21 replies to this topic

[Tristan Winter]

I may be biased here, since most of my experience from skill trees comes from MMORPGs*. But I was hoping that the skill tree would make it easier to play according to a certain style. I'm used to thinking in terms of specialization. DPS spec warrior, stealth spec rogue, tank spec commando, healing spec jedi consul or whatever.

In MWO, the skill tree is obviously a lot deeper, making it harder to define cookie-cutter roles like that. But I would still imagine certain paths to take, like equipping AC20 and pulse lasers and then focus on all the mobility skills that let you move fast in a brawl, while turning and torso twisting to distribute damage, for example. Or equipping dual gauss and focus on all the skills that let you attack from long range with good long range sensors, good accuracy and high rate of fire, for example.

I feel like it's a bit too hard to carve out a niche like this. I understand that PGI wants to make it harder to get a level 2 seismic sensors, for example, but does the two seismic sensor skills have to be on the opposite side of the branch? This forces players to just level up a generic "sensor" role, instead of tuning sensors to fit their exact need.

Do you like that the position of the nodes are "scrambled", forcing you to invest more fully in a branch in order to get all the best skills, like seismic sensors and speed tweak? Or would you prefer a more traditional MMORPG approach, where it's easier to identify a path to achieve a certain specialization?

* I'm not strictly basing this on MMORPGs, however. If you play Star Conflict, you'll see that the skilltrees in that game are designed to encourage certain roles in much the same way. For example, if you want to play a frigate as either long range fire support or short range drone repair ship, you can specialize in either role without being forced to take certain skills that you don't really want.

Edited by Tristan Winter, 09 February 2017 - 04:48 AM.

[Ibrandul Mike]

For getting a bit of everything the system as it is on the PTS is better.
For specialising your system looks better. You could argue with the distribution of classes but yeah.

I would prefer the "easy" system with linear specialisation, especially without having to get not needed skills.
Which btw. neither system guarantees. If I am for what ever reason interested in the Speed and Ammo I still would need to take a lot of unnecessary points. even more so in the proposed system. (10 vs 13 to get all speed and ammo points on the shortest route)

There will be no way to get the skills without fillers that you don't really want, which is ok.
The system I would prefer is the "get what you want and ignore the rest" system... which would probably a really bad idea. THAT would be cookie cutter build land.

[Old-dirty B]

I would like a very simple system or structure where nothing is obstructed by anything (mostly useless / less desirable) but allows the user to make specific choices what and thus what not to get. To put my idea or view into the very extreme you would get 5 skill points that you may distribute over the available branches:

• Firepower
• Survival
• Mobility
• Operations
• Infotech

You may dump them all into one branch or 1 on each or anything in between. 5 points in a branch get you the full effect, one point gets you the least amount of effect that branch has to offer.

Examples:
* 5 points in firepower will get you the most cooldown, speed and ammo capacity, one point will gain a 20% of the total improvement.
* 5 points in survival will get you the best armor, one point will gain just a 20% of the total improvement.
* 5 poins in mobility will get you the fastest, most mobile chassis, one point will gain just a 20% of the total improvement.
* 5 points in operations will get you the most heat capacity and dissipation etc, one point will gain just a 20% of the total improvement.
* 5 points in infotech will get you the most sensor range, seismic range, rader derp time etc, one point will gain just a 20% of the total improvement.

Basically you determine how big of an effect / gain you will get by spending points.

Thats what i would like to see.

Edited by B3R3ND, 09 February 2017 - 05:06 AM.

[Tristan Winter]

Ibrandul Mike, on 09 February 2017 - 04:53 AM, said:

For getting a bit of everything the system as it is on the PTS is better.
For specialising your system looks better. You could argue with the distribution of classes but yeah.

I would prefer the "easy" system with linear specialisation, especially without having to get not needed skills.
Which btw. neither system guarantees. If I am for what ever reason interested in the Speed and Ammo I still would need to take a lot of unnecessary points. even more so in the proposed system. (10 vs 13 to get all speed and ammo points on the shortest route)

There will be no way to get the skills without fillers that you don't really want, which is ok.
The system I would prefer is the "get what you want and ignore the rest" system... which would probably a really bad idea. THAT would be cookie cutter build land.

