Messages

1

General / Re: Let's talk about new PSI System!

« on: April 13, 2015, 03:16:00 pm »

Maybe have a PSI regen of 5* per turn if under 50%*, +1 for every point of will above 5*, the latter being tripled when above 50%*.

*The values would have to be balanced properly.

It would result in rather mundane chars having a very limited effective PSI pool that for most purposes doesn't regenerate in combat, while high PSI characters will have an interest in not "burning out".
Low Will chars would use PSI boosters to be able to use the better abilities at all, or repeatedly, while high will characters would use PSI boosters to sustain their powers during prolonged fights.
The above could also be reached by having a lower regen bonus per point of will and a feat that regenerates a % of the remaining PSI points, meaning you can easily sustain cheap abilities if that's all you use.
Sadly, the optimal solutions tend to be complex, and I suck at explaining.

If that level of complexity is being considered, then I would suggest a reservoir system as probably a lot easier to understand.

The basic problem being fixed here, as I understand it, is throttling how much psi can be spent rapidly in combat.  Purely on that metric, 20psi per turn regen isn't really all that bad at any level of Willpower (3 all the way to 10).  What needs to be addressed however is the totally free psi regen this enables, particularly for low Will characters.

Keeping a psi total reservoir at the same level as the old psi pool (i.e. before the most recent changes) and then using the 100psi/20psi regen per turn to throttle how quickly you can burn your total psi pool fixes the free psi, because all psi characters would still need to use psi boosters to refill the reservoir.  Meanwhile, a Will 3 character would have quite a low psi pool, and be subject to all the limitations as they used to be.

2

General / Re: Let's talk about new PSI System!

« on: April 03, 2015, 01:18:23 pm »

I think we're talking cross purposes here: I think that free psi regen puts psi characters outside of the game economy for all intents and purposes at low levels, whereas other characters have to spend money on bullets and/or item degradation.  This acts as a balancing factor, e.g. switching to .44 or 9mm ammo too early is tricky both in terms of availability and in terms of cost.  There is no such issue for psi characters under the changed system, and worse, at low levels something like cryokinesis outdamages any realistic gun option available to low level characters, and it does this for essentially no resource cost (plus it can't miss and has a huge range for the icing on the cake).

Tweaking psi regen rates is aimed at balancing how psi is used in combat, but it doesn't really address the economic issue at all.  Furthermore, making psi characters essentially unplayable unless you go with a starting will of 10 is bad design; you don't need to start a gun or xbow character with 10 Per, nor a melee with 10 in str or dex (if anything, this may well be suboptimal with the changes made to the feats) and I don't see why psi users should labour under that burden.

In some ways, its an attempt to inflict a classes on a skills-based system.  Requiring such heavy investment into a single attribute such that it locks out hybrid approaches is just creating a pseudo- class based system in my view.

3

General / Re: Let's talk about new PSI System!

« on: April 03, 2015, 09:22:15 am »

Its a horribly inelegant idea, but if psi costs are to be meaningful, then one approach would be to have a 'psi reservoir' that works like the old non-regenerating system.  Your current psi is refilled from that reservoir, at whatever rate is considered balanced (20 per round may be fine under this system).  However, when your reservoir runs out, you no longer get the top up each round.  Under this set up, psi boosters would work by first refilling your active psi (up to the max of 100 or 115 with appropriate headband) and then whatever was leftover would go to refill the reservoir.

This way, you'd limit the amount of psi that could be thrown around round by round, but keep psi as a costly resource that has to be filled up by using credits.

The psi reservoir could be set to something like 50 x Willpower say, and if you wanted to reward high Will you could perhaps make psi boosters provide Psi equal to 20 x Willpower rather than the flat 100 (or whatever it is, I forget offhand) as it currently stands.

4

General / Re: Let's talk about new PSI System!

« on: April 03, 2015, 08:14:05 am »

A couple of things - EMP grenades can be dropped at your feet without harm as long as you're not wearing a powered-up shield.

