Lt. Commander
Join Date: Dec 2007
Posts: 120
As I have a better grasp on the skill system, I'm seeing various issues that jump out at me.
  1. Skill point cost
  2. Front loaded skills
  3. Over-concentration
  4. Relative value of skills

Skill point cost

The overall cost of skills at tier 4 and 5 are exceptionally high. Relative to Holodeck skills, you simply can't afford a proportional amount of skills. I detailed this in another thread, the ratio of skill costs to the available number of skills is disproportionately high in the new system.

On Holodeck, Tier 5 was where you could buy the T5 skills you're after and then the rest would go fill out any spots you wanted to fill in the previous tiers. On Tribble? Good luck with that. You can, at best, max out 3 T5 skills and that's it.


Front loaded skills

On first glance, this would seem to counterbalance the problem to the cost issue. However, the skill costs are simply too expensive to afford specializing heavily for 9 ranks in a skill. This means that these two issues together far too strongly incentivize a smattering of skill points in a lot of places. Consider that 2 ranks in any skill gives you 48% the effectiveness of the fully ranked skill. 4 ranks gives you 72% of the fully trained skill.

This is extremely silly and punishes anyone going for either maximum effectiveness, or for training the tier 3 BOFF power, far too much because of the much higher skill point costs. And by extension it rewards anyone splashing far too much with a good, broad range of effectiveness across the board that the specialist can't match. To put this in explicit detail, I had 8 ranks in Weapon Performance, that gave me 9 points of bonus weapon power. Moving down to 4 ranks, I only lost 2 points of power, but saved 3000 skill points.


Over-concentration

This issue is chiefly a problem in the science skill tree and not so much anywhere else. The problem is that so many BOFF abilities are dependent on the same skill that it over focuses spending towards those skills in lieu of any others. Flow Capacitors, Shield Emitters, Hull Repair, and Particle Generators are the biggest offenders. Graviton Generators is similar, but it offers much lower value effects.

However the over-concentration also makes these skills very easy to skip for the same reason. An escort captain gets no value from Particle Generators, and thus the skill may as well not exist for them. The opposite issue exists for Attack pattens. It has little value for science ships, thus they have no reason to skill it.

The net result of over-concentration is going to make too many cookie-cutter builds, where the customization of one's skills is essentially rhetorical, and the cost of higher tier skills are impossible to justify. It will also have the secondary effect of doing exactly what the skill tree was supposed to "fix" which was forcing a respec if you want to change ship classes. With skills having zero value to certain ship types, they are essentially wasted points if you then want to switch ships.


Relative value of skills

This heavily ties into the above issues and the skill point cost issue. Not only are some of the skills over-concentrated, but they are extremely cheap, given the concentration of abilities. It is a cost-benefit issue that makes it far more worthwhile to skill up these skills than the others. You get too much for too little.

A quick glance at the extremes ought to make this issue quite obvious. Flow Capacitors vs. Subspace Decompilers or Countermeasure Systems. Flow Capacitors affects all subsystem targeting, energy siphon, tykens rift, and tachyon beam, among other miscellaneous abilities. Subspace Decompilers chiefly affects Viral Matrix, and the not particularly important secondary effects for a few other powers. Countermeasure Systems affects Jam Sensors and Scramble Sensors, and while the value of Scramble is still high for PVP, the cost-benefit is now skewed heavily against it being worthwhile. Jam sensors and Viral Matrix are still very underwhelming, and certainly do not deserve the exorbitant costs associated with their skills.

The engineering performance skills suffer from the same questionable pricing. Why are aux and weapon performance more expensive, a tier higher, and thus far less accessible than engine and shield performance?

Counter-skills are also questionable in their placement. Why do they appear lower in the tree than the skill they resist, in some cases far lower, like subsystem repair vs subspace decompilers?

These issues combine to make some skills not worth taking for the far easier to gain resistance to their effects, or for the simple issue that they cost too much for their effect, or that other skills are far too valuable to pass up.



So how to fix these issues?

Skill point cost


This is one issue with a lot of potential solutions. One simple suggestion is to move skills down a tier, or fix their pricing at lower costs. Tier 5 is too expensive by far and I would simply remove it entirely to tier 4. A few other skills should be moved down as well. Warp core potential to tier 1 is not a bad idea since it used to be there before and doesn't do much.

