42 replies to this topic

[EvAbsence]

Rationale Behind Proposed Changes

Woefully simplistic and completely arbitrary, the current state of the Mech Skill Tree is universally agreed upon to be one of the least polished aspects of the game. Since closed beta it has only undergone a single alteration - the reduction of percentage bonuses awarded from each skill. (Ok, two alterations if you want to include the admission that Pinpoint doesn’t actually do anything…) In the meantime, the Pilot “Skills” menu has gotten more crowded with the inclusion of weapon specific range and rate of fire modules for both Clan and IS, modules to improve consumables, and modules to improve your Mech’s battle-awareness capabilities (sensors).

The Issue: Pilot Skills = Mech Modules
This is a fundamental problem with the current system. It implies that a pilot skill (something that a pilot could theoretically improve through experience or practice) is the same thing as plugging in a new video card to your computer or swapping out fans for liquid cooling. Sure, you can argue that it takes some experience to know how to plug in and install a new video card or heat block, but it is also a well-documented process that anyone with a little time, money, and the ability to read could figure out. A pilot skill should be something that is representative of the pilot’s improved ability piloting Mechs.

A second issue is that the Mech Skill Tree is supposed to represent the pilot’s familiarity with the Mech they have chosen to pilot. It is safe to assume that with increased experience in a specific Mech a pilot would have more experience aiming, maneuvering, and… making the Mech move beyond the previously established mechanical maximums?...

A good example of this is Speed Tweak. Before you have unlocked Speed Tweak, if you put your throttle to maximum it shows 100% and you go a set speed. Once you unlock Speed Tweak, if you put your throttle to maximum it shows 100% and go a different set speed… What happened there? If Speed Tweak is supposed to be a piloting skill representative of familiarity with the Mech, shouldn’t it read 107.5%? Or to make even more sense, the initial, non-Speed Tweaked maximum should have read 92.5%, representing that the machine has the capability to go faster, but you, the pilot, can't handle it. I'd be all for this if you could override this limit and risk your Mech randomly falling over, but this isn't in the game currently.

Previous games like Mech Commander or the BT days, where you were not directly responsible for the details of targeting or movement, represented pilot skill in increased AI competency or numeric values that allowed the Mech move efficiently, target quickly, and improve RNG weapon accuracy.

Summary of Rationale and Conclusions

THE EXISTING SYSTEM DOES NOT WORK FOR MWO
The player is the pilot. I am the pilot. We don’t need a Pilot Skill tree, and we don’t need the Mech Skill Tree to determine the potency of, or the pilot’s familiarity with the mech.
My improved piloting ability is represented in the game through MY IMPROVED PILOTING.

With that cleared up, what do we have left? MECH UPGRADES

Proposed Mech Upgrade Paths


****Disclaimer*****
Obviously the upgrade paths displayed above are a draft, and were created solely on my own opinions with refinement from my friends in C4. I don't claim to know every intricacy in the game, but I feel I am well enough informed to make some suggestions. I welcome questions or concerns in the comments, so long as they are constructive, and will monitor this thread and update with answers as needed.

As previously discussed, these represent physical or software upgrades to each Mech and not piloting efficiencies. It incorporates virtually every ‘skill’ from the Mech Skill Tree as well as every Pilot Skill that can be unlocked with GXP. (Notable exceptions are Enhanced NARC, Cool Shot 9x9 and Cool Boost, which I feel should be removed and their bonus values set as the standard.)

How It Could Work

This concept is universally recognizable to anyone who has played a game with established player ‘Roles’, and that is a term I know I’ve heard Russ mentioned he wanted more of in MWO.

You are awarded a set number of upgrade points acquired through playing the specific Mech variant in much the same way that Mech XP is acquired now. The idea is that you will be given enough points to make it most of the way down two upgrade paths – Offense with most of Defense for instance – but that to do so you would have to forgo access to desired mobility and information bonuses. You’d be tanky and hit harder than others in the same variant built the same way, but you are less maneuverable have less battle awareness when alone. This trade-off is what will create defined roles in MWO.

