74 replies to this topic

[Bishop Steiner]

The Lobsters, on 02 March 2017 - 04:42 AM, said:

That new skill tree.....

Well there goes my dream of syncing up my ppc velocities with my autocannons.

which would be part of the idea.... since they spent the last three years intentionally tryign to desync them.

[CMDR Sunset Shimmer]

Zergling, on 02 March 2017 - 06:02 AM, said:

Don't put words in my mouth, or falsely attribute motivations.

I want a working system that makes sense, and can be balanced. Right now you there is a bunch of 'must have' picks that suck of almost all points and prevent any choice from occuring, while players have to spend excessive amounts of points for small gains due to all the junk nodes.


Like, look at how unbalanced the skill points to benefit ratio is in the Firepower tree; 20% Velocity costs 8 points, versus -15% Laser Duration costing 14 points, or -7.5% Missile Spread also costing 14 points.

Range? Ha, 28 points to get +15%.

In an ideal system, each of those bonuses would cost the same amount to max out, and have equal value to builds that make use of them.
But with the trees, that isn't possible; they are simply too complicated for it.

no, you want to beable to pick and choose, wihout wasting points... it's the min/maxer's mentality.

And one of the things PGI is attempting to Stop.

[Duke Nedo]

CMDR Sunset Shimmer, on 02 March 2017 - 06:29 AM, said:

no, you want to beable to pick and choose, wihout wasting points... it's the min/maxer's mentality.

And one of the things PGI is attempting to Stop.

You can't Stop it... if you have choice, people will try to pick the best. The only thing you can do is to balance it so that every pick is a good one.

[Zergling]

CMDR Sunset Shimmer, on 02 March 2017 - 06:29 AM, said:

no, you want to beable to pick and choose, wihout wasting points... it's the min/maxer's mentality.

And one of the things PGI is attempting to Stop.

You are falsely assigning motivations to me because you don't understand what I'm saying. Stop it.

People will min-max whatever PGI does; the simpler system I'm proposing will make the system more balanced so that min-maxing will have LESS benefit than it does in the PTS.

Right now in the PTS, there are obvious picks that cost less points for more benefit than others, which means people that are min-maxing will take those picks.
In the simpler system I'm proposing, each bonus will be balanced directly against each other, so whatever points the player allocates towards, they'll gain a similar amount of benefit; the picks will simply come down to player preference and choice.

Edited by Zergling, 02 March 2017 - 06:54 AM.

[meteorol]

El Bandito, on 02 March 2017 - 02:23 AM, said:

With set of 8 quirks and TCMK1, Jade Kite can actually get 50% CERPPC velocity, in exchange for bit lower energy hardpoints. That equals to 1950 m/s CERPPCs.

You can actually run way bigger TCs on the JK, because it will most likely run out of critspace before you run out of tonnage if you are using a TC1. With 3 tons of gauss ammo, you will end up with something like 2.5 tons leftover if you use a TC 1. You can skip on one heatsink and add a TC 4.

[CMDR Sunset Shimmer]

Duke Nedo, on 02 March 2017 - 06:34 AM, said:

You can't Stop it... if you have choice, people will try to pick the best. The only thing you can do is to balance it so that every pick is a good one.

Except then nothing shines.

Min/Maxers are attempting to do things such as sync velocity's of weapons, [believe me I understand the desire, it's one of the reasons I fire the SRM and AC20 together on my atlas, because similar range and velocity profiles.]

Which PGI has been trying hard to stop, because it's one of the things causing TTK being far too short. Granted, the "Easy" fix to their biggest issues with TTK is to put in a damned Cone of Fire mechanic... but they'll never do that because you all whine too damned much about it.

So, this is the decision, and because of it, some things just won't be "The best choice" that's never going to happen, it's just unrealistic to even think it CAN happen.

I've been looking at the skill tree, and while yeah, there are some nodes that I groan over having to take, it's not the worst thing ever, and I feel it's an improvement over the initial iteration.

[Zergling]

Just look at how many points it costs to get LBX spread; 16 points for -10% spread, versus 14 points for -7.5% missile spread, or -15% laser duration.

Or 17 points for 5% UAC jam chance, which is a pathetically weak bonus for that many points.

Like, just look at the 28 points it takes to get +15% Range; the problem there is a lot of other bonuses in such a range build, some of which are worthless for some builds, but useful for others.
But the tree system can't differentiate between worthless and useful bonuses!


Even cross tree comparisons show how skewed the 'benefit to points' ratio is.

