I want to start this off by saying “thank you” to PGI for delaying the skill tree implementation until it will be ready. I do think there are some major issues with the current skill tree (mainly: the Rule of 3, which I thought in early beta was bad for new players especially, and the Pinpoint skill), and there are other issues that people would like to solve with a skill tree (such as TTK) that can be addressed with or without the tree.
I have seen several ideas promoted for alternatives to the skill tree that was presented. The two biggest issues with the proposed tree is the web-like structure (which players feel limits their choice potential) and the price (which module-poor, mech rich players like myself feel is unfair). I am going to propose several different ideas, some of which are congruent with each other, some of which are standalone ideas, which will give options to consider when looking for a new version of the skill tree.
Some of these ideas I have really fleshed out, others I’m sort of thinking out loud…in text.
At the end of this post I will also share ideas for new modules (because half these ideas keep the module system), my thoughts on costs of the skill tree, how it can affect balance regarding Quirks and the 3060 update, share my order of preference in terms of the trees (I am presenting them in a more logical order from no skills to current skills to proposed skills to deviations from the proposed tree), and then discuss any follow-up.
Option 1: No Skill Tree
Option 1 is simple: remove the skill tree and remove XP from the game entirely. The steps in this process include:
- If possible, refund any MC that was spent on XP->GXP conversion. This isn’t necessary (as it paid for a convenience at the time), but this is from the perspective of someone who hasn’t spent too much MC on the conversion. If not possible, or determined that the quicker progress was worth it, then don’t bother.
- Eliminate XP. There could be a C-Bill refund based on XP (i.e. every XP spent or banked on each Mech, every GXP spent on pilot skills or modules could grant 1 C-Bill. It wouldn’t be a huge refund, but it would be something).
- Make all current Mech skills baseline when you buy the Mech.
- Make all current modules fully unlocked, but still require the C-Bill purchase price.
- Give Champion mechs another bonus. Options I can see here include a 10% or 15% C-Bill bonus, extra reputation bonus when used in Faction Play, or free upgrades (i.e. endo, artemis, ferro, DHS).
- Change any XP or GXP rewards to C-Bill rewards
How New Players Would Appreciate Having No Skill Tree
- No more Rule of 3, so the first 4 Mechs you get are ready to go into Faction Play
- They wouldn’t start the game behind in leveling, and could jump in with Mechs that are as capable as any other Mech
- They wouldn’t have to worry about unlocking the wrong module with GXP, even though they have to buy it still
How Veteran Players Would Appreciate Having No Skill Tree
- Removing Rule of 3 means I can buy a new Mech without having to buy 3, so trying new Mechs is less of an investment
- I consider XP to be a gate to buying new Mechs, in that if I’m leveling things, I don’t want to buy new Mechs that will either sit in my Mechlab or will make others sit in my Mechlab. It feels wasteful. If I don’t have an XP gate, I might buy more Mechs or Mech Packs
- When I buy new Mechs, I’m not dropping into Tier 2 matches with a Mech that’s just starting on its basics
Why PGI Would Benefit From Removing Skill Tree
- I believe more new players would stay as it would eliminate the startup costs and some of the learning curve
- I believe veteran players would be more willing to buy Mechs (including Mech packs, hero mechs, and MC for Mechbays)
- The only lost MC expenditures would be from XP->GXP conversion, but that would be made up for with more players playing and players buying more things
Option 2: Fix the Current Skill Tree
When is it time to buy a new car? When you have a dent and need a new set of tires, or when you’ve sunk thousands of dollars into all sorts of engine problems? Currently, I think the skill tree fits the former category. There’s 2 major issues with it, but those are both solvable without a new skill tree system. So, rather than getting a new “car” and going with a whole new skill tree, we could just fix what’s wrong with the current one. This could even be a stopgap measure.
Option 2 is REALLY simple (actually even simpler than removing XP and skills entirely). Here’s the steps:
- Eliminate Rule of 3, and simply unlock Elites if the variant has unlocked Elites
- Change the Pinpoint skill to a 10% armor buff.
Option 3: The Proposed Skill Tree, With Modifications
The tree proposed by PGI certainly had problems, but it also had a lot of promise. I’m going to break this up into several suggestions I’ve seen that use the basic model presented by PGI, but attempt to fix the issues with it, or give it a little umph that would make the players feel more in control. This option will include several sub-options, some of which may be compatible with each other. All of these options keep a web design with a lot of points for the players.