Using a gated access system, you can invest your points selectively to pick specific upgrades that compliment your build or your role, diversify your build to be a jack of all trades but a master of none, or you can focus your efforts on one path with a smattering of other upgrades from other paths. Access to these tiers would rely both on the number of points invested into the chosen path, as well as how many of the three required variants have invested points to unlock the next tier. (Think the current basic to elite skill access restrictions.) The last tier of each upgrade path should hold the upgrades that are most powerful in order to restrict access on that Mech unless they have invested enough in the path. To reinforce this, reaching the last tier of a path using the minimum point investment will gain you access to ONE of the three ‘Master Upgrades’, while maxing out the path will gain you access to two. The center-most master upgrade is the most in line with that path, while the two that overlap with adjacent paths can be accessed by the respective overlapping path.

Currently, it takes 57,250 Mech Experience points to fully master a Mech and unlock the Master Module slot. Conveniently, a quick starting calculation of a limit to how many upgrades a Mech should be able to hold ended up at 57 Upgrade Hard Points. An easy transition would be to award the maximum of 57 Upgrade Hard Points to anyone with 57,000 or more XP on a given mech. For XP values over that amount, the remaining value could be converted to GXP (at some exchange rate) to be used specifically on that Mech for returning invested Mech Hard Points so you can ‘Re-Upgrade’ your Mech. This would be done using increasingly expensive scaling to a cap (TBD) depending on how many times you reinvent your build. This will help prevent less experienced players from suffering GXP losses they don’t have when they are learning how these skills effect gameplay. GXP currently awarded in game can be used in the same way, but from a central pool you can draw from for all Mechs.

C-Bill costs associated with in-game module purchases may also want to be incorporated somehow, but that is a question for the PGI devs to figure out with respect to the game economy as a whole.

The other benefit of this system that I see is the removal of quirks from the game as we know them. If a Mech is deemed to be an under-performer due to loadout limitations, the variant base values can be modified by PGI and displayed in a clear and comparable way vs other similar Mechs. (Can we get some bar graphs please?) Special variant specific traits, like the Hellslinger's 50% environmental heat transfer could be a unique 'Attribute' to the Mech and displayed with whatever specific functionality upgrades you have unlocked (like seismic sensor). Alternatively, take that 50% heat transfer off the Hellslinger and make it an all new Mech Upgrade node.

Currently Mechs are pigeon-holed into specific loadouts because of their assigned weapon quirks. The Hunchback 4SP should never be run as an LRM boat in the current system, and why would you when you have the 4J sitting there two bays down? The reasoning for such things were pulled from lore, but don't make much sense in a game where you should be given full freedom to build how you want. If you want to encourage a specific loadout in a particular variant, change its stats to make it less capable in the role you don't want it in. Don't give it a LRM buff - give it a mobility nerf. Now it doesn't make sense to build it as a brawler/striker, because it will pale in comparison to others in that role, and rightly so. This will prevent the ridiculous boom and bust in weapon specific quirks we have seen over the years, not to mention make it a lot easier to anticipate how fast that AC10 is going to fly (because it isn't different for every freakin' Mech anymore!)

Build Examples

Using the draft Upgrade Tree linked above, I have compiled a list of 4 examples that can be found below. Hopefully this will demonstrate how this system could be used to further define Mech roles.