Another solution is to have escalating costs. The more ranks you have in a skill, the more the next rank costs. This however, would certainly have to be coupled by removing the frontloading issues. A variant might be to adjust costs dynamically based on how many total skills and ranks you have, though this might be difficult to balance.


Front loaded skills

This issue is relatively easy to fix I think. Depending on how you want to balance it, you could flatten the curve completely so that each rank gives the same bonus of about 3.66. I think it would be fine, however, if you just shaved some off the front and put it onto the end of the skill. For example, instead of 10-6-4-4-2-2-2-2-1, make it 6-5-4-3-3-3-3-3-3. This would flatten the curve and give more incentive to deeper investments in the skill, while still allowing some amount of front-loading. Also perhaps 6-4-3-2-2-2-3-3-4-4 would be interesting, in making it a little more expensive to front-load, but provides better value to get the full nine ranks.


Over-concentration and Relative value of skills

These issues really need to be solved together, and it requires some wide reorganization of the skill tree. Partly it also depends on how the skill costs are dealt with.

The most important facet however, is that the counter needs to be rolled into the main effect skill. That is, improving gravity well's repel, and resisting gravity well's repel should come from the same skill. Why is this? It goes hand in hand with the whole purpose of the skill revamp, to simplify and remove the reason to respec for changing ships. It also makes very logical sense, understanding how sensors work should help you evade and defeat them better when cloaking, for example.

Of course just combining the counter into the relevant skill would cause a problem by shrinking the skill tree even further and making certain skills even more important than they are now. Because of that, there needs to be splits from many of the over-concentrated skills.

Tachyon beam should be pulled out of Flow Capacitors, and could probably combine well with Feedback pulse in a new skill. Tractor beam can be taken out of Graviton Generators, and is probably annoying enough to have its own skill all to itself along with any other hold/slow type powers, while Graviton Generators focuses only on repels. Sensors should have the resistance to scramble/jam removed from it and it should be sensors/cloaking. Attack patterns needs to be rolled into some other skill, like targeting systems, or have a secondary passive effect added to it.

Training for skills also needs to be locked to the primary effect, not some other random skill. CPB cannot be trained by Subspace Decompilers, that is just silly. It has to be part of Particle Generators, or whatever skill increases the damage for the shield damaging effect. Jam Sensors has no business being part of sensors if it isn't improved by it, and so it belongs in Countermeasure systems at least.



Reworking the system is necessary, no matter what is actually done.
Lt. Commander
Join Date: Dec 2007
Posts: 120
# 2
11-13-2011, 01:43 PM
Thanks Foxrocks. I wanted to write something on the skill system, but I didn't find the motivation to put all my thoughts together. I think your observations are spot on (at least for the most part).

I think it is not necessarily problematic that the first 3 ranks basically buy you 50 % of a skill. This only is a real problem when you want to be able to train a lot of skills. But the plus side is that you can actually get a good return on your investment with less skill points.
But maybe it is a problem, because the cost of raising skills in the first place is off, and the distrubtion across the tiers as well. The cost for Tier 4 and Tier 5 skills is too high. I noticed that I didn't feel "finished" with my Tier 3 and 4 skill selection when I got to to the next tier of skills. And I didn't even try to cover ground skills, and I did optimize as good as I could taking the 50 % bonus for 3 ranks rule in mind.

Unfortunately I do not know if this is in the real of possibilities, but if it is - I would try to change the tier-related skill caps a little and then adjust the skill point cost.

Tier 1 Skills should cost 100 skill points as before.
Tier 2 Skills cost 300, Tier 3 skills 400, Tier 4 skills 500 and Tier 5 Skills 600.
Adjust the level cap for Tier 1 somewhat (by about 1/3), and people can max just as much skills as they can now with the system before they get access to newer skills.
Overall results:
  • People can spec more into higher tier skills.
  • Tier 1 - the most boring part of the game, basically - is over sooner and people are ready for their new ship sooner (including the possiblity to buy a Tier 2 ship).
To compensate a little, a flatter skill curve would now probably be advised. It's really hard to device a diminishing curve with integer numbers and 3 different improvement steps without ending at something where the first 3 skill points buy you ~50 % of the bonus.
Lt. Commander
Join Date: Dec 2007
Posts: 120
# 3
11-13-2011, 04:16 PM
Part of the problem I have with the front-loading of skill benefits, while its very nice to get a lot of benefit for less than half the total cost of the skill, there is very little benefit to maxing it out or continuing to invest beyond the cheap 2-4 first ranks.

Again, 4 ranks = 72% improvement in that skill. Double your investment and you only get 24% more improvement. I think the diminished returns penalize specialization too much with the new costs. The tradeoff of course is 96% performance with one skill or 72% with two.

Specialization doesn't give enough performance improvement in most cases to warrant the full investment, even for training, and in some cases, like Charged Particle Burst, it is clearly better to respec out of the skill after training. And of course the reason for this is that is not just the marginal improvement to a full investment, but also the fact that there are many other good places to spend skill points.

And many people are finding this to be true. Much of the common advice in the new system that I have read is just to put a few ranks in, and you can do pretty well at everything. I think that means it fails overall, because this ends up making the last 4-5 ranks rhetorical. Yes they exist, but why take them when it is so expensive and they give so little?
Lt. Commander
Join Date: Dec 2007
Posts: 120
# 4
11-13-2011, 04:26 PM
I think the biggest problem is the seemingly arbitrary placement of skills, some of which make sense, some that do not. Although, without a doubt the cost of the higher tier skills needs to go down to some degree, I also being the skill tree should follow this style:

(T1/T2) Foundation Skills: Mostly passive broad bonuses.
(T2 to T4) Focused Skills: Specialize you into a particular area.
(T4/T5) Universally Powerful but Focused Skills: Very powerful/useful no matter the ship.

At the moment many T5 skills are very powerful, but only with a specific ship type or setup. Just as an example, a universal skill that buffed ALL weapon's critical chance and damage bonus would be a good T5 skill, but not one that only worked with specific weapons, those should be lower in the tree. A skill that boosts all systems energy levels should have its effect increased and moved to T5, while the specific system power boosts should be slightly weakened and moved down in the tree. Also, not a single boff skill should be effected primarily by a T5 skill.

Only by doing that can one truly spec for a 'style' of play instead of a ship.
Lt. Commander
Join Date: Dec 2007
Posts: 120
# 5
11-13-2011, 05:00 PM
While in theory, cutting the number of skills and doubling the cost seems like a balanced move, it isn't. They should have just ripped the ship specific skill branch out of the tree, and inserted the specific skills that were boosted in that branch - something that I never was able to fully determine. Then tweak costs, and/or add some other skills and been done with it.
Lt. Commander
Join Date: Dec 2007
Posts: 120
# 6
11-13-2011, 05:20 PM
You can now spend 50 more skill points after hitting the VA limit as a cumpensation. However, I am quite sure they're gunna add a new ground skill tree as well. It might take sum weeks though.
Lt. Commander
Join Date: Dec 2007
Posts: 120
# 7
11-13-2011, 05:42 PM
We are assuming that the Devs wish for Players to be able to heavily specialize in a role or "build".
This may not be the case, they may have designed the skill tree to favor a more open build over a specialized one via skill point costs.


Quote:
Skill point cost

The overall cost of skills at tier 4 and 5 are exceptionally high.
I agree, the skill cost of T4 and T5 need to be lessened slighlty or the whole skill tree made a little looser.

Quote:
Front loaded skills
Assumes a skill tree designed to allow heavy specialization.
Given the high PvE to low PvP player numbers the new skill tree may be designed to favor the PvE'er in a general build over the highly specialized of the PvP'er and its function in a team based enviroment.
To keep teams for overly specialized ships from dominating in iether a PuG or Premade enviroment, this would be a similiar move to cull that.
Allow specialization to a point with diminishing returns past a preset point of acsensiom up to max for any skill, thus allowing the minimax'er and the like to build to a specialized function but not the level of specialization we once had in hopes of allowing possibly less sophisticated builds by more casual players to be viable in a combat setting.

Quote:
Over-concentration
Cookie cutter builds already exist and have always existed. Somebody makes a build and others copy it to a point. Very few of us run pure builds.
And to be honest while the Science has the hardest time spending skill points ( as the CC guys, wizards, what ever form they take in agame usaually do) it may not be as hard for the Tacs but I can gaurante you we want to spend lots of points in SS Energy Weapon Specialization, SS Projectile Weapon Specialization and SS Weapons Performance just as badly as science or engineering wants thier T5 skills highly pointed.