I'm not sure there's any way to avoid cookie cutter builds, really. After all, min-maxers are gonna min-max, and even if you tax them with more unnecessary skills, they are still going to find a way to reduce as many unnecessary skills as possible. For example, they might use the whole survivability branch but carefully leave out the AMS skills, even though the AMS nodes are spread far apart, or they may find the way to get as much weapon cooldown as possible and as little extra heat bonus as possible, if they're playing a gauss build.

You're definitely right that neither model presented above gives the player total freedom to pick any combination of skills. After all, then it wouldn't be a "tree", as such. It would just be a bucket (lol) of available skills you could pick. You still want some kind of branch structure if you want to encourage variety.

Now, I don't want to look too closely at the quick model I knocked up above, because it wasn't meant as a literal alternative to the current ballistic branch. But the trick would be to make it easy for the players to group together useful combinations. It's not a problem if it's impossible to have max range bonus and max ammo bonus if you decide that (1) this doesn't fit the concept of a max range sniper or (2) few people will want that particular combination anyway.

It's a bit tricky for weapons, because I'm sure a lot of players will want to max out the whole weapon skill branch anyway. Other branches have more skills that aren't as useful to combine.

[ForceUser]

Having played around with the current one I've already run into multiple cases where I had pretty interesting choices to make regarding how I would get to a specific skill I wanted, how to avoid unnecessary ones and getting the most out of points I had left.

See the point is that this isn't a 'real life problem'. What I mean by that is in the real life you could just tell the designer to change the layout or structure to be exactly how you want it so you don't have to make any sacrifices. Think for example of a game puzzle. You can just tell the dev to implement a button that instantly solves the puzzle. The end result is what the dev wanted right? The puzzle is solved right? But that's not the point though now is it? The point is that this is a puzzle that you need to solve withing the way it's designed. And the best part is it's possible to solve it in multiple ways depending on what you have or need at that moment.

A practical example in the current system (I have like 3 from mastering just 4 different mechs btw). Say I only have 10 points left and I want to get the bonus ammo skill at the bottom of the one tree. Now I can go left or I can go right down the tree. However the one side has an AMS (iirc) node while the other has the Gyro node. That means say it's for a light or an AMS focused mech, I then want to go left down the line but say it's for a heavy brawler I want that pretty nice gyro node so I go right.

That right there is the game presenting a puzzle, and me solving it in a way that gives me a slight advantage over someone who didn't solve it correctly for that particular situation. The game is rewarding me for being clever (or doing research). It's easy (and boring) to just tell the dev to make it so I can have my cake and eat it too but that;s completely missing the point of it being a GAME.

Edited by ForceUser, 09 February 2017 - 05:28 AM.

[Old-dirty B]

i hate puzzles when im here to move around a "vehicle" that shoots.

That aside, once this puzzle is solved we all know how to get seismic and radar derp most efficient and the "fun" of this puzzle is gone. Then we all know what nodes to choose and what not and all result down to the same builds and variety we already know.

I want to see some basic fundamental choices that have a dramatic effect on the outcome, in short you choose for a tank, a scout, a glass cannon or support or a bit of everything. Slap that on top of the mechs / weight classes we already have and you have a great way to setup mechs for a specific way of play, to emphasis its strong and / or fix its weak spots.

Edited by B3R3ND, 09 February 2017 - 05:40 AM.

[Tristan Winter]

ForceUser, on 09 February 2017 - 05:27 AM, said:

That right there is the game presenting a puzzle, and me solving it in a way that gives me a slight advantage over someone who didn't solve it correctly for that particular situation. The game is rewarding me for being clever (or doing research). It's easy (and boring) to just tell the dev to make it so I can have my cake and eat it too but that;s completely missing the point of it being a GAME.

I definitely agree that there are some interesting choices to be made right now. It's not just a matter of setting the course and levelling up a pre-determined path. In time, this will probably change a little bit. For example, I thought it would be interesting to decide how many skillpoints to spend on the jump jet branch. Turns out that the best answer is probably "none".

