More complex armor calculations

Users who are viewing this thread

bryce777

Count
Well, I really don't like the way armor works.

For example, it's been mentioned a few times that helmets are worthless.  The reasoning is that headshots damn near always kill you, anyhow, so what's the point?  I don't think helmets are worthless, but they are definitely not worth all that much and only the heaviest helmets will save you from a headshot and that's only if it didn't do too much damage.

Now, you could do a kludge and make headshots do slightly less damage or make helmets have a higher rating than body armor (which does make sense since helmets are usually MUCH thicker than body armor) but I think that a much better solution is possible.

So the simplest armor system possible is to just take a few points of damage off.  This is what happens now, and it is simple, but it does not really model amor very well at all.

A heavy blow can easily bounce off the side of a facegaurd on a helmet from a certain angle doing no damage, and just as easily instantly kill someone if it is an inch to the left by going through the eye ino the brain.

In a system like DnD, you either hit or don't hit, and that's the end of the story.  In M&B you never miss this way - some say it's unrealistic for your arrow to hit on the screen and do no damage, but it's also unrealistic to never have a blow deflected, as well.

Both of these should be used sparingly, however, because their results are too dramatic. 

Some systems use damage resistance as a percentage.  So, if you have 100 points of damage, armor that absorbs 30% of it will leave you taking 70 points of damage.

Some systems also have coverage area, which is basically how likely you are to bypass the armor completely.

An ideal system would combine them both in a sort of GURPSlike manner:

Damage resistance: Even clothing would dampen blows a bit and give 5% or so resistance, whereas chainmail might get 50% resistance and plate armor 70% resistance

Coverage:  Usually this will be a high number like 85-90% for body armors and for some body armors even 95%  However, this allows you to model things like openfaced helmets that might only cover 60% of the head, or breastplates which have extremely high protection but do not cover the whole upper body, or plate armor that does not cover the rear in the legs.  If there is 80% coverage, there is a 20% chance a blow bypasses the armor completely, which could be very serious for a head shot.

Deflection:  this is a chance that a blow will flat out be stopped cold no matter how hard it is delivered.  Basically, curved armor, especially plate mail, can have even a severe blow simply slide right off if it does not contact the armor very directly.  Armor like leather would have an extremely low deflection chance such as 5% or less, whereas plae armor could have a much higher chance, perhaps as much as 30-40%.

Soak:  Soak is what's used now, and it should be used extremely sparingly.  In more sophisticated systems it is used sparingly indeed because it is so powerful and hard to balance, but it does have its place.  Without balancing it youe asily get into situations where you either totally deflect blows or get killed instantly if you don't have enough armor, which is what happens here and is not terribly realistic - a man in plate armor is not invulnerable to a mob, nor is he probably going to die from one glancing blow of a scimitar either.  Both can happen in M&B, though.  What soak IS good for is to represent the ability of an armor to not be phased by very weak blows.  A weak peasant with a rake should not be able to do much or maybe any damage unless he gets lucky and hits an unprotected spot.

 
Overall I like the idea.

Deflection could be made by custom tailored hitboxes. I don't know hard this would be to implent but IMO would be a lot better than a chance of deflecting. And in other cases too per cent chances should be avoided: I don't want to hit the enemy into the face and have him receive the armour because there is a random chance of it happening (or vice versa to the back of the head, for example [now it is always the same, no matter what]).
 
Yeah, i'd like to see many more hitboxes. I wouldn't like to see random stuff, as it take away from the skill factor.
 
It would use virtually zero memory or resources of any kind.


*******************************


As for hitboxes, that is just not practical.  I see what you are saying and in an ideal world you would be able to calculate exactly which link of mail the arrow hits and whether the arrow hit or not.

The truth is that as it stands now, the CURRENT hitboxes are not very good at detecting collisions.  Making hitboxes is not an exact thing at all and takes a great deal of work.

