152 replies to this topic

[Kael Posavatz]

Issues, lots and lots of issues.

1) Mechs that can boat a single weapon type (such as Back Knight, Grasshopper, any Omni) are at a distinct advantage over mechs that must bring multiple types (Marauder, Atlas, Orion IIC). The former's weapon nodes are more efficient (or at least able to be used more efficiently) than the latter

2) Having Standard and ER lasers on the same tree is confusing. No other weapon tree affects two different weapon systems

3) Some nodes are oddly placed (NARC is buried underneath UAV and capture assist? Really?)

4) Consumables are a danger of becoming pay-to-win. Arty/Air/Coolshot lack nodes. Assuming nodes are put in (and have similar tree sizes as UAV), then 20 nodes (player can only carry 2 consumables)--22% of available nodes--will be needed to bring these in line with MC consumables. Meanwhile, a player willing to spend money will get max-performance consumables and effectively 28% more nodes than someone who maxed ot consumable nodes

5) No nodes to increase arm-yaw

[sitting target]

1. We need linear trees if we are doing trees at all!

Can anyone tell me any scenario when you want arm speed?

[ingramli]

sitting target, on 09 February 2017 - 06:25 AM, said:

Can anyone tell me any scenario when you want arm speed?

Perhaps an assault is left alone, and he need to deal with a locust closing in

Edited by ingramli, 09 February 2017 - 06:44 AM.

[Kuaron]

Maybe an assault is falling behind and has a has a perfect view on a marching Nova’s admirable lower torso back.

[Bullseye Bob]

Simple solutions to salvage new skill tree system:

1. Draw a horizontal line through the middle of the skill trees and have all skill nodes below the line cost two skill points instead of one. This would reduce the boating issue since it would cost significantly more skill points to max-out one tree, whereas those points could "basic" three trees for the same cost. Also, this would front-load the benefits allowing newly purchased mechs to be competitive quicker.

2. Allow each mech variant to have multiple skill point distribution trees, kind of like a re-spec macro. Each additional configuration would cost c-bills/MC's to be made available, like unlocking additional drop decks. This would allow a player to easily switch between configurations on a single chassis and encourage experimentation. This would be especially important for Hero's, where a player would not have multiple copies of the same mech.

3. Reduce the XP and c-bill cost of buying skill points on additional (different) variants of the same mech type. For example, I buy one Centurion variant and it costs the stated 100k c-bills and 1500 xp for each SP. I buy a different Centurion variant, and this one costs 65k c-bills and 1000 xp for each SP (as long as the total SP's are less than the first variant). This would encourage diversity and the purchase of additional mechs.

[Horseman]

sitting target, on 09 February 2017 - 06:25 AM, said:

Can anyone tell me any scenario when you want arm speed?

If your primary weapon mounts are in the arms rather than torso.

[Fox2232]

Kael Posavatz, on 09 February 2017 - 06:20 AM, said:

Issues, lots and lots of issues.

2) Having Standard and ER lasers on the same tree is confusing. No other weapon tree affects two different weapon systems

PPC tree affects both PPC and ER PPC in very same fashion. And on top of that it does not even state ER PPC effect...

[Kael Posavatz]

Fox2232, on 09 February 2017 - 07:44 AM, said:

PPC tree affects both PPC and ER PPC in very same fashion. And on top of that it does not even state ER PPC effect...

^6) This

7) cost of respecing will discourage experimentation and diversification. How many are willing to spend more than 25 million c-bills on a mech they don't know will work relatively well?

8) Repeated statements from PGI staff about module-cost reimbursement either overlook or willfully ignore how this will negatively affect the new-player-experience. Right now someone can purchase Radar Dep once and swap it around as many mechs as he wants/has patience for. With skill trees this is no longer the case

9) Magazine expansion is nice, but overlooks MGs, and all missiles, and is buried at the bottom of a skill tree that has beans to do with weapons.

Edited by Kael Posavatz, 09 February 2017 - 07:46 AM.

[Carminus]

I think the poll is missing an option. It should have an option to pick "helps meta but not the overall health of the game". Skill tree limits build options without respec. For meta players example tier 1/2 they don't change their builds nearly as much compared to tier 4/5 players or players who like to tinker.

I find the skill tree forces you to put in points into places you do not want or need. They should of mimic the one used by Witcher 3

[1453 R]

Having run a few matches last night, my initial impressions are mixed.