Front-Line Pusher:

42 Defense, 2/3 Master

15 Offense, 0/3 Master

Profile: High defense, medium damage sustainability, low mobility, low battlefield awareness

Defense

Arm/Leg Armor 5/5

Arm/Leg Structure 5/5

Side Torso Structure 5/5

Side Torso Armor 5/5

Reduced Crit Chance 5/5

Center Torso Structure 5/5

Center Torso Armor 5/5

AMS Overload 5/5

Glancing Blow 1/1

Improved Gyros 1/1

Offense

Cool Run 5/5

Fast Fire 5/5

Heat Containment 5/5

Back-Line Sniper:

26 Offense, 1/3 Master

26 Mobility, 1/3 Master

5 Infotech, 0/3 Master

Profile: Medium-high offense, medium-high mobility, low defense, low battlefield awareness

Offense

Cool Run 5/5

Fast Fire 5/5

Heat Containment 5/5

Improved Range 5/5

Improved Cooldown 5/5

Pinpoint 1/1

Mobility

Kinetic Burst 5/5

Twist Speed 5/5

Arm Reflex 5/5

Improved Jumpjets 5/5

Speed Tweak 5/5

Shock Absorbance 1/1

Infotech

Target Info Gathering 5/5

Front-Line Striker:

17 Offense, 0/3 Master

30 Mobility, 0/3 Master (One Available Unused)

10 Infotech, 0/3 Master

Profile: Medium offense, high mobility, low defense, medium-low battlefield awareness

Offense

Cool Run 5/5

Fast Fire 5/5

Heat Containment 5/5

Improved Cooldown 2/5

Mobility

Hard Break 5/5

Kinetic Burst 5/5

Anchor Turn 5/5

Twist Speed 5/5

Twist X 5/5

Speed Tweak 5/5

Infotech

Target Info Gathering 5/5

Adv. Seismic Sensor 5/5

Light Scout:

20 Mobility, 0/3 Master

37 Infotech, 1/3 Master

Profile: Low offense, medium mobility, low defense, high battlefield awareness

Mobility

Kinetic Burst 5/5

Twist Speed 5/5

Anchor Turn 5/5

Speed Tweak 5/5

Infotech

Target Info Gathering 5/5

360 Target Retention 5/5

Adv. Target Decay 5/5

Adv. Seismic Sensor 5/5

Capture Accelerator 5/5

Adv. Sensor Range 5/5

Radar Deprivation 5/5

Improved UAV 1/5

Advanced Targeting 1/1

Some Fine Details and Answers to Your Questions (Will be Updated)

Some aspects of this system will need adjustments and refinements. Certain upgrades as I have them won't necessarily work after testing. Others may be reduced or increased in strength. This will all need to undergo balancing against the current Mechs in game to find out where it breaks. Decisions will need to be made: Adjust the paths or adjust the Mechs that break it. I will continue to address questions or concerns posted in response to this thread in this section, assuming I have a good answer - and if the question/concern has merit.

Bye-Bye Weapon Modules:

No more weapon modules needed. You want to upgrade that medium laser range? Go down the Offense path. Obviously this will change how modules will be handled on each Mech, and the quantity of those modules. In a perfect world of my creation, the only modules you would have to equip would be consumables.

Energy Draw:

None exist now, but any implementation of this system should also, obviously, include the energy draw system.

Pinpoint:

This has been suggested by many other people, and in the context of this upgrade system it is even more relevant... Make pinpoint do something, such as allow for weapon convergence at a sighted distance even without weapon lock (ie. how targeting works now.) If you don't have pinpoint, have a cone of fire that your weapons might point in independently of the rest, or just have they fire straight from the Mech as if you were aiming at infinity. The cone of fire option is already implemented to some extent with screen shake while jumpjetting, while the infinity option would still allow skilled players to put damage where they want, but it will be harder for you the wider your Mech is. You could still put damage on target without it, but not specific components at significant range unless you really know your Mech (ooh pilot skill there!). A must-have for snipers, but a not-so-necessary for brawlers. It would also increase the value of scouts and spotters for those without the upgrade.