Option 3A: Keep modules as items, and just use XP on the skills
Two of the big issues I had with the proposed skill tree was the cost associated with it, and the fact you could get all the sensory modules. The suggestion here is:
- Remove Weapon Modules and maybe a few others and refund the C-Bills for those
- Keep most Mech modules as unlocks that you purchase for your Mech
- Have a skill tree that only includes three “branches”
- Firepower
- Defensive
- Mobility
- Firepower
Essentially, it’s half of the proposed skill tree, with the modules intact. So a hybrid of the module system and the PTS build.
Option 3B: More pathfinding options in the trees
One problem a lot of people had is they felt like specific filler skills were requirements to get other skills. It might not be too much of an issue in some trees, but (going back to the one I had the most problems with) the sensory tree seemed to give you too much. If you wanted Target Decay or Radar Derp, you pretty much got them both.
There’s a few different ways that pathfinding (navigating from one point to the next) could be improved, which would still require you to choose “filler” skills in order to progress down the tree. I’ll propose a linear option later.
Option 3B-1: Vertical Freedom
Allow us to move up or down the tree, instead of only down. This will still require an investment on our part, but would give us more freedom in which filler to get
Option 3B-2: Different Starting Points
Another option is to have a different starting point or shape than what we currently have, and obviously something a little bit easier to navigate from point to point. This could manifest as:
- A web that you start in the middle of, similar to Path of Exile, but obviously a LOT smaller in scope, where the juicy skills are on the perimeter
- A web that you can start on multiple edges of, similar to the Pandemic mobile game (you could start in the corners or on the top or bottom and work your way towards the center), where the juicy skills are in the center
Option 3C: Combine Weapon Skills That Are Similar
There are two levels this could take:
- Combine weapon skills like SRM spread and LRM spread, which are literally just the same skill applied to different weapons.
- Combine skills with similar purpose, for example:
- Velocity and Laser Duration, which both increase hit chances
- Spread and UAC Jam chance, which both increase reliability
- Velocity and Laser Duration, which both increase hit chances
Option 4: Linear Tree With Gating Built In
Have a linear tree, but have a requirement to move from tier to tier (similar to how in the original WoW trees you had to invest 5 points in a tree before moving to the next tier).
Option 4A: Linear, Gated Tree; Simple
The simple version of this could work if you had several skills in the Firepower tree that each had 5 ranks, such as Cooldown, Heat Generation, Velocity, UAC Jam chance, SRM/LBX Spread, Laser Duration, High Explosive, Ammo, Weapon Range. In order to get Rank 2 of anything, you must have spent 4 points in the tree. In order to get Rank 3, you must have spent 8 points, and so on. Things that require a greater point investment (like laser duration, range, or cooldown) might have 2 skills every tier.
In this way, you couldn’t simply spend 10 points to get 10 cooldown skills. Maybe you start off with 2 CD skills and 1 each of Laser Duration and Heat to unlock Tier 2. Then you do the same for Tier 3. So by the time you get 10 ranks of Cooldown, you also have 4 ranks of Laser Duration and 4 ranks of Heat. It’s linear, but it’s gated so that you still need filler.
Now, obviously the numbers don’t have to be the same, but they can be.
Option 4B: Linear, Gated Tree; Complex
The more complex version is to have different skills at different tiers. Mobility would be a prime example of this. Here’s what the different tiers might have available:
- Arm or torso twist, pitch, speed; hill climb
- Arm twist, torso twist, pitch, or speed; braking; hill climb
- Torso twist, pitch, or speed, braking or accelerating
- Torso speed, braking, accelerating, or anchor turn; speed retention
- Braking, accelerating, anchor turn; speed retention
- Accelerating, anchor turn, and speed tweak
- Speed tweak
This is just an example of how this can work. As you can see, things that are less popular are easy to get, while things that are more popular would be buried further in the tree.
