9 replies to this topic

[BearFlag]

TL;DR
1) linear trees
2) number of nodes in tree decided by value
3) reduced total nodes/reduced SP, presumed
4) last node in path greatest effect
5) combine Duration/Velocity into an 'all weapons' buff
6) separate idea, remove model quirks, roll in as skill bonus

I still strongly support the idea of a Skill Tree system and appreciate PGI's efforts. The improvement to decel is welcomed. A couple of matters were not taken on in PTS 2. The sheer number of nodes and having to traverse "undesirable" nodes were part of the original concerns expressed by many players. Offered here is an example remedy. Some of the ideas are mine, many have been suggested by forumites.

(Almost) All trees become a straight progression. The popularity/effectiveness of a skill determine the number of nodes in its tree. Speed Tweak or Range might have five or six nodes while Hill Climb may have only one or two. "High value" skills cost more points; low value skills less. As can been seen in the example, this approach is more direct and intuitive. Overall the number of nodes would be reduced, perhaps by about a quarter. The limiting factor would still be available Skill Points which would be reduced in line with fewer nodes. In most cases, as an irritating enticement, the last node in a tree is most potent.

Following on PGI's idea to consolidate weapons charateristics in general, and in order to more strongly support diverse buildouts, Duration and Velocity are combined here into a buff for all weapons (except flamer). All Weapons Basic nodes would benefit all weapons. Mechs which are nearly compelled to mixed loadouts should appreciate this.

Individual weapon type improvements are moved to "Weapons Special" which is also a rare example of gating in this system. A number of nodes of Duration/Velocity must be traversed before weapon specialization is opened.

Trade-offs would come into play through SP exhaustion. There's nothing to stop a player from bee-lining straight to those buffs he most adores. But with the number of nodes/SP being adjusted, he may have to give up Torso Yaw for full Speed Tweak.

An entirely separate idea is folded into the chart. The notion is to eliminate all model quirks and replace them with Skill Tree bonuses. In the example, the Victor 9K is receiving a Weapons Basic bonus of 10%. So if he Skills his range to +10%, he actually receives 11%. The bonus would apply to all Weapons Basic buffs. The upside to this is that quirks can removed entirely (I think). The downside is that lackluster mechs would lose those stock compensations they get through quirks. An under-performing mech would be a real dog until Skills kick in. However, with strong Skill Tree bonuses such Ugly Ducklings could grow into really mean Swans.

Notes about the chart: The numbers are mostly from PGI's total buff number in a skill. The example is not focused on exact numbers. Not all Skill Trees are included and those included may not be complete. The chart is IS, of course.

[R Valentine]

Why eliminate quirks? Clans are already a step ahead of IS. Quirks is all that keeps IS afloat.

[BearFlag]

Kiran Yagami, on 04 March 2017 - 03:15 PM, said:

Why eliminate quirks? Clans are already a step ahead of IS. Quirks is all that keeps IS afloat.

Just an idea to play with and not essential to the tree. The quirks would be replaced with Skill Tree bonuses - skills would be amplified for those mechs that need it. As mentioned, stock mechs would be hurt until Skilled up. In a sense, specific quirks would be replaced with category quirks. So not really eliminated. I should have said "changed."

Edited by BearFlag, 04 March 2017 - 03:47 PM.

[Cato Phoenix]

Essentially my idea of using quirks inherently in the skill tree, although a bit more generalized whereas mine was more individualized per variant.

So range using yours using a weapons basic bonus of 10%:

1 node = 2%+(*.1)= 2.2%
2 node = 4%+4*(.1)=4.4%
3 node = 6%+6*(.1)= 6.6%
4 node = 8%+ 8*(.1)=8.8%
5 node = 10% +10*(.1)=11%

However, that difference is pretty marginal compared to an unquirked clan mech, say. But the margin is just math, really.

My system proposes, say, 1.5x modifier to victor's range:

1 node = 2% *(1.5)= 3% (1% buff over good mechs)
2 node = 4%*(1.5)=6% (2% buff over good mechs)
3 node = 6%*(1.5)= 9% (3% buff over good mechs)
4 node = 8%*(1.5)=12% (4% buff over good mechs)
5 node = 10%*(1.5)=15% (5% buff over good mechs)

So you are rewarded for building encouraged / lore builds, and rewarded for buying further into the skill tree.

My system proposed enfolding the current quirks into the skill tree - so rather than just a weapon basic boost you could have a gauss rifle cooldown boost, a laser weapon duration boost, etc.
https://mwomercs.com...23#entry5631123

Now, it doesn't give unskilled mechs a baseline boost, true, but it does allow them the choice of spending points where they want, rather than just having a perhaps unused quirk. And quirks currently just add a flat bonus, rather than an additive one - again, nothing necessarily against that, just a philosophical difference.

Using quirks:

1 node = 2%+5%= 7%
2 node = 4%+5%=9%
3 node = 6%+5%= 11%
4 node = 8%+5%=13%
5 node = 10%+5%=15%

I think that using quirks - you get a bigger benefit of spending less SP in the tree, and multipliers - each point increases your benefit.

Edited by Cato Phoenix, 04 March 2017 - 05:06 PM.

[Alistair Winter]

I agree with most of this, but you're effectively just replacing quirks with skill bonuses. They would still need to be listed and balanced for each mech. And PGI would need to write code for that.

I think the current system does the job just as well. That being said, I'd love to replace all minor quirks (5% improved cooldown or 10% missile speed) with extra skill points.

The list of quirks on each mech is a god damned mess, it's like 50 different bonuses, of which most are basically inconsequential. It's messy, it's ugly, it's annoying. I think PGI should keep the major quirks, like the mobility quirks for the Locust and the armor quirks for the Victor, and replace all the minor quirks with extra skill points.

