Skill Tree Public Test Session
#541
Posted 11 February 2017 - 01:32 PM
#542
Posted 11 February 2017 - 01:33 PM
I am looking at the skill trees as implemented like the costs of custom modding a car for a specific driver. Sure a good driver can extract every ounce of performance from a chassis, but the mods tailored to a specific drivers style, push those performance levels over the top. So the "costs" associated with the "hot-rodding" of a mech can be justified in that way and the experience to "drive" the new chassis. (The learning curve)
The costs to reset as they stand are way too expensive and all the costs should be more relative to the cost of the mech itself. There should also be a recoup of costs/xp on respec as some customized parts may be sold for new parts. (Time for that bigger turbo and injectors!)
The fact that you can add UAC jam chance mitigation to any mech now makes UAC boating stronger than ever. (More or less for those that don't use "chain-firing" scripting/macros) This is in addition to the fact, already mentioned by many others, that boating in general is already greatly buffed with the new skill tree and omnimechs exacerbate the point. (Nothing new there, though I chuckle every time I see someone post "KD3 Online")
The boating issue can be solved with what most other games do and that would be increasing the cost to go down the line for the enhancements further down the tree. This way, the mechs with varied hard points are not overly punished and there is an associated cost with boating. This will also keep the mech skill experience the same without diversifying by mech and adding more complexity. Also the torso and arms should be kept on opposite sides of the same skill tree to help offset the additional cost.
From a personal standpoint with the amount of mechs I have already mastered, (70% of a little over 300 mechs) I can very much agree with the argument of collecting mechs stops here and the selling off the less favored. There is no way I would go through that again as you are not given the resources to remaster a mech that was previously mastered and this is pretty low by any game standards to do this. Sure, power creep happens. But taking away previously gained progress is just wrong.
The other negative points were covered ad nauseam in this thread so hopefully PGI in the end takes the constructive criticism to heart and does not have another "failure to implement" on their hands.
P.S. I feel bad for any mech in the torso angle of a DW-UV with DWF-B ST's boating UAC-2's in the PTS. (like it wasn't a minigun before...)
#543
Posted 11 February 2017 - 01:46 PM
OldOrgandonor, on 11 February 2017 - 01:24 PM, said:
Exactly this. As usual not overthinked, unnecessary complicated. PGI please start to play your own game more and then you will realize whats where to do and how to do. thats what the Community tells you since years
#544
Posted 11 February 2017 - 02:28 PM
snipercam7, on 11 February 2017 - 11:30 AM, said:
Honestly, I'd like to see a split in the current skill-tree plan.
Mech-tree and Pilot tree.
Mech tree: Mobility, Defense, Heat Management, Weapons.
Pilot: Consumables, Sensors.
Then we can start inventing new Pilot skills, and new Mech skills that have both positives and negatives.
The issue there is, again, if there is any kind of cap on pilot skills the same situation would exist.
Someone willing to pay real money would have an in-game advantage (max performance mech, consumable, and pilot skills) over someone who did not.
That is the definition of a play-to-win game mechanic. It is something that PGI has thus far been very good at avoiding. And I find the fact that they are apparently considering it to be rather sad.
#545
Posted 11 February 2017 - 02:37 PM
Diminishing returns, start medium and then forward, (helps newbies getting into the game)
Easier "entry" costs, and then higher costs for higher levels, (helps newbies getting into the game)
More real choice between skills, trade-offs, (as seen in Fantastic Tuesday video that was posted earlier here)
Not calling this thing skills but litterally quirks since you spec your machine to your liking not the pilot! (skills should be pilot dependant not mech dependant)
Also, skills, as they are right now, are just a series of upgrades, there's no sidegrades, penalties when selecting a skill.
Cheers!
#546
Posted 11 February 2017 - 02:52 PM
#548
Posted 11 February 2017 - 02:55 PM
Kael Posavatz, on 11 February 2017 - 02:28 PM, said:
Someone willing to pay real money would have an in-game advantage (max performance mech, consumable, and pilot skills) over someone who did not.
That is the definition of a play-to-win game mechanic. It is something that PGI has thus far been very good at avoiding. And I find the fact that they are apparently considering it to be rather sad.
Oh frankly I agree, I'd leave it uncapped. If you cap out your GXP, you're a very experienced pilot, and that shows because you have literally every pilot skill. If you cap out a mech, it's at peak performance. I would not impose a cap in terms of how many upgrades you can buy. If someone wants to spend their entire time in one specific mech, metamech or not, let them and let them continue to improve it until there's nothing left to improve.
#549
Posted 11 February 2017 - 02:56 PM
mad kat, on 11 February 2017 - 11:39 AM, said:
They never said that.
The GXP you used to unlock Pilot Skills will be refunded as GXP.