On the broader point though about the ramifications of the changes to psi, specifically regen vs. having to use psi boosters:

I can understand why this was done, but it has thrown the concept of economising out of the window for psi builds.  I played the earlier alpha versions quite a bit, and ran quite a few psi builds back then.  The psi booster requirements were actually pretty much equivalent to ammo requirements for guns users and/or the cost of grenades for throwers.  Consider that each 7.62mm bullet costs something like 28cr (iirc) and these costs quickly add up.  In the past, throwing out a cryokinesis bolt was roughly in the same ballpark costwise as shooting .44 ammo iirc.  This made sense from a balance perspective, because you got similar levels of damage per money unit.

I just tried out a fresh psi build and the difference cost wise is night and day - you can throw around whatever stuff you like, rarely have to use psi boosters, and can easily sell all the ammo you collect because its no use to you.

Perhaps in the long term though this makes no real difference - you end up swimming in cash after a point anyway.  However, whereas in the early game a guns user has to balance the ammo he uses with cost in mind, the psi user is now completely free of this type of thinking.  I think maybe its a step too far in the other direction really.

5

Suggestions / Re: Suggested newbie builds

« on: April 03, 2015, 07:56:54 am »

Attack skills increase your damage (idk the exact formula, never bothered to check it)

I think it's Base Damage x (0.7 x effective skill)%

at least for melee weapons and guns, I think it scales differently with psi

I'm assuming that should read base damage x ( 1 + (0.7 x effective skill)% ) in which case 8.5% increase per stat point equates to around an extra 3% per stat point if I've done the maths correctly.  E.g. at 100 effective skill vs. 108 effective skill, you end up with:

1+0.7x1 vs. 1+0.7x1.08 as the damage multiplier to base damage, i.e. 1.7 vs 1.756.  1.756/1.7 => 3% increase in damage

Therefore, you'd need to spend 3 points in Per to get the same effect as 'Gun Nut' on its own, all other things being equal.  Sure its nice to boost Per for an extra 3% damage per stat point, but only if you had literally nowhere else to spend the points.

Pretty sure this is only true for skills = 100.  Consider as skills approach infinite, the 1 makes no difference.  But yeah, much less than 10% anyways, more like 5 by the end of the game

In any case, I think this mere discussion proves the need for some form of guide or suggested builds

True, but skills don't approach infinity.  I used 100 as a reasonable 'base' skill level, as it assumes you're level 18 and have fully invested in the skill each level up.  As I understand it, the highest 'base' skill is 135 at level 25, which doesn't make too much difference to the derived % increase - perhaps its around 4% per stat point at level 25, around 3% at level 15, and lower than this in the early game.  Note that here I'm considering the 'base' as the points you invest in the skill each level up, with stat bonuses then applying on top of this at the flat rate of 8.5% per increment.

I'm not aware of any items adding to guns / throwing / crossbow / melee skill, so the base is fairly constrained in the game.

The 5% you've assumed wouldn't apply even at a base of 200 (base 200 => 4.7% increase I think) which would be at the equivalent of level 38.


Anyway, the main point of this is not to get sucked into the maths, but to point out that Underrail is not a game modelled around intricate calculations on dps (games like WoW are the model for this as far as I know, where developers spend a lot of time looking at damage output / damage mitigation possible for builds and even quite small increments in additional dps can be the difference between winning and losing a boss fight).

Rather, in Underrail being short 10% dps isn't usually what gets you killed - its being in a situation where your build doesn't have the right options to easily defeat the given encounter.  As such, versatile builds tend to be better than one-trick ponies.  As a slight digression, the main reason that Psi builds were completely overpowered back in the day was that they were a simple way of having single target damage / aoe damage / mech resistance bypassing damage / crowd control with just a single attribute and 3 skill investments out of your total 8.

With the changes to the feats, particularly in this final alpha version, my review of the details on the wiki makes me think that its now much harder to qualify for a broad mix of feats (i.e. attribute requirements have been increased for many of the really good feats) which means that the days of max'ing out your primary damage attribute are gone, unless you only want that one-trick pony. 