Quote:
Relative value of skills
Change in how the Devs percieve balance that should be to make ships tougher, still deal decent damage (including spike) but allowing for a slower pace of cambat with less reliance on certain CC skills?
I agree the cramming of some skills full with the number of powers they interact with seems disportionate.

The engineering performance skills suffer from the same questionable pricing possibly to counter act the games wide use of Auxillary for so many powers (which many augment with batteries hence why Batteries is a skill now) as well to keep damage slightly lower in comparison to Holodeck stats to once again slow combat down.

As to counter skills, they have the resist lower to give theopportunity to the player to be able to more effectively build to resist those CC powers. For months there have been complainst on SS, AMS, SNB, SScan and the other CC powers in STO (much like the Damage gripes and anti-1shot cries of a year ago) that this may well be how the balance out those powers to make combat less about CC and more about the ship, damage being flung and then what tricks everybody brings to combat via science.

All the skills that directly effect hull, shields, power levels, turn, speed and direct applied damge have always been in high demand they just appear to have spread that thinner with the removal of teh ship specific skills.


Of course these are just my thoughts though.
Lt. Commander
Join Date: Dec 2007
Posts: 120
# 8
11-14-2011, 06:02 PM
I enjoy the 1st part of the new skills tree, however, at max level players meet a wall, they run out of points pretty easily and will have none for the ground skills when they finally make it on Tribble, which I noticed on STOked will probably be about a month. To balance out the two skill sets, we either need to increase the skill points beyond the current 72700 (which will probably only come when they increase the level cap) or lower the cost of skills (I'd prefer half).
Lt. Commander
Join Date: Dec 2007
Posts: 120
# 9
11-14-2011, 09:15 PM
Quote:
Originally Posted by Foxrocks
Part of the problem I have with the front-loading of skill benefits, while its very nice to get a lot of benefit for less than half the total cost of the skill, there is very little benefit to maxing it out or continuing to invest beyond the cheap 2-4 first ranks.
Exactly. That is the point. In fact, there is a new change coming out soon that will further emphasize this. Space Skills will be:
18 (+18)
36 (+18)
54 (+18)
64 (+10)
74 (+10)
84 (+10)
89 (+5)
94 (+5)
99 (+5)

We hope to get the UI to reflect the 3 "blocks" of "front-loadedness".

The point is, you don't need to put 9 points in each skill. Going broad in a lot of skills, medium in some, and deep in just a couple will actually gain you more effectiveness than what you can do currently on holodeck.

To emphasize this more, Training Nodes will now unlock at Rank 6 insead of 9 - further trying to emphasize that the last 3 points are just for min maxing.

This is just another iteration, so we look forward to your feedback once these changes are live

Quote:
Originally Posted by CrustyMac
While in theory, cutting the number of skills and doubling the cost seems like a balanced move, it isn't. They should have just ripped the ship specific skill branch out of the tree, and inserted the specific skills that were boosted in that branch - something that I never was able to fully determine. Then tweak costs, and/or add some other skills and been done with it.
Let be clear. We didn't cut the number of skills in half and double the cost. We cut the number of skills in half, only raised the cost by 50% (a net gain in buying power), and increased the effectiveness of the early points (front-loading). When we finish the ground skills, those too will be cut down in half or even to 1/3 - giving you even more buying power.

We understand this takes a little getting used to. We appreciate your diligence in testing. Try different builds that are more diversified, and only specialize in a few skills. I think you will find you are quite effective.
Lt. Commander
Join Date: Dec 2007
Posts: 120
# 10
11-14-2011, 09:39 PM
Quote:
Originally Posted by CaptainGeko
To emphasize this more, Training Nodes will now unlock at Rank 6 insead of 9 - further trying to emphasize that the last 3 points are just for min maxing.

This is just another iteration, so we look forward to your feedback once these changes are live
I am really looking forward to this above all the rest of the skill changes, to be honest. I've always tried to unlock all the captain training skills, so I've put 9 points in those skills, even though I've felt it was a bit of a waste (especially in the skills I don't utilize much). Changing them to 6 instead of 9 will be very welcome by me, and I can't wait to see it happen.
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is Off
HTML code is Off


All times are GMT -7. The time now is 05:06 PM.