Overall? A positive step, if a shaky one. In order:

1.) Fluff vs. Choice

91 points to spend is a LOT, and allows players to easily max out or effectively max out several branches. That said, each individual node is all but useless. Spending 1,500 experience and 100,000 C-bills on a 0.8% bonus just does not feel good, and it means you HAVE to max a given branch in order to get bonuses you can actually feel in a match. My experiences fiddling with the new tree were basically "What's the most efficient way to get all the armor/structure nodes, all the Infotech nodes, all the (useful) nodes for my chief weapon, and all the leg nodes?" There wasn't a ton of point in trying to carefully balance a mix of individual nodes for individual 'Mechs because most of the branches' individual nodes provide such insignificant bonuses that either you get EVERY such node the branch offers, effectively maxing the branch given how Piranha distributed the nodes, or you skip the branch altogether as a waste of points.

You can't really 'splash' into most branches, invest a small number of points for one or two useful benefits without having to go whole hog into it. The sheer number of nodes you get exacerbates the issue since there's not really any reason to splash or go for a multiple-options approach instead of maxing out the most metalicious branches. Is there anyone who's not going to just soak up a couple of fall damage filler points to get every armor/structure node in Survival, when you have nearly a hundred points to throw around?

HOW TO SOLVE: It'd be difficult and require a rejiggering of the branches, but I'd like to see a much smaller number of nodes distributed amongst branches with much stronger per-node benefits. Make player choice matter, and make it an honest decision whether to max out a branch or whether you can skip some of the nodes in a given branch in order to invest them elsewhere. Right now a player can comfortably get just about everything worth getting if they invest in only a single weapon type in Firepower. That doesn't make for satisfying choices or role specialization, that's just kinna what we already have now with master'd 'Mechs except they take a billion times more experience to grind up. Speaking of...

2.) Costs

Unlike some, I'm not crazy bent out of shape over the 9.1M C-bill price tag on a mastered 'Mech. This system is replacing modules as well as skills, and the day you spend only nine million C-bills on a full module set in the current game is a very good day indeed. It'd be nice if they were less costly, but eh. BUUUUUUUUT...asking us to sink nearly a hundred and forty thousand experience into any single 'Mech for bonuses as weak and limp-wristed as 'bout two thirds of the current Skilltree system is a serious downer, and massively debilitating for newer players who generate a lot less experience than more advanced players do. There's incentivizing premium time and GXP conversion, and there's "f*** you, pick two or three 'Mechs and never pilot anything else." Yes, this is an artefact of the removal of the Rule of 3, I get it, but still. C'mon, Piranha. C'mon. You're not even letting identical chassis share experience anymore. Be nice.

HOW TO SOLVE: Cut the experience costs of new nodes. You should have to play multiple winning games in order to afford a new skill node. One good, solid victorious match in which you were helpful and Credit To Team should be enough to earn the experience for a single skill node, if we're sticking with this ALL THE NODEZ 91-point system. That's still ~91 good, solid victorious games a player needs to max out a 'Mech, which is a real goddamn lot.

3.) Variety

Currently, the Skilltree values are basically just the old pilot efficiencies and a handful of the more popular weapon modules all cut up into pizza-square slices and set up like Jenga towers. There's not a lot of new here, not a lot of interesting abilities to try and dig for. InfoTech (my favorite of the new trees, for a few reasons) has a few "new thing you do" nodes, but even those are just rehashes of older modules. It's kinda boring, and makes me a little sad. The 'Magazine Capacity' node in Operations is a good example of the sort of thing I'd like to see more of, as well as Advanced Zoom in InfoTech. Stuff that changes the rules on you, even if only a little bit, or offers new capabilities instead of just being "+1.2% bettah [X]", where [X] is a thing you already do anyways.

Modifying behaviors, or introducing new capabilities, is harder than just +1% [X], but it's also a lot more satisfying for the player to work with and enjoy, and adding capstones in the skilltrees that allow for these behavior modifications can make all the junk +1% [X] nodes you have to take to get to them a lot less sadness-y.

OVERALL

I think Piranha's actually onto something with this system. The bones are good, the underlying systems and interfaces are solid. It's the numbers overlaid on those bones that feel really off to me after my experiments. I'll be fiddling with it more as I can, but as of right now? I feel like this is roughly equivalent to putting your shoes on before taking your first step in the right direction.

[TorinZ]

My Opinion so far.
I like that I can skill up a mech to provide bonuses to things that are important to me. I do not need to be tied to weapons a mech was originally quirked for and can build out to what I want as a player. For Offense or Defense, so a good thing. And tied per mech, so build out duplicates differently. All good.

CBill Pricing - I don't mind paying CBills and XP for the nodes. But I think they need to reduce the pricing some. I get that players will have some options to level that we didn't have today with modules, and could potentially boost more weapons than the old 2-3 weapon module only setup. But I think the price for 100k per node is a bit much for new players. Maybe reduce the pricing to 75k (or less) per node. Yes, I know we didn't buy a ton of modules for all our mechs and swapped things around, but I am sure weapon mods and such would come up more frequently, especially for all of your favorite mechs. 6.8 million isn't too bad versus getting the bonuses in the new skill tree.

Promotes Weapon Diversity- well, you COULD diversify your build with this skill system, but human nature won't let us. More than likely we will top out skills for primary weapons, maybe a few nodes for secondary and then build out on the other trees to increase our defense and mobility. Time will tell how that will work out. My initial thought was, some of my better mechs feel stronger, but, I a betting so will my foe's mech. So a wash. Even one of my mechs I chose to skill up 2 weapon trees seemed like I still came out ahead from how it is today. I had to reduce some skills elsewhere, but that seemed to work out well.

Particular Trees I wonder about their worth- Upper Torso, seems like I was able to skip this one and not care. Jump Jets, was hard to tell the difference, I probably need to remove it, try it, put back on and retry. And then do one side or the other and see if those vectoring nodes are wanted or not. Or if points are better somewhere else. UAVs was another one I found hard to justify paying points into since they typically get shot down quickly. I think I would buy in more if the higher levels provided more UAVs. So give us a +1 UAV to launch during a match at higher levels in that tree.

UI- I had hoped we could play around and place points and see the bonuses before submitting, and I was pleased to see that.

HXP- I had hoped these would be more like GXP for purely selfish reasons.

Overall, Iike what I see, needs some tweaks, but is going in the right direction. Now...how will future mech packs be sold...And let one of the IS mechs be an Argus to introduce RAC5s...

[GenJack]

I think its good but needs some love in the organization of skills and maybe a system like this.

https://mwomercs.com...xperimentation/

[Horseman]

Carminus, on 09 February 2017 - 08:06 AM, said:

I find the skill tree forces you to put in points into places you do not want or need. They should of mimic the one used by Witcher 3

Yes, that's probably the way to go.
For those who don't know: Witcher 3's system gates higher tier skills by requiring you to have spent at least x skill points on lower tier skills within the tree.

[Lolo van Trollinger]

if one thing the exessive costs of speccing a mech will make monoculture-boating all the worse.
AC+Laser boat ? to expensive. just go full laser or AC. mech hasnt got the hardpoints ? might as well delete the chassis.

oh, and the new skill tree PUTS AN END to any mech CHASSIS that might have been sellable before that that is NOT A LASER VOMIT.

Edited by Lolo van Trollinger, 09 February 2017 - 09:24 AM.

[Admiral Brad]

I find that the skill tree is a step in the right direction as I love customization, but my main concerns with the new skill tree are:
Too many nodes - I am not even talking about scatterbrained webs, just that there are so many nodes to click to get any where. This means that every match you will be going into your skills to pick new nodes. This also makes any real decision making hard because there are so many of them. There is simply too much pointless clicking.
Too complicated and Expensive for New Players - It already takes a long time to get your first mech, getting to your second shouldn't be hindered by mastering your first. Burning all your c-bills on mastery (and it does look like a close to a perfect match on your c-bill to xp earning rate) means having to decide between getting a second mech, having fun trying new stuff configs or levelling your mech to help win. The number nodes doesn't help as there are just too many nodes and new players will not know which ones are good for their mech. Perhaps a free "basic tree" would be helpful. This would be very close to identical to the original 8 basic nodes.
Promotes Boating - This system hasn't done anything to alleviate boating as the huge number of weapon tree nodes make it impossible to make multi-hardpoint mechs as powerful as single weapon boats such at quad AC mechs or laser-boats. Multi-Hardpoints get to either have default firepower or virtually no other points. i.e. The Atlas has LRM, SRM, AC, and Lasers by default. That's four trees leaving only 11 points left over if they max out firepower, where any boating mech would have 71 leftover. That's a huge difference.

[Ben2m]

It needs changes, big changes.
I think the above posts go into detail enough so i wont repeat most of the negative arguments ive read.

I feel you are making the same mistake a lot of other game makers have been making in the past years: first you nerf all kinds of stuff, like srm spread, lrm range etc and then you give it back through skills and you have to work hard for the experience and cbills to only get a small part back of the abilities and quirks you once had.

Content / balance patches should be about adding something, giving new exciting stuff to grind for. Not taking away and then giving a small part back. And as mentioned above the costs are unbalanced, they will both hurt new players and veterans with 100 mechs alike.

I understand the need for additional income as a company but, in my humble opinion, this is definitely not the way to go and in the long run it will cost you a lot of players.

[Johnny Z]

Its great. It will take a while to figure out.

I don't use unlocked arms hardly so that's not useful, but that's ok because it fits into the tree well.

For some reason the apply button greys out after I go a little ways up the tree. Not sure why that is yet.

- Side note, I see some complaining going on about nothing, the mech tree needed to be done properly and this is an excellent way to do it. The best really.

What is there to possibly complain about at this point? No one has mastered the balance of this tree yet to complain about anything.

The finer points of balance arguments will be ongoing for the entire rest of the year at least and that makes some sense unless its very badly done. But no one even knows that yet.

I looked over the tree quickly and didn't see any problems with it. If someone sees an actual mistake or problem somewhere then go ahead and point it out.

Edited by Johnny Z, 09 February 2017 - 10:20 AM.

[paws2sky]

I just did a lengthy post about my experience with the system, but after reading this, I saw a couple things I hadn't thought of.

Weapon Grouping could look like something that mirrors the current hard point distributions:

Energy (Standard Laser, Pulse Laser, ER Laser, PPC, ER PPC)

Ballistic (AC, LBX, UAC, Gauss)

Missile (SRM, LRM, Streaks)

Apply the bonuses across the board to each weapon in the category. Include node branches that specifically apply to some of the unique weapon types.

For example, left side of the Energy tree has a branch for laser duration, while the right side has a branch for PPC velocity. Left side of Ballistic tree has UAC Jam Reduction, while the left side has LBX Spread reduction, and maybe there's a node or two that specifically buff Machine Guns. Missiles might have an outlying branch for NARC and another for Spread reduction. The central branches focus on core things like cool down, range, and heat generation.


-paws

[Malrock]

Ibrandul Mike, on 09 February 2017 - 04:22 AM, said:

2. How does it work for new players?
Same complexity as what we already have

Which is not really my opinion. It is not worse or better if you think a bit about it. Your choices are more meaningful. Which might lead to a lot more problems with wrong skilling and then having to pay too much for a simple mistake.
More options means more complexity but it is easier to understand all in all.
What you have as a problem is the point that making the "right" choices is not as easy as it was before. You go cookie cutter or you go potatoe... at least after a few months. Regardless if there are perhaps more viable meta choices... you might have a hand full of "good" skilled mechs and a lot of bad skilled mechs.

Turning the skill tree into how good of an internet warrior are you in order to find the optimized skill build is completely stupid, and creates merely the illusion of choice. What is actually going to happen is an optimal path will be established and anyone who deviates from said path will be running sub optimal builds and gimping themselves. We already see this in the mech module choices people make, it is widely accepted that radar dep, is the way to go, with seismic as second, the only option that gets used other than these two is the zoom for people who like to snipe, and then the wall hack goes for zoom, anything else was a false choice / waste of cbills / general exp. This is the same thing WoW had, everyone went to elitist jerks found out the best new build and just copied it into the game, there is no diversity, or actual options you are just going to have bad players wasting c bills and mech exp because their google foo wasn't up to par. Completely stupid complexity for the sake of complexity which makes the game harder for new players, and puts them at a disadvantage to older players.

It is only a matter of time before some one math's out the best route. This is why I think it should be super simplified into about 10 available skill points that would be used to purchase module replacement effects. Call it the quirk tree, and keep the first two levels of the skill tree as they currently are.

[Johnny Z]

Malrock, on 09 February 2017 - 10:20 AM, said:

Turning the skill tree into how good of an internet warrior are you in order to find the optimized skill build is completely stupid, and creates merely the illusion of choice. What is actually going to happen is an optimal path will be established and anyone who deviates from said path will be running sub optimal builds and gimping themselves. We already see this in the mech module choices people make, it is widely accepted that radar dep, is the way to go, with seismic as second, the only option that gets used other than these two is the zoom for people who like to snipe, and then the wall hack goes for zoom, anything else was a false choice / waste of cbills / general exp. This is the same thing WoW had, everyone went to elitist jerks found out the best new build and just copied it into the game, there is no diversity, or actual options you are just going to have bad players wasting c bills and mech exp because their google foo wasn't up to par. Completely stupid complexity for the sake of complexity which makes the game harder for new players, and puts them at a disadvantage to older players.

It is only a matter of time before some one math's out the best route. This is why I think it should be super simplified into about 10 available skill points that would be used to purchase module replacement effects. Call it the quirk tree, and keep the first two levels of the skill tree as they currently are.

Your long winded wall of text is an argument about balance that isn't even present. Yes there will be balance fine tuning or what ever, so what?

Sum this entire reply up next time with one word, "balance". Thanks.

Edited by Johnny Z, 09 February 2017 - 10:24 AM.