Eg, for 22 points I can get -15% laser duration, -5% heat generation, -3.2% cooldown, +4.5% range in the Firepower tree.

But for 21 points in the Operations tree, I can get +10% heat dissipation, +15% max heat, -35% startup duration, -35% screen shake, +10% hill climb, +20% speed retention.
The Operations tree bonuses are clearly better; 10% heat dissipation and 15% max heat just can't be beaten by -5% heat generation, -15% laser duration and -3.2% cooldown.

Survival tree? That's 20 points for ridiculously good armor and structure buffs. It'd be stupid not to take those.


PGI can go ahead and increase the costs again for those trees, but that is a counter-productive method to force players to not select the attractive options, as it just restricts the total number of choices the player can make.

Edited by Zergling, 02 March 2017 - 07:05 AM.

[Paigan]

Zergling, on 02 March 2017 - 07:04 AM, said:

Just look at how many points it costs to get LBX spread; 16 points for -10% spread, versus 14 points for -7.5% missile spread, or -15% laser duration.

Or 17 points for 5% UAC jam chance, which is a pathetically weak bonus for that many points.

Like, just look at the 28 points it takes to get +15% Range; the problem there is a lot of other bonuses in such a range build, some of which are worthless for some builds, but useful for others.
But the tree system can't differentiate between worthless and useful bonuses!


Even cross tree comparisons show how skewed the 'benefit to points' ratio is.

Eg, for 22 points I can get -15% laser duration, -5% heat generation, -3.2% cooldown, +4.5% range in the Firepower tree.

But for 21 points in the Operations tree, I can get +10% heat dissipation, +15% max heat, -35% startup duration, -35% screen shake, +10% hill climb, +20% speed retention.
The Operations tree bonuses are clearly better; 10% heat dissipation and 15% max heat just can't be beaten by -5% heat generation, -15% laser duration and -3.2% cooldown.

Survival tree? That's 20 points for ridiculously good armor and structure buffs. It'd be stupid not to take those.


PGI can go ahead and increase the costs again for those trees, but that is a counter-productive method to force players to not select the attractive options, as it just restricts the total number of choices the player can make.

You don't pay 17 points for a 5% bonus.
You get other stuff along the way.
[Redacted]

Another [Redacted]:
-35% startup duration is a higher number than -3.2% cooldown. But guess which one is better.
Buffing firepower results in increased "processing power" of a Mech, while stuff like hill climbing and startup durection are nice gimmicks at best.
Thus, the 22 firepower points you listed are CLEARLY better than the 21 Ops points you listed.
[Redacted]
Also consider:
Weapon stats do tactically multiply each other. More cooldown PLUS less heat makes a weapon way better than just summing up both modifiers.

Edited by draiocht, 02 March 2017 - 10:41 AM.
insults

[MOBAjobg]

Zergling, on 02 March 2017 - 04:50 AM, said:

Want velocity buffs? Put skill points into that, directly, without having to waste any points on junk you don't want.

Just have a set of categories, where each skill point in one bonus is worth balanced against the bonus gained from a skill point in another category.
Keep limits to the maximums to how many points can be put into a specific bonus to prevent it from being stacked too high, of course, but otherwise give players the freedom to put points into the things they want.

There doesn't have to be anything like diminishing returns either. The system can be incredibly simple and still allow for a ton of customisation/choice if the bonuses and number of skill points available are balanced right.


Example of this at work, Diablo 3's Paragon points system:


Although in that case, the bonuses are segregated into 4 categories. Such segregation between firepower/mobility/etc categories could also work in MWO.


Further example of how it'd work (don't take this as anything other than a rough example):

Weapon Cooldown, 0.5% per skill point, maximum of 10 skill points / 5% CD
Weapon Velocity, 5% per skill point, maximum of 5 skill points / 25% Velocity
Acceleration, 5% per skill point, maximum of 5 skill points, 25% Acceleration

Just let players put points into whatever bonus they want, with no restrictions of having to spend points in other stuff first.
The trees in the PTS are completely unnecessary, complicating the whole system and frustrating players with all the junk nodes.

My P1302 character knows what you're talking about.

Edited by MOBAjobg, 02 March 2017 - 07:31 AM.

[Zergling]

Paigan, on 02 March 2017 - 07:11 AM, said:

You don't pay 17 points for a 5% bonus.
You get other stuff along the way.
[Redacted]

Another [Redacted]:
-35% startup duration is a higher number than -3.2% cooldown. But guess which one is better.
Buffing firepower results in increased "processing power" of a Mech, while stuff like hill climbing and startup durection are nice gimmicks at best.
Thus, the 22 firepower points you listed are CLEARLY better than the 21 Ops points you listed.
[Redacted]
Also consider:
Weapon stats do tactically multiply each other. More cooldown PLUS less heat makes a weapon way better than just summing up both modifiers.

