36 replies to this topic

[Fantastic Tuesday]

TL;DW: (although I'd really prefer you watched the video)

The Skill Tree fails to meet the stated design goal's in PGI's writeup.

It massively makes boating more effective. Many fun-but-good builds are now dead.

It utterly kills the ability to change builds around between different weapon types.

The proposed fixes:

Skill nodes should be hard choices. Ie: you either buff your acceleration or you buff your leg structure. You either increase your beam range or you increase your beam duration. Etc.

[Gas Guzzler]

Fantastic Tuesday, on 09 February 2017 - 04:12 PM, said:

It massively makes boating more effective. Many fun-but-good builds are now dead.

Dead? Why dead? Less optimal sure... but dead?

[Fantastic Tuesday]

Gas Guzzler, on 09 February 2017 - 04:13 PM, said:

Dead? Why dead? Less optimal sure... but dead?

I was typing this quickly. The video explains it far better.

[Champion of Khorne Lord of Blood]

I'm really hoping they don't scrap the idea entirely. I'm actually liking the general concept, but it could do much better with some changes.

Lowering the costs, giving weak mechs a multiplier to the bonuses they can buy (basically instead of 1% per node they get like 3% or more or less depending on how good or bad they are), giving people more weapon skill nodes so that boating and not boating allows for you to still upgrade the weapons you have fully and leave the same number of remaining points, and giving people the option to fully specialize their mech by getting rid of the junk nodes no one wants, giving people straight lines to go down, then making those lines longer would go a long way to improving the system.

This way someone could pick to be generalist as they are now with a mix of weapon, structure, mobility, and info tech, or someone could go fully into defense and have a super tanky mech or they could go fully into offense and have very powerful weaponry yet die quickly. People could even go full mobility or full sensors and be great pokers or scouts. I feel that would give people a lot of choice on specialization and make them happy while also giving people real decisions on whether they want to have a stronger weapon, tank more, or dodge better and picking one weakens the others potential.

[Gas Guzzler]

Okay, I get it... you can do a ridiculous amount of skills when you only have one weapon tree.

[slodeath]

I agree with your observations. I looked at the Stormcrow. I have 3 variants, all mastered. I will receive 57,291 exp refund per variant. this gives me 38 points once converted. that is enough to do the lower chassis tree except turn 1,2 and 3 and the Mech Operations tree except improved gyros with 1 point left. this leaves me nothing for torso or arm speed or degree of movement which was in the original skill list. now I know I didn't have as much hard brake in the original skill list as I do now, but you have to take it to get to the 7.5% speed tweak I do have now.

I think PGI needs to move the points for the things the mechs get now in the current skill list up to where they can be achieved with the xp being refunded and leave all the other stuff further down the tree to be taken as desired. maybe put them in a tree all by themselves that must be unlocked before any other skill trees can be used and make the first 38 no c bill cost.

I am not talking about weapon, consumable or sensor points here as they were never in the original skill list.

[Flying Blind]

Yup this video pretty much nails it

[DAYLEET]

I dont think forcing a choice between cooldown and duration and range is the right way to go but they should definitely not be linked.

Look at it this way, the other mech tree are really powerfull, much more than the weapon buff and you will always want them more than weaponry in this iteration because spending for cooldown when i want less duration might not be worth it. It would be reasonable to unlink weapon buff so i can cooldown my STD laser and range my Pulse so i can sync them a bit. Alternatively i could velocity my ac20 and range my SL. Linking them cost too much for too little and just encourage boating.

The rest of the tree is the same but i would keep it linked for now. If you unlink it with the weapon then you can really pick and chose only the best. While this open some possibility to specialise into some rare stuff like Fall Damage, i dont think it will be useful for more than 10 of my 161mech and ill just take the very best all the time. I like that this game will be more about strenght and survivability than creeping firepower, well i hope so.

[Lovas]

The big thing is one of the things you said, is that this tree is very lackluster/unimaginative/very little decision making is needed by the player (NO ONE is gonna touch the UAV tree, maybe 1% will touch the JJ tree).

No way an actual game designer made this; I think their interns took all the possible skills and modules and put them in these "trees" - although these trees have very little to no branching - they should be called "skill telephone poles".

Edited by Lovas, 09 February 2017 - 09:22 PM.

[Fantastic Tuesday]

Lovas, on 09 February 2017 - 09:21 PM, said:

they should be called "skill telephone poles".

I love this.

I really hope they programmed it in such a way that making exclusive choices is even possible. Otherwise I'm not sure how this Tree will even be made to work.

[McMurl]

I think tuesday nailed it. New players will suffer tremendously, boating is not only encouraged, but is basically the only viable option, there is no real decision making or trade offs in this current iteration, the list goes on.

Tuesday is also right to assume that THIS SYSTEM IS INCREDIBLY PROMISING, PGI IS HEADED IN THE RIGHT DIRECTION. HOWEVER IT DOES NEED SOME CHANGES.