#550
Posted 11 February 2017 - 05:37 PM
#551
Posted 11 February 2017 - 07:30 PM
Overall I think that a revamp of the skill tree is a great and needed move in the game and thank you guys at PGI for making that move. After reading through the proposal and referencing statements that I've heard and read about this game in the past it looks like you are changing your business model away from mech sales as a way of keeping the lights on. I applaud your efforts on keeping the game FTP for all who choose and "pay as you go" for those that do fund your efforts. This being said here's my concern.
I currently have exactly 200 mastered mechs. Many of them don't really fall into the skill tree that currently is in place so I welcome the concept of being able to tailor their efficiencies better to their chassis and build. That being said I spent years of free time getting to this point, and I don't have 1.8 billion cbills to put them back right again. Selling the modules I have won't scratch the surface of this deficit either, as I found them expensive and easy to swap out so I only bought a couple of each kind in order to focus my earnings on weapons and chassis.
If Cbills are required to skill up a mech, then what is the point of XP? I feel that your skill currency should still be your skill currency, and your game currency should be spent on things like weapons and chassis, not on skills. If cbill payouts need to be reduced to reflect the fact that we will have less to spend them on I'm fine with that. I just don't want to throw out the years of grinding on this game that I've done.
I'm going to reserve my final thoughts until after the pts testing that I do, but currently this is my main reservation.
#552
Posted 11 February 2017 - 09:03 PM
#553
Posted 11 February 2017 - 09:18 PM
Kael Posavatz, on 11 February 2017 - 08:09 AM, said:
I can sort-of see that working for some mechs, but you're talking about potentially 30million+ c-bills on skill-nodes. Yeah, not every mech can field every weapon type, or has jump jets, but omnis (and with a timeline advancement IS omnis are on the horizon) can accommodate radical payload shifts that do make the vast majority of nodes at least not a total waste.
To be more clear ... as it is right now:
- you spend 100,000 CB and 1,500 XP to unlock one node,
- you spend 25,000 CB or 10 MC to re-spec one node,
- you spend another 100,000 CB to unlock a different node (you don't have to pay the XP, but you do pay full CB cost).
So, when (random hypothetical example) PGI nerfs LRMs into the ground and everyone wants to swap out their super-meta LRM skills for Streak SRM skills, they will be required to pay 25,000 CB per LRM node and 100,000 CB per SSRM (or other) node purchased, all because of a balance patch.
... or, if a player gets smart and realizes that LRMs are bad, and SRMs are awesome ...
... or, if a player just wants to experiment with different builds ...
... we get screwed with the bill.
#554
Posted 11 February 2017 - 09:46 PM
Kageru Ikazuchi, on 11 February 2017 - 09:18 PM, said:
... or, if a player gets smart and realizes that LRMs are bad, and SRMs are awesome ...
... or, if a player just wants to experiment with different builds ...
... we get screwed with the bill.
our you can get multiple mechs spec'd in different areas
#555
Posted 11 February 2017 - 11:39 PM
BLOOD WOLF, on 11 February 2017 - 09:46 PM, said:
..and how exactly does that make it cheaper to try out different builds?
I agree: go for the paid=unlocked model, allowing more build diversity, less boating, less meta, but only once you indeed have sunk in the full skills money anyway.
Also, really love the many suggestions I read about exponentially increasing the weight of higher level skill nodes. That way it really becomes a choice between specialist to the max (expensive lvl5 unlocks) or general all rounder (lots of different cheap lvl1 and lvl2 nodes). Love that idea!
#556
Posted 12 February 2017 - 12:24 AM
My solution was to revert each tree, one-by-one, until I found the Skill Points I had bought with XP instead of GXP.
An indication on the "flipped/green" skill point hexes of whether XP or GXP had benn spent for the Skill Point would have made it reasonable to figure out where my erroneous XP was expended.
#557
Posted 12 February 2017 - 12:35 AM
MechaBattler, on 08 February 2017 - 03:41 PM, said:
I get the modules were expensive. But we didn't all buy a set of them for each mech. I bought maybe 6 of them. : /
How about you just make the former modules more expensive and make the other skills cheaper?
This is going to be rough for new players who are not only at a disadvantage because they're only starting to collect equipment, but now they have to pay for basic skills. So they have to decide do I get that engine and PPC or do I skill up?
9.1 million cbills not 7.
Best thing for the new system....THROW IT IN THE TRASH!
#558
Posted 12 February 2017 - 01:30 AM
First, the current architecture of the system forces you to take useless or unwanted nodes to drill deeper in the tree. I do understand it is important to prevent specialisation from being too powerful and keep mechs with different weapon systems on an equal footing. As it has already been said in many constructive feedbacks, making the cost of the nodes increases as you go deeper would be a better solution. First it allows you to un-melt the stats (it is a bit of a mess right now) since maxing a stat would be possible but expensive compared to a more mixed approach. It gives the decision back to the player. Secondly it would be more effective and easy to balance to put weapon boating and other builds on an equal footing.