Access to Current Mech Modules:

The tree mentions Adv. Seismic Sensor in the Infotech path, but I'd suggest it be a tiered access to Seismic Sensor as a whole, as well as other similar Mech Modules. The reason Infotech doesn't exist in the game right now is because you can put Infotech equipment on EVERY MECH, save for equipping ECM. There is NO REASON to be a light spotter in the game when any streakcrow can sub out a SSRM 6 for a NARC and help the whole team out. You have to limit information on the battlefield if you want scouting to mean anything, and you have to provide adequate rewards for doing so.

Closing the Gap: Leveled vs Unleveled

Because the opponents you face are determined by the skill tier system in the game, you could be facing someone in the same Mech with the same loadout as you that is literally better than you in every way. The difference in capability is significant too. By limiting access to every bonus you previously saw under the current skill tree you would be reducing the disparity in power between like Mechs. Now isn't that more fair for new players, and a benefit to new player retention?

Adding New Modules/Upgrades:

Who wants to see Mines? Maybe the ability to call in a squad of elementals? Orbital lasers? All possible under this new system too - as new Master Upgrades. They could take the form of upgrades to provide They would be placed in the path that makes the most sense for the upgrade. Mines - Defense. Elementals - Offense. Orbital Lasers - Infotech. Upgrades can be added to the paths to create interesting effects if the devs so choose... like, high efficiency ammo storage for a 10% ammo capacity boost down the offense path, or refractive paint to help deflect some of the damage from lasers in the defense path.

Edited by EvAbsence, 16 May 2017 - 04:46 PM.

[Kyrie]

I like the concept, and the pretty graph, but the values in it might be a tad broken. As an example, 50% bonus heat containment is... stronk. Very Very stronk.

Things like skill trees are in my mind highly problematic in a pvp centered game for two reasons:
-It places new players at a serious disadvantage
-They become indispensable
-Unless really really carefully implemented they create less variety than would seem as people figure out the min-max choices offered by the system.

At the present time the current skill tree grants some significant but not necessarily soul-crushing advantages. Your suggested values would indeed be soul-crushing. :-)

I am honestly not sure how a proper pvp skill-tree system should be designed. I am just very concerned about implementing a system that forces certain choices to be made that act to limit rather than expand game-play choices. I am also quite concerned about how this would create additional gaps in the new player experience, or further issues in match-making (i.e., separating people by mech class, skill and now also mastery levels?!).

[El Bandito]

Have a like, just for the effort.

[EvAbsence]

Good comments, though like I mentioned, all values are completely up for modification and I do not claim to be an expert on game balance.

I honestly feel that the concerns you've expressed are already present in the game: This would only serve to narrow the gap between a fully leveled mech and an unleveled one by preventing a fully leveled one to outclass an unleveled one in every regard; Our min-maxing environment already exists in the form of metamechs, and would only deepen in complexity with an upgrade tree like the one proposed; and matchmaking couldn't be affected any more than it already is, since the amount a mech is leveled is not considered with pairing you with another player in your tier range.

Currently every mech is a do everything mech, but if you wanted, in this system, you could build something completely ridiculous under the current meta and it could be viable enough that you could get away with it. At least that is what I believe.

Thank you for the comments!

[Suko]

Anything beyond the shell of a system we have now is good. You get a like from me just by bringing an educated alternative solution to the table.

[RestosIII]

He expects them to announce something that isn't a hack job or them releasing a new game instead of giving MW:O content?

[Juodas Varnas]

I, for one, just want the entire system GONE.
Just GONE and DEAD (preferably with concrete shoes sitting on a bottom of a lake).

And the consumables alongside them (if not, then at least merged into the actual equipment, BAP, Command Console, TAG, etc).
To hell with modules too. In my opinion, EVERYTHING that affects the mech and its performance should cost tonnage and critslots. If it doesn't? It shouldn't exist.

Edited by Juodas Varnas, 07 September 2016 - 09:17 PM.

[kesmai]

Skilltree needs to be bound on the player not the mech.
It needs to be based on matchtime/performance on a weight class base for the player.
All weapons modules need to go.
All other modules need to be changed to equipment with weight and slot cost.
Consumable should be changed to equipment with weight and space too.
Strikes have to be be bound to the command console or a tc of similar weight. I think something like a illumination device or even the existing tag would be a nice prerequisite for striking.

