72 replies to this topic

[1453 R]

So. new Dev Vlog hit, here for those who are both curious and extraordinarily bad at finding information. In it we got some details on the Great Skill Rejiggering and the Event Queue. The latter of which sounds like a great deal of fun, but it's a subject for a different thread that already exists.

This one's for Skill Tree.

Short summation for the terminally attention-deprived:

-Skills are no longer arranged in a single interconnected web for each sub-tree, like Firepower, Mobility, and the like. Instead, each type of node - Heat Gen, Kinetic Burst, Armor Hardening, and the like - are in their own column and can be allocated freely without having to take any other nodes.
-This is specifically NOT the case for the Operations tree, which is weird and bad and interacts with consumables in an awkward, interdependent way, and also for certain high-value skills traditionally balanced by needing investment to get to. This list includes Radar Derp, Seismic Sensor, ECM Make-Gooder, and Speed Tweak. Each of these is at the end of a different skill, and so still requires investment to reach.
-The total number of skills remains the same, as does the 91-point limit. This change is not adding, removing, or changing any existing skill nodes, it's just changing how you are allowed to allocate skills.
-There will be NO refunding of skill points, XP, or the like, because the intent is that your 'Mechs all still work. There's edge cases where you may have bought-but-unallocated nodes, mostly in Firepower, but the system as a whole is designed to maximize compatibility with existing skill allocations.

The thread linked above has useful screenshots I will not transpose here. Check them out.

What do folks think the impact of this will be?

Frankly, I think this is gonna be pretty huge. The Firepower tree was a gigantic snarled pain in the hindquarters, and it was effectively impossible to truly maximize things like Range, Heat Gen, or Cooldown. There were 'practical' maximums because getting the last few nodes of each required a disproportionately punitive point investment, but that's not an issue now. if you want every Heat Gen node there is and literally nothing else in Firepower, you're allowed to do that. If you want the two ammo booster nodes for your specific munitions now and nothing else, you're allowed to do that.

The same is true, if to a lesser extent, of Survival and Mobility as well, with players able to maximize Armor Hardening without having to take one single waste-of-life Reinforced Casing or Shock Absorber node, and players able to pick and choose the specific movement patches they need from Mobility. You can maximize Anchor Turn and Torso Speed on assault 'Mechs for only eleven points, increasing your ability to track targets on the horizontal axis by an additive sixty percent. "Shoot The Damned Fish" has never been easier.

The nigh-complete elimination of 'dead' nodes you take because you have to opens up a lot more options for taking things people normally discount. The Jump jets tree is gonna be a lot more alluring now because you'll have a lot more points to spend after getting your essentials done in stuff like Firepower and Survival, and people can take some of the good stuff in Sensors much more easily. Yes, Seismic and RaDerp are barred behind other stuff, but you can get 100m of Seismic with exactly four points - three TIG nodes you'll like anyways and the one for Seismic, and maximizing RaDerp requires ten points rather than "The Entire Sensors Tree". A lot of the stuff people found nice to have but not essential is spun off into penny-packet groupings you can take with all the points you save by not having to put points into dumb stuff like Shock Absorption or Quick Ignition, or having to put 45+ points into Firepower just to get to the ~25 points of things you actually want.

It's gonna be a big overall power spike, but I think it's the sort of power spike that'll be welcomed. You can actually use the Skill Tree to try and give your 'Mech a lean towards one specialization or another, rather than there being only a few ways to make good use of all your points in the Skill Minor Erdtree. There's a lot more room to take the things you want to take, instead of feeling constrained to take what's 'Good' because it's naturally point-efficient to get to. Gonna be more variety in skill loadouts, which can only be a good thing.

...and Operations still sucks donkey rocks because combat consumables are a pox on MWO that should never have been introduced in anything remotely like their current form, but you can't win everything with one rejiggering.

Anyways. What do you all think? Fantastic flavorful rejigger, or just a bunch of blatant pandering power creep?

[Meep Meep]

Just browsing the images I can tell there are already going to be some very powerful builds on certain mechs if the skill point allotment is still 91. This is fine though as everyone will have access to those builds and any op builds can be tweaked. Gonna be a whole new game on some levels if this hits more or less how its being presented. I'm looking forward to this change the most of all.

[FupDup]

The cost reduction is undebatably good.

The lack of refund is understandable because of how convoluted the HSP system already is, and it's not like you can walk into Walmart and demand a refund for the thing you bought 2 years ago just because it went on sale yesterday.

The ability to directly pick nodes is definitely a lot less annoying than having to take filler nodes, but it will really put a spotlight on just how incredibly bad those filler nodes were (Cauldron buff plz).

It will definitely be a bit powercreepy overall but I don't think it'll be too bad. It's certainly not going to creep nearly as much power as some of the other Cauldron changes have...

One possible upside is that it might be a lot easier to add more nodes in the future now that you don't have to rearrange an existing tree to fit them in (because they can be in their own area).

[Meep Meep]

FupDup, on 07 April 2022 - 04:04 PM, said:

It will definitely be a bit powercreepy overall but I don't think it'll be too bad. It's certainly not going to creep nearly as much power as some of the other Cauldron changes have...

One possible upside is that it might be a lot easier to add more nodes in the future now that you don't have to rearrange an existing tree to fit them in (because they can be in their own area).

The powercreep is going to be real. Been doing some builds using the images and now that you can pick and choose your skills you can get the same core build as before but no filler and saving about 10~15 points on average to put on other relevant stuff. The first few days of this are going to be a mad laboratory of min/max fits.

I agree about the upside. This allows them to add in more skills and maybe 'elite' skills that only open up after a certain point investment and cost extra points per activation.

[SFC174]

Sounds like they've still missed an opportunity. I think skill decisions should have a heavy dose of this OR that, with less of this AND that.

IOW, make people make actual choices. Choose weapon range? Get locked out of a significant number of heat generation nodes. Choose jump jet duration, get less peak thrust. Real tradeoffs. This wouldn't work for everything (what do you lose access to if you take armor hardening for example?)

While it would up the importance of mech smithing in the lab (and that's not good for beginners), I think it would increase build diversity and also add more rock-paper-scissors interplay on the battlefield. The more you specialize the more you also create weaknesses that others can exploit.

YMMV, but I've been asking for this for years.

[YueFei]

SFC174, on 07 April 2022 - 04:33 PM, said:

Sounds like they've still missed an opportunity. I think skill decisions should have a heavy dose of this OR that, with less of this AND that.

IOW, make people make actual choices. Choose weapon range? Get locked out of a significant number of heat generation nodes. Choose jump jet duration, get less peak thrust. Real tradeoffs. This wouldn't work for everything (what do you lose access to if you take armor hardening for example?)

While it would up the importance of mech smithing in the lab (and that's not good for beginners), I think it would increase build diversity and also add more rock-paper-scissors interplay on the battlefield. The more you specialize the more you also create weaknesses that others can exploit.

YMMV, but I've been asking for this for years.

I don't think it's necessary to include built-in drawbacks for the nodes. The trees just need to be arranged such that there are sufficient opportunity costs, where some of the juicier nodes are gated by a requirement to invest sufficient points into prior nodes, and doing so prevents you from also grabbing other very juicy nodes in other trees, due to limited number of total points allowed.

Like, that's generally how WoW talent trees work. Each point you pick up just plain makes your character better. They did not make it so that picking up 5 points in +Parry% would also reduce your chance to dodge by 5%. It's just that as you reached for the juicier talents deeper in one tree, you'd run out of points and be unable to invest deep into another tree to pick up its best talent(s).

If it were up to me, I'd like to see MWO's skill trees revamped along the lines of various roles / archetypes, instead of organized around attributes like firepower/survival/agility.

