majority of Mechs wouldn't need them.

IS vs. Clan could be addressed by bonus points or by different values. It's not very well addressed in the PTS skill tree, either.

Quote

I like the idea of linear skill trees however the weapon specific nodes should INDIVIDUAL not consilidated. This is even worse than pgis implementation because at least with pgis method you can choose some weapon attributes individually.

Consolidating them gives you free choice by making them more palatable for a variety of builds, instead of just useful for a UAC build. However, the consolidation was more about making it so it doesn't encourage boating.

Quote

What we dont want is the abiliy to get firepower, survivabilty , and maneuverability all at the same time.
Idealy you want to be limited to 2 fully decked out areas and half another. So full firePower, full Survivabilty and the rest spread amongst 1 shot skills like ammo capcity,and few points in maneuverability maybe.

If you can deck 2 areas fully out, you can get 3s or 4s in most things, with the cost creep for specializing in a skill. This is part of why I think maybe a higher initial cost would be better.

---

Thanks for the feedback, guys. Those who have said "other" would be acceptable, I'm curious to hear your ideas, too!

[Skribs]

It seems based on the small number of votes so far that the consensus is that linear trees with increasing costs per node in the same tree is preferable to the web.

[Kmieciu]

I'm against the elimination of quirks. They add a lot of character to mechs.

Just look at IS Hunchbacks: we all know that the 4G is the AC20 one while the 4H is better suited to an AC10+lasers. Without quirks 4H is outright better for every build that uses a single ballistic (LBX10, AC10, AC20, Gauss).

4SP is the SRM brawler and 4J is the LRM boat, despite both having 2 missile hardpoints.


Same goes for the Centurion-D (LBX autoshotgun) or the Enforcer-5P (UAC5s).

Then there are mechs that have almost identical variants, like Crab, Grashopper, Black Knight. Without quirks those variants are pretty much redundant.

Edited by Kmieciu, 08 March 2017 - 12:52 AM.

[Reno Blade]

I had posted this tree below with similar increasing cost (per tier).
In the end, the layout is the only real difference between the Hex and the linear UI, but it might be easier to use "Role presets" with the hex layout.

If there would be a different cost in different "tiers", it could already limit the amount of total nodes (even with same # of SP available) and also provide more give/take for going deep into one branch vs only few nodes in all branches:

Reno Blade, on 06 March 2017 - 03:07 PM, said:

In general the radial all-in-one tree is not much different than the current split into different branches...
I added the tier-rings, but that can also be done by lines in the current branches.

theoretically similar to current full Elite = 2x basic skills


I would actually take no diminishing return, but 2x basic bonus for previous tier once unlocking next tier.
Let’s take the example here.


Skirmisher preset uses
5 red (operation) nodes unlocking tier3 operation bonus
3 white (sensor) nodes unlocking tier2 sensor bonus
10 purple (firepower) nodes unlocking tier4 weapon bonus
12 blue (mobility) nodes unlocking tier4 mobility bonus

you could have such levels:

Unlock first skill point (tier1) cost 1 SP
1% (node1) x 1 node = 1%

Unlock second skill points (tier2) cost 2 SP
1.5% x 1 (node2)
+ 1% x 1 (node1) x 2 (tier2 bonus)
= 3.5% for 3 SP

Unlock third skill point (also tier2) cost 2 SP and then the forth (first node in tier3) for 3 SP
2% (node4) x 1 node
+ 1.5% x 2 (node2 and node3) x 1.5 (tier3 bonus)
+ 1% x1 (node1) x2 (tier2 bonus)
= 8.5% for 11 SP

Now unlocking two more nodes (#6 in tier3 and #7 in tier4) for 3+ 4 SP would unlock the Tier4 bonus.
2.5% x1 (node7)
+ 2% x 3 (node4, 5 and 6) x1.2 (tier4 bonus)
+ 1.5% x 2 (node2 and node3) x 1.5 (tier3 bonus)
+ 1% x1 (node1) x2 (tier2 bonus)
= 16.2% for 18 SP

This is already some kind of diminishing return even with the bonus for tier unlocks.
Any further and you would spend a lot of SP to gain only 2% for 4SP (tier4).

Ofc there are more than one skill attribute to skill in each section which would make the larger % per node actually look fairly good, as it’s split between multiple attributes (e.g. operations using heat capacity and heat dissipation).

using current weapon tree, it could have tier levels, so it could lok like this:

Spoiler

[Skribs]

Kmi, there are two different examples you provide: 4G vs. 4H and number of hardpoints, and 4SP vs. 4J vs. types of weapons quirked..

The 4G vs. 4H issue would be solved with the bonus point system. The 4G would receive additional bonus points to compensate for the lack of 2 energy hardpoints compared to the 4H. So while the 4H may receive 10 bonus points (for example), the 4G would receive 18.

As to the 4SP vs. 4J, I don't like the fact that on the 4J the only weapon I can use and feel like I'm making use of the Mech is LRM10s. I don't like the fact that the choice in variant basically makes my choice in weapons for me. What if I want to run LRMs on the 4SP? What if I want to run SRMs in the 4J and use my left side as a shield? They have different numbers and locations of energy hardpoints as well, where the 4J has more firepower potential and more high-mounted weapons in the hunch, but the 4SP has more energy on the arms.

So, there's reasons to own different variants besides the quirks, and my system would replace the performance boost provided to Mechs that currently need the quirks. You also wouldn't see a Hunchback 4J and immediately go "LRM boat" because you actually have a choice in what you put on it.