358 replies to this topic

[-Skyrider-]

Something I noticed when unlocking skills down a tree, it is now harder to tell which ones you have unlocked and which ones are available to be unlocked based on this new color system. I think it work be more intuitive to have a different color for nodes you can unlock, and a node which you have already unlocked.

[Void Angel]

Hrm. Initial response is generally favorable; still trying to get in a decent amount of testing.

Anyhow, the system is a definite improvement. The impact for new players is less (though I still strongly recommend eliminating the c-bill costs for skills on that front; I think new players will feel caught on the horns of a dilemma as they try to decide whether to save up for a new 'mech or actually make the 'mech they have worthwhile.

The trees themselves are an improvement, though the problem of junk skills still persists. Whether it's a King Crab or a Locust, some 'mechs can't even use Arm Mobility - much less actually want it. It's annoying to need four Quick Ignition nodes just to get the heat skills I really want. Nor does reduced falling damage feel like anything but a wasted point. I realize that some of this sort of thing may be inevitable because not all quirks are created equal, but the tree could probably still do without its most egregious offenders.

On the subject of mobility, I find that 'mech handling is a bit too cumbersome - especially if you don't buy any of the mobility tree. My Spider is only slightly less agile in most ways than it used to be, and the jump jets feel a bit more responsive. And they should - I took every point but the center row of heat reduction skills in that category, with similar investments in Mobility. On the other team, some poor Battlemaster Large Laser boat didn't seem to have any Mobility quirks; he couldn't even bring his guns to bear on other Assaults, much less smaller targets. My own King Crab, with 21 points in Mobility, was so unresponsive it was difficult to hit targets with literally every applicable point on the Autocannon side of the Firepower tree. And I ran into heat difficulties as well.

Some quirks are still practically required in order to feel competitive, particularly the entire mobility tree (except Arm Pitch,) or the way the skill tree expects us to pick up heat reduction quirks to compensate for the overall nerf to the cooling skill. Having played matches in both Light and Assault 'mechs (and performing dramatically better in a Light,) my impression is that Mobility and Operations are far too useful to allow players to forgo a major investment in either tree. This makes the other tree options decidedly weak, and causes players to feel that their skill choices are artificially constrained.

Finally, while the skills are (for the most part) pretty clear as to what thing each one does, the value of said thing won't be terribly clear to new players, particularly given the learning curve of MWO - and the massive number of skill points ultimately available is likely to greatly exacerbate the issue. Players will always have to learn to focus what they want to do with their builds (you saw that mistake a lot in Diablo 2&3,) but we should if at all possible make their choices transparent.

So, my recommendations are this:

• Restructure the skill tree and give players more meaningful choices. For example, use a sort of radial snowflake instead of the Plinko cascade pattern we have now. Each category has its own ring, and players will spend skills separately between them just as they do now. Don't link the skills together in chains, but simply require a certain number of points be spent in the inner ring before moving to outer rings. Alternatively, you could implement a Green/Veteran/Elite tier system with different starting points available at each tier, based on a theme: Once enough points are unlocked in Green, the player can mix and match from separate Ballistic, Missile, and Laser themed sub-trees (much like the corresponding skill tree areas now.) This will be much more intuitive for the player, and will reinforce the idea of spending skill points with an ultimate goal in mind. Combine less useful skills together into better options and/or simply require some of them to be bought in order to unlock the higher/outer tiers.
  
• Reduce the number of skill points (a lot) and make the skills more powerful - and expensive, if you must ;] - to compensate. This will reduce the complexity of 'mech builds to a level more manageable to new players, make planning builds much easier, and create less variance in 'mech builds - leading to clearer telemetry and feedback on future balance.
  
• Consider putting the critical must-haves, superior options, and clear favorites like Radar Deprivation, Improved ECM, and Seismic Sensor into their own category. These should be selected unlocks, rather than true skill points, from which the player selects as their total skill points spent increases. Players should not be able to get all (or possibly even most) of these options, and so will have to select a focus for what they want to specialize in for that 'mech - possibly even have them pick only or twice, with additional capabilities unlocking as they skill up the 'mech. Options should include Spotter, Scout, Brawler, Sniper, etc - rhymes with "mole red hair."
  
• Give some love - not too much - to the torso mobility of very large Battlemechs. Currently, they just feel too cumbersome and slow. While it's good for huge, lumbering war machines to be huge and lumbering, they need to be able to torso twist to track targets with rapid-fire weapons. Right now, that's not happening for my King Crab and others (if the plight of enemy Assaults is any indication,) so we're going to have to transition to high-alpha, slow-firing weapons - as if that wasn't meta enough already. Increasing base torso twist rates should give big 'mechs enough mobility to function without being able to turn on a dime and invalidate the existence of Lights on the Battlefield.

That's all I can think of for now; overall, the improved skill tree is an improvement, and I look forward to seeing its final incarnation on the live servers. =)

Edited by Void Angel, 02 March 2017 - 03:02 PM.

[Void Angel]

OH! And add a solid, colored border around skill hexes that you've activated, so you can better see at a glance which skills you've picked.

[-Pik-]

Maybe highlight similar skills when you hover over a node? Search function?

[Joe Decker]

DAYLEET, on 02 March 2017 - 02:02 PM, said:

Yeah its kind of a bad idea. But it wouldnt be the worse thing we've seen in mwo. The only problem i see is that its not free but that can be fixed easily. I dont think when PGI decide to release it it will be the end of the testing. With everyone playing it there will be more changes to be made im sure.



Imagine if the attribute nodes required to get to your skills were stuff you dont need and dont want and it wouldnt change the game in any ways. Suddenly that tree would seem like its crap and be hard to navigate, even if the gameplay didnt change. In poe, you spec, in mwo, you take a bunch of stuff.