[Gas Guzzler]

BearFlag, on 04 March 2017 - 03:38 PM, said:

Just an idea to play with and not essential to the tree. The quirks would be replaced with Skill Tree bonuses - skills would be amplified for those mechs that need it. As mentioned, stock mechs would be hurt until Skilled up. In a sense, specific quirks would be replaced with category quirks. So not really eliminated. I should have said "changed."

Not a fan of garbage tier mechs pre-mastery that get massive boosts in the skill tree to be competitive with the good mechs...

[BearFlag]

Cato Phoenix, on 04 March 2017 - 05:00 PM, said:

Essentially my idea of using quirks inherently in the skill tree, although a bit more generalized whereas mine was more individualized per variant.

So range using yours using a weapons basic bonus of 10%:

1 node = 2%+(*.1)= 2.2%
2 node = 4%+4*(.1)=4.4%
3 node = 6%+6*(.1)= 6.6%
4 node = 8%+ 8*(.1)=8.8%
5 node = 10% +10*(.1)=11%

However, that difference is pretty marginal compared to an unquirked clan mech, say. But the margin is just math, really.

My system proposes, say, 1.5x modifier to victor's range:

1 node = 2% *(1.5)= 3% (1% buff over good mechs)
2 node = 4%*(1.5)=6% (2% buff over good mechs)
3 node = 6%*(1.5)= 9% (3% buff over good mechs)
4 node = 8%*(1.5)=12% (4% buff over good mechs)
5 node = 10%*(1.5)=15% (5% buff over good mechs)

So you are rewarded for building encouraged / lore builds, and rewarded for buying further into the skill tree.

My system proposed enfolding the current quirks into the skill tree - so rather than just a weapon basic boost you could have a gauss rifle cooldown boost, a laser weapon duration boost, etc.
https://mwomercs.com...23#entry5631123

Now, it doesn't give unskilled mechs a baseline boost, true, but it does allow them the choice of spending points where they want, rather than just having a perhaps unused quirk. And quirks currently just add a flat bonus, rather than an additive one - again, nothing necessarily against that, just a philosophical difference.

Using quirks:

1 node = 2%+5%= 7%
2 node = 4%+5%=9%
3 node = 6%+5%= 11%
4 node = 8%+5%=13%
5 node = 10%+5%=15%

I think that using quirks - you get a bigger benefit of spending less SP in the tree, and multipliers - each point increases your benefit.

Credit where credit is due. A good system. Yes the numbers could be adjusted. I suspect, but can't be sure, that PGI is aiming at some level of general nerfing to up TTK. Numbers would be easier if we knew all their design goals. More refined bonuses are good way to buff and buff in a certain direction. A general buff allows players to choose any weapons system to focus on. Again, a matter of what end result if preferable.

Alistair Winter, on 04 March 2017 - 05:07 PM, said:

I agree with most of this, but you're effectively just replacing quirks with skill bonuses. They would still need to be listed and balanced for each mech. And PGI would need to write code for that.

I think the current system does the job just as well. That being said, I'd love to replace all minor quirks (5% improved cooldown or 10% missile speed) with extra skill points.

The list of quirks on each mech is a god damned mess, it's like 50 different bonuses, of which most are basically inconsequential. It's messy, it's ugly, it's annoying. I think PGI should keep the major quirks, like the mobility quirks for the Locust and the armor quirks for the Victor, and replace all the minor quirks with extra skill points.

I can see replacing small quirks with an SP or two reducing quirks overall. For that matter, extra SPs could be the means of "balancing" disadvantaged mechs either in whole(?) or part.

Gas Guzzler, on 04 March 2017 - 09:25 PM, said:

Not a fan of garbage tier mechs pre-mastery that get massive boosts in the skill tree to be competitive with the good mechs...

Isn't that what balancing quirks ~try~ to do now? I'm for PGI making money by selling mechs. But right now they're doing it with the Mech of the Month Club. Shouldn't the poor Victor be sellable? To me, it seems the way to do that is to make poor mechs competitive.

Edited by BearFlag, 05 March 2017 - 08:59 PM.

[Kmieciu]

I'm not sure if PGI is actually capable of creating separate skill trees for every mech variant.

What I'm certain of is, they don't want to do this - because that would make the system even tougher to balance than quirks.

Edited by Kmieciu, 06 March 2017 - 12:23 AM.

[BearFlag]

Kmieciu, on 06 March 2017 - 12:23 AM, said:

I'm not sure if PGI is actually capable of creating separate skill trees for every mech variant.

What I'm certain of is, they don't want to do this - because that would make the system even tougher to balance than quirks.

Guessing you're talking about the bonus thing. Like I said, not necessary to the Tree idea. But, to clarify, the trees would not be unique, rather the bonus. For balancing it would reduce the number of parameters from [however many quirks there are] to a few categories.

I would think (could be wrong) that fewer parms would make balancing easier.

[Kaeb Odellas]

BearFlag, on 05 March 2017 - 08:58 PM, said:

Isn't that what balancing quirks ~try~ to do now? I'm for PGI making money by selling mechs. But right now they're doing it with the Mech of the Month Club. Shouldn't the poor Victor be sellable? To me, it seems the way to do that is to make poor mechs competitive.

No, what we mean is that mechs shouldn't be garbage when you buy them and only usable after grinding up its skills. A Storm Crow is still plenty powerful with no skills unlocked, whereas a Kintaro with no quirks and no skills is worthless. Giving a Kintaro extra skill points is not going to fix that.