Option 5: Linear Tree With Expanding Costs
Like the tree I proposed here, but with some updates I’ll share in this post (based on discussions in the thread):
https://mwomercs.com...ear-skill-tree/
There would be 3 types of skills: major skills, minor skills, and modules:
- Major Skills would start at a cost of 6 SP, and increase by 1 up to 10 SP at Rank 5. This way, a single skill at Rank 5 would cost 40 points, while you could get 6.67 skills up to Rank 1 with 40 points. (In the original design, where it went from 1-5, Rank 5 would cost 15 total points, where you could get 15 skills up to Rank 1).
- Minor Skills would be skills that are either less useful or less universal. Jump Jet skills could be one, because not all Mechs have jump jets, so you don’t want people to devote too many points to that. Others could be things like arm quirks or quick ignition, which aren’t as potent. These might start at 5 and go up to 6 and 7 SP, and only be 3-point skills.
- Prime Skills could be another one, where you take big skills like cooldown, heat, or armor boosts and increase them up to 7 skills.
- Modules would be binary yes/no options and would cost a little or a lot depending on which module.
- Hill Climb would be pretty cheap, probably around 5-8 SP
- Advanced Zoom would be moderately expensive, maybe 20-25 SP
- Seismic Sensor or Radar Derp would be expensive, at around 40-50 SP
- Hill Climb would be pretty cheap, probably around 5-8 SP
- Total skill points would probably be around 200 or so (I haven’t really built and tested this, so I don’t know the right number) to allow for a few max-rank skills and a few big modules.
While filler can be a boost compared to investing more skill points to just get what you want, players generally feel better about this. While this isn’t my personal favorite suggested option, it seems to be a community favorite based on what I’ve read and the polls I’ve seen (use my linked post as an example).
Option 6: Big-Choice Skill Tree
I proposed a modern WoW-style tree here:
https://mwomercs.com...ll-tree-design/
As I said above, we don’t need detailed customization in the skill tree, because we already have it in the MechLab. This would give us a new skill tree where each skill feels like a huge boon.
I don’t think I need to speak too much on this one, and I’ll let the graphic in that thread speak for itself. If you’re too lazy to follow the link, the general idea is that you have 3 or 5 possible points in each of several categories, and you can spend 1 point or 3 points (respectively) in each category as you level up. This way you could choose a specific firepower boon (easier hits, faster hits, or more cooling), a survival boost (radar derp, armor boost, or component and equipment health), sensory boost (target decay, seismic, 360 targeting, advanced zoom, target info + sensor range), etc.
This one has the benefit of being the simplest option (besides no skill or the current no-option tree) for new players to learn, or for returning players to re-learn. It’s the simplest number of clicks to respec. It should be easier to balance as well.
Option 7: Sliders with a Sidegrade
The last option would be based on games like Need for Speed (2015) tuning style, rather than a pure upgrade skill tree. This could work as either an XP-less system (which we would need to do something similar to Option 1), or it could work with XP, but I’ll explain the XP system in a minute.
The Basic Slider System
The slider system would work by giving you the option of tuning specific aspects of your Mech. There would be base stats you could adjust, as well as sliders for every piece of equipment. Now, this isn’t a fully fleshed out idea at the moment, and these are just examples. So if a specific slider seems dumb, please don’t condemn the idea because of a specific detail that could be ironed out. Anyway, here are some ideas of what you could have base on the Mech:
- 10% Armor vs. 15% Acceleration + 25% Deceleration
- 10° Torso Twisting vs. 8% Speed
- 20% Hill Climb vs. -30% Fall Damage
- 10% Heat Dissipation vs. 20% Internal Structure
- -50% Improved Gyros vs. 10° Torso Pitch
- Something that is less popular can be balanced against several other things. For example, Improved Gyros might get up to -80% bonus, but it would be balanced against 3 different things to get that far.
- Something that is very popular may only get small bonuses, and require a lot of sacrifice to get there. For example, you may only get 3% speed for going all the way towards speed on a slider, but there’s 4 speed sliders, so you could sacrifice a lot to get a 12% speed boost.
The kicker is that every piece of equipment you add to the Mech (besides the engine) would give you additional sliders. Those sliders would be specific to each weapon class (so a 6x ML boat would only get 1 set of ML sliders that would affect all 6 medium lasers), but could be different based on the weapon.