You did not read the Rest of my Post... So you misunderstood what I wrote.

[- Pestilence -]

-Skyrider-, on 02 March 2017 - 02:53 PM, said:

Something I noticed when unlocking skills down a tree, it is now harder to tell which ones you have unlocked and which ones are available to be unlocked based on this new color system. I think it work be more intuitive to have a different color for nodes you can unlock, and a node which you have already unlocked.

Specially for color blind people.

[catspider]

Mechs that cannot possibly equip a certain type of weapon - ie Black night cannot have balistics - should not have to unlock balistic nodes to get past them

[Ricbro85]

PGI

Can you make the skill trees so you can just take single skills insted of having to go down and unlock skills you dont want to get to the skill you want.

then you are able to pick a skill from the trees you want insted of having to waste arm pitch to get to speed if your in a mech with no arms.

And is it best to have a player level or a XP pool for all mechs as it bit of pain if have so many then having to look at each mech

XP pool insted of XP on a each mech.

[shameless]

you learned pretty much nothing from the first test. Still taking the whole survival tree, and the weapon bonuses are worthless. it takes most of the mobility tree to get full speed tweak, and all of the sensor to get radar deprivation. dozens of worthless nodes we are forced to take The extra consumables is interesting, but really, we're gonna go back to everyone having 2 coolshots, arty and air....

[shameless]

You nerfed the Victors and the phoenix hawk.. why did you do this?

[Felicitatem Parco]

I kind of understand the branching trees and the "have to get this skill before you get that skill" theme.

As you gain "experience" you generally get better at piloting. You don't get better at just one particular aspect of piloting.

[Boss Hawg]

catspider, on 02 March 2017 - 03:19 PM, said:

Mechs that cannot possibly equip a certain type of weapon - ie Black night cannot have balistics - should not have to unlock balistic nodes to get past them

The new resistance heroes have a mix up of hardpoint types. There is a Black Knight hero that has ballistics now. I'm sure PGI introduced these new mech packs with the new skill tree in mind.

[Felicitatem Parco]

Tater Chips, on 02 March 2017 - 03:55 PM, said:

The new resistance heroes have a mix up of hardpoint types. There is a Black Knight hero that has ballistics now. I'm sure PGI introduced these new mech packs with the new skill tree in mind.

The black Knight was given an AC/20 instead of a hatchet.

[Bud Crue]

Any word on a hot fix for the PTS and the needed fixing of the broken movement profiles?

[DAYLEET]

Joe Decker, on 02 March 2017 - 03:13 PM, said:

You did not read the Rest of my Post... So you misunderstood what I wrote.

I was just transposing what we have now into the poe tree. I dont see what the rest of you post has anything to do with it, i wast criticising it or anything,

[KrocodockleTheBooBoxLoader-GetIn]

How much of this skill tree is simply a result of pgi focusing on mw5?

[Zephrym]

Void Angel, on 02 March 2017 - 02:59 PM, said:

Hrm. Initial response is generally favorable; still trying to get in a decent amount of testing.

Anyhow, the system is a definite improvement. .....

Couldn't have put it better (no really...). THIS!

[Bulletsponge0]

step in the right direction, but still needs work

but the new player experience was brutal before, and the skill tree, whenever it goes live in whatever form it takes, will make it exponentially worse

[Nutta88]

I don't mind the cost of the skill "tree" but don't like the maze like nature of it. Just fitting the current set of skills and modules into a skill maze feels broken and an unnecessary change.

If you want the skill matix to have relationships then the “skills” should effect each other. Link the effect of the skills within the current branches and between them.

e.g. have speed tweak increase max speed but also decrease acceleration and have acceleration decrease max speed.

Though all of this needs for the current balance at least to be at least maintained. The lights on the test server feel like tanks. I am guessing that this is either a deliberate “test” or an oversite.

On live a mist lynx has +60 accel +60 decl +45 turn

On test this becomes +25 +25 and +20… after you have all the skills.

No one would ever pick a light that is performing like a heavy. It is bad enough testing light performance in 4v4 but this makes them silly.

The skill maze feels overly complicated, not adding to the game. The skills could be linear choices but with drawbacks. The drawback could start low and move high as you progress down a path.

Increase in Sensor Range - Increases the time to gather target info
Target Gathering info - decreases sensor range.
Target Decay Duration - decreases sensor range.
Radar Deprivation - Reduces Seismic sensor range
Advanced Zoom - ? Maybe increased screen shake.
Seismic sensor range - Reduces Radar deprivation
360 Target retention range - decreases sensor range.
ECM target range reduction - decreases sensor range.

So getting advantage creates a disadvantage, a trade off...

Reduce laser Duration - Increases Cool down
Reduce cool down - Increases Duration
Increase weapon range - Increases Heat gen
Reduce weapon heat gen - Reduces range

uac jam chance reduction - increases cool down
mag cap - ammo explosion chance or reduced damage per shot?

Normally there is a downside to improving something in a certain aspect, the more you improve a particular metric the more you decrease another. A lighter engine may be more fragile etc. It would be nice to make it so a non-levelled machine is not completely out of the game, just different.

I don't mind change, I just would like it to enhance a game that I am enjoying.

Edited by Nutta88, 02 March 2017 - 04:57 PM.

[Draglock]

Prosperity Park, on 02 March 2017 - 03:59 PM, said:

The black Knight was given an AC/20 instead of a hatchet.

Is it right arm getting a shield like the centurion, it had a massive shield in lore.

Edited by Draglock, 02 March 2017 - 05:03 PM.