Go do some calculations for -5% heat generation versus +10% heat dissipation combined with +15% max heat.


I went and did some myself: IS mech with triple large pulse lasers, 10 engine heat sinks, 8 external heat sinks.

-heat generation mech: fires every 3.92 seconds, generates 19.55 heat per salvo, dissipates 3.2 heat per second, has 62 maximum heat threshold.

0 seconds: fires first salvo, +19.95 heat, current heat is 19.95, 32.18% of max heat
3.92 seconds: -12.544 heat, current heat level is 7.41, 11.95% of max heat.
3.92 seconds: fires second salvo, +19.95 heat, current heat level is 27.36, 44.12% of max heat

Time to cool from 19.95 heat to 0 heat = 8.55 seconds


+heat dissipation mech. fires every 3.92 seconds, generates 21.00 heat per salvo, dissipates 3.52 heat per second, has 71.3 66.5 maximum heat threshold.

0 seconds: fires first salvo, +21 heat, current heat is 21, 29.45% 31.58% of max heat
3.92 seconds: -13.80 heat, current heat is 7.20, 10.10% 10.83% of max heat
3.92 seconds: fires second salvo, +21 heat, current heat is 28.20, 39.55% 42.41% of max heat

Time to cool from 28.20 heat to 0 heat = 8.01 seconds


As is plainly obvious, the mech with +10% heat dissipation and +15% max heat is far superior to a mech with -5% heat generation, to the point that even with -15% laser duration and -3.2% cooldown, it is still inferior.


And what about builds that don't benefit at all from laser duration, like PPC builds? They still have to pay 17 points to get the 5 heat generation nodes on that side of the tree.
Hell even -8% heat generation is still inferior to +10% heat dissipation and +15% max heat, and that costs 25 points.


EDIT: going further with the PPC build; it costs 26 points to get -8% heat generation and +20% velocity. Alternatively, it costs 27 points for +10% heat dissipation, +15% max heat AND +20% velocity. Just 1 more point, for far better heat management with the same velocity bonus.

Edited by Zergling, 02 March 2017 - 09:01 PM.
Quote Clean-Up

[CMDR Sunset Shimmer]

Zergling, on 02 March 2017 - 07:31 AM, said:

Go do some calculations for -5% heat generation versus +10% heat dissipation combined with +15% max heat.


I went and did some myself: IS mech with triple large pulse lasers, 10 engine heat sinks, 8 external heat sinks.

-heat generation mech: fires every 3.92 seconds, generates 19.55 heat per salvo, dissipates 3.2 heat per second, has 62 maximum heat threshold.

0 seconds: fires first salvo, +19.95 heat, current heat is 19.95, 32.18% of max heat
3.92 seconds: -12.544 heat, current heat level is 7.41, 11.95% of max heat.
3.92 seconds: fires second salvo, +19.95 heat, current heat level is 27.36, 44.12% of max heat

Time to cool from 19.95 heat to 0 heat = 8.55 seconds


+heat dissipation mech. fires every 3.92 seconds, generates 21.00 heat per salvo, dissipates 3.52 heat per second, has 71.3 maximum heat threshold.

0 seconds: fires first salvo, +21 heat, current heat is 21, 29.45% of max heat
3.92 seconds: -13.80 heat, current heat is 7.20, 10.10% of max heat
3.92 seconds: fires second salvo, +21 heat, current heat is 28.20, 39.55% of max heat

Time to cool from 28.20 heat to 0 heat = 8.01 seconds


As is plainly obvious, the mech with +10% heat dissipation and +15% max heat is far superior to a mech with -5% heat generation, to the point that even with -15% laser duration and -3.2% cooldown, it is still inferior.


And what about builds that don't benefit at all from laser duration, like PPC builds? They still have to pay 17 points to get the 5 heat generation nodes on that side of the tree.
Hell even -8% heat generation is still inferior to +10% heat dissipation and +15% max heat, and that costs 25 points.


EDIT: going further with the PPC build; it costs 26 points to get -8% heat generation and +20% velocity. Alternatively, it costs 27 points for +10% heat dissipation, +15% max heat AND +20% velocity. Just 1 more point, for far better heat management with the same velocity bonus.

Did you ever think, that maybe the point of this system, is to force you to diversify your builds? You know, like the Canon tends to?

boating is an issue, and PGI's been trying to push you AWAY from boating weapons... this skill system actually seems to reward a spread of weapons, instead of specialization, to make the most use of the skill tree.


