Two important points about shields that you might not know:
In order to damage a shield, you need penetration equal to or higher than the shield’s resistance. This means only Reflective shields can stop beams with 24 pen, like the Federation Fusion beam (which, IMO, makes this particular weapon way too good - at least the Proton beam sucks aside from it’s boosted penetration). Beams with 24 penetration will go straight through High Recharge and basic Cruiser shields. This is less a problem and more a confusion issue - logic says you’d need to exceed the target’s resistance, not equal it.
Only your highest-resistance shield matters when calculating penetration. So by all means, load up on Multiphasic shield generators - it’s not like they’re actually costing you anything as long as you add a single Reflective shield generator! This makes no sense to me considering how damage is normally apportioned (being divided into four clumps that hit random shields). A strike from a weapon that can penetrate some of the target’s shields, but not all of them, should be able to damage the target’s low-resistance shields rather than being shrugged off completely by the single high-resistance generator.
I’ve actually suspected these for a while, but finally did some empirical testing today to be sure. The former is slightly unintuitive, and the latter takes away a lot of the potential complexity from shield selection.
This also seems as good a time as any to comment on how crappy the low-resistance shields are - or perhaps how strong beams are! I honestly can’t imagine building a Cruiser without a max-resistance shield generator right now, because any ship that’s vulnerable to beam lasers is going to get horribly mutilated by beams due to their range (the biggest weakness of Frigates is that they cannot resist beam lasers). The boost in shield strength and recharge for playing with lower resistance shields needs to be much, much higher for them to be even worth considering (beyond their cheesy current role of backing up a reflective shield).
Wow, is that really true? It doesn’t seem to match up with what Cliffski has mentioned in some of his other posts about shields, but perhaps I misunderstood him. In various posts he’s said that each hit gets split into four parts, which each pick one shield at random to take one quarter of the damage. I assumed that each of those four hits would test the weapon’s penetration against the resistance of the selected shield modules the damage was going to go to.
Was your testing based off of the results of a long battle between fusion beam ships vs multiple-shielded ships or was it based on watching the graphical effects from a few hits? It might be that the effect will show a beam being reflected even if some of the shield segments get damaged. I know that when you have part of your shield system knocked out you’ll see further incoming fire stop at the shield even when a substantial portion of the damage is going through the gaps to the armor (you won’t see a plasma torpedo actually hit the hull in this situation).
Nope. That’s how it should work, IMO, but it doesn’t work that way currently.
My test methodology was to challenge myself with a ship bearing only federation fusion beams for weapons. First test I deployed a ship with only fast recharge shields, and watched as it took damage and eventually died under fusion fire. Second test I deployed a ship with one reflective and one multiphasic shield - it took no shield damage at all when under constant fusion fire.
Is that the stat that those shield disruption missiles strike at? If so, no, I haven’t, and I probably won’t even if it varies between shield types - like any missile, point-defense seems the most proven way to handle disruption missiles.
This discussion raises some good points. My thinking behind picking the top resistance was by visualising the multiple shields as multiple layers, like a layer-cake of positronic energy fields. If a single layer of the cake (perhaps the chocolate cream) was strong enough to deflect the beam, then its ability to penetrate the other layers (such as the chocolate sponge) was moot.
However, you make a good point in terms of my other explanation about incoming fire scattering across non-overlapping shields. In this case, the shields are best visualised not as a chocolate layer cake, but more like a battenburg cake, with different shields covering different areas.
With this in mind, I shall change the code for the next version so that individual impacts are reflected (or not) based on which shield they hit (and its resistance).
And now I need cake.
why not do simple stuff like stacking with penalty:
4 shields with 24 resist with first shield module giving effectiveness of 100%, second 50%, third 25% and forth 12.5%:
1-(1-0.24)(1-0.240.5)(1-0.240.25)(1-0.240.125) ~ 0.39 resist
Hmmm… The way I see it, the problem is that it behaves like layer-cake for resistance, but like the battenburg cake for recharging / total shield “buffness”. Perhaps if making the code completely “battenburg” is too complex, making the code completely “layer cake” would be simpler. Then the shields would layer most reflective first. This would also have the advantage of reducing the shield stacking benefit (each shield gets a turn, but is the only shield during that time). Currently, if I stack 4 or 5 fast recharge shields, it takes some serious combined firepower to really have much impact.
On the other hand, the recharge shield stacking encourages the use of the disruptor bomb, so maybe this is one situation where the best change is simply a matter of updating the space manual.
