-
Notifications
You must be signed in to change notification settings - Fork 19
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Change: Adjusted Battle, Hellfire and Scout Drone armour and health. #1463
base: main
Are you sure you want to change the base?
Conversation
On the table for "number of shots to destroy a drone" you can also note the percentage of change. Quad can now shoot down drone 2.35 times quicker than before. Are we confident that such massive Nerf/Buff is okay for starters? Note also, you do not need to use |
could a new armor fix the weirdness like jet missiles? |
Elaborate. |
Stubