News Discussion with Mark Allen on damage and defenses

[Frenotx]

Greetings Commanders,

Recently and over time, we've seen a fair amount of discussions regarding game mechanics, such as damage and defenses. Mark Allen joins us today to discuss such!

I definitely like what I see here. The process looks very solid, but I think there are a few specifics that need some tweaking. The big one that stands out is step #3. I was actually just talking about this to my boyfriend the other day. Was trying to think of a way to make hull more relevant, but also keep module-sub targeting possible. The current implementation using the remaining hull % to bias an RNG is decent, but I think no matter how well it's balanced, will always feel a little clunky. With this, you have the chance of shots just sailing through a more-or-less untarnished hull totally unimpeded. Additionally, you have the chance for a barely-together swiss cheese wreck of a hull to completely and absolutely block a shot from making any progress. On average things will work out, but on a shot-by-shot basis, things will occasionally look really silly. These "critical hits / critical fails" can also be very frustrating to the player, not unlike the situation you mentioned re the hull reinforcement packages. If I land a shot dead on target, from the right angle, on a target with a crippled hull, I am confused and irritated when I watch it do 0 internal damage.

What I propose instead would be to remove the RNG from step #3 (have weapons always technically "penetrate"), and have the remaining hull percentage instead modify the base penetration depth mentioned in step #5. The more hull remaining, the smaller the base penetration depth. Once you've got the values adjusted right, you can get things to behave on average identically to your current implementation, but the shot-by-shot case will be a lot more consistent and predictable. Being at nearly full health would lead to only the most armor piercing weapons being able to travel a significant distance through the hull, and being at almost no health would mean most weapons can get to your modules with relative ease. I think this would feel a lot smoother and reliable, and take away the effective "random critical hits / critical fails" that the current system gives rise to.

Another unrelated and much more minor change I suggest is to adjust the "chance to be hit" of modules described in step 6 to be 0 when the module is completely destroyed. As it is, if your shot hits an already destroyed module, a lot of it is effectively wasted. ~20% goes to the hull, and the remaining ~80% dumps into the already-at-0 module. It would be nice if such shots instead hit some other eligible module, and if none are available, just did full damage to the hull. Since you can't see that you did any damage to an already-destroyed module, you once again run into the issue described re the hull reinforcement packages having hitboxes. Your shot appears to randomly do drastically reduced damage for no apparent reason.

 

[nrage]

I definitely like what I see here. The process looks very solid, but I think there are a few specifics that need some tweaking. The big one that stands out is step #3. I was actually just talking about this to my boyfriend the other day. Was trying to think of a way to make hull more relevant, but also keep module-sub targeting possible. The current implementation using the remaining hull % to bias an RNG is decent, but I think no matter how well it's balanced, will always feel a little clunky. With this, you have the chance of shots just sailing through a more-or-less untarnished hull totally unimpeded. Additionally, you have the chance for a barely-together swiss cheese wreck of a hull to completely and absolutely block a shot from making any progress. On average things will work out, but on a shot-by-shot basis, things will occasionally look really silly. These "critical hits / critical fails" can also be very frustrating to the player, not unlike the situation you mentioned re the hull reinforcement packages. If I land a shot dead on target, from the right angle, on a target with a crippled hull, I am confused and irritated when I watch it do 0 internal damage.

What I propose instead would be to remove the RNG from step #3 (have weapons always technically "penetrate"), and have the remaining hull percentage instead modify the base penetration depth mentioned in step #5. The more hull remaining, the smaller the base penetration depth. Once you've got the values adjusted right, you can get things to behave on average identically to your current implementation, but the shot-by-shot case will be a lot more consistent and predictable. Being at nearly full health would lead to only the most armor piercing weapons being able to travel a significant distance through the hull, and being at almost no health would mean most weapons can get to your modules with relative ease. I think this would feel a lot smoother and reliable, and take away the effective "random critical hits / critical fails" that the current system gives rise to.

I mostly agree with this. I can see why they went with RNG, it makes sense given that the hull with damage model is not precise enough to determine if a shot flies through an existing hole or hits something in the way. But, as you say this random nature makes the result somewhat unintuitive/frustrating to players.