Or you could make it so that each shield generator adds another shield circle around a ship so players could see that there really are, for example, 4 different physical shield circles around the ship come from those 4 separate shield generators and make it so that weakest shield would be uttermost and strongest one would be innermost. Then each time the shield health bar goes down a certain amount (in our case we have 4 shield generators so that would give 1 shield generator 25% room of the health bar) starting from the uttermost shield, the shields would start to go down. (Or just make it so that all shields go down at the same time when the shield health bar reaches zero.)
Why this kind of setup? That way players could again see that strong beam lasers could penetrate the weak uttermost shields but they get reflected away from the strong innermost shield. Then if some lasers are very weak, player could see that they bounce off already from the weakest uttermost shield.
(Maybe even give the shields small tone change based on their strenght. Strongest one would be dark blue and weakest one would be like light cyan or so.)
And if all shields were equally strong, they all would be same colored and if a laser is capable of penetrating the uttermost shield, it still would have to 3/4 chances that it gets blocked by the remaining shields. (If the game works like that atm, I don’t know that well yet.)
i think that the way i would do shield resistances would be to multiply the number of shield modules (x) by a shield stacking variable (s) to get (y) and then take the sum total of all resistance from all shield modules (t) and divide it by (y). t/y = your resistance. this would allow shield resistance to be the average output of all modules, but also it would have a stacking benefit that scaled with the inverse of (s). if s was 1, then you would just get the average of the modules, but if it was less then 1, you would get some benefit from additional shield modules that stacked.
i would calculate this value during loadout and display it in the loadout screen, so that people could work to fine tune their shield resistance values before combat, and then during combat i would only do 1 resistance test for every shot against this value multiplied by whatever math you use for shield stability.
i think that’s how i imagined shields being done before i read otherwise, just my 2 cents.
EDIT: i think my argument for this type of math stems from viewing shields as being 1 field supported by many projectors. resistance is how strongly the field can hold the energy in place and hitpoints is the amount of energy out there. if you think about it that way then adding more energy from multiphasic emitters would mean that the reflective emitters need to work harder to hold the greater amount of energy in place, hence less resistance. but more reflective emitters = more field strength at the cost of there not really being much energy (hp) out there. of course both regeneration and hardening of shields draws energy, and as such shields aren’t going to gain resistance as they absorb incoming fire and lose hitpoints.
I’d just like to point out that I tried shining a laser pointer at a chocolate cake, and very little of the beam was reflected. On the other hand, the reflection was quite a lot higher for a battenburg, so I guess we should probably use battenburgs for our ships’ shields. Or maybe christmas cakes? Their icing has even higher reflection.
For the record, my objection is not particularly a consistency one (though that’s also true). It’s a gameplay problem: the current system essentially removes almost all choice from shield module selection because every shield beyond the first is just plain better as Multiphasic. So I very much hope you’ll change this behavior at some point, even if it’s non-trivial to implement.
Hmm. I’d expect the top layer to matter most, and then the others. If you can’t get through the chocolate sprinkles, the chocolate cream’s going to go unscathed. And even if you do get through that, the damage to the chocolate cream will be lessened by the sprinkles, until all the sprinkles are gone; and so on all the way down to the plate.
This way, all layers have some effect, no matter how thin and weak. The problem is it requires some means of ordering the layers; either as a player driven interface, or an in-game mechanism.
You really need to have different health bars/stability for each shield. The strongest shield should be the first one to try to eat/reflect/fail a hit.
As shields are reduced in strength/stability, other shields will naturally take their place in front.
I feel that a persons skin is a much better analogy than a cake, because skin naturally recovers slowly over time. It also has layers that are more/less resistant to damage and damage types. You can hit somebody in the arm with a hammer, barely break the skin but completely shatter a bone (a ships armor!), or stab them in the same place with a knife - it will barely scratch the bone, but the skin (and tendons and arteries) are all sliced up.
I was thinking about a solution along the lines of Splad’s suggestion, but I like how Splad worked it out better. The idea that resistence gets “dilluted” among the total energy contributions of the multiple shield modules is intuitive and it would solve the game play issue so that each shield module selection would represent a real trade-off between resistance and strength.
Well, fast recharge v. multiphasic is still an open question since after damage is split up each shield recharges independently. All forum reading players are now going to use exactly one reflective shield and then load up on one of the other types. Is this a bad thing?