Now, in regards to your last point (GAME), I would say that there are two ways of looking at this. If you look at my illustration above, the skill tree on the left may make it more entertaining to pick skills. But it's possible that the skill tree on the right makes it easier to customize, which may (depending on your perspective) make it more fun to play the actual game (as B3r3nd alluded to). Instead of diluting each "role", it could make it easier to differentiate between snipers, brawlers, strikers, support mechs, scout mechs, or whatever other roles one can think of.

[Ibrandul Mike]

Get away from the roles!!!!!!!!!

They might take it and the skill tree and just give you the choice between the roles you just named!

With their choice of skills "appropriate" for the roles!

[Tristan Winter]

Ibrandul Mike, on 09 February 2017 - 05:50 AM, said:

Get away from the roles!!!!!!!!!

They might take it and the skill tree and just give you the choice between the roles you just named!

With their choice of skills "appropriate" for the roles!

Like this?

http://mwomercs.com/...le-warfare-cont

[ForceUser]

B3R3ND, on 09 February 2017 - 05:38 AM, said:

i hate puzzles when im here to move around a "vehicle" that shoots.

That aside, once this puzzle is solved we all know how to get seismic and radar derp most efficient and the "fun" of this puzzle is gone. Then we all know what nodes to choose and what not and all result down to the same builds and variety we already know

Only I gave an exact example of how there isn't just one solution to the puzzle. That's the point. Before there was NO choices AT ALL. At best you got to choose what to get first but that's IT (seriously, the meta guides list this because that's all there is to list lol), everyone got the exact same skills. Now you can get what, less than 1/3 of the total available points and what you want to do with the mech radically changes what you pick up.

[Kuaron]

Just finished preparing and launched a quite related thread to this one, with a more global and versatile approach:
https://mwomercs.com...weapon-boating/
Mentioned this one there.

[ForceUser]

Tristan Winter, on 09 February 2017 - 05:47 AM, said:

I definitely agree that there are some interesting choices to be made right now. It's not just a matter of setting the course and levelling up a pre-determined path. In time, this will probably change a little bit. For example, I thought it would be interesting to decide how many skillpoints to spend on the jump jet branch. Turns out that the best answer is probably "none".

Interesting thing here with the jumpjet thing. I put together my PPC warhammer first and then I had a look at my mostest favorite PPC jump sniper the (loyalty)summoner. The difference between what skills I picked between mechs that are very, VERY close in terms of build was rather funny. I focussed on fall damage 'subnodes' for the summoner, I actually picked up the JJ skills. Where I picked out both legs of the speed tweak tree for the warhammer (for 7.5% speed) I only had enough points for half the tree (still 4.5% speed) for the summoner. Yea I hope they buff the JJ skills a bit but since I'm spending waaaay more of my time jump jet sniping and taking leg damage in the summoner and since it already has a big engine, I need the mobility skills for the warhammer waaaaay more than the summoner.

This is the kind of stuff that has me excited about this skill tree.

[Cy Mitchell]

I would prefer a more linear tree and a lot less skill points to invest. I do not like the grind and I do not play games that require a lot of grinding to avoid being at a significant disadvantage over and above my own lack of skill.

[Buehgler]

I think the layout for the Firepower and the Lower Chassis mobility trees is quite reasonable, they each offer a strategy for specialization and progression that makes sense. However, the other trees really seem to force together too many choices in a bit of a haphazard manner. Do we really need/want to force points into Hill Climb and Speed Retention in order to unlock two of (what have been) the most basic mech skills -- Cool Run and Heat Containment? Similarly the number of arm speed and pitch unlocks required to secure torso yaw and speed seems a bit problematic. I am still on the fence about the Sensors tree. Specifically because Seismic and radar Dep were modules that an experienced pilot could install in a new mech with zero grind/cost, and now they are very deep in a complex tree. I suspect this is going to significantly increase the pain/frustration associated with the initial grind of a new mech.

[sitting target]

Rip brawling. I guess its time to quit the game. There are too many useless skills mixed in with the things I need to be a good brawler. They are basically forcing you to buy the modules nobody ever used in order to get the things you want. This causes you to miss out on many of the things you used to have.
I was hoping to be able to specialize my mech to whatever build I want to run but it seems like your only option is get zero modules and focus on weapons or get zero weapons and get one module you actually want.