Another unrelated and much more minor change I suggest is to adjust the "chance to be hit" of modules described in step 6 to be 0 when the module is completely destroyed. As it is, if your shot hits an already destroyed module, a lot of it is effectively wasted. ~20% goes to the hull, and the remaining ~80% dumps into the already-at-0 module. It would be nice if such shots instead hit some other eligible module, and if none are available, just did full damage to the hull. Since you can't see that you did any damage to an already-destroyed module, you once again run into the issue described re the hull reinforcement packages having hitboxes. Your shot appears to randomly do drastically reduced damage for no apparent reason.

I disagree with this. A totally destroyed module will still be a lump of metal which gets in the way of shots and inadvertently protects other modules. It basically becomes a piece of less effective hull.

 

[noldunar]

So if I am understanding correctly what I have read, armor and hull reinforcements do protect sub-systems in contrast to what everybody thought. Just maybe not as much as was hoped for.

So let's simplify this a little with some examples:

Example 1, no armor:
Your hull has an unmodified strength of 100 and you have light alloys (no armor). You have no shields. You are hit for 10 points of damage.
1.) 10 points are multiplied by your armor, which is 0 so still 10 points. 10 points are subtracted from 100, your armor is 90 now (90%).
2.) Now the check is made to see if the shot penetrated the armor, which is a percentage check based on current hull healt. So let's say at 90% hull the chance is 45% (it was stated that 100% hull has more or less 40% chance).
3.) If the check indicates that armor is pierced, damage is applied to modules.

Example 2: With armor and hull reinforcement:

So if I have military armor, which is said to have 50% reduction to both damage types and let's say 100 points hull reinforcement ist would be like this:
1.) 10 points are multiplied by 0.5 (50% armor reduction) leaving 5 points of damage. 5 points are reduced from the reinforced hull value of 200 (100 base + 100 hull reinforcement). Your armor value is now 195 which should be a loss of 2.5 % so your hull is at 97.5 %.
2.) Now the check is made to see if the shot penetrated the armor. But since your hull percentage is now 97.5% instead of 90 as in example 1 the chance for that is lower. Let's say 42% instead of 45.
3.) If check indicates that armor is pierced, damage is applied to modules.

So basically, armor and hull reinforcement lowers the chance of module damage BUT the chance can never go below the stated 40% (at 100% hull), HOWEVER, you can keep the chance at 40% or close to 40% by adding armor and hull reinforcement for a longer period of time.

Did I get that correct?

So I think you could really argue about if 40% chance of internal damage at 100% hull is to high a chance. I think it is, but your mileage may vary, but armor does provide a little protection.

 

[Bunkerkind Anni]

your example is not fully correct, as you deduct hitpoints from hull already at step 1, while that will only happen at step 8.

at step 7 the penetration rating of the weapon fired at you will determine how much damage is dealt to the hull (at step 8), and how much of it to the module

so Tl;dr version would be:
Different Bulkheads will lower total incomming damge, and their additional hitpoints will keep the chance to have internal hit at a lower chance.

"Hull Reinforcement Modules" will keep the chance to have internals hit at a lower chance, but not lower the damage.

 

[Frenotx]

I mostly agree with this. I can see why they went with RNG, it makes sense given that the hull with damage model is not precise enough to determine if a shot flies through an existing hole or hits something in the way. But, as you say this random nature makes the result somewhat unintuitive/frustrating to players.



I disagree with this. A totally destroyed module will still be a lump of metal which gets in the way of shots and inadvertently protects other modules. It basically becomes a piece of less effective hull.

With the current system, a destroyed module is basically a MORE effective piece of hull. It's a thing on the ship that is effectively invincible, and can absorb ~80% of shot. Just imagine if there were a module with that description. "Invincible module that when hit, reduces the damage of the shot by 80% and prevents further module damage." Don't you think people might find that a bit ridiculous?