Now, as far as player skill goes, I don't think this will change that at ALL.  The current system does not reward you for hitting other areas either, except when it comes to headshots with a bow.  In fact, it is really bad about not bypassing shields when it should, bypassing them when it should not, etc.  You also do not have targets which do things like duck, nor can you duck, and there is no realistic way to program this in in a controllable fashion, either.  So, there is always some level of luck or character skill which can never be transferred to complete control of you the player.

 
I like this. It would not need much more memory (the code would be larger however), and it would require more calculations. I doubt that any of this would be noticible, however. Nearly every decent armor (except helmets) should have some soak value however, reflecting that there is layering. Even if you get a perfect strike on a joint, there is going to be some reduction due to this under-armor layer.
 
When you count billions of ops per second and measure memory in megabytes and even gigabytes, you don't need to worry about optimizing a method that will get called only a few hundred or few thousand times per battle or a few extra bytes on the definition of each item(which would be stored in exactly one place in any sensible coding scheme).

Believe me guys, when I say not to ever worry about perormance in suggestions and let armagan worry about that unless you have a degree in programming or lots of experience.

Anyway, not addressed at you calandale, or even arab archer, jsut as a generalization.
 
bryce777 said:
Believe me guys, when I say not to ever worry about perormance in suggestions and let armagan worry about that unless you have a degree in programming or lots of experience.

Anyway, not addressed at you calandale, or even arab archer, jsut as a generalization.

That's ok, I qualify on both counts. And it is an issue that we all should worry about when thrashing discussions out. The fact that there are a significant number of programmers on these boards helps weed out some really bad choices; it may also just give Armagan a glimmer of how to do something, as all of us know that sometimes even the obvious can elude us.
 
Thanks guys - I am glad to see some others agree with me on this.

I think it would also be neice because it would give the modders the ability to make just about any types of armor system they choose to.

It would also allow the creation of real differences in armors and helms - so that armor type could matter beyond mere weight and protection factor - one armor might be better in certain circumstances such as against large groups of weak enemies, whereas another might be better against stronger ones.
 
In real life war you did not aim for an knights head, that would just be stupid, instead you aimed for his horse since then he might get crushed when he falls down  :twisted:

But if i see someone runnin' at me with an Ak-47 and full body kevlar without helmet screaming Allah Ackbal! then i would take an revolver and aim for hit head, but thats another story  :wink:
 
Don Jacobo said:
In real life war you did not aim for an knights head, that would just be stupid, instead you aimed for his horse since then he might get crushed when he falls down  :twisted:
So are you telling me that knights beat enemy knights' horses till one of them dropped, when they could aim their strikes at potential attackers? (and btw. it's an ancient topic)
 
More qualities that can be given to armour is very good from a modding point of view as well, particularly if certain types of attack (such as blunt etc) ignore or alter one of the qualities, such as soak.

Maelith.
 
Depends on how it's done. If your running a script which is checking for impact constantly, it's incredibly bad performance wise. Only checking when contact occurs isn't too much of a resource hog, but isn't as accurate. It also depends on the number of poly's, obviously.
 
If armor gets overhauled, I'd say it would also be a good idea to do the same with weapons. It would be nice to be able to tweak all kinds of settings. Armor penetration, knockdown and weapon lethality could all be made into seperate scales rather than being tied to damage type - a bodkin could have better penetration than a barbed arrow and some polearms should have a good chance of knockdown/unhorseing despite not being blunt. Weapon lethality could be done by having a negative number, and if a blow from the weapon reduces an enemies health below this then they die, but if it goes below 0 without going under the lethality number they just get knocked unconscious. A club could have -100, making it never get kills and a sword something like -10 so that it rarely kills people in heavy armor.
 
Ilex said:
Don Jacobo said:
In real life war you did not aim for an knights head, that would just be stupid, instead you aimed for his horse since then he might get crushed when he falls down  :twisted:
So are you telling me that knights beat enemy knights' horses till one of them dropped, when they could aim their strikes at potential attackers? (and btw. it's an ancient topic)

Archers did, it is hard to hit an horse from another horse due to the armour that sometimes existed. Although i am certain it varied from knight to knight
 
Back
Top Bottom