I agree, each node should be occupied by 2 different skills. for this node, do you want Cool Run or Heat Containment? In this node do you want Laser Duration or Laser cooldown? Srm Spread or Srm Cooldown? Hell you could even replace the Range nodes with Optimal range or Fall off range, where falloff range is lets say 2% and optimal is 1%. There needs to be trade offs. Also, there should be a specific amount of points per tree, not just a giant pool. Like you get 30 points to spend in the weapons pool, 30 points in the movement tree, etc. Tuesday is also right in saying that it should be front loaded, for 2 very simple and logical reasons: It doesnt promote boating if you get more points early in each tree. Lets say you have Small Pulse and SRMs. Why would you want to get 1% extra SRM range when you could get 3% SPL range? Boating becomes less attractive, save for mechs that do very well when only bringing 1 weapon system, IE nova or Black Knight. Also because it lets new players have the ability to actually contribute to a team fight much sooner than later, because they arent gimped by not having the uber skills that veterans have. Getting the later tier skills should be an end game thing for that chassis, requiring you to really hone the desires of your play style ETC.

PGI, as much as Fantastic Tuesday can be nails on a chalkboard trolly, In this instance, he has absolutely nailed what this system should be. Its a good start, but it desperately needs changes if you want the community to accept it.

I am also in total favor for mechs having exclusive skill nodes. Vindicator is a PPC mech, why not give it, idk, a special node, an end-of-tree node that makes its PPCs do more crit, or when hitting an ECM mech with a PPC, that mechs ECM is turned off for a longer duration?

Highlander is a jumper, why not allow it the ability to jump much higher than it currently can, make the JJ tree more powerful, or even let it regenerate jump juice mid air. Special abilities exclusive to each mech chassis or even variant to really give it an edge in specific instances, or even to distinguish mechs from each other, IE grasshopper and black knight.

PS: Fekkin nerf the damn missile trees... between the uber crit chances now and the super pinpoint alpha, SRMs are just ridiculous, and critting the crap out of something at 1000m with LRMs isnt cool either.... Either that or lower the crit chances...

Edited by McMurl, 10 February 2017 - 01:59 AM.

[AngrySpartan]

As with a previous PTS you nailed that! That's the best wrap up of my concerns and thoughts on a new skill tree and thank you for doing that video FT.

For whoever read's this, I encourage you to at least leave a comment in this topic to grab PGI's attention. Hopefully they will listen to at least some of those thoughts and go back to a drawing board.

One addition - mech's should have some unique nodes in their respective trees. E.g. Highlander&Victor should get something for JJs and MLX should be a supreme scout mech with it's hardwired BAP, etc. That's hardly a topic for early PTS, yet it's something that I hope skill tree will achieve eventually.

[Kuaron]

You are late, Fanta, I already linked your video in another thread.

[Fantastic Tuesday]

McMurl, on 10 February 2017 - 01:49 AM, said:

Fantastic Tuesday can be nails on a chalkboard trolly.

You just crit my emotions.

Edited by Fantastic Tuesday, 10 February 2017 - 08:56 AM.

[InsaneRotta]

Many of the points FantasticTueday made, mirror my thoughts about the new skill trees. There are a lot of bottlenecks in the trees that don't make sense. In Tuesdays video at 2:29 you can see that Fall Damage 3 node lead to AMS Overload 2. How is that tree logical? Those nodes have nothing in common, but yet you need to take this if you would desire to take full ams overload.

Another wild card is the arm speed / torso speed conundrum - To get full torso speed quirks, oh sorry SKILLS, you also need all of the arm speed nodes which are 8% each making it 40% in total. With that speed my mechs arms were flailing around wildly and I had to reduce my mouse sensitivity so low I couldn't torso twist properly anymore. Kinda felt like my mech was this guy



-.-

Edited by InsaneRotta, 10 February 2017 - 11:01 AM.

[Mcgral18]

Fantastic Tuesday, on 10 February 2017 - 01:18 AM, said:

I really hope they programmed it in such a way that making exclusive choices is even possible. Otherwise I'm not sure how this Tree will even be made to work.

Adding a flag should be nothing at all

[Taxxian]

That is not Torso SPEED

0.8%

So instead of 100 degrees/second I get 100.8 degrees per Second? What framerate do I need to theoretically beeing able to at least notice the difference? (without lab equipment^^)

Edited by Taxxian, 10 February 2017 - 01:38 PM.

[cazidin]

Taxxian, on 10 February 2017 - 01:30 PM, said:

That is not Torso SPEED

0.8%

So instead of 100 degrees/second I get 100.8 degrees per Second? What framerate do I need to theoretically beeing able to at least notice the difference? (without lab equipment^^)

About 100.8 frames per second?

[Fantastic Tuesday]

Mcgral18, on 10 February 2017 - 12:01 PM, said:

Adding a flag should be nothing at all

You say this. But the skill tree tanks my fps to about 60 below what I get in a match. Anything is possible.

[Gas Guzzler]

I don't really like the exclusive choice thing tbh. I don't see what agility and durability should be exclusive. What if you want to be agile and durable? What if you want agility and pew pew? What if you want pew pew and durability? What if you want all 3 and no infotech, but also no upper chassis agility?

I don't know, in my personal opinion, I'm not a fan of exclusive choices, but I do think some of the nodes need to be nerfed, including accel/decel, velocities, durations, jam chance, and some others.