Edited by YueFei, 07 April 2022 - 04:54 PM.

[SFC174]

YueFei, on 07 April 2022 - 04:53 PM, said:

I don't think it's necessary to include built-in drawbacks for the nodes. The trees just need to be arranged such that there are sufficient opportunity costs, where some of the juicier nodes are gated by a requirement to invest sufficient points into prior nodes, and doing so prevents you from also grabbing other very juicy nodes in other trees, due to limited number of total points allowed.

Like, that's generally how WoW talent trees work. Each point you pick up just plain makes your character better. They did not make it so that picking up 5 points in +Parry% would also reduce your chance to dodge by 5%. It's just that as you reached for the juicier talents deeper in one tree, you'd run out of points and be unable to invest deep into another tree to pick up its best talent(s).

If it were up to me, I'd like to see MWO's skill trees revamped along the lines of various roles / archetypes, instead of organized around attributes like firepower/survival/agility.

Perhaps I wasn't clear. I'm not saying make it so that taking a range node increases heat (although that would seem logical if game physics had any consistency . Rather I'm saying that after the first few nodes you have to choose a branch to go down. Heat or range. Choose one branch and the other one is locked out. You don't generate more heat but you can't spend to reduce heat.

Minor distinction, but an important one. And if memory serves there are a lot of games that do this too. Perhaps its in line with the archetype branching you suggest.

[Ballistic Panicmode]

The only thing that I find out of place is the single point investment for advanced zoom, considering how powerful ranged sniping in the current game. Consider moving Radar Deprivation 1 up to branch off of Sensor Range 3, making the first 20% available sooner. Then bring Advanced Zoom off of Sensor Range 5, which is where it logically belongs imo.

[YueFei]

SFC174, on 07 April 2022 - 05:20 PM, said:

Perhaps I wasn't clear. I'm not saying make it so that taking a range node increases heat (although that would seem logical if game physics had any consistency . Rather I'm saying that after the first few nodes you have to choose a branch to go down. Heat or range. Choose one branch and the other one is locked out. You don't generate more heat but you can't spend to reduce heat.

Minor distinction, but an important one. And if memory serves there are a lot of games that do this too. Perhaps its in line with the archetype branching you suggest.

Oh OK, yeah we're on the same page.

To expand on what I was saying about roles / archetypes:
I think it'd be cool if we had more types of nodes that led to abilities you had to actively use, particularly at the end of various Skill Trees (so it requires a heavy investment). So not just more passive bonuses. For example, a "Combat Support" type of Tree might have Info War type stuff, like:

• The ability to deploy smoke bombs (which also block thermals). Size/duration, cooldown/number, these can be tuned for balance.
  
• The ability to utilize a magnetic anomaly scan mode, but this requires the user to actively scan. A wider scan has a shorter range, a narrow scan has a longer range. Again, duration/cooldown/recharge (or even heat gen) can be tuned for balance. This is essentially a very transient wall-hack that requires skill on the user to make the most of it. I'd also make it asymmetric for different mech tonnages, so that a larger mech is more easily detected by this than a smaller mech.
  
• A short-duration directional jamming ability that disrupts the enemy's C3 network (so affected enemy mechs lose target info sharing). Again, tune it via any number of parameters (such as the width of the jamming cone, the heat/sec it generates).
  
• etc...

And then make it so that investing deeply enough into that tree would make it so you could not pick up some really cool abilities from a different Tree.

Problem is that given the current programming resources, a major overhaul and introduction of new features and mechanics like the above are probably not feasible.

Edited by YueFei, 07 April 2022 - 05:58 PM.

[pattonesque]

YueFei, on 07 April 2022 - 05:57 PM, said:

Oh OK, yeah we're on the same page.