I mean, fitting a ship with tons of weak, cheap, modules with the express intent of them getting destroyed and acting like damage sinks is actually a reasonable thing to do. The more internals you have filled with little "damage sink" modules, the harder it is to do damage to the hull (shots will keep counting as having pierced the hull, then just waste their game on an already destroyed module), and to other more important modules (shots can wind up striking one of the destroyed modules instead of the one you're aiming at). Depending on the number of internals and their placement in a given ship, you may get more survivability from filling all your internals with class e1 afmus than you would if you'd filled those slots with hull reinforcement packages.

 

[nrage]

With the current system, a destroyed module is basically a MORE effective piece of hull. It's a thing on the ship that is effectively invincible, and can absorb ~80% of shot. Just imagine if there were a module with that description. "Invincible module that when hit, reduces the damage of the shot by 80% and prevents further module damage." Don't you think people might find that a bit ridiculous?

I mean, fitting a ship with tons of weak, cheap, modules with the express intent of them getting destroyed and acting like damage sinks is actually a reasonable thing to do. The more internals you have filled with little "damage sink" modules, the harder it is to do damage to the hull (shots will keep counting as having pierced the hull, then just waste their game on an already destroyed module), and to other more important modules (shots can wind up striking one of the destroyed modules instead of the one you're aiming at). Depending on the number of internals and their placement in a given ship, you may get more survivability from filling all your internals with class e1 afmus than you would if you'd filled those slots with hull reinforcement packages.

It only absorbs 80% if it's hit, and as no-one is actively targetting it it's not very likely to be hit. Especially if they are actively targetting something else like the power plant. By the time a module is at 0% or two of them are at 0% the main hull will be all but gone, so the worst that happens is that their death is delayed by a few shots/seconds. It doesn't seem like a huge problem to me TBH but it does make sense for shots to penetrate through the module and potentially get the next one in line, so I'm all for a change that makes that happen.

 

[Zenow]

Great post! Now I want to get some ideas off my chest concerning damage/combat, but (and I hope I don't disqualify myself now) I haven't got the game yet. The Elite I know is.. well, from the 80s and 90s. I still need to get my hardware in order before I can get it. Still, I've seen video's, and I've played games in the past where subsystem targeting was needed (freespace, wing commander), so I do have some thoughts..

- I saw ships still moving around even though engines were gone.. so does shooting subsystems work well enough? Although I like all the explosions, I would love to see less ships destroyed, and more disabled - simply because that could open op a while range of scenario's and missions:

Hijacking
- boarding a damaged ship: if it is large enough, by docking to it (requires airlock on both ships). Could result in simply introducing something like sleeping gas into the air system of the conquered ship to get past the game dynamics of that. You can then haul the crew into your cargo bay and sell them as slaves (meaning other player loses his ship and can buy his freedom. If not: stranded on space station until he can work at a ship that is hiring crew (as pilot of course) or borrow money for a new ship - resulting in the pressure to make regular payments on the loan - with all the risks (and bounty missions) that entails. Amount you can borrow can depend on your rating. (by the way: debt collection could be great reason for not destroying ships, if the debt is too low)
You could even make sure most vessels have an airlock, and allow even small ships to either be boarded or the pilot being rescued by his friends if the ship is disabled (resulting in a secondary pilot in your cockpit for a while). When docked, both ships move as one (graphics challenge, of course). On the other hand, if the propulsion of the conquered ship still works - you might even steal it: more missions (retrieve my ship! Bounty is XXX: players could then set bounties)

- towing a damaged ship: if a ship does not have an airlock, it would have to be towed to a space station. That could result in other wonderful missions:
- your wingmen can rescue you by towing your disabled ship. Of course, that makes you sitting ducks Maybe you can even send out a distress signal for someone to tow you - with the risk of being conquered - or someone to help protect your friend's damaged ship as you tow it: an impromptu convoy mission, where those whose aid you accept, have to get paid, of course.
And then, of course, if you are towed by the enemy, he would have to tow you to a system where he can actually get your ship repaired while taking it from you, without being arrested - so that means only specific systems: giving your buddies time to come rescue you yet again.