These are just a few vague concepts, but I would like to see the change done along these lines.

[C E Dwyer]

Personally I don't want to see this, because any addition P.G.I have added to the game, or plan to in the last 18 months has made the game worse.

Hit detection got better, but that's not an addition.

[Funkin Disher]

El Bandito, on 07 September 2016 - 08:34 PM, said:

Have a like, just for the effort.

Ditto.

[Alistair Winter]

The main challenge with the skill tree revamp is promoting role warfare. I.e. making each 'class' attractive and getting the players to use synergy between classes in a match. So ideally, you would have Defense mechs tanking while Offense mechs lay down support fire and Infotech mechs provide targeting data, for example.

In this scenario, the question is whether the chosen skills and values are actually enough to create synergy (e.g. is the Defense Atlas really that much tougher than the Offense Atlas?) and also whether some skills are generally a lot more useful than others. For example, if you're playing Skirmish on HPG Manifold, how valuable is a light mech with Master Mobility vs Master Infotech? Mobility in itself is not that useful in MWO, beyond a certain point, because the maps are so small.

I'm not sure you could have 4 skill trees like this that would be equally valuable for all weight classes. In this case, I think Mobility would be the least popular one.

[El Bandito]

Alistair Winter, on 08 September 2016 - 02:58 AM, said:

The main challenge with the skill tree revamp is promoting role warfare. I.e. making each 'class' attractive and getting the players to use synergy between classes in a match. So ideally, you would have Defense mechs tanking while Offense mechs lay down support fire and Infotech mechs provide targeting data, for example.

In this scenario, the question is whether the chosen skills and values are actually enough to create synergy (e.g. is the Defense Atlas really that much tougher than the Offense Atlas?) and also whether some skills are generally a lot more useful than others. For example, if you're playing Skirmish on HPG Manifold, how valuable is a light mech with Master Mobility vs Master Infotech? Mobility in itself is not that useful in MWO, beyond a certain point, because the maps are so small.

I'm not sure you could have 4 skill trees like this that would be equally valuable for all weight classes. In this case, I think Mobility would be the least popular one.

Mobility (specifically + movespeed) could be vital in Scouting mode, for example. My Locust-1V would appreciate it.

PGI can make Mobility more competitive by increasing its bonus numbers.

Finally Mobility doesn't have to be its own tree, and can be incorporated into others.

Edited by El Bandito, 08 September 2016 - 03:15 AM.

[Asaru]

Great post. Really curious too see how something like this compares to what PGI is planning to announce at MechCon.

[EvAbsence]

Alistair Winter, on 08 September 2016 - 02:58 AM, said:

The main challenge with the skill tree revamp is promoting role warfare. I.e. making each 'class' attractive and getting the players to use synergy between classes in a match. So ideally, you would have Defense mechs tanking while Offense mechs lay down support fire and Infotech mechs provide targeting data, for example.

In this scenario, the question is whether the chosen skills and values are actually enough to create synergy (e.g. is the Defense Atlas really that much tougher than the Offense Atlas?) and also whether some skills are generally a lot more useful than others. For example, if you're playing Skirmish on HPG Manifold, how valuable is a light mech with Master Mobility vs Master Infotech? Mobility in itself is not that useful in MWO, beyond a certain point, because the maps are so small.

I'm not sure you could have 4 skill trees like this that would be equally valuable for all weight classes. In this case, I think Mobility would be the least popular one.

Thank you for the post, and the constructive input. I agree completely with the synergy you speak of, and feel this system would help to instill a little role clarity in MWO. It is worth noting that the mobility bonuses seen in my draft were not modified to reflect the doubled values you see when eliting a mech in the current system. As far as the valuenof mobility goes, i feel very inadequate when stepping into an unleveled mech, in part because it is sluggish as hell. Lol
Wow. I'll make b sure to proofread my posts when on my phone next time. Many typos.