- Lasers might get multiple sliders to balance range, heat, and cooldown. You would have range vs. heat, range vs. cooldown, and heat vs. cooldown. You may focus on range and end up with a 20% range boost, but 10% more heat generation and 10% longer cooldown. Or you may sacrifice one of those 10% nerfs for the other, and end up with 20% range, heat neutral, and 20% longer cooldown. Or maybe you just take a 10% cooldown buff at the cost of a 10% range nerf, and leave heat alone. Maybe you keep everything the same as it came. Maybe you only adjust a couple sliders 5% each instead of 10%.
- Lasers might be grouped individually (i.e. only Medium Lasers), by size (Small/Medium/Large), or by type (Pulse or Regular, or Pulse/Regular/ER). Similar for auto-cannons. Missiles would most likely just have SSRM, LRM, and SRM groups (until MRMs and ATMs come out).
- PPCs may be able to trade minimum range for effective range (i.e. +20% minimum range to get +10% effective range, or -20% minimum range for -10% effective range).
- ER PPCs (and PPCs) may be able to trade maximum range for heat.
- An AC5 may be able to trade range for ammo/ton.
- A UAC5 may be able to trade jam chance for cooldown.
- An LBX10 may be able to trade spread for critical hit chance.
- Machine guns may be able to trade range for critical hit chance.
- LRMs may be able to adjust the trajectory for higher angle or faster time to target.
- SRMs may be able to trade spread for velocity.
- SSRMs may have a toggle that makes 1 missile from each launcher focus on CT, but increase cooldown by 20%.
- Jump Jets may have a slider for jump height vs. jump speed.
- ECM may have a slider for disruption power (i.e. disrupting enemies you are near) vs. coverage power (i.e. hiding or providing ECM cover for assaults)
- NARC may have a slider for velocity vs. duration
- TAG may have a slider for range vs. uh…I’m not really sure here.
- Seismic may have a slider for max seismic range vs. max speed you can move while seismic is active
- Advanced Zoom may have a slider for size of the PIP scope vs. zoom level
The Slider System, as Locked by XP
I was using Need For Speed as my example, and in NFS, your level determined how far you could move sliders. For example, a low-level player with a stock part may have no tuning options for acceleration vs. top speed, or may be able to make small modifications, but a max-level player with end-game parts could potentially get huge differences.
Similarly, if you have something that gives you the option to balance 10% range bonus on one end and 10% less heat generation on the other, at 0 XP (just bought the Mech) you would get no option to trade off. At 8000 XP, you might unlock the ability to change it 1%. At 16000 XP, you might unlock 2%. All the way up to 80K XP, where you can change the whole 10%.
This would allow players to step into a Mech with 0 XP and be theoretically balanced with everyone else. They may not have the Mech tuned the way they want, but at least they’re not going in underpowered. The experience points you earn with the Mech allow you to move the sliders further and give you more customization.
It would also give new players the ability to learn the system slowly and make bigger decisions once they have more literal and game experience.
New Modules
A lot of these systems, particularly Options 1, 2, 3A, and 7 still include modules. Ideas that found their way into the proposed skill tree on the PTS could find their way into modules, including:
- Magazine Extension (probably Mech Module that would affect Missiles and Ballistics)
- Jump Jet module to increase JJ distance and/or speed
- Armor module to grant 10% armor
- More robust Weapon modules, to include Heat, UAC Jam Chance, Spread, Missile Crit Chance, Laser Duration, Velocity, and possibly others
- Generic Weapon Modules (i.e. 5% weapon CD instead of 12% to a specific weapon)
- Incoming Crit Reduction
On Costs
I understand that PGI needs us to spend money on the game in order for the game to survive. Not only do they have to eat and pay for their programmers to eat, but there’s also the utility costs to keep the servers running. I get that. However…
- The costs should be costs that encourage us to pay for perks, rather than feel like a punishment if we don’t. For example, it’s fine for a car dealership to charge extra for premium decals or for more horsepower, but I think most people would be upset if you saw a car listed for $15,000, and the drivetrain was sold separately. If the skill tree feels too expensive, players will feel the grind is punishing, not rewarding. Too many C-Bills tied into the skill tree will push players away.
- There is a difference between increasing the XP cap and making us touch all of our Mechs again to top them off, vs. locking our XP behind a C-Bill paywall and making us play a handful of moneymakers to re-level all of our Mechs. There is also a difference between increasing the XP cap to maybe make us touch some of our Mechs again and significantly increasing the XP cap so that we’re stuck in a hole.
