(Outdated) Damage mechanics test 4 explosionScale (impactDamageScaler, explosionScale, muzzle / projectile velocity, diameter / caliber, collision, partCollisionResponse)
Latest version - mk3.2
This one has 8 cannons, and each cannon will automatically de-activate when the respective block is destroyed. Automatically calculates how many rounds were fired & average damage of the gun
https://www.simpleplanes.com/a/e7vKx5/Cannon-Damage-Tester-Mk3-2-average-damage-calcurator
SP cannon experimentation
This contraption is for testing cannon (weapon) damage mechanics in the game. Goal of this experiment is to find out how cannon damage, HP, and aircraft's connections work. (If anyone's done file digging and have all the answers, you may be able to save my time)
Relation between explosionScale/explosion damage, HP of the block hit, HP of the root block connected to the block hit, and when the connection between these blocks get destroyed.
Testing method
Shoot the blocks and see how the change in cannon's relevant XML attributes affects when/how the blocks are damaged upon impact.
Each block is a 2.5 meters big cube, has collision and 1000HP, partCollisionResponse set to None, and individually connected to, and ONLY to the sole cockpit. Cannon is aimed at the top left block (in the screenshot). Since blocks aren't connected to each other, only explosion or cannon shell physically large enough (in diameter) can affect the rest of the blocks.
As for cannon, I will use diameter of 100mm/muzzle velocity of 1000m/s/ExplosionScale of 1/impactDamageScaler of 1 as the base value (I will forget about this/not really relevant).
Do take note of each relevant attributes when recording the damage.
My findings so far
-Condition where "Critically Damaged" state occurs seems to be when the plane lost about 1/3 the starting number of parts.
-Cannon explosion does deal damage, even when impactDamageScaler is 0.
-Explosion can destroy/severe the connections between blocks, even when all the relevant blocks can survive the explosion & has partCollisionResponse set toNone
-Cannon's caliber affects the explosion size (seems to be a multiplier).
-Cannon's muzzle velocity (projectile velocity) affects impact (kinetic) damage it deals. Seems to be a multiplier for impactDamageScaler.
-It seems that Cannon shell with no explosion can only damage one block at a time, regardless of how big the shell physically is in diameter.
Example: https://www.simpleplanes.com/a/35UoqM/Damage-test-4-physically-large-shell-cannon-diameter
If you have suggestions or information on the topic, feel free to post that in the description (but do read the description first!)
Specifications
General Characteristics
- Created On iOS
- Wingspan 164.0ft (50.0m)
- Length 249.3ft (76.0m)
- Height 69.7ft (21.3m)
- Empty Weight 4,734,385lbs (2,147,483kg)
- Loaded Weight 4,734,385lbs (2,147,483kg)
Performance
- Wing Loading N/A
- Wing Area 0.0ft2 (0.0m2)
- Drag Points 0
Parts
- Number of Parts 28
- Control Surfaces 0
- Performance Cost 78
Even without the explosion, the projectile does damage to several blocks. This was found out when I was testing a made of glass ERA for a tank. Since the glass breaks from any damage regardless of health, it is even possible to estimate approximately the area of damage from such an explosion.
link
I'm interested (information is always a good thing (as long as it's not information about me, lol)), but i don't have much info.
HP of connected parts affects number of shoots needed to destroy "this" part, but i was testing it with miniguns (average of ten attempts to destroy a block) and it was very much random.
About "one block at a time": perhaps shell disappears when it touches something, and since it's spherical or sharp-nosed it may be hard to make it touch two blocks simultaneously, so it only touches the block in front of cannon while it's alive.
My concept about connections was "HP pools": similarly to your concept, i was trying to use few blocks with high HP and standard blocks connected to them and them only (see link above).
@Kendog84 The thing is... from what I've seen from gun tests, say we have a
cockpit - wing - wingchain or acockpit - block - blockchain, and the second the end of the chain gets destroyed/blows up (which seems to have its own AOE effect) it would sometimes detonate the surrounding parts until it reaches the cockpit. My personal creations usually have locked rotators (range of motion = 0, speed = floppy) to break up the chain.Also, I'll be pinning comments that I think has especially useful/new info, but it ruins the order of the comments and will make it hard to follow the discussion later, so I'll probably pick and compile them into a more easily-readable format later. Suggestions or help are welcome.
.
Speaking of which, is there a game mechanics equivalent of SimpleCheats or Snowflakes' Funky Trees Guide? I think it would be nice to have these useful info in a easily readable and accessible format. Right now, as far as I'm aware, these info are scattered all over the site, discord and such, and can be removed by accident or the whim of the creator of the post.
Built a new test gear (for non-explosive ammo).
https://www.simpleplanes.com/a/Epj6OF/Damage-test-gear-cannon
This one has 4 cannons, and they automatically stop firing when the respective block is destroyed (they are enabled when green cannons above each block is connected to the cockpit). You can re-enable them and count the rounds fired with AG1.
@11qazxc @ThomasRoderick
@ThomasRoderick
Thanks. So that part of the mechanic has nothing to do with explosionScale value, when it's Basic.
@11qazxc
That's a really cool observation, didn't think the value was rerolled with every map restart.
I can't do the testing myself right now, but I wanna see if the game applies the same amount of boost/nerf for 1) all cannons 2) all cannons of the same name 3) all cannons with the same Xml characteristics (so same muzzle velocity and impactDamageScaler), or it's different for each cannon every time.
