17 replies to this topic

[Mycroft000]

There seems to be enough people who are okay with the web layout of the skill tree as currently designed, that it's leading me to ask, am I missing something?

Does the tree currently actually promote real meaningful choices? As far as I can tell it really doesn't. The tradeoffs for taking high value skills aren't steep enough to keep me from accessing them, and when I take the high value skills I want, I get LOTS of extra functionality.

I know it seems counterintuitive that I think that's a bad thing, but I really do. Knowing I pretty much always want Seismic and Radar Dep means that I will always be getting some very good sensor benefits as well. I feel like taking the defensive sensor skills should come at the cost of not being able to take the offensive sensor skills. I'm not actually having to make a choice of "Do I want to know when someone is nearby when I'm standing still, or do I want to know what components to aim at when I'm firing at them?" I get both, and I get the info gathering effectively for free since I am going to take Radar Dep and Seismic anyway.

Why isn't there something that makes it so that if I want to be much more capable of taking damage(increased structure and armor), I do so at the expense of Agility?

This system has nothing built into it that makes you actually have to weight cost vs. benefit.

If I'm wrong, explain it to me, I want to know.






Also this has nothing to do with the need to rethink the GSP refunds, we need a better refund solution than that.

[Capt Deadpool]

I understand where you are coming from, as currently designed, the give and take is between different trees, not within them. You spend the points on ecm, derp/ and/or seismic, that is a good chunk of point you will not have to put into another tree.

That being said, I agree that the choices aren't very meaningful for a different reason: the relative weakness of the firepower tree to the other trees. Increasing the potency of the firepower nodes would make choices. More difficult and meaningful. I do not see that happening, however, as the unstated underlying purpose of the skill tree is to increase time to kill.

I do feel like the PTS build will be pretty decent as soon as they re-quirk the underpeforming mechs to bring them into line with new newly buffed (via skill tree) clan mechs, and increase agility on the mechs that rely on it (locust, other lights, linebacker).

[Mycroft000]

As far as I'm concerned, with rare exceptions(meaning very specific builds) the firepower tree simply doesn't exist.

[Capt Deadpool]

Yep, that is the reason there is an illusion of choice, not really meaningful choices.

[Mycroft000]

Yeah, I've boiled my "default skill loadout" down to 87 nodes that are going to be virtually mandatory on 90% of my builds leaving me 4 points to choose how I use. And I'm betting that my template is going to be fairly close to what ends up as the "solved" tree. It gives the maximum benefit while using the fewest number of extraneous skills.

[Mycroft000]

By the way, I REALLY want dissenting opinions here. Show me I'm wrong.

[Ruar]

There are a few who disagree, but not many. Most people fully recognize the failings of the web system and have suggested many other options that make more sense. The few who disagree typically see the web as the only way to force people to take low value skills and are afraid if there wasn't a forced design then people would just cookie cutter a min/max setup. The irony is the web design forces a min/max setup even more than other methods because the gate nodes make it feel like you have fewer options thus forcing people into very similar paths.

As you said, the only real difference is going to be which section you choose and I think even that is going to solidify into avoiding the firepower tree in the main and then a matter of whether or not you fill in survival or agility. Those choices will just lead to a min/max for a tank build or a mobility build.

[Capt Deadpool]

You aren't gonna get them because you are correct Except from players who mistakenly believe the firepower tree has more value than it does.

The REAL reason for the skill tree is increasing TTK, not to provide players with meaningful choices. (I am fairly ambivalent about increasing or decreasing TTK, BTW. I will be effective regardless.)

[Mycroft000]

I think the optimal choices end up giving you a strong mix of both agility and survivability. I've posted my Optimal Skill Layout for those interested.

I'm just frustrated because I feel like a stronger design would truly make us weigh the pros and cons between deciding if Seismic is more important or Radar Dep. The combination of those two are part of what has made me turn the Nova Prime into such a devastatingly broken chassis.

[MustrumRidcully]

The "meaningfulness" of the choice seems to be currently that to take a desirable skill, you might have to spend a lot of points on skills you don't necessarily need or want, and then cannot buy other skills. That's a very indirect form of meaningful.
It's further hampered by the individual skills being rather low buffs to stats, so you have a lot of mini-chocies but not a single, clear choice to make.

It lacks a certain degree of transparency in the decision making, because it's all so indirect, and you only see the small values.

A sparser tree with more explicit choices with larger bonuses but overall less picks would convey meaningful choices better, I think.

Edited by MustrumRidcully, 28 April 2017 - 10:06 AM.

[BodakOfSseth]

Or you're going to get people who don't care enough to really speak up. While silence = consent when voting, the dynamic is much more nuanced when talking social interaction.

For some people boiling their choices down to a standard selection for most of their builds is going to be fine - mainly those people who run basically the same mech with different geometry.

Other people will find greater variation in their choices.
Going into survivability for a light mech is probably not that meaningful, since getting a few points of armor on their CT isn't going to mean much against high alphas.
OTOH,
Going into Mobility for a 100t assault may likewise not very meaningful since engine decoupling will cause them to be a bit less mobile.

It's all a numbers game, depending on the player and the build. Adding 10% range to a SPLas build can be fairly meh unless a player really thinks an extra 11m is valuable, while adding 10% to an ERPPC build will get you almost 100m extra.

As previously mentioned, selections to fill out an area means you are less able to fill out other areas. If those areas are not valuable to you, then so be it. Others may worry and stress over the options. For me, the nodes I select for my Raven are wildly different than the ones I select for my Battlemaster.