I think the game requirements of this are not that high, except for the docking variation. A towed vessel could work almost just like in a wing formation, except you need to visualise a tractor beam, and exchanging money for your hired convoy is easy. People losing their ship and being sold as slave/stranded on a space station are also simple interface exchanges, no 3d graphics needed.

A final one: I don't know about the reasoning being offered for all the sounds you hear in the cockpit, but I can imagine they are generated for the benefit of the pilot by some kind of environmental audio system. That takes care of the reality of 'In space, no one can hear you scream', but also: that would mean that system could become damaged - making you deaf to the outside world. That may not sound like much, but I bet it is a distinct disadvantage in combat. If the game becomes boring without any audio, you could just make it glitch, have annoying squeeks, alerts, etc.
Other subsystem malfunctions could result in random (or not) launching of missiles (or all of them), defective 'gear down' indicators, failing comms, etc.

---
Anyway, all these are just ideas of how the game can become more interesting if ships are not so easily destroyed. Some of the best moments in space movies, are those when a ship is falling apart, and there is time enough to feel and appreciate the tension that brings.


(sorry for the length of the post)

 

[Morbad]

Point #3 was particularly enlightening and was one of the key things I think we hadn't been able to figure out from experience.

#5 is also pretty interesting.

For example, if your ships armor is 100%, then a module should take no damage and your hull would take the full damage amount. If your armor is at 80%, then a module takes 20% of the damage and your hull takes the remaining 80% damage. Basically the more damage your ship takes, the more exposed modules become as your ships armor is peeled away.

This isn't how it works. This is not how it should work. This is not more plausible than the current system.

Armor is not impenetrable and making something behind it automatically immune to damage is absurd.

can we expect that, with the correct calculation, that hitting a particular subsystem actually takes effort to fire from the best angle? For example against an Anaconda, imho targetting the power plant shouldn't be enough, you should also have to align yourself so that you are facing perpendicular directly onto the hull closest to the power plant, and otherwise only do very insignificant damage to it at all. In other words, can we expect the volume of the ship itself, where it does not contain any particular module, to also effectively obstruct modules, especially internal ones?

It's worked this way in the past and will probably work this way again.

This is genius, please, FD, you should consider this!

The penetration roll is already based on hull strength. Making internal modules outright immune to damage until the hull has been damage is idiotic.

I have to admit, the RNG nature of this bothers me. Why not make it a fixed percentage of the damage making it through, with this percentage starting small (say, 10%) and then ramping up as your armor takes damage? It feels antithetical to the skill-based nature of the game to have the damage calculation have this random element.

I disagree.

The abstractness of random tables can result in additional plausibility without increasing the detail that needs to be modeled to absurd levels. Skill still applies.

They could have made missile racks and torpedo pylons only fit on large and bigger hard points. That would have given larger ships the power they deserve and keep it away from smaller ships that should not have room to fit large weapon systems in the first place.

They should have just left them the way they were so pilots would have learned to move rather than eat scores of missiles.

With this, you have the chance of shots just sailing through a more-or-less untarnished hull totally unimpeded. Additionally, you have the chance for a barely-together swiss cheese wreck of a hull to completely and absolutely block a shot from making any progress. On average things will work out, but on a shot-by-shot basis, things will occasionally look really silly.

And this will produce the most plausible end result until you can model the entire game universe at the molecular level.

If I'm in my pristine armored car, there is a small chance of a rifle round fired at it hitting the comparatively unarmored gun port. Likewise, even if I've just run over a massive IED and it's not a flaming wreck with the bulk of the armor spread over a 30m radius, that same incoming round could still hit a piece of framework and be deflected.

Not everything should work as expected every time. The very act of removing the possibility for the unusual and unpredictable reduces plausibility.

I mean, fitting a ship with tons of weak, cheap, modules with the express intent of them getting destroyed and acting like damage sinks is actually a reasonable thing to do.

Yes, and it's rather silly.

I'd like to see a module that absorbs sigificantly more damage that what's needed to destroy it simply stop being registered. It's fine for a disabled module to still absorb some damage, but sooner or later it's going to be so ruined as to not provide any appreciable cover.