Edit: one interesting fact though is that each additional point in an attribute adds more to damage output than the previous one, i.e. 15 vs. 14 is a bigger damage increase than 14 vs 13 and so on.  Its quite a nice system in many ways, in that extreme specialisation does create a clearly apparent change rather than being prone to diminishing returns as many systems are.

6

Suggestions / Re: Suggested newbie builds

« on: April 02, 2015, 10:26:52 am »

Attack skills increase your damage (idk the exact formula, never bothered to check it)

I think it's Base Damage x (0.7 x effective skill)%

at least for melee weapons and guns, I think it scales differently with psi

I'm assuming that should read base damage x ( 1 + (0.7 x effective skill)% ) in which case 8.5% increase per stat point equates to around an extra 3% per stat point if I've done the maths correctly.  E.g. at 100 effective skill vs. 108 effective skill, you end up with:

1+0.7x1 vs. 1+0.7x1.08 as the damage multiplier to base damage, i.e. 1.7 vs 1.756.  1.756/1.7 => 3% increase in damage

Therefore, you'd need to spend 3 points in Per to get the same effect as 'Gun Nut' on its own, all other things being equal.  Sure its nice to boost Per for an extra 3% damage per stat point, but only if you had literally nowhere else to spend the points.

7

Suggestions / Re: Suggested newbie builds

« on: April 02, 2015, 10:14:58 am »

Primary stats don't affect damage, but they effectively do because skills affect primary damage.  Since primary stats bump skills by 8-10% (don't know the number) I'm assuming it's a 10% increase.  Possibly not, but I know it's pretty decent.  The extra damage is not shown on your guns, although it is shown on your psionic skills.

I agree that thats how it works for psionics, I'm just questioning if the same damage formula applies to guns/throwing/crossbows.  Also, the impact will be affected by the 'base' number that the skill modifies.

E.g. if the damage calc is (100+effective skill)/100 for the damage multiplier, then you'd see e.g. a 4% difference at effective skill level 100 vs 108 (i.e. the increment from one extra stat point).

Edit: I just saw Epeli's note about the combat stats page, I will check this out - thank you.

8

Suggestions / Re: Suggested newbie builds

« on: April 02, 2015, 10:02:57 am »

Well.. sort of.  Every point in your primary attack is about 10% more damage, so that adds up.  While most builds don't benefit from going all the way to 16 in a stat like they used to, most of the time you should go pretty high.  My current assault build is 8 str, 10 con, 11 perc, and 7 int, and I probably didn't need to go so high in int, the feats really didn't pay off.

But even then, a lot of players don't even get that feat access and damage is the primary purpose of stats.

I didn't know that each point was 10% damage - are you sure about this?  How does the damage formula work, is it based on total attack skill for all damage types?  When I look at item stats (e.g. gun damage range) they don't seem to be affected by the player's skill total; does one have to monitor the combat logs to see the impact of the additional skill on damage?

Gun nut on its own is at least 10% extra damage on average, and that would be multiplicative with any primary stat increases (it may be better than 10% on high damage threshold targets).  Plus the 7 int also opens up power management along with ballistics or armor sloping (and it makes crafting skills cheaper skill point wise to achieve a certain target) - as such, I would argue that there is a good reason to go with 7 Int on most gun builds unless you absolutely needed the points elsewhere (e.g. a high dex pistol build).

But my main point was really that blindly putting 9 or 10 into the 'primary' stat may not be the best build advice any longer, and that it may be better to think of builds as collections of feats that you want to access, with stat distributions primarily being about how to accommodate that.

9

Suggestions / Re: Suggested newbie builds

« on: April 02, 2015, 08:49:27 am »

A guide is simple to do. You don't so much need a pre-made build so much as you need the players to have some vary basic notions.

First thing. Always start with a stat at 9 or 10. That stat should be the stat affecting your main form of attack. Either perception for guns, strenght for melee or will for psionics. Similarly you shouldn't be afraid to get some starting stats at 3. A psion probably won't have a problem with str 3 and even constitution which non psi characters normally have no need for will. Think of it like playing D&D. Maximise your best combat stat and have a dump stat.

Second thing. Use the wiki... Seriously, just use it. You may not want spoilers and that is all nice and dandy, but consult the perks and learn which stats you need for the perks that will make your character functional so that you can better create the character.

Third. Some almost universal usefull skills. Stealth is great. Aside a heavy armor character which will have too much penalty it's always a solid investment as it allows you to start combat on your own terms. Hacking and Lockpicking. They allow to bypass some stuff but above all you don't lose a ton of extra loot. ombat skills. well yes, you'll always need some combat skills but do not forget to always have at least 2 different types of attack skills! If you limit yourself to just one type of attack you will find yourself in a situation where you will probably regreat it.

I think this is enough for any newbies to start learning the ropes.

Whilst this advice seems reasonable, I don't think its really true anymore.

The major impact that stats have is over feat access rather than skill enhancement.  For instance, its entirely fine to run a guns build with a perception of just 4, or perhaps 6 if you really want aimed shot.  The additional build points into perception don't really help your damage output all that much, all they do is open up further feat choices you may want depending on your chosen weapon.

The majority of builds will also want access to decent AoE options, which is probably going to come from either throwing (dex 6 req for grenadier) or from metathermics (wil 7+ is probably a good idea in this case).

The new crafting feats mean that Int 6 or 7 is very useful, again depending on exactly what gear you want to run with.

Pistols now have an effective str requirement of 6 (for steadfast aim).

And so on.  Basically, a build which looks horribly 'unoptimised' [e.g. str7/dex6/agi4/con4/per8/wil10/int7 @L24] may in fact be necessary to provide access to all the feats you want to have on your character (in this case, the listed stat distribution would allow for full assault rifle damage, grenadier, locus of control, gun nut - i.e. a very good balanced hybrid character that would output almost 100% of the damage a 'pure' assault rifle build would do along with access to crowd control via thought control, AoE damage via throwing, and even stealth if desired if you tack on ballistics to give you very good tactical vest options).

10

Development Log / Re: Dev Log #38: Version 0.1.14.0 released

« on: April 02, 2015, 08:32:44 am »

My only question about the new minimum ap cost is whether it should be 4ap or 3ap (and I'm just throwing this out there, I don't know which is better really).

Dex 15, lightning punches, tabi boots previously gave an ap cost of 3 for unarmed (10 - 4 - 2 - 1) which now would be bumped up to the min of 4.  Dex 11 I think is the minimum for 3ap reduction on fists, which along with the other factors would hit the minimum 4ap cost.

I'm not sure if this is good or bad to be honest: 4ap / attack = 12.5 attacks per round, 3ap = 16.7 attacks per round.  The real cheese (in my opinion) kicked in when you got down to 2ap.

The upshot is that a pure fists build now has 4 stat points freed up (the additional benefits of going from 11 to 15 dex are much harder to justify when you're not getting the extra attacks per round imo).  Conversely, total damage output has gone down but then again once you factor in feats like expertise its quite possible that combat is unbalanced even at the 12.5 attacks per round level.

Given that other melee weapons gain relatively from this change (e.g. ap reductions from taste for blood are unlikely to be useful now for fist builds), its possible that fist base damage may need to be adjusted to compensate.  Again, I'm not really sure about the full impact here, but I recall that fists used to be considered poor damage relative to other options, and these changes may undo some of the other balancing that was done.

11

General / Re: Chance to land psi-based status effects dependent on your skill?

« on: June 09, 2014, 03:52:06 pm »

Thank you for the confirmation Styg, thats useful to know from a build perspective.  I guess I will have to alter my build plan of investing minimally in psychokinesis to obtain a cheap stun.  Still, I'm glad that this is how the mechanics work from a game design point of view, the skill should definitely require ongoing investment as you level up.

12

General / Re: Chance to land psi-based status effects dependent on your skill?

« on: June 07, 2014, 08:05:10 pm »

Reason I asked here is that I did a bit of testing but the results are inconclusive to say the least.

I rolled up a character with will 4 to start, and increased psychokinesis to just 40 (so I could take premeditation).  On this guy, telekinetic punch and electrokinesis both worked quite reliably all the way to the end of Junkyard, and into the warehouse and rail crossing sections.