You, are trying to make the square peg [the skill tree] fit into the round hole [the current meta]. And it's not going to work.

[Zergling]

CMDR Sunset Shimmer, on 02 March 2017 - 07:56 AM, said:

Did you ever think, that maybe the point of this system, is to force you to diversify your builds? You know, like the Canon tends to?

boating is an issue, and PGI's been trying to push you AWAY from boating weapons... this skill system actually seems to reward a spread of weapons, instead of specialization, to make the most use of the skill tree.


You, are trying to make the square peg [the skill tree] fit into the round hole [the current meta]. And it's not going to work.

Did you ever think that having certain build options being completely superior to others, results in zero choice and diversity?


See, the trees are the problem here; the Devs can't just buff the -heat generation nodes to be in balance with +heat dissipation and +max heat in the Operations tree, because to get the -heat generation nodes requires taking a bunch of other nodes along with them.
For builds that make use of the bonuses from those other nodes, it results in an overpowered build. Yet if they leave it as it is now, for the builds that don't make use of those nodes, it is an underpowered build.

The only way to achieve balance is to allow players to select nodes without prerequisites, balance them directly against each other only by point cost/bonus value and without the need to consider other bonuses that may or not be useful to the build.

Edited by Zergling, 02 March 2017 - 08:11 AM.

[Mawai]

Skill systems in which the higher levels of skills cost more are very common and are more or less self-balancing.

If you want the best in each category then you end up with less in other categories.

The example above of 1 point/skill level so that level 5 costs 5 is one way to do it and would work.
Another way is to double the point costs/level so it goes 1, 2, 4, 8 which makes the last level cost more than the rest combined but you could also increase the bonus ... maybe on a linear scale.
1 = +1%
2 = +2%
3 = +3%
4 = +4%
5 = +5%

Maximum available would be 15% but that last 5% would cost you proportionally more if each level cost 1,2,4,8,16.

(I am not suggesting this is a good system but just using it as an example of the truly huge number of skill allocation systems that are possible. EVE online uses the linear skill increase with an exponential time requirement to learn the skills).

However, I think the pure linear reward/linear cost does not do enough to limit specialization. From a balance perspective, in my opinion, really narrow specialized bonuses should require less depth across the skill tree so that the player is sacrificing more to get that last level in a skill.

Finally, I generally agree with the comments on the "skill tree" ... I think it is better if the players can decide what skills they want for particular mechs and not have to try to optimize picks from a skill tree to minimize the bonus values they don't want in order to get some of the bonus values they do want. This isn't the kind of design that works well at keeping players involved or encourages folks to try alternate builds. I think the skill tree approach leads to one or two mostly optimal ways through a tree while scaled point allocation leads to uncertainty about whether the cost of the last level or two of a skill is worth the expenditure.

P.S. I like the fact that velocity skills apply to all ballistics ... someone mentioned trying to synchronize the velocities of all their projectile/energy weapons ... I don't think the skill tree should facilitate that.

[Gas Guzzler]

DaZur, on 02 March 2017 - 05:09 AM, said:

As I see it... Fine, you want to commit to a PPFLD build? Understand "all" your other attributes will be stunted to afford you that luxury.

So you want mechs to really only specialize in one thing? No thanks.

Honestly as it is I had to leave 3 of the trees completely empty after only filling in a 3rd of the weapon tree.

You also have to realize that things like PPCs are hotter now, and ACs have much lower cooldowns, even you manage to spend all the points required to get up to 8.8% cooldown which takes a TON of points.

You also have to realize that most mechs feel like they are walking in molasses as it is, so you can't skimp on agility.

El Bandito, on 02 March 2017 - 06:07 AM, said:

Correct. Rifleman-3N for example, got its ballistic velocity quirk reduced from 40% to 20% in the new PTS patch, and you have to make that up with the skill tree. Meanwhile, KDK-3 can get 20% velocity, something it could never have before, just from skill tree alone. Which is bullcrap.

At least my ARC-5W can lurm with +30% missile speed cause its velocity hasn't been nerfed...

Agreed on the Clan vs IS part... hopefully they will take this opportunity to address the tech imbalance

Edited by Gas Guzzler, 02 March 2017 - 08:17 AM.

[El Bandito]

Zergling, on 02 March 2017 - 07:31 AM, said:

As is plainly obvious, the mech with +10% heat dissipation and +15% max heat is far superior to a mech with -5% heat generation, to the point that even with -15% laser duration and -3.2% cooldown, it is still inferior.

