Development > Artwork

Alien bestiary

<< < (9/10) > >>

maackey:

--- Quote ---Frankly, I'd completely change the damage types for something that's imspler and more intuitive.
--- End quote ---
This. A thousand times this.

I've looked around at the weapon balance a few times before and was stopped dead in my tracks by the dauntless numbers of damage types and resistances. Not to mention the wiki weapon tables are woefully out of date. (I've been working on a little lua script to parse the data to better compare weapons/armor... I don't suppose anyone has a working program to do this already? or perhaps it isn't very convenient to use...)

The convoluted amount of damage types are detrimental for both developers and players. On the developer side, with so many variables it is practically impossible to properly balance weapons, armor, aliens, etc. On the player side, it is incredibly confusing when your weapons one shot ultra powerful brutes in one mission, and are completely ineffective against fragile twig-men the next. Not to mention the disconnect in expectations (ever play a point and click adventure with those "simple" puzzles that you needed a manual for?) "... what do you mean bloodspiders are resistant to shock!? they are completely electro-mechanical! No amount of electrical shielding can stop yadda yadda"

It would be really nice if we had something like the taylor knockout formula for ballistic projectiles. It is roughly:  damage = mass * velocity * diameter of bullet
There are other factors for sure, such as bullet shape and composition,but its a good guideline.

I dunno if I should post here or somewhere else, but some initial thoughts for a proposed change:

ballistic (physical)
 - physical projectiles
 - cut/stab/bludgeon
 - blast (concussion)
energy
 - electric
 - plasma
 - laser
 - fire
 - blast (eg. from plasma grenade)
hazmat
 - stun gas
 - poisons
 - blinding flash (from stun grenades?)
 - mind controlling spores

Three damage types may be a pretty extreme change from the current 30 (hush... its hyperbole), but look at it this way: damage isn't the only variable we have to play with. It makes no sense to have sniper bullets have a different damage type to pistol bullets. Anything that a .22LR can hit a .50 BMG will hit harder. No exceptions. But that just means that 22s can have less recoil (less TU), cheaper ammo, more capacity, etc.

The example is a bit weaker for energy & "hazmat" but the principle is the same: KISS. Reduce the extraneous variables and only work with ones that give meaningful results. I can probably think up counter examples and reasons not do put flash/poisons together but is it really worth it to have them separate? How many weapons use the mechanics of the flash grenade? or stun gas? Is there a vital reason why an alien would be weak to stun gas but not flash grenades (other than *arbitrary* story fluff)?

I would also love to have more varied ammo types, so for example there would be your standard ball, but also armor penetrating, AP incendiary, AP plasma etc. AP rounds would get a bonus vs ballistic armor, incendiary would apply energy damage over time, plasma applies large instant energy damage. I dislike using stock alien weapons. I would much rather take their technology and turn it into something human. I know there were a few threads discussing franken-weapons: alien weapon models with human modification on them which would be awesome. (If I ever get into modeling again I might have a go at some)

heh, thats a lot of text. I've got to stop making these giant walls...

H-Hour:
It's not as difficult to properly balance the existing weapons as it appears at first. Once you dive into it, you will realise that a lot of damage weights have only one or two weapons associated with them, so you can have pretty fine control. Adding new weapon concepts is difficult, though, without adding new damage weights (this was done for the Encased Plasma Ammo).

I have toyed with the idea of a new damage model as well. But I'm not yet convinced that it is actually simpler for the player -- what seems simpler from a conceptual point of view (ie - the developer) is not necessarily simpler for the player. If I have to calculate damage by analysing an armour penetration rating, mass, velocity or other elements, that may just be more ambiguous than a simple damage number. I'm not yet sure about this, though, and my primary concern was to expand the ability to better model a weapon's interaction with armour.

The real problem we face with the current implementation is that the player is not given information on the real damage weights protection/resistance for armour and aliens. So he's unable to make a simple addition/subtraction comparison with a weapon. Including this information as the result of an autopsy would be a nice incentive for completing autopsy research, and the same information should be made available upon researching alien armour.

maackey:
Hooray, another wall of text from your bedsick fellow complaining about more stuff :) I do it because I love.
(should this be split to a separate thread? I don't really want to drive the thread off-topic)


--- Quote from: H-Hour on December 12, 2012, 11:23:25 am ---It's not as difficult to properly balance the existing weapons as it appears at first. Once you dive into it, you will realise that a lot of damage weights have only one or two weapons associated with them, so you can have pretty fine control. Adding new weapon concepts is difficult, though, without adding new damage weights (this was done for the Encased Plasma Ammo).
--- End quote ---
Some damage weights make sense, and I'm pretty sure it would be fine to have a half dozen or more if it meant keeping fine control, but in the resistances graph you linked to there are types like normal light, spray, medium, heavy, fire light, medium, heavy, flamer etc. I can't imagine why there would be a difference in *TYPE* for a light/medium/heavy damagetype.