To expand on what I was saying about roles / archetypes:
I think it'd be cool if we had more types of nodes that led to abilities you had to actively use, particularly at the end of various Skill Trees (so it requires a heavy investment). So not just more passive bonuses. For example, a "Combat Support" type of Tree might have Info War type stuff, like:

• The ability to deploy smoke bombs (which also block thermals). Size/duration, cooldown/number, these can be tuned for balance.
  
• The ability to utilize a magnetic anomaly scan mode, but this requires the user to actively scan. A wider scan has a shorter range, a narrow scan has a longer range. Again, duration/cooldown/recharge (or even heat gen) can be tuned for balance. This is essentially a very transient wall-hack that requires skill on the user to make the most of it. I'd also make it asymmetric for different mech tonnages, so that a larger mech is more easily detected by this than a smaller mech.
  
• A short-duration directional jamming ability that disrupts the enemy's C3 network (so affected enemy mechs lose target info sharing). Again, tune it via any number of parameters (such as the width of the jamming cone, the heat/sec it generates).
  
• etc...

And then make it so that investing deeply enough into that tree would make it so you could not pick up some really cool abilities from a different Tree.


Problem is that given the current programming resources, a major overhaul and introduction of new features and mechanics like the above are probably not feasible.

I would love things along this line for mechs like the -3L, or equipment like BAP. It'd be cool to introduce some kind of skill-based element to every piece of equipment

[Extra Guac]

SFC174, on 07 April 2022 - 04:33 PM, said:

Sounds like they've still missed an opportunity. I think skill decisions should have a heavy dose of this OR that, with less of this AND that.

IOW, make people make actual choices. Choose weapon range? Get locked out of a significant number of heat generation nodes. Choose jump jet duration, get less peak thrust. Real tradeoffs. This wouldn't work for everything (what do you lose access to if you take armor hardening for example?)

While it would up the importance of mech smithing in the lab (and that's not good for beginners), I think it would increase build diversity and also add more rock-paper-scissors interplay on the battlefield. The more you specialize the more you also create weaknesses that others can exploit.

YMMV, but I've been asking for this for years.

I see what you're saying, but I support the new skill tree changes. You say that you want specialization, but I think the new system actually increases the level of specialization in the game - it's easier to select ballistic-focused skills, laser-specific skills, etc.

There's no point in making people go through a bunch of unnecessary nodes to get the stuff they really want - that's just a waste of time & a flawed system.

If the power creep is too high under the new system, then they can reduce the maximum # of skills from 91 to 75, or whatever it takes to achieve the metrics that they are looking for.

It's clearly a step in the right direction because it streamlines the game & makes it more intuitive, with less time spent on boring/mundane activities.

[An6ryMan69]

Watched the dev log.

I have to say, from what I can see there, and in this thread, everything I see seems good to me.

The Event Queue seems like a great way both to run events but also to maybe give players who dislike standard Quickplay another way to play, perhaps?

As someone who is often wary of changes to something I already like, I'm looking forward to this all coming on line!

[Dollar Bill]

I like the new Skill Tree. But, you know, that's funny. Back when the current 'Skill Maze' **** got pushed out, someone came up with a better Skill Tree, kind of like this one. But, PGI/Russ didn't give it a chance and just said it wouldn't work. When those in the community complained, they were arrogantly brushed off by PGI. So thousands left. Now after the mass exodus of players from MWO, this type of Skill Tree is suddenly a good idea, PGI?

Well, lets see what really happens. So far so good. However, PGI has a history of failed implementation, disappointment, and one step forward then two steps back.

[MrMadguy]

Yeah, I said, that for now SPs are the biggest bottleneck just yesterday and devs' reaction was pretty quick. I guess, new skill trees will be better, because I won't need to spend some "junk" points to get to needed ones and will be able to spend them elsewhere.

But still. It's not game's major problem now. Low population, causing extremely terrible matchmaking - is.

[Dauntless Blint]

Looks good to me. Might get me 1/4 of the way to resurrecting an old build or 2.