 

[Chaodiurn]

Great! Thanks! I'd love to see more precise numbers on what affects what how. It it 20% more resistance against kinetic weapons by reactive armour on hull damage (taken from 20% more damage of thermics against shields), or both hull and module damage? If it affects the hardness versus damage types it's the second surely? Are the hardness values of 20 and 65 fixed for small / large ships? If so, are small / medium / large ships defined by their landing pads? Or has every ship its unique hardness? In any case, what are the other values? Etc.

 

[Kastor]

Thank you for this information. I do have a question regarding step 7, particularly this section.



My question is why the ratio is not determined by the armor? For example, if your ships armor is 100%, then a module should take no damage and your hull would take the full damage amount. If your armor is at 80%, then a module takes 20% of the damage and your hull takes the remaining 80% damage. Basically the more damage your ship takes, the more exposed modules become as your ships armor is peeled away.

Please forgive me if I misread what you said.

This definitely this. Also can we make sure hull reinforcements add on top of the 100%. So that while people are whittling down your reinforcements you take no module damage.

 

[n13L5]

Is there a source link for Mark Allen's original post?

 

[OgreWookie]

My impression from what I understand is that making improved armors harder to penetrate would solve many woes (such as making them worth their weight and cost)
and reduce PP sniping. Would it be possible for hull reinforcement packages to increase the armor value of internal modules ?

The way I read it hull reinforcement is saving your modules a lot of damage because the shot is highly reduced in effectiveness.
Quote from original post:

With that in mind, when you hit a shield the process is fairly simple - the damage is multiplied by the shields' defences and health subtracted. If damage spills over after bursting the shield it will be applied to the hull but otherwise stops there.

If you hit the hull then a much more complex process kicks off:

1) The first step is to multiply the damage dealt by the armours' defences as for shields.

So massive armour rating equals everything from that point on being at greatly reduced possible damage.

 

[cluelessforumuser65578]

am i right in thinking that changing the values in step 3 you could make it so you cant penetrate armour until whatever % hull the target has left regardless of weapon?

 

[Bunkerkind Anni]

Is there a source link for Mark Allen's original post?

this topic IS the original post... or the newsletter

The way I read it hull reinforcement is saving your modules a lot of damage because the shot is highly reduced in effectiveness.
Quote from original post:


So massive armour rating equals everything from that point on being at greatly reduced possible damage.

Armor defence = resists = what bulkheads grant
Hull Reinforce packages only add additional hitpoints that matter in later steps when it comes to the CHANCE to hit an internal.
this was explained in post #32
https://forums.frontier.co.uk/showthread.php?t=170205&p=2601924&viewfull=1#post2601924

 

[EUS]

I can't believe I've missed this!
Thanks for sharing, it's what us players have needed.

 

[Tripfy]

My question is why the ratio is not determined by the armor? For example, if your ships armor is 100%, then a module should take no damage and your hull would take the full damage amount. If your armor is at 80%, then a module takes 20% of the damage and your hull takes the remaining 80% damage. Basically the more damage your ship takes, the more exposed modules become as your ships armor is peeled away.

Great idea, in my opinion the game mechanics should work this way, it would support the idea of using aditional armor and bulkheads, give the armor some meaning

 

[Gran Solo]

Hi! Would it be possible to explain how the Retributor is supposed to work? I know this would only benefit a niche audience, but it would help a lot to see the design process behind how this weapon is meant to be used - I cannot see it and a lot of players pledged to Mahon are equally confused. I thought the weapon was bugged and filed a bug report, when I got the reply that it was working as intended, which begs the question what I'm doing wrong. Any dev feedback would be greatly appreciated.

 

[GluttonyFang]

Greetings Commanders,

Recently and over time, we've seen a fair amount of discussions regarding game mechanics, such as damage and defenses. Mark Allen joins us today to discuss such!

Now that is the communication I'm talking about, good work/transparency FD.

 

[Zenow]

I think that with modification of step 2 and 3, as well as a reversal of the order of the steps (doing step 7 first), it will become possible to get more specific module hits. The bonus of the chances I suggest would be that ships are less often completely destroyed, and more often just disabled, leaving options for enhanced gameplay.

note: It seems that step2 determines damage, step 3 whether or not it occurs. However, harness seems only to be taken into account for step 2? That would be extremely odd, I am probably misreading this.

damage/piercing: In step 2, it appears that while hardness does take into account the specifics of the hull versus the weapon, the simple ratio of the equation would mean that any weapon does at least some damage to the hull. That sounds unreasonable to me. If I shoot an arrow at a 10 inch steel plate, nothing will happen, no matter how often I shoot it. So in Elite, that would mean that if the difference between piercing and hardness values is large enough, there should be no damage at all. Maybe scorch marks, denting, but certainly no real damage. One way to do this, is take the difference between the two values into account, and applying a threshold. I'd also like it if there was an additional variable: thickness. Whether or not piercing occurs, could depend on hardness * thickness: that way, even with the best hull material, if it is thicker, it will be harder to penetrate. That could differentiate between smaller and larger ships, as the thickest hulls couldn't be fitted to the smallest ships - it would make them too heavy and leave no room for cargo, e.g. Also, the thickness helps reduce damage: the thicker the hull, the more at least projectile weapons will be slowed and damage reduced (hence the multiplication of harness * thickness)

In step 3, it appears that hull penetration is determined yet again, now based on the condition of the hull, and a weird percentage roll is used for that. But didn't we determine that in step 2 already? It seems to me step 2 is gives us both the damage and whether or not it occurs, no roll of the dice needed, this is not dungeons and dragons after all

Now, after damage and whether or not it occurs, damage is divided between modules and hull. This is where I'd like to suggest a second significant change. Right now, the hull is seen as one thing. However, that is not really realistic, is it? If I hit the hull at the point of module x, the hull at that point is damaged. If my next hit is on the other side of the ship, it should not encounter a damaged hull plate, but one at 100%. So I would suggest that FIRST we determine whether or not a shot is aimed at one of the blue orbs, and if so, hull damage to that module's hull quality is applied. That means giving each module its own hull quality; total hull quality is thus only reduced by a part, even if that part of the hull is completely destroyed. If the hull around your drive is completely destroyed, she ship is then not destroyed, but the engine is. That would automatically make module failure far less random. The way things are now, because the hull is one value, consistently aiming at hitting navigation, could result in the destruction of entirely different module if the next hit is accidentally in a different spot - that makes no sense. Giving hull quality values to each external module, makes one particular system vulnerable, not the entire integrity of the ship -as it should be. A secondary effect is that players will be far more likely to target specific subsystems instead of just generally shooting at the ship, as that will have only a random effect, even if the hull is breached.

Now if you want to make it even more interesting, you could give different parts of the hull (hull plates) different hull quality values as wel. That would mean consistently hitting the same hull part will result in quicker destruction of that part and open up a ship's inside. Of course, the more variables, the more work, so I can imagine the number of hull plates. Breaching the hull that way would of course result in life support problems (oxygen) depending on the ship, and that could mean a pilot could have to flee even before his entire ship was destroyed. Ship destruction, the exploding of the entire thing, should, imho, be reserved for very specific hits: if the fuel tank is hit and destroyed.

These changes should make it more likely for players to want to target modules, and for those exact modules to be destroyed that are being targeted. It would also make it harder to destroy a ship, while still being able to disable it. A side effect of this is that bounty hunting could become more differentiated: 'dead or alive' instead of just 'dead'. And if we re-introduce the escap pod (haven't seen that in-game) an escape pod could be needed to survive ship disabling; scooping up the escape pod would then enable arrests - or abductions. New missions, yeah!

 

[Pete Clements]

Greetings Commanders,

Recently and over time, we've seen a fair amount of discussions regarding game mechanics, such as damage and defenses. Mark Allen joins us today to discuss such!

Would so love to see a slow mo replay (camera which i can control) of my ship post combat, obviously no real time in game, to see what actually happened and how, any chance of some thing like that in the future?