--- Quote ---I have toyed with the idea of a new damage model as well. But I'm not yet convinced that it is actually simpler for the player -- what seems simpler from a conceptual point of view (ie - the developer) is not necessarily simpler for the player.
--- End quote ---

I suppose... do you have any specific examples though?


--- Quote ---If I have to calculate damage by analysing an armour penetration rating, mass, velocity or other elements, that may just be more ambiguous than a simple damage number. I'm not yet sure about this, though, and my primary concern was to expand the ability to better model a weapon's interaction with armour.
--- End quote ---

I agree, there is a balance between realistic simulation and fun intuitive gameplay.

Lots of games have AP, its an easy enough concept to understand. Damage range falloff also makes sense: it realistically models projectile drag, laser light scattering, etc. I'm curious on what your thoughts are for weapon armor interaction. I've done a lot of searching on the wiki and forums and have only ever found really old out of date discussion on the topic (which frustrate me to no end*). In my view, armor is there to reduce damage. Either by subtracting a flat amount, or reduced by a fraction, or both. I dislike the general idea of minimum damage (eg. armor blocks 10000000 damage and pistol damage is 10 -- minimum 5 goes through) I'd rather have better reduction/damage values. (eg. armor blocks 15+50% and pistol damage is 20 -- 2.5 damage goes through) and I dislike* pistols,shotguns,rifles etc. having different damage reduction types (on top of different damage values) as it contributes to the confusion of how much damage does my weapon actually do?

I play a lot of Zero-K and its motto is no special damage types -- and it works wonderfully. Armor is modeled by more HP, weapons and units are balanced according to the physical interactions and limitations, it is dead easy to determine how much damage a unit will do, and how well it will stand up to punishment. I'm not suggesting going to that extreme, just trying to give perspective.


*The one thing that bugs me the most and I would like to see changed is the light/medium/heavy damage types -- heavy plasma can be modeled by shots doing 100 damage, light plasma can have shots do 20 -- they are both plasma and it doesn't make sense to have them separate.


--- Quote ---The real problem we face with the current implementation is that the player is not given information on the real damage weights protection/resistance for armour and aliens. So he's unable to make a simple addition/subtraction comparison with a weapon. Including this information as the result of an autopsy would be a nice incentive for completing autopsy research, and the same information should be made available upon researching alien armour.
--- End quote ---

With so many damage types I wouldn't want to compare values with every alien/armor/weapon combo even if I had them automagically pop up side by side. Having fewer damage types would mean it would be easier to naturally infer resistances/weaknesses, which could counteract the benefit of autopsy research, but that means that I don't need to make tons of addition/subtraction comparisons! To compensate for that we could give the player a small 5% increased damage buff vs that type of alien -- or some other fun little perk. Either way it would be nice to have that information in the autopsy, I wholeheartedly agree.

Anarch Cassius:
When I first saw this system I completely cringed. What's listed in UFOpedia seemed simple and elegant like Fallout, which uses modifiers to handle penetration.

However I now think this system is the way to go. Rolemaster used a similair super detailed system and the idea is very realistic if you can have a computer doing the math. The downside of any reasonable complex damage system is it is nearly impossible for the player to calculate damage on the fly. On the other hand if the general system is intuitive not being able to get exact results in your head can be a good thing.

I think what's needed is a more formulaic approach to the subtypes and adding a few rather than trying to simplfy the system. With a little clean-up and continuing attention to detail this probably is the best way to model the interaction of weapons with various armor types.

H-Hour:
@maackey: the main reason we have the damage weights (spray, light, medium, heavy, etc.) is to be able to model armour. The SMG, for instance, uses normal_light, and can be effective on Tamans early in the game. But as soon as armour is introduced it becomes pretty obsolete, because armour's protection against normal_light drastically reduces the damage potential. Having separate damage weights allows us to model the effectiveness of armour differently for different weapons. Assault rifles (normal_medium) get a lot weaker against armour, but the sniper rifle (normal_heavy) still packs a powerful punch. If we tried to do this just with higher damage values, we'd end up under-powering or over-powering weapons in the distribution.

As I said, I'm not entirely happy with the damage weights system, but I'm not in any position to change it at this time. I've also just finished a pretty comprehensive rebalancing of the weapons. I'm interested in riding the system we've got for a while and seeing how it plays out.

Personally I like the idea of modelling more abstract weapon parameters which define the interaction with armour and the wounding process, but I'm only half-way through my campaign with the new weapon balance and I'm really happy with the weapon balance we've got (pre-Needler, though, this could get rough!). Maybe the system does work, even if it doesn't appear as elegant under the hood, and I'm not going to beg a coder to rewrite a system that works unless I'm really convinced the benefits will be worth taking their time away from other features.

Navigation

[0] Message Index

[#] Next page

[*] Previous page