Finally, this is a test - it sounds like pretty much everyone is poo-pooing the value of the weapons tree. To make it more attractive, PGI may change the values there, boosting it to make it more attractive. We'll see when they release it into production.

Edited by ScottAleric, 28 April 2017 - 09:47 AM.

[Ruar]

ScottAleric, on 28 April 2017 - 09:46 AM, said:

Or you're going to get people who don't care enough to really speak up. While silence = consent when voting, the dynamic is much more nuanced when talking social interaction.

For some people boiling their choices down to a standard selection for most of their builds is going to be fine - mainly those people who run basically the same mech with different geometry.

Other people will find greater variation in their choices.
Going into survivability for a light mech is probably not that meaningful, since getting a few points of armor on their CT isn't going to mean much against high alphas.
OTOH,
Going into Mobility for a 100t assault may likewise not very meaningful since engine decoupling will cause them to be a bit less mobile.

It's all a numbers game, depending on the player and the build. Adding 10% range to a SPLas build can be fairly meh unless a player really thinks an extra 11m is valuable, while adding 10% to an ERPPC build will get you almost 100m extra.

As previously mentioned, selections to fill out an area means you are less able to fill out other areas. If those areas are not valuable to you, then so be it. Others may worry and stress over the options. For me, the nodes I select for my Raven are wildly different than the ones I select for my Battlemaster.

Finally, this is a test - it sounds like pretty much everyone is poo-pooing the value of the weapons tree. To make it more attractive, PGI may change the values there, boosting it to make it more attractive. We'll see when they release it into production.

Good points all.

What I don't understand is the lack of interaction by PGI on their own forums. You'd think they would offer some kind of response in regards to what is being said. Instead there is silence while we try to figure out where to focus on testing, argue over possibly moot issues, and give up testing because it seems like it doesn't truly matter.

I don't expect a lot of commitment while the testing process is happening, but I do expect some kind of "we are aware players aren't happy with the current design and are looking at how it can be changed" or "We hear the comments about the tree design but please provide some feedback on how decoupling is affecting mechs 65ton + and 35 ton -".

Instead we just get crickets making the test so much less effective than it could be.

[Capt Deadpool]

Ruar, on 28 April 2017 - 10:01 AM, said:

Good points all.

What I don't understand is the lack of interaction by PGI on their own forums. You'd think they would offer some kind of response in regards to what is being said. Instead there is silence while we try to figure out where to focus on testing, argue over possibly moot issues, and give up testing because it seems like it doesn't truly matter.

I don't expect a lot of commitment while the testing process is happening, but I do expect some kind of "we are aware players aren't happy with the current design and are looking at how it can be changed" or "We hear the comments about the tree design but please provide some feedback on how decoupling is affecting mechs 65ton + and 35 ton -".

Instead we just get crickets making the test so much less effective than it could be.

I agree. I have always been surprised and disappointed by the lack of developer communication on this forum compared to almost every other game I have played. Then again, it seems to me the average poster on this forum is very negative towards the game and immature compared to other game forums, dissuading developer interaction...

[Karl Streiger]

Capt Deadpool, on 28 April 2017 - 10:31 AM, said:

I agree. I have always been surprised and disappointed by the lack of developer communication on this forum compared to almost every other game I have played. Then again, it seems to me the average poster on this forum is very negative towards the game and immature compared to other game forums, dissuading developer interaction...

Up to today almost every white Knight had surrendered, the roll of the fallen and turned Knights is much longer as the numbers still in defiance to the obvious.

There was never much 2 way communication in the Forums.
Sry for the Off topic
+1 OP

[Ruar]

Capt Deadpool, on 28 April 2017 - 10:31 AM, said:

I agree. I have always been surprised and disappointed by the lack of developer communication on this forum compared to almost every other game I have played. Then again, it seems to me the average poster on this forum is very negative towards the game and immature compared to other game forums, dissuading developer interaction...

Don't go to the WoT forums then. That community is brutal. And I spent a few years playing EVE online, which is actually polite compared to WoT.

This forum is pretty nice as well for the most part. Biggest gripes I see is the lack of communication and listening to the player base on major changes. Which is common in a lot of game devs I've seen over the years. They tend to think they know best in everything instead of focusing on big picture and using player ideas to fill in the details to match that big picture.

[Mycroft000]

Okay so I'm comparing maxing out the Agility tree to my template. What my agility skill selection gives vs what max gives:



Does using the extra 22 nodes give enough performance increase in areas that actually matter in real gameplay to go ahead and use?

[Ruar]

mycroft000, on 28 April 2017 - 11:11 AM, said:

Okay so I'm comparing maxing out the Agility tree to my template. What my agility skill selection gives vs what max gives:



Does using the extra 22 nodes give enough performance increase in areas that actually matter in real gameplay to go ahead and use?

Honestly, that is up to you if it's valuable enough. I assume that decision will be more about what mech you are using and the intended role than anything else.

My Roughneck absolutely needs added agility, my Assassin not so much.

[Mycroft000]

I understand there will be exceptions, but I see this selection of nodes in the agility tree as a fairly optimal across the board set. It gets you to the bulk of the most critical agility skills without spending more than 2x the points on fairly incremental gains. Meaning this set gives me 22 nodes to spend elsewhere at a loss of 3% top speed, 5% deceleration, and 15% turn rate; the arm and torso nodes I consider to be among the "useless" nodes that are a tax on my agility choices.