Now I want to see installable armor types, so there is a multiplicative defence stat in the game to go with "weapon doors closed" damage mitigation.

+50% Defence Lasers
or
+50% Defence Ballistics
or
+50% Defence Missiles

Could have half value Armor options for half the critical slot investment.

[1453 R]

MrMadguy, on 07 April 2022 - 08:58 PM, said:

Yeah, I said, that for now SPs are the biggest bottleneck just yesterday and devs' reaction was pretty quick. I guess, new skill trees will be better, because I won't need to spend some "junk" points to get to needed ones and will be able to spend them elsewhere.

But still. It's not game's major problem now. Low population, causing extremely terrible matchmaking - is.

Spoilers: you fix low population by making improvements to the game that make old players want to come back and try out the changes. Things like making the skill tree less of an obnoxious performance tax and allowing people to take things they want to take whilst changing the tree in a way that allows them to better expand and alter it in the future.

[TheCaptainJZ]

I don't see how they would do this, but they should decrease the number of nodes down from 91 to master a mech since you can min/max a lot better. The reason the old system was so convoluted was to directly nerf min/maxing. Now that you can min/max, it would make more sense to reduce how many skill points you actually get in totality since you don't have to blow any on "filler." That would require some kind of refund to be fair, though. Short of that, they might reduce the bonuses of some of the nodes, or, as has always been the case, certain nodes are expected to be allocated on all mechs and other balance changes are made universally to reign in the buffs by the change in skill tree.

[pbiggz]

TheCaptainJZ, on 08 April 2022 - 06:42 AM, said:

I don't see how they would do this, but they should decrease the number of nodes down from 91 to master a mech since you can min/max a lot better. The reason the old system was so convoluted was to directly nerf min/maxing. Now that you can min/max, it would make more sense to reduce how many skill points you actually get in totality since you don't have to blow any on "filler." That would require some kind of refund to be fair, though. Short of that, they might reduce the bonuses of some of the nodes, or, as has always been the case, certain nodes are expected to be allocated on all mechs and other balance changes are made universally to reign in the buffs by the change in skill tree.

They did it so that they dont have to refund any mechs. When the new system drops your existing skills will be applied 1-1 to the new trees. Every node that exists now will exist then, they just moved them.

[TheCaptainJZ]

pbiggz, on 08 April 2022 - 06:46 AM, said:

They did it so that they dont have to refund any mechs. When the new system drops your existing skills will be applied 1-1 to the new trees. Every node that exists now will exist then, they just moved them.

I know that, but once you change them in the new system, you will be able to really min/max like never before. About half the node people would never buy if they didn't have to so now you'll see a lot of nodes everyone picks because you have to, and the rest are ignored. To make the new tree more of a give or take, you'd have to reduce the number of nodes you get to use, but that's out of scope for this project because it'd require giving something back to make that fair for spent SP, cbills, xp.

[ScrapIron Prime]

I'm someone who always takes Seismic Sensors and 60-100 radar Deprivation, on the theory that this saves more armor for later so it is effectively akin to the survival tree. The new skill system keeps these skills expensive, so there must be something to that line of thinking (at least in the minds of the Devs).

So how does it compare? Current system, to get full seismic and radar derp costs is 20 points. In the new system, it is 17 points, as you have to take the entire Target Info tree and nearly all of the Sensor Range tree. If I want only 60% radar derp plus full seismic, it is 15 points in both systems.

So the new system saves me a few points if I want the full package, and is the same if I want 60% derp. Might as well spend the other 2 points for 100% radar derp, because the stuff is magic. The five point tax in front of radar deprivation is a lot less than it used to be, so LRM boats will have an even harder time under this system.

Of course, if you're not into seismic sensors (why WOULDN'T you want to see the piranha behind you before it fires???) you can save yourself 4-7 points, instead of the original system which saves only 1-2. So there IS a real savings there.