Second, the costs. I'm actually very ok with the intial cost. It may be more expensive since swapping is no longer possible, but it is not a bad thing. I am this kind of swapping guy, and I am sometimes quite annoyed when doing faction play to not own enough modules for equip my whole drop deck. The current system is also better for new player. HOWEVER the respec cost is just crazy. Respec is very important to adapt to meta and balance changes. I do agree it should not be free, but making it cost C-Bills for removing and re-adding a skill point is far too much. Actually I think that removing a skill node should give you back at least 50% of the cost you initially paid (I think 25% would be better, but ... compromise you know ). It would make a respec cost 50% of the initial price versus the crazy 125% currently in place. It is especially important since it is not possible to test a spec choice before locking it.
Some more (less important) ideas:
- The defensive tree looks a bit too strong. increasing the TTK is definitely nice, but making a tree part almost mandatory is not very good. Maybe increase a bit the tankiness of all mechs in the game (or increase cooldowns ?) and reduce the bonuses of the tree accordingly ?
- Give an XP bonuses if you are playing in a non-mastered mech of which you already own a mastered variant (bigger bonus if it is the same variant). It would encourage people to own multiple variants for the same mech or the same variant multiple times with different builds. Or allow XP transfer (with a % loss or a C-Bill/MC cost) between different mechs of the exact same variant, or maybe even between different variants. It would give a interesting use for excessive XP (converting to GXP is very expensive currently, and I bet it is almost unused).
Thank you for continuously trying to make this game better, I really hope the skill tree become as incredible as it could be. And please please please, give some love to the infowar to give lights the role they deserve, this tree is a good start !
EDIT : A small edit to say that for player that have a lots of mechs and used to swap modules a lot, the cost of the skill tree is just too high, and I didn't take that into account enough. A simple solution if you do not want to lower C-Bill prices of unlocking skills: give the money needed to spec the mechs that players already own. It is a lot of C-bills, especially for players that will also get refunds for modules. However it would be the only way to prevent mass departure from experienced player that would feel ripped off by the process.
Edited by Altaille, 12 February 2017 - 05:46 AM.
#559
Posted 12 February 2017 - 01:58 AM
This "skill tree" is awful. You have to spend xp on skills you dont want/need to get the full skill of one you want/need.
If I want to have full radar deprivation an full seismic sensors for example, I have to put 22 of 27 points into the Infotech / Sensors skill cluster, inculding sensor range, target decay and target info gathering. Maybe nice to have, but to me of second, third or fourth priority.
99% of all my builds will look like:
19 points Infotech / Sensors
21 points in ´Mech Operations (with beeing forced to skill hill climb and speed retention for absolutely no reason, and btw these are the most useless skills i would never put money/xp in if there was a logic and real skill tree)
14 points in Defensive (and again beeing forced to spend xp an money on a skill (fall damage) i dont want to spend any xp on)
16 points in Mobility / Lower Chassis to get speed tweak. Maybe some light mechs dont need this, because they are fast enough, but for the others, i have to spend 11 points on not first priority skills to get speed tweak.
The remaining 21 points (19 if I have a ´Mech with AMS) will go to either weapon, magazine capacity or faster cap and enhanced narc, depending on the individual mech an its purpose.
Some facts if not yet clear:
1.500,00 xp per skill
100,000,00 C-Bills per skill
91 skill points to spend = 136.500,00 xp and 9.100.000,00 C-Bills for a full mastered mech.
At this point i have 129 mechs, which will cost me 1.173.900.000,00 C-Bills to master all of them. In case you dont know how much this is:
onebilliononehundredandseventythreemillionninehundredthousand C-Bills
With an average of 150.000,00 C-Bills per game I have to play 7.826 games to get it. 15 Minutes per game (With match-searching and redicilous long returning-to-mechbay-time (why?)) I can do 4 games per hour. That makes 1.956,50 hours. Playing 2 hours per day makes 978,25 days to play, which is more than 2 1/2 years.
Why not doing these calculations BEFORE doing this system? And there are players with more `Mechs than that! And players whot don´t play two hours a day and have no premium time.
Is this how PGI tries to force people to spend MC on C-Bills to master their ´Mechs? Won´t happen! Why spend money on mastering your ´Mech in the first place? To reset your skills ok, maybe you have to pay for it. But to master them?
Spending xp on skills you dont need, spending money on skills. I so hope this gets some more work.
Edited by CptnHero, 12 February 2017 - 02:03 AM.
#560
Posted 12 February 2017 - 07:11 AM
I think 100.000 c-bills for 1 skill is a lot maybe is better to will be cheaper.
Edited by STEEL JUSTICE, 12 February 2017 - 07:13 AM.
9 user(s) are reading this topic
0 members, 9 guests, 0 anonymous users