Pts Skill Tree Feedback
#121
Posted 13 February 2017 - 04:28 AM
#122
Posted 13 February 2017 - 06:52 AM
First of all the bad:
Xp conversion and leveling costs :
This is bad for both new players and to veterans alike.
New players will be stuck leveling mechs a lot longer and as such will have the process of finding their most suited / enjoyed role pushed back. This will likely lead to a lot of new players very quickly leaving as the time and resources you have to invest in a single mech will mean a lot of people will be stuck playing a mech that they simply do not enjoy for a long time purely to see if they actually like it as a fully quirked mecjh is significantly different to a stock one.
For veterans , particularly those who have invested a lot into mech bays and acquiring a large assortment of mechs are going to be sorely disappointed that their previously mastered mechs now are less than 50% of the way through their grind. I know of people with 100+ mechs they have mastered and even if they were to convert all their extra xp using MC , they would not be able to afford to skill up even a half of their mechs due to the astronomical c-bill cost.
Skill placement
There are skills that are just placed in the wrong trees / locations. This is a fairly common gripe so not going to cover it too much. Biggest thing for me is the amount of filler skills that are spread far and wide, even to trees they don't belong in just to make skills like radar deprivation or cool run further down the tree. I have no problem with prerequisite skills for more powerful abilities, but at least group abilities so choosing a powerful skill doesn't require maxing out a whole half a tree just so you can get to one node. I would like to see a variable point xp cost put in so that average or next to useless skills are cheap but powerful skills aren't.
Mech Vs Skill balance
It is fairly plain to see that there are mechs that are just plain better than others. Weapon mount positioning, geometry, maneuverability and free space contribute to a mechs effectiveness as much as its skills and perks set. As such mechs with advantageous factors should not have as high a quirk cap as those with horrible geometry, internal setup, poor movement or minimal weight. For instance a spider-5v should have more quirks than any arctic-cheetah as they are not even on the same planet as far as power level at stock.
Skills Vs Other skills
At the moment there are some skill trees that are simply overshadowed by the the other trees. Once you pick up all the important skills for survival like radar dep, speed tweak and targeting abilities you have only a handful of points to spare. Some of the more filler abilities or skill trees need to just be made baseline or removed all together. Removal of some of the trees like the operations tree and making the abilities baseline or simply no longer existing would go a long way to freeing up points that would increase variance in builds. Just about every mech requiring 21 points just to be at maximum heat efficiency is just limiting diversity for the sake of a point sink. As almost all mechs will be getting these skills any way why not give it to all and promote diversity or simply remove it all together and just set all mechs heat performance at a lower base line and force people to adapt to smaller heat pools and slower dissipation. This would also help combat the high precision alpha / dissipation meta we have at the moment.
The meta
The problem with the current meta is mostly the symptoms of 3 major challenges you as dev's face for this game. inter Mech balance, weapon balance and the real time nature of game play.
Mech balance is probably the easiest of the issues to resolve. Simply penalize extremely powerful mechs or buff low powered mechs as far as skill pools are concerned, or do both. The end result of this is similar to the current quirk system. High power mechs don't have many skills and low powered mechs have heaps.
Weapon balance is particularly tough as not only do you have the lore to stay true to but also have to balance damage output to weight / space to balance. For my mind the problem here is precise damage like gause, ppc and short duration laser vomit is far too powerful. Longer cooldowns for gauss and ppc as well as longer laser duration would encourage people to stop the current high damage / low duration poke that is so prevalent. I suppose the extreme example of this is "poptarting" that is so prevalent.
Finally you have the nature of the game as a first person shooter. This adds not just mech and load out as differentiating factor in a fight but also time. Mechs that can strike at targets with out revealing a lot of themselves are innately more power than those who have to show a significant portion of the mech to get any meaningful amount of firepower on target. This one is needs to be tackled from a few directions. Firstly forcing mechs to stay in the open longer while firing would give those with average or poor geometry a chance to at least return some fire even if it a smaller target. The cooldown of single shot firing weapons such as gauss and ppc's and burn in time of lasers need extending so weapons that are meant to be precise by lore don't allow mechs to stay in cover 90% of the time and still offer a firing line that is just suicide to try and break cover and close in on. Finally mechs that can bring large volumes of firepower to bear with minimal exposure need to have some weakness introduced to them, be it less skills , more difficult heat management or some sort of negative quirk that brings them into line with other mechs.
The Good
I know i have talked about the faults a lot but there are some things i am really liking.
The ability to no longer be forced in to leveling three mechs just because I want to add a particular mech to a CW drop deck or want to play a certain play style on occasion.
The potential for a greater variance that could come from the skill tree if some of the filler was removed. Being able to give any mech any weapon quirk greatly increases the possibility of different play styles and loadouts.
The ability to customize my mechs to my play style. Having always enjoyed the extreme range sniper role, its now nice to know there is the possibility of playing that role in most mechs.
All in all I think the change is a good one, but there are currently too many nodes and too many filler skills. Meta mechs get even more powerful while the weaker mechs get more so and people are given no incentive to grow their mech collection as the leveling system is simply too costly.
Edited by Maelware, 13 February 2017 - 07:01 AM.
#123
Posted 13 February 2017 - 07:54 AM
-reduce / increase nr of points per variant, to boost underperformers and reel in the overperformers.
-tree path changes so we can truly choose which skills to invest in.
-weighted costs when going into the deeper lvls, so you either go 'allrounder' or 'full specialist' without promoting boating too much.
#124
Posted 13 February 2017 - 01:38 PM
Odanan, on 12 February 2017 - 07:02 PM, said:
The only problem I see with different number of skill points for different mechs is this: balance tweaking. If PGI decides to decrease the number skill points for some mech, she will have to refund what the players spent on it.
True, some mechs are not born equal.
I would try and steer away from bonus points, for the reasons you have pointed out. It's a lousy method to try and balance through and it's not really much different to leaving quirks on mechs or even having some of the skills pre-bought.
I think the better option is to have a unique tree per mech chassis, maybe not variant, but something that we have the choice to spend our points on that would stack with some of the other bonuses. ie. If the Atlas has a reputation of being a tough as nails mech, it has a unique 'lore' tree that has some additional defensive options. So if a player really wants to live up to that reputation they have that choice to put their points into that tree, but it's at the expense of something else.
Same for all the mechs. This would be a good spot to put the Set of 8 bonuses for Omnimechs as well.
For mechs that have the vaunted high weapon mounts, like the Kodiak 3, these mechs have to be poor in other areas such as having a lousy mobility. We can change the base stats of these mechs so they don't have that same level of mobility as mechs that have arm focused weapons, but this doesn't really deal with the weapon boating part that also goes along with it. The best way to apply balance for mechs that will boat one type of weapon is simply have less skill points over all.
Investing heavily in one tree needs to be done at the expense of being able to invest in another. We can't have mechs having a full weapon tree, a full upper and lower mobility tree and a full sensor tree.
If we do another iteration of the PTS I reckon reducing the skill total down to 60 would be worth giving a go.
Odanan, on 12 February 2017 - 07:02 PM, said:
Well, it's pretty easy to navigate, we've got the zoom and can use the group selections to jump to the different locations.
Selecting and undoing skills is pretty easy.
Yes the XP conversion stuff is a bit messy and there are multiple types but this will be necessary until everyone has cleared out their historic xp and we see it clear up a little.
Personally I felt it was ok.
Odanan, on 12 February 2017 - 07:02 PM, said:
It's unlikely to help the bigger mechs much. I did a few tests with a Viper and tested out the jump jet tree. It made a noticable difference for a mech like that. I imagine a mech like the Spider would benefit as well.
Depends on what you want to use the mech for though. Maybe it's not a great option for an Assault or even a Heavy, but Mediums and Lights that would like a bit of extra mobility who already have a good high end speed, it creates a point of difference.
#125
Posted 13 February 2017 - 02:11 PM
ex. Billy gets 50k cbills match one, and 500 exp.
match two 50k cbills and 500 exp
match three 50k cbills and 500 exp.
then he gets and experence point and raises the skill of his mech. (Unless i misunderstand the system).
this sounds good.
#126
Posted 13 February 2017 - 03:31 PM
Emperorcleon, on 13 February 2017 - 02:11 PM, said:
ex. Billy gets 50k cbills match one, and 500 exp.
match two 50k cbills and 500 exp
match three 50k cbills and 500 exp.
then he gets and experence point and raises the skill of his mech. (Unless i misunderstand the system).
this sounds good.
So you say it is a good thing that a new players takes 3 matches to improve a skill point. That would mean it would take a new player 273 matches to max out their mech? Which doesn't include any actual hardware upgrades, if they tried to change the engine (86 more matches), or any of the weapons then many many more. That is a terrible system. Won't get any new players as a 300 plus match grind just to get one mech going is insane. To fix the new player experience, they need to allow them enough c bills to buy 3 mech's so that they can really get a start on their collection, and interested in improving them. Not put the goals for one mech so far away that they will never get there.
#127
Posted 13 February 2017 - 06:38 PM
Otherwise, if the number of points to complete a mech is significantly reduced the cost is subsequently reduced.
#128
Posted 13 February 2017 - 09:46 PM
#129
Posted 13 February 2017 - 10:54 PM
Things they need to fix:
1. Faction is still practically dead.
2. Finally, they are adding some weapons, no new heat weapons besides flamers yet... For Flamers, there is no reason for why they have exponential heat, especially since they only overheat a mech if the enemy fires a heat generating weapons, flamers NEVER shut a mech down, only bring it close because of the 90%. and 90m range is not even lore, it is supposed to be the same as small laser.
3. Skill tree still leaving flamers and mg completely out.
4. Further, skill tree cost 100,000 c-bills to unlock skills, that might be easy for the great player, but horrible to the new player. (been updated to less c-bills I hear.)
5. Connection problems and performance issues still persist.
6. Decals are nice, but still want to see critical issues solved, although I think they are getting it abit since the last version of this post.
Edited by Independence MK2, 08 April 2017 - 12:25 PM.
#130
Posted 13 February 2017 - 10:59 PM
#131
Posted 13 February 2017 - 11:46 PM
Here >>> https://mwomercs.com...il-on-the-head/
To summarize:
- Less skill nodes, less grind, make it more simple - why so many nodes of 1% with such a clusterf*** on the UI? Follow principle less = more. Make every choice in the node compelling and impacting, not with grindy increments of 1%.
- Make the player choose between A or B, don't let him have both - create logical choices (laser range vs duration), example from another game; less skill nodes, compelling A or B choices
- Diminishing returns, discourage mono boating meta, reward mixed loadouts
- Try reducing weapon categories, also reducing incentive to mono boat
- Equalize armor and buff structure across tonnage (look at Kanna's graph, 75 tonner has more hp than a 90 tonner and a 50 tonner)
- Things like arm movement quirks on mechs without actuators, skill trees need to be redone overall into somethjng more sensible, again: lock the routes and group them up, don't force the player to pick everything from a category + the stuff he doesn't want, make him choose.
+ Other stuff he mentioned, but these make the gist of it.
Edited by NeoCodex, 13 February 2017 - 11:47 PM.
#132
Posted 14 February 2017 - 01:22 AM
The PTS system has no tradeoffs and encompasses way too many different skill trees with only marginal number increases per skill. The current iteration also generates a huge inflation of XP and C-Bill requirements.
My suggested system below gives people a way to specialize in either their own playstyle, or a role in particular, over general mech enhancements. It's also visually comprehensive, because everything fits on just 1 skills page. The 3-variants rule is removed because every mech individually gives points for a global (customizable per mech) skill system, and have a reason to buy/master multiple mechs and then keep the ones you really like.
It works as follows:
Each mech has a Basic, Elite and Master level based on XP gained in that mech. Each time you complete a level, you get a skillpoint and you unlock a new tier of skills you can put points in for that mech. EACH MECH in your garage can use as many skill points as you've unlocked, though certain skills will only be available after basic/elite/master level.
The skills are organised in a grid, starting in the middle and extending in four directions:
UP: Scout (radar-depr, speed, targeting, UAV, ECM, NARC)
DOWN: Tank (Armor buffs, structure buffs, component integrity, damage reduction)
LEFT: Brawler (SRMS, lasers, MGs, big ACs, flamers, heat dissipation)
RIGHT: Support (LRMS, PPC, small ACs, range, cooldowns, Ammo)
Unlocking a node in the grid will open up all adjacent skill nodes, which may branch into other trees, because of skills that can fall in both roles. Each node costs a certain amount of skill points (and Cbills) to unlock, and the further you are from the center, the more skill points they cost and the more powerful the skills become. Brawlers that also want to be tanky, will have to sacrifice their more powerful skills.
Examples:
Decreased LRM-lock time / NARCs fall in Scout and Support categories.
AMS range/ROF falls in Support and Tank categories.
Decreased (S)SRM spread applies to Brawler and Scout
Increased Component Health applies to Brawler and Tank
Some nodes include XP/Cbill gain and underpowered Mechs, could come with freely unlocked skills. beforehand, which would normally be locked.
Similar (but different) skill systems can be found in succesful F2P models such as Payday 2, Path of Exile and League of Legends.
Edited by Excalibaard, 14 February 2017 - 07:10 AM.
#133
Posted 14 February 2017 - 09:27 AM
#134
Posted 14 February 2017 - 10:49 AM
mycroft000, on 14 February 2017 - 09:27 AM, said:
I've no idea, but you have a good point, would have thought number of skill points would balance the losses.
Chances are they probably forgot.
I'll be honest, this mess we've been presented with is getting me down.
Players most of them not involved in game development, have taken a few days to display all the flaws in this and previous big game changes, and not a small number in less than a week have come up with far better idea's that could be coded with minimal changes than it's taken P.G.I's team or professional months working at it.
My confidence is very shaken by their lack of vision
Edited by Cathy, 14 February 2017 - 10:53 AM.
#135
Posted 14 February 2017 - 11:15 AM
This is an extremely niche game, and they likely don't have enough play-testers that have enough distance from the design to spot the flaws that are found when putting this out into the wild.
I don't know how many developers they have on staff, but even if it's in the double digits, that's a fraction of the number of players who are passionate about the game that are willing to dig in once test servers go live to discover the good and the bad from potential updates.
I understand many of the complaints but honestly many of them are overblown, some are not, and some are absolutely and completely correct. I hope PGI can sift through all of the feedback and take the best and implement them, while not giving too much weight to the complaints that are solely based in the mindset of "but this isn't how it works now".
#136
Posted 14 February 2017 - 12:25 PM
mycroft000, on 14 February 2017 - 09:27 AM, said:
There has been a little bit of discussion on this. As they specifically offered sales for special mechs with extra module slots, and people made purchasing decisions with those in mind. However there has been no word of any type of buff to those mechs to compensate them for the lost power. Those people are just getting hosed it seems.
#137
Posted 14 February 2017 - 12:28 PM
#138
Posted 14 February 2017 - 12:56 PM
This has no end game, each mech is a new grind. It also penalize the tinkering aspect heavily.
As such, I have proposed this, the exact values may be changed, IE up the grind factor or lower the cbill re-spec cost or increase it to force more people to buy extra mechs / spend mc:
This is the proposal to follow industry standard for skill trees, and to still have carrot and stick for spending money for MC/premium/mechs
This is designed to allow players with large number of mechs to experiment, ease of respec, and eliminates the modules system.
It will allow players with medium amount of mechs to better acquire new mechs and to invest in liked mechs by promoting ease of acquisition (only needs 2-10 games once you hit level 30 to get a mech to its max). While if they loved the mech, can master it by getting 3 variants to the max to remove / reduce the respec costs to better experiment with it. If they need to fill a role in cw, they can now do so much easier
For new players, they have a path of progression that is clear, similar to most other mmos and mobas, and will hopefully have their own queue to play in. This rewards play time, regardless if they were super good or bad, they have something to look forward to at the end of each match.
COST: Simply put, you earn xp on any mech that goes towards a pilot level, each level gets 1 points of specialization, starting at 0. After level 50, you unlock all 50 points of specialization. Exp needed scales non-linearly, and should take around 15-40 hours to complete. Assuming a win each game, worth around 1k exp, takes 10 minutes that comes out to 60 matches or 90k exp. Maximum would hopefully take 150 games for 15 minutes each for 40 hours. After which it is done. Expected that most players will take around 20-30 hours to do this. Champion mechs and premium time speeds this up.
To spend such point per mech, you need to have at least 20 mastering point under your belt (sped up by premium, or wins). You start off with 10 skill points to spend, and gets 2 per mastering point. After each game, you unlock 2 points worth of mastering pts for losing - 10 games, 3 for losing with premium / hero mech / champion mech / brought mech with MC - 7 games, 6 for winning - 4 games, and 10 for winning with premium / hero / champion / brought mech with MC - 2 games.
GXP becomes a way to unlock this faster. 1k gxp per mastery point to spend anywhere. 20k per mech, champions may need to be tweaked to earn more gxp in this new world
To reset you need to spend 5 million cbills, regardless of how many you want to spend. But this is eliminated or severely reduced (to say 100k) by mastering a mech by owning 3 variants and unlocking their tree.
For newer players (< level 30), respec cost is either 0, or starts very low and ramps up towards 5 mil only at the very, very end (IE the last 5 levels to go from 10k to 5 mil). They should have their own queue in qp, and if they must be put against level 30s the matching system should notify the player and give enhanced rewards (maybe count it as a win regardless) and strive to have balanced amount of non 30s around. Skilled players under 30 (smurfs or just naturals) may be promoted to play with 30s, with accelerated leveling speeds, but will otherwise enjoy the same respec reduction.
Trial mechs will always have free respec and unlocked mastery, same unlocks and everything once purchased, but equipment cannot be changed. Rotates every week.
BALANCE: 50 point means that you have to pick and choose your path. You can max out approx 2 tress with this. Be it tanky and ganky, or sensors and speed, or a random mix of everything. I would put diminishing returns on WEAPONS only. IE for weapons first point gives you 40 % of bonus, second gives additional 30 %, third gives additional 15 %, then fourth 10 % and fifth 5%. Everything else is equal progression.
This should help prevent boating.
If possible, revamp system to not use a tree, but rather individual un-linked choices. But this would mean major re-do of the number of points, a rebalance to ensure things like the JJ bonus is actually up to snuff (only have say 1 point vs 5 and gets all the benefits).
GENERAL:
This is again, something of a standard industry practice, by hooking players in, you increase the pool of potential players. With the nag of the 5 mil respec fee, that should help convert players towards paying members, and it offers a very tangible reward for veteran players.
#139
Posted 14 February 2017 - 01:41 PM
1. Its a step in a new direction but its odd that skills are attached to a mech and not the mechwarrior. Why would a yori flyswatter on my atlas work any differently than the yori flyswatter on my catapult. Same with a defiance ppc why would i have to relearn weapon systems? Mech piloting ect makes sense being tied to a mech but weapon systems make no sense at all to me.
2. Newbros like me to a point will be confused and may bury themselves investing skillpoints into a bad mech. Before bad mech justified thier existance with mastery. And the gxp ro mech xp is terrifying high like insanely expensive. My brother who plays much more than me would have to pay 50$ just to convert the excess xp on his Timberwolf alone. It looks like they are doubling down on cash transactions with this.........
3. Meta will even more well meta.. now you will pick the 91 best points on the most boat-able mechs in the game.
#140
Posted 14 February 2017 - 02:07 PM
50 50, on 13 February 2017 - 06:38 PM, said:
Otherwise, if the number of points to complete a mech is significantly reduced the cost is subsequently reduced.
What if instead of increasing the payouts, the final bonus came with a voucher for 1 mech and mechbay?
1 user(s) are reading this topic
0 members, 1 guests, 0 anonymous users