45 replies to this topic

[Fetherator]

How about using a skilltree like this:

lets say we have 10Skills Laser Range,
1-4 Skills use 1 Point each = 4 Points used
5-7 Skills use 2 Points each = 6 Points used
8-9 Skills use 3 Points each = 6 Points used
10th Skill use 5 Points = 5 Points used
It uses now 21 Points for full Laser Range example.

So what does PGI do?
They do Excact the same thing, but they give u some Extra on Top for using more points u dont want.

So the tree could be way smaller in size and besides more easy to handle, but hey...its their way to make it.

[HANGMAN1962]

how about just giving the chosen mech what it all ready has in game and maybe a few extra points to custemize it ?!!!! need to stop hiding those skills behind other skills.
stream line it.
soo you have 3 levels of a skill put then in a row and not in the middle of the frikin skill tree either i shouldnt have to by a whole skilltree to get ecm skills are you kidding me?!!
you cant get the skill effects your mech allready has for less then 120 to 130 skill points becouse of hiding them behind other skills.
soo arange them in an orderly manner to let us actualy have some points left over to custemize each mech and not gimp every mech we own!
thats a big issue for meny people as they own literaly hundreds of mechs!!!!!!!!!!
here's another way to look at it: that 50 doller boars head premium mech wont work as well as the way you spent 50 to buy it in the 1st place?!!

[KodiakGW]

Made two suggestions during the first PTS. One for Sensors, and one for Mech Operations. Completely ignored by PGI. They didn't touch Sensors at all, and when they changed Mech Operations they still left skills buried under Hill Climb and Improved Gyros. Two skills I have no interest is wasting SP on. So, I'll give you a link to my original post, but I'm not going to waste my time updating the Mech Operations tree. They are just going to ignore it again because I'm not one of their "inner circle."

Link (copy paste if it won't direct link)
https://mwomercs.com...99#entry5611099

[AWOL 01]

Psst...

Alistair...

What program did you use to make your skill tree? It looks pretty great...

[Alistair Winter]

AWOL 01, on 03 March 2017 - 07:17 AM, said:

Psst...

Alistair...

What program did you use to make your skill tree? It looks pretty great...

[Alistair Winter]

Fetherator, on 03 March 2017 - 06:01 AM, said:

How about using a skilltree like this:
lets say we have 10Skills Laser Range,
1-4 Skills use 1 Point each = 4 Points used
5-7 Skills use 2 Points each = 6 Points used
8-9 Skills use 3 Points each = 6 Points used
10th Skill use 5 Points = 5 Points used
It uses now 21 Points for full Laser Range example.
So what does PGI do?
They do Excact the same thing, but they give u some Extra on Top for using more points u dont want.
So the tree could be way smaller in size and besides more easy to handle, but hey...its their way to make it.

I see a lot of people talk about using diminishing returns, in different ways. Either make skills increasingly expensive, or make each skill upgrade less and less powerful (5%, 4%, 3%, 2%, 1% buffs).

But what does this accomplish, exactly? There needs to be a philosophy behind it. If it's applied universally, the only thing you accomplish, from what I can tell, is that you are stopping people from specializing. Generally speaking, people will take a little mobility, a little survivability, a little extra firepower and a little extra jump jet ability, because the lowest skills are so much cheaper than the max skills.

I'm not against diminishing returns, but there needs to be some sort of purposeful concept behind it, beyond simply "let's nerf the specialists". I'm not really convinced that the most OP mechs are the ones that max their skills, necessarily.

[1453 R]

Is art required? I have the artistic talent of a prepubescent capibara, but have otherwise been ruminating on possible rebuilds of the skill tree for a good bit of time here (when not trying to convince people that mobility is good for MWO and we should try and keep it).

Admittedly most of my ideas are more in the line of somewhat Mass Effect-ish semi-branched trees with mutually exclusive choices and possibly limited point budgets per branch, which I'm not sure most others would care for. But the overall intent is to allow a player to take something from a branch if they want that something whilst making it painful to take everything from said branch, and dispensing with this "you get either everything or [X]% of Everything because it's impossible to specialize in the current skill tree" nonsense. Also to introduce more interesting benefits and effects than simply +2% Butt Scratching Speed. Because nobody is really excited to finally earn the cash to lay out for their next +2% Butt Scratching Speed node.

[Alistair Winter]

1453 R, on 03 March 2017 - 08:06 AM, said:

Is art required? I have the artistic talent of a prepubescent capibara, but have otherwise been ruminating on possible rebuilds of the skill tree for a good bit of time here (when not trying to convince people that mobility is good for MWO and we should try and keep it).

Admittedly most of my ideas are more in the line of somewhat Mass Effect-ish semi-branched trees with mutually exclusive choices and possibly limited point budgets per branch, which I'm not sure most others would care for. But the overall intent is to allow a player to take something from a branch if they want that something whilst making it painful to take everything from said branch, and dispensing with this "you get either everything or [X]% of Everything because it's impossible to specialize in the current skill tree" nonsense. Also to introduce more interesting benefits and effects than simply +2% Butt Scratching Speed. Because nobody is really excited to finally earn the cash to lay out for their next +2% Butt Scratching Speed node.