And what about builds that don't benefit at all from laser duration, like PPC builds? They still have to pay 17 points to get the 5 heat generation nodes on that side of the tree.
Hell even -8% heat generation is still inferior to +10% heat dissipation and +15% max heat, and that costs 25 points.

EDIT: going further with the PPC build; it costs 26 points to get -8% heat generation and +20% velocity. Alternatively, it costs 27 points for +10% heat dissipation, +15% max heat AND +20% velocity. Just 1 more point, for far better heat management with the same velocity bonus.

Just wanna tell you that if the heat dissipation and max heat skills work the same as current skill tree, it only counts the base value of 30 heat that comes with the engine, not the external and internal heatsinks, which makes the skills weaker than they appear.

Edited by El Bandito, 02 March 2017 - 08:17 AM.

[Felicitatem Parco]

Wait... OP says it only takes 8 nodes to get massive velocity Buffs, and that's a bad thing because you should have to invest more points to get that much value...

then the next Super Popular idea presented in their thread is a "direct point system" that would let you get your full velocity Buffs with *fewer* points to invest because you can bypass the unwanted points...

Make up your da** minds, people.

Edited by Prosperity Park, 02 March 2017 - 08:20 AM.

[Zergling]

El Bandito, on 02 March 2017 - 08:16 AM, said:

Just wanna tell you that if the heat dissipation and max heat skills work the same as current skill tree, it only counts the base value of 30 heat that comes with the engine, not the external and internal heatsinks, which makes the skills weaker than they appear.

Derp, my bad on that mistake. Thanks for catching it.

Here's the redone +heat dissipation / +max heat calculated data:

Fires every 3.92 seconds, generates 21.00 heat per salvo, dissipates 3.52 heat per second, has 66.5 maximum heat threshold.

0 seconds: fires first salvo, +21 heat, current heat is 21, 31.58% of max heat
3.92 seconds: -13.80 heat, current heat is 7.20, 10.83% of max heat
3.92 seconds: fires second salvo, +21 heat, current heat is 28.20, 42.41% of max heat

Time to cool from 28.20 heat to 0 heat = 8.01 seconds


That is still better than this:

Zergling, on 02 March 2017 - 07:31 AM, said:

-heat generation mech: fires every 3.92 seconds, generates 19.55 heat per salvo, dissipates 3.2 heat per second, has 62 maximum heat threshold.

0 seconds: fires first salvo, +19.95 heat, current heat is 19.95, 32.18% of max heat
3.92 seconds: -12.544 heat, current heat level is 7.41, 11.95% of max heat.
3.92 seconds: fires second salvo, +19.95 heat, current heat level is 27.36, 44.12% of max heat

Time to cool from 19.95 heat to 0 heat = 8.55 seconds

-8% heat generation is about equal to +10% heat dissipation / +15% max heat. Heat % levels are very slightly higher in the +heat dissipation build (about 0.5% higher at each time), but time to cool off is very slightly faster.


EDIT: -8% heat generation is inferior with builds that have higher heat dissipation, like 6x ER Medium Laser with 10 internal + 20 external heat sinks.

Edited by Zergling, 02 March 2017 - 09:01 PM.

[Metus regem]

This is going to highlight again, that not all quirks skill nodes are created equal.... Just like Hard Points and hard point locations.... I foresee this going sideways really fast....

[Gas Guzzler]

Bandito, you realize that the velocity affects LRMs as well right? I honestly don't think this is a big deal, and weapons still won't sync up any better because you can't just boost your PPCs to be close to your Gauss like you could last time. Also, I thoight velocity only took like 6 nodes last time through the skill tree... soo its more costly now, espeically because survival and mobility are much more costly.

Edited by Gas Guzzler, 02 March 2017 - 08:34 AM.

[Zergling]

Prosperity Park, on 02 March 2017 - 08:19 AM, said:

Wait... OP says it only takes 8 nodes to get massive velocity Buffs, and that's a bad thing because you should have to invest more points to get that much value...

then the next Super Popular idea presented in their thread is a "direct point system" that would let you get your full velocity Buffs with *fewer* points to invest because you can bypass the unwanted points...

Make up your da** minds, people.

No, my idea was for velocity bonuses to be balanced in points requirements with other bonuses.

The actual value of the points in my examples are irrelevant as they are only placeholders figures; it could just as easily be +20% velocity for 20 points as it could be for 5 points, so long that it is balanced against other bonuses.

Edited by Zergling, 02 March 2017 - 08:42 AM.