I then tried a will 3, psychokinesis 0 guy.  He couldn't even land the stun against rathounds.

A will 3, psychokinesis 15 guy could comfortably land stun against the initial rathounds, and the fight with the woman with 2 pet rathounds.  I didn't take this build further at this stage, as I thought I'd ask here in case someone's already done the investigation or the mechanics have been detailed already.  I'll probably run this build with the psychokinesis kept at 15 to see when it starts to feel unreliable.

So at the moment, I'm really not sure if this is a threshold thing (i.e. there is a hard minimum that you need for the stun to take effect, but then raising your skill higher doesn't make it more effective) or if it scales linearly with increasing skill but is subject to a hard cap. 

I'd certainly have expected the will 4 / pscyhokinesis 40 guy's effectiveness to be very low by the end of junkyard for example, but it was still quite reliable at that stage - certainly it didn't feel qualitatively worse than my maxed out psi guy.

Also, nothing in the description of resolve suggests that its a contested roll (vs attacker's skill) - this doesn't mean to say that it isn't contested of course, but I could easily see it being a straight attribute check for the defender independent of the attacker's skill.

13

General / Chance to land psi-based status effects dependent on your skill?

« on: June 07, 2014, 08:52:05 am »

As the subject title: does increasing your psi skill and/or Will attribute increase the chance to land things like stuns?  I know it increases the damage component, but am curious specifically about the status effect success probabilities.

14

Suggestions / Hot key to swap weapons

« on: June 06, 2014, 09:15:37 am »

I'm not sure if maybe I haven't found what key does this, but it would be good to have a keybinding to swap your active weapon.

15

Suggestions / Re: Shroomhead alternative?

« on: January 22, 2014, 10:05:03 am »

Started playing Underrail a few days ago (had tried the alpha a few months ago but didn't really get into it then), and wanted to make a few comments about psionics based on my experiences so far.  Logged about 25hours according to Steam so far, which includes trying several different builds all the way to getting to Junkyard.

First, I have no experience of the old shroomhead feat, but the new one is worthless imo - I would not take this for any character, as ultimately it translates to receiving approx 75 credits per mindshroom picked up (using 700 credits cost of psi booster / 170MP x 15 psi per mindshroom picked up).  I've no issue with the feat being as it is for flavour purposes, but as a selectable feat it looks extremely suboptimal to me.

Having said that, I encountered no difficulties whatsoever with a recent build that was pure psionic - skills were the 3 psionic disciplines, stealth, hacking, lockpicking, dodge and evasion (I wanted to try out a completely pure psionic user, so no other weapon skills whatsoever).  Stats were str 4, dex 4, agi 7, con 6, per 5, wil 10, int 4.  I played on classic xp (purely to test out the difference, as my first playthroughs were all using the oddity system).  By the time I'd finished the outpost quest and the hopper quest, I had enough credits from sold equipment to purchase a galvanic vest(! - if I hadn't bought this, I could have purchased another 15-20psi boosters), 6 or 7 psi boosters plus a few other bits and pieces.  I also found that despite having 0 in guns, I was able to equip 5mm and 7.6mm pistols which provided cheap attacks for when fighting easy or almost dead mobs.

I got through the entire GMS quest by using up all but one of the psi boosters I had, but of course once you've finished that quest you have a lot of loot that can then be sold for more psi boosters.  I had enough credits under the new economy system to pick up all the spells I could learn at that stage, and had a few nice guns and other loot left over to sell at Junkyard when I got there, plus a purchase of another 10 or so psi boosters.

I'm really not seeing any difficulty playing the early game as a pure psionics user, provided you make use of the starting pistol with zero skill invested for very easy fights where you really don't need to burn MP.  I suspect that once you've finished GMS, the cost of buying psi boosters is no longer a real issue, so the fact that it was straightforward to do so tells me that the economy isn't overly unforgiving even for a pure psionic build.

The only caveat to the pure build in my mind is how the early game plays on oddity - I intend to restart under that option to compare the two xp systems, but I don't think this will materially change the difficulty.