EDIT -> PGI Please give linear skill trees!

I don't want to buy a bunch of target decay and retention skills in order to get seismic sensor thats worth anything. Give Atlas some love here!

Edited by sitting target, 09 February 2017 - 06:18 AM.

[riffraff777]

Correct me if I'm wrong, but GAMES are supposed to be FUN, right?
Why does this skill tree have to feel like doing TAXES ?
And why 91? Why not 100 ? 100 would feel a lot better to an obsessive/compulsive like me

Edited by riffraff302, 09 February 2017 - 07:08 AM.

[Tristan Winter]

Buehgler, on 09 February 2017 - 06:13 AM, said:

I think the layout for the Firepower and the Lower Chassis mobility trees is quite reasonable, they each offer a strategy for specialization and progression that makes sense. However, the other trees really seem to force together too many choices in a bit of a haphazard manner. Do we really need/want to force points into Hill Climb and Speed Retention in order to unlock two of (what have been) the most basic mech skills -- Cool Run and Heat Containment? Similarly the number of arm speed and pitch unlocks required to secure torso yaw and speed seems a bit problematic. I am still on the fence about the Sensors tree. Specifically because Seismic and radar Dep were modules that an experienced pilot could install in a new mech with zero grind/cost, and now they are very deep in a complex tree. I suspect this is going to significantly increase the pain/frustration associated with the initial grind of a new mech.

It's definitely better for some branches than for others. For me personally, it's definitely more fun when the progression is more linear, however. If I have to upgrade everything on the lower chassis branch to get the full speed tweak, then PGI has basically forced my hand and eliminated most of the choices for all my speed dependent mechs anyway (hypothetical example).

Rampage, on 09 February 2017 - 06:07 AM, said:

I would prefer a more linear tree and a lot less skill points to invest. I do not like the grind and I do not play games that require a lot of grinding to avoid being at a significant disadvantage over and above my own lack of skill.

Definitely a lot more grind. I was already ignoring most of the mechs in my mech bay (84 on my IS account and 60 on my Clan account) and now it looks like an even higher number of mechs will be left collecting dust, as I simply won't have the C-bills to compensate for the lack of quirks and make them competitive.

riffraff302, on 09 February 2017 - 07:07 AM, said:

Correct me if I'm wrong, but GAMES are supposed to be FUN, right?
Why does this skill tree have to feel like doing TAXES ?
And why 91? Why not 100 ? 100 would feel a lot better to an obsessive/compulsive like me

Both valid questions. They're definitely taking the grind too far for veteran gamers who have already sunk a lot of time into this game. As for 91, I have no idea. If you factorize 91 you get 13 * 7, both prime numbers. I wonder if there was any logic that got lost on the way, or if there's some logic we can't see at a glance.

[RuptorJoB]

Just finished to mess around in the mechlab...

So many nodes are unwanted and you are forced to "lose points" to get what you want.
For example, if I want to maximize my torso twist speed iI have to "buy" arm pitch and speed then torso pitch......to gain access to all the torso speed nodes. WTF

Icing on the cake, those unwanted arms modifiers render the mech movement horrendous (like you have mouse sensitivity to 10 instead of 1) if you don't use arm lock (who s using it anyway)!!!

Same on weapon , if I want only the cooldown nodes, i dont need to lose point on range nodes or velocity.

The idea is good and i managed to get an AS-7 to a max CT armor of 170 , but the general design looks like is was design by "a trainee during unpaid overtime"

LINEAR SKILL TREE PLEASE !!!!!!!!!!!!!!!!!!!!!

Edited by RuptorJoB, 09 February 2017 - 08:43 AM.

[PraetorGix]

Well, PGI did say they want us to customize our mechs according to our tastes in a manner never before seen, and tbh your proposed systems let us do just that much better than PGI's own.

[amenophis]

What I missed is a definition of whatever you mouse over. An example would be the skill of skeletal density. I think I know what is meant there but without a definition I cannot be sure.
Other than that the node is fine. I also wish there was a decent way to use all the exp ounce you get to the max. I know you can convert it but that costs money to do that. You need to buy MC.

Edited by amenophis, 10 February 2017 - 03:09 AM.