- Grind without content is just more grind. MWO doesn’t really feel like an RPG, and in many ways is different from an expansion being released in an MMORPG. MWO already has plenty of grind, especially for us players who play the game like Pokemon (gotta catch em all). Don’t punish us for the sake of giving players who are content with what they have more to grind.
- Even if the refund we get doesn’t cause us to lose progress, if it slows down our progress that will still hurt our playstyle.
- Module prices were considered excessive when released, but the fact that they can be swapped is the only reason the community was okay with them.
Instead, let new players get started, and let them pay for conveniences and faster progress. I’m sure far more players would play the game in this case.
On Balance
Looking back at the Quirks and ahead at the 3060 Tech Update, it’s clear that MWO is in for some changes. The weapons themselves will probably take at least a few passes to get right, and a lot of quirks will have to be redone, if nothing else than to make the new weapons fun. Bigger picture, IS seems to be getting a lot more power relative to clan. This isn’t a bad thing, as unquirked IS would generally be weaker, but will require a reworking of the quirks to balance them out again.
With that in mind, the nerfs to the quirks in the PTS make sense, but I don’t think it was well communicated to the players that this was the idea. Whatever the case, there are two areas of balance that need to be preserved:
- Inner Sphere vs. Clan, for the fundamental purpose of Faction Play
- Underperforming vs. Overperforming, for the fundamental purpose of making the Mechs that aren’t as good worthwhile to bring
Personal Preferences
I presented 7 options, although including sub-options it’s more depending on how you’re counting it. So which options would I prefer? In order from least favorite to favorite:
#8: Proposed skill tree with more pathfinding options (Option 3B). I feel this leaves the most to be desired, as it leaves the majority of the problems with the skill tree without fixing new ones. This one in combination with keeping the modules is a good idea, though.
#7: Fix the Current Skill Tree (Option 2). I think this should be done in the interim, but if I’m going to grind XP I want some choice in how I spend it. It would still be the easiest for PGI to implement, and isn’t a bad stepping stone towards fixing the new tree. It would also eliminate all of the “yeah, but Rule of 3 is gone” arguments from the threads and we could focus on the tree itself.
#6: Linear Tree with Gating (Option 4). Like the proposed tree, this one leaves a lot to be desired in terms of
#5: Keep Modules as is, and simplify the PTS Tree (Option 3A). I like this one for three reasons: 1) it eliminates the question of what to do with the C-Bills spent on modules, 2) it keeps the trees within the triad of Firepower-Defensiveness-Mobility, and 3) it allows for the tree to be a bit simpler than what is currently on the PTS.
#4: Big-Choice Skill Tree (Option 6). This tree is sleek and easy to understand, and there isn’t a lot in it to balance. It does have some criticisms and might take a bit of work to get right, but would also be easier to build and test than any of the other options (except for Option 1 and 2).
#3: Linear Tree with Expanding Costs (Option 5). This seems to be the most popular choice, and it would give players a reason to not go all-in on a single tree. That seems to make it more of a gray area choice instead of a yes/no choice, which makes it feel a lot more like the player is in control, rather than picking a cookie cutter made by PGI.
#2: Sliders with a Sidegrade (Option 7). This one is nice because Mechs aren’t underpowered when underskilled, they are simply given less choices. It will work out well for new players, and like Option 3A it is not incompatible with the module system. I also feel it’s what most players who say “the skills should have tradeoffs” want.
#1: No Skill Tree (Option 1). This is honestly my favorite one, because it eliminates one grind completely from the game.
Conclusion and Invitation for Follow-Up
I have included links to Option 5 and 6 because I have already done those mockups. I haven’t done one for option 1 or 2, because there isn’t a whole lot to show there, and option 3A, 3B-1, and 3C change too little from the PTS. Option 3B-2 is kind of beyond my power as a graphic designer. I could probably show option 4 or 7. If you want to see a visual representation of either of those, let me know.
If you have any other ideas, I’d be interested to hear them. This may be a bit of a long thread, but I thought it was worth it to put all the ideas I’ve seen into one place.
Once again, these are macro-level ideas. I don’t need feedback like “you should have 1% on this skill instead of 2%”. I need feedback like “this is the kind of system I like” or “here’s how they could make it work.”