Art is not required, but it's generally easier to communicate this sort of stuff with diagrams than with walls of texts or, even worse, 2000-word posts with bulletpoints and 30 different chapters and sub-chapters.

I like the Mass Effect skill tree, however I'm not sure if global diminishing returns is better for the game.

Edited by Alistair Winter, 03 March 2017 - 08:26 AM.

[Appuagab]

I would make something like this. All values are obviously placeholders.



A few notes:

1. Well, it's not exactly linear and still has some fillers, but I tried to make them reasonable for further nodes. Like you need more precise dense shots at higher range or better cooling efficiency when you have lower cooldowns.

2. I think that skill tree shouldn't have any nodes increasing heat capacity at all because big heat capacity is one of the roots of alphastrikes meta issue. That's why I didn't include them and put some cooling nodes instead. Yes, I think that cooling efficiency belongs to weapons branch rather than mech operations. As well as advanced zoom.

3. As you may have guessed, my Gauss nodes imply that chargeable Gauss does not explode. Because it shouldn't. At least when it's not charged.

4. I believe that various nodes should have different SP prices so I included them as a placeholder in case I bother myself making other branches later.

[Trev Firestorm]

Random thought: See Armored Core 4/for Answer tuning system, its all we really need.

Maybe it wouldn't translate well but its a simple layout with good potential variety.

Edited by Trev Firestorm, 03 March 2017 - 08:20 PM.

[MischiefSC]

Here's my tree -

All Skills have 6 nodes and are cumulative.

Each branch is separate because you can then quirk mechs by giving them bonus nodes to apply to set branches. So, for example, the Atlas gets + 3 nodes in Survival Branch.

SURVIVAL
======================
Armor:
1: + 1%
2: + 1%
3: + 1%
4: + 2%
5: + 2%
6: + 3%

Structure:
1: + 1%
2: + 1%
3: + 1%
4: + 2%
5: + 2%
6: + 3%

Crit Reduction:

1: + 1%
2: + 1.5%
3: + 2%
4: + 3%
5: + 4%
6: + 6%
=======================

ALL WEAPONS - by having one set of quirks that impact all weapons you promote mixed and varied builds because you're not forced to drill down on a single weapon subset.
=======================

Cooldown:

1: + 1%
2: + 1%
3: + 2%
4: + 3%
5: + 3%
6: + 5%

Range:

1: + 1%
2: + 1%
3: + 2%
4: + 3%
5: + 3%
6: + 5%

Velocity (ballistics and missiles)/Crit Chance (for lasers)

1: + 1%
2: + 1%
3: + 2%
4: + 3%
5: + 3%
6: + 5%

=================

Make sense? I don't have time to work out sensors and the nifty stuff like ammo buffs here. Maybe have a selection of single node perks that require 2 pts each, like an ammo bonus perk.

JJs would need 2 or 3 x the bonus currently provided and it needs to be flat values, such as 3m bonus jump range, etc. It needs to be as valuable as comparable weapon quirks.

Then give players enough points to max out 45% of the total quirks.

[Elizander]

PGI can easily set unmarked tiers in the skill tree based on gains and points spent to balance it out with other trees. This can work and it won't matter how many or how few points you have in the skill system.

This means that getting 25% of the agility bonuses will cost (x) points, getting the next 50% of it will cost (x+y) points and getting the last 50% for 100% effectiveness will cost (x+y+z) points.

Now getting (x) points should have equivalent value among all the available skills that you can get with (x) points and the same should go with (x+y) and so on. This doesn't have to be exactly equal, but relatively close enough to make it a choice for some players.

A good choice does not mean that the player-base is split 50-50 between a certain path on a tree. You can have it 80-20 and it can still be a good choice to make.

The problem with these mixed nodes is that it messes up and dilutes the values of the points spent to get something. Separation in the tree is also necessary (at least for the initial design) to see how far a player is willing to go to get something before picking off items in another branch. This does not mean that it is impossible to value the current tree as it is, but it just puts more factors in that you have to balance out. A question to ask yourself is, "Is spending 22 points in survival returning an equal value when I spend it on mobility or weapons?" If not, then the separate trees are not equal and should be re-evaluated or further categorized into tiers of superior and inferior trees which just makes the whole thing more complicated to balance out.

A simple way would be to break the 91 points down into 10 point segments with an 11th point spent down one path being exceptional compared to the rest while getting 10 points normalizes your mech in some way (just as an example). Meaning the more 11 point goodies you pick up down branches means you will be more deficient in other aspects.

A tree should have a default setting that is considered the balanced setting and this would be your baseline battlemech and picking more 11 point items will reduce your effectiveness in one area or other.

Numbers can change, of course. Can be 5-6 or 20-21 or whatever, but PGI should also set their tree and tell (at least themselves) that "This is the default tree setting for all mechs" and adjust changes from there.

Edited by Elizander, 03 March 2017 - 10:16 PM.

[Antares102]

+1 to OP

[WolvesX]

Still like this one very much.

[Chound]

KodiakGW, on 03 March 2017 - 07:01 AM, said:

Made two suggestions during the first PTS. One for Sensors, and one for Mech Operations. Completely ignored by PGI. They didn't touch Sensors at all, and when they changed Mech Operations they still left skills buried under Hill Climb and Improved Gyros. Two skills I have no interest is wasting SP on. So, I'll give you a link to my original post, but I'm not going to waste my time updating the Mech Operations tree. They are just going to ignore it again because I'm not one of their "inner circle."

Link (copy paste if it won't direct link)
https://mwomercs.com...99#entry5611099

I think the idea is If you want that Seismic sensor you have to pick up three or four skills you don't want to add to the "cost" of getting the node. XP, gxp, C-bills and other nodes that you could have used for something elsee. you can't get all the best items because there aren't enough nodes available.Selecting 91 out of 220 isn't enough for their big bosses I guess.

[C E Dwyer]

Oberost, on 03 March 2017 - 03:28 AM, said:

I posted on another topic during the PTS1...

This is a tier based skill tree:



In this tree you can choose only one skill from each Tier, and every skill is balanced with the other skills from that Tier. The result is this:



If you balance things to be almost equal in usefulness you will have to make REAL choices, because picking one skill will prevent you from picking the other two useful skills on that Tier.

Do the skill trees different AT LEAST for each weight class and you'll have some more balance and variety.

Hell, you can even use hexagons instead of rectangles if this fits you better...

This or something very similar is what I hoped to see, obviously there would be more categories, and more choices.

There would be a series of columns each level unlocked would give you a choice of skills you pick one from the level and the rest are locked, the second level reached and again a set of choices and only one can be picked until all 'levels' for that category are filled, or you chose not to complete all 'levels'.

There would be a different tree for all weight classes, with different possible skill choices. Each choice if it's common skill between two or more weight classes would have separate values.

[Pyed]

Alistair Winter, on 02 March 2017 - 01:13 PM, said:

To everyone who feels that PGI should actually use some sort of skill tree to promote role warfare and give players more freedom to customize... can you do better than PGI did?

Yep. Have done it twice. (One only covered weapons).

Quote

But if you want to see some sort of skill tree that resembles what PGI is working on today, can you actually do better than them?

Yep. One of them resembles PGI's tree a lot more than the other though.

Quote

Note: The goal should be to avoid giving every mech in the game all the best skills

We're of one mind here.

Quote

I'd love to see people post some suggestions in this thread, and then I will edit them into this opening post.

Here ya go:

https://mwomercs.com...age__p__5639154

https://mwomercs.com...age__p__5614417

[Leroy the Walrus]

My big criticism of the current skill tree is the mixed "skill-salad" nature of the nodes results in very generic skill setups, even for drastically different builds. The example I've used is a laser-brawler vs a laser-sniper/skirmisher - your firepower tree right now ends up looking the same for both, even though they're dramatically different mech builds. There's not really a lot of real choice in the tree, and by choice I mean things that involve real trade-offs.

My point was to allow people to improve the way their mech performs in certain situations, but with a definite cost to how it performs in others. You should be able to specialize your skills for the function you want, but costs/trade-offs increase as you try to get better at everything in a category, so it starts cutting into the points you have for other categories.

The suggestion was to put all related skills in a "group" (ex: laser weapons), and let you put points into any sub-group you want (range, duration, heat, cool-down, each having several increments), but point costs increase as the total points invested in the group increases - i.e. the first five skills you allocate in the laser group cost 1 point each, whatever they are, but the next five skills cost 2 points, then three, etc. You could do the same with the "mobility" group (accel/decel, turn speed, twist speed, twist range, max speed).

This allows you to focus on the 1 or 2 aspects of a tree that fit your build best (i.e. laser range) with relative ease, or get modest bonuses in several categories, but to do everything laser-related better (range, duration, heat, cooldown) cuts into the points left for other trees.

Original post on the topic here:
https://mwomercs.com...via-trade-offs/

[MechaBattler]

I wonder if PGI will budge on their non linear spider web setup. I'm sure they've noticed people wanting a more linear design. So they seem pretty intent on their setup.

[Alistair Winter]

MechaBattler, on 04 March 2017 - 07:02 PM, said:

I wonder if PGI will budge on their non linear spider web setup. I'm sure they've noticed people wanting a more linear design. So they seem pretty intent on their setup.

The feedback seems surprisingly unified. From what I can tell, most people are saying:

• We want a simpler system, not 90+ skill points and a skill tree with hundreds of nodes
  
• We want a more linear skill tree, not this scrambled web of nodes in random positions
  
• The current system with so many different skills + quirks per mech is too complex. PGI should make the system more easy to understand, and more easy to balance.
  
• Weight class balance is not good enough, light mechs and medium mechs are suffering too much from mobility nerfs.

I'll be very interested to see if PGI will listen to this. To me, there seems to be a fairly solid consensus on certain major points.