We can test this by simply copy-pasting the cannon, place them so they each shoot at different block, and take note of how many shots each one requires to destroy the block. (Blocks should all have 1000 HP, and are individually connected to cockpit and only the cockpit--I'm not sure if the HP-pool thing you mentioned will have any effect on the result). Hopefully so we can get more data for average/range of damage a given combo of muzzle velocity + impactDamageScaler will do.
Guys, do we at least have weapon that deals same damage each time?
UPD: i downloaded this craft and changed explosion scale to zero and impact damage scale to 0.2 and tried to destroy that block 50 times.
12 times i needed 6 shoots to destroy it
20 times i needed 11 shoots to destroy it
17 times i needed 16 shoots to destroy it
and 1 time i needed 17 shoots to destroy it.
It's very strange results.
If damage is choosen randomly for each shot it will need something between (HP/min damage) and (HP/max damage) shoots to destroy a block. But it gives just three peaks (11 shoots plus-minus 45 percents it seems) as if damage is choosen one time at level start.
@Kendog84 See that "puff" of fire? That's the "explody" part. It somehow functions differently than when the shell is set to "explosive", but it does still have a small blast radius and can still unravel planes. That said, given the whole "unraveling" process is caused by a part receiving too much damage instead of any inherent process...
@ThomasRoderick
Kinda going back a few comments (harkening back to Astro's comment), but basic shells do explosion damage even when the explosionScale is set to 0? No obvious explosion visually and damage wise, certainly, but I want to be sure (will do test myself)
@Astro12 Just ran another test, turns out the best way to use cannons as aircraft weapon (aside from "just say no") is that... make it explosive and then nerf both the impactDamageScalar and the explosionScalar.
@ThomasRoderick
I've seen that too, and your hypothesis, theory, whatever's the right word here would make sense.
(It's been bugging me as it happens with my Sinkable Ship when I blow it up with explosives, scatters the light buoyancy blocks all over the place and kinda ruins the immersion/fun)
@Astro12
Yeah, this thread alone has shown me that there are so many mechanic to it that I wasn't aware of. I just gotta hope I can make a decent HP/survivability& damage system for different class of armor/ship types.
@Astro12 Yeah, that "crack" sound is an explosion as you can first, damage yourself with it; and second, make smaller by literally making the round explosive and reduce the explosionScalar. I have managed to (somehow) unravel a ship with a .50cal "shell" with both impactDamageScalar and explosionScalar at zero, and after that I just quitted using cannons altogether. My rule of thumb? If the ammo explodes and destroys a part then something's getting unraveled. But retarded bouncyballs are really unfun so I just switched to guns.
@Kendog84 What I have also seen is that sometime rockets/bombs/torps/missiles would not damage something it could clearly destroy if that thing was formerly separated from a cockpit. Say, My first bomb destroyed a target and the target's parts (checked to have 100 health) are scattered on the ground. When I drop my bombs again on the scattered parts sometimes literally nothing happened. Does the broken chain also prevent a part from being damaged by explosions?
@ThomasRoderick Base projectiles deal damage in the same way as explosive projectiles, but with a very low Exp.Scal. This is what I can tell you from experience. One time I shot at a pile of unconnected pieces of glass lying on the ground, breaking several at a time.
@Kendog84 I've noticed that, too. In general, the weapons in SP work very strange. You'd have to write a serious scientific paper to describe it all.
@Astro12 Yeah... from what I have gleamed from my (only) design with actual glass I can attest to that: my glass canopy was shot out; not a scratch on the frames it's connected to. Does connecting a part with multiple other parts help prevent the chain from affecting the block?
@Astro12
I do have some experience where a hit to a part of the tank messing up a rotator and such located elsewhere (tends to "lock" the tator in place, and makes it unable to move it--without destroying the connection)
Can't be sure 100%, but I think all the parts involved has partCollisionResponse of None
@ThomasRoderick
If this chain does work, it does so in an extremely strange way, because the glass is not connected to each other. In that case, the armor block would probably be destroyed too, but it is intact. In addition, the gun was damaged, and it has nothing in common with the ERA units except close proximity and a very long chain of parts, which goes through two rotators.
@Astro12 If one pane of glass breaks the "chain of unraveling" would still happen under some circumstances, so... I guess it could also be used to gauge if the radius of the "chain" have a finite radius or connections?
@11qazxc
Wow... this is quite interesting, never figured out (or even noticed) this mechanics.
As for cannon, I think your theory on shape might be correct. I'll see if it loses damage attribute once it hit a block.
@Kendog84 Probably yes if the AOE is large enough to cover multiple load (read: connection) paths, or perhaps the number of parts between the cockpit and the damaged one could also be a factor.
Still, I pretty much stopped using cannons after I managed to disintegrate a battleship using a .50cal HMG with a single "explosive" (in quotation marks as the explosionScalar is literally set to zero) round.
@ThomasRoderick
Huh... did this happen regardless of partCollisionResponse?--wait, scratch that. I know it happens even when PCR is None for all parts involved. Do you think the physical distance between these parts matter?
@ThomasRoderick
Thanks, lemme try these examples
detacher damage model
@Kendog84 Say, this for example, if your weapons hit the parts on a roatator, the rotator would snap but the remaining parts would be fine.