Voronoi-Based Armour

Kodey

Veteran endo
Joined
Jun 13, 2020
Messages
193
#1
Currently, using plates smaller than the biggest size is a terrible choice for any ship, and is the equivalent of space butter. However, I feel that it'd be possible to mitigate this problem with Voronoi-based armour.

This a visualization of what I mean, one big plate with a single Armour value, Vs. one plate with a lot of tiny armour values, where the Av depletion is instead localized to where the shot was - instead of turning the entire plate to butter. Here's what the Av of each Voronoi could look like, keep in mind the actual value of the armour isn’t the point.
 
Last edited:

five

Master endo
Joined
Jun 15, 2020
Messages
293
#2
could u perhaps make a video with a damage test and demostration, to see if it works?
 

Kodey

Veteran endo
Joined
Jun 13, 2020
Messages
193
#3
could u perhaps make a video with a damage test and demostration, to see if it works?
The screenshot is just to show that the Voronoi fractures exist and they're easily visible on glass, I otherwise couldn't demonstrate it
 

Kenetor

Master endo
Joined
Aug 9, 2019
Messages
326
#4
ive heard this idea a few times and does indeed seem like a better way to mitigate the large plate meta, I just wonder about the overhead.It would also take a long time to get the balance right for new armour values that it would need across the board.
I'd love to hear the devs chime in or at least give us an idea on what they are thinking for an armour overhaul to, but I doubt they will
 

pavvvel

Veteran endo
Joined
Aug 31, 2021
Messages
222
#5
The problem has to be solved.
Making square ships is too dreary.

Another suggestion for improving the appearance of ships is to make glass with different armour levels
 

Kodey

Veteran endo
Joined
Jun 13, 2020
Messages
193
#6
I should probably post this here, but I have tried asking Ville about this! Though, I would like to try to ask Ville again, as this was a long time ago. I would try to, but I would suspect Ville has their DMs closed, same with Lauri and other devs.
1637043538222.png

1637043505484.png
 

VilleFB

Frozenbyte Developer
Frozenbyte
Joined
May 6, 2020
Messages
4
#7
I should probably post this here, but I have tried asking Ville about this! Though, I would like to try to ask Ville again, as this was a long time ago. I would try to, but I would suspect Ville has their DMs closed, same with Lauri and other devs. View attachment 3707
View attachment 3706
The armor topic is quite interesting and is in my mind almost day to day basis. Right now the design focus is elsewhere, but there are some ideas and solutions to current state of armor mechanics that are being considered. I don't go into that much detail here right now, but on top of my head, there are few things to consider when it comes to future of armor mechanics.

Firstly, how our plates work in game right now. They come in predetermined 12cm thickness across the board and only decorative plates and triangle plates offer other thicknesses (of course you can also layer plates in other angles, but that's not always possible). I'm not saying that it's impossible to introduce other thicknesses, but without also including features like allowing plate thickness to be adjusted in editor; lets say in increments of 1 centimeter. That would open door for thickness based armor system, but it also brings up other issues if we go fully realistic armor model. Mainly problems where weapon fire would be deflected after certain threshold of armor thickness is achieved. And with current limited amount of weapon options available, it would mean that rail cannons and plasma cannons would become only options to penetrate heavily armored ship cores and such complex armor system would require us to introduce more ship weapon variance to solve different armor thicknesses (and armor material variations). Not impossible future, but there's a lot of variables and fronts where change would be needed and not even starting with testing if such system would even be fun in practice.

Secondly, some closed alpha testers might remember the early day "floor mat meta" where tiny plates made out of Oninum and Charodium were the king of armor. This would probably be the end result of changing armor mechanics to "wrong" direction. Main problem being the effectiveness of small plates how they absorb voxel damage, especially weapons like rail cannons and plasma cannons which have quite wide fracture damage profile when penetrating. Smaller plates would make up much smaller hitbox for said weapons and causing smaller holes to ships. Current armor mechanics are indeed compromise between gameplay and optimization.

With current armor mechanics, small plates can actually provide much more better protection when heavier materials are used as armor penetration multiplier reduces the voxel penetration depth a lot. However obvious downside is that such plating will be heavy, but there are ways to armor up those parts of the ship that can't fit larger plates to them.

edit: To add to the suggested voronoi system, it's quite interesting take. Probably something that might work out as long as tech is built up to support it, if it's feasible.
 

Venombrew

Master endo
Joined
Aug 9, 2019
Messages
369
#8
@Kodey dude i like this idea, i think you got something here. quick thought how much of the alloy changes with furnaces will have an impact on an idea like this? do you think it would still be possible with different alloy mixtures on the final product after the release of furnaces?
With current armor mechanics, small plates can actually provide much more better protection when heavier materials are used as armor penetration multiplier reduces the voxel penetration depth a lot. However obvious downside is that such plating will be heavy, but there are ways to armor up those parts of the ship that can't fit larger plates to them.
@VilleFB so right now layering like scales on a dragon/serpent or a phalanx with smaller plates overlapping provides more defense?
 
Joined
Feb 28, 2020
Messages
17
#9
I wonder if it wouldn't be advantageous to remove the voxel system entirely and switch to just using the fracture system for performance/flexibility. Tweaking the size of the fractures would probably have an identical effect to what the voxels are doing.

Recapping so I understand the issue, the goal is that the size of plates shouldn't change projectile effects (damage/destruction). Hitting a 12x12 grid of 12x12x24 plates should function nearly identically as two stacked 144x144 plates creating the same damage pattern. If a hole is formed it should look more or less the same and/or happen at the same time.

I'm going from the premise that the current voronoi regions are precomputed per part and this design probably won't change. This could be an issue if the 12x12x24 don't fracture in half as they'd be completely destroyed forming a hole where the stacked 144x144 didn't.

Naively one would say "why not make 12x12x12 sized voxels? Then when a projectile hits it destroys the 12x12x12 voxels identical in both plating configurations. That would make the fracture system more complex as each part would have a ton of fracture parts. (It would just be the voxel problem, but with 12x12x12 voxels, and probably have the same performance issues).

I don't really see a system using fractures that is perfect, but that's probably not a deal breaker. This could probably be remedied by transferring some damage to adjacent plates. So if a projectile impacts a fracture region doing 10 damage it attempts to transfer some of that damage to adjacent regions - regions within a sphere). It queries for all the fracture regions adjacent (including behind) and say it finds 7 regions then the fracture region that was hit has a weight of 3. So with 10 damage it would take 3 and all the adjacent ones would take 1 damage. In the original setup two stacked 144x144 would have the top plate taking all the damage. In this change the top 144x144 plate still breaks first, but a hole forms at more or less the same rate as the 12x12 grid setup. Changing how the damage propagation functions by analyzing some examples should create a system where the plating sizes don't affect the outcome. (Not sure if this would handle the edge cases people have talked about, so I might be missing issues).

some closed alpha testers might remember the early day "floor mat meta" where tiny plates made out of Oninum and Charodium were the king of armor
Can someone explain the exact scenario that caused this? Was it that their armor value was too high for their volume? Was the penetration check limited so it couldn't damage multiple stacked ones? Trying to understand if what I described above would handle that.

Off-topic, but a question to the devs would be if removing the voxel system would allow more plates/beams to be added or is the inherent memory issue still there?
 

CalenLoki

Master endo
Joined
Aug 9, 2019
Messages
741
#10
wonder if it wouldn't be advantageous to remove the voxel system entirely and switch to just using the fracture system for performance/flexibility. Tweaking the size of the fractures would probably have an identical effect to what the voxels are doing.
I thought about the same. They look just as good, or maybe even better than voxels. At the very least it would allow devs to focus on one destruction system, instead of 2 (or 3, if we count invisible HP calculations)

Can someone explain the exact scenario that caused this? Was it that their armor value was too high for their volume? Was the penetration check limited so it couldn't damage multiple stacked ones? Trying to understand if what I described above would handle that.
Floor mat meta was actually based on thin Bastium plates, not charo/onin.
Back then the projectile damage was reduced by armour value before doing voxel damage. Armour value was a constant number for each material, completely unaffected by plate size or thickness (unlike now). Already back then armour value was much more important than voxel resistance.
So by using floor mats (3cm thick) you could stop AC/LC hit with just 3 layers (9cm), while standard plates also required 3 layers for that (36cm - 4 times heavier)

Oninum was only available in the form of turret armour.

Later thickness multiplier started affecting armour value, so the best strategy was using tiny vertical plates - they had too much thickness to be penetrated with a single hit (48cm) and were too small to hit the same spot twice (12x12cm each).
At this time plasma cannon got AoE splash damage which completely ignored armour value, chewing through oninum and ajatite at the same rate. That kept the vertical tiny plate meta at bay for some time, until it got splash got removed. For a short time well armoured ships were able to soak up so much that matches ended due to ammo shortages.
 
Top