Edited by EvAbsence, 08 September 2016 - 09:49 AM.

[Khobai]

I see no reason to ever put points in infotech

and speed tweak is so easily accessible in the mobility tree that most people will just take speed tweak then put the rest in offense and defense

unless you can make all four trees equal theres no point in offering players choices

[WildeKarde]

I think there should be skills that go against each other. Taking a defensive skill removes the opportunity to take an offensive one might be an option

But the skills also need to vary on mech type as well to promote roles, so some assaults might not get all the mobility skills but may have more point they can put in an offensive skill for example.

For example taking atlas vs cyclops vs highlander you'd have more offensive / defensive but less mobility on atlas, more info warfare and offensive but less defensive on cyclops with the highlander maybe being mobility and defensive vs offensive.

[Kyrie]

Asaru, on 08 September 2016 - 06:26 AM, said:

Great post. Really curious too see how something like this compares to what PGI is planning to announce at MechCon.

I am afraid that there will be a powerpoint presentation with much handwavium.

[Khobai]

Quote

I think there should be skills that go against each other. Taking a defensive skill removes the opportunity to take an offensive one might be an option

thats the whole point of having more skills than you have skill points. so you cant take every skill.

the problem is the best skills like speed tweak and radar derprivation need to be buried much deeper in the trees. and infotech needs to offer something much more powerful and worthwhile on the whole compared to the other trees.

I personally think damage for all weapons should be reduced by 30% against targets you dont have a sensor lock on because that would make infotech actually matter. ECM should also not provide stealth anymore because its not a cloaking device.

note thats not at all the same thing they already tried on the PTR because what they tried on the PTR only affected lasers not all weapons and they never fixed ecm so it made ecm into antilaser tech which was just silly. What im suggesting is much more refined.

but the reality is theres NO other way to make sensors/infotech matter other than to tie them directly into damage. Because sensor information in of itself is simply not valuable enough to justify an entire infotech skill tree.

Quote

I am afraid that there will be a powerpoint presentation with much handwavium.

most likely were going to get ghost skills 1.0

the more skills your mech has the less skills your mech has

we all know its coming because... PGI

Edited by Khobai, 08 September 2016 - 10:15 AM.

[JC Daxion]

A for effort.. But honestly skill tree is about as far down the list as you can go on my wants or needs. To me it works, mechs get a boost when you master them, and is a C-bill sink, and gives a bit of variety especially IS side..


But IMO all that dev time should be working on engine issues, and game play updates.. that is a lot of coding, for what i think in the end will be not much of an effect.


To me i like the mech modules, as i can tweak my mech towards my builds, and quirks help balance.. I just don't need an RPG skill-up. Now if this was for HBS, and your pilots.. of course this would be great, but this is a shooter. Not sure i would want to make a mech tankier, or more fire power or what ever.. To me that is why you buy another mech.


I liked re-scale, I am encouraged about ED, I can't wait for the next balance pass because things have really hit a bad spot atm..

Spend time making better chat features, work on game modes, and back end issues, like even more improved hit detection, and other coding features like improving command wheel for example... But i guess this is already far in the works, so my thoughts are kinda pointless..


I also like being able to change builds, on mechs.. How would this be effected? buy a new mech? pay for a respec? I dunno.. To me the skill system is fine, and quirks are fine.. the only thing they really need to get rid of is Jam-chance, and velocity, the others are just for improved balancing..

Edited by JC Daxion, 08 September 2016 - 10:14 AM.

[Charlie Grant]

Make these mech improvements (and the mechs they are attached to) even more special and interesting by having unique, mech-specific, choices interspersed amongst the universal ones. For example, the unique improvements available on the Shadow Hawk would help to set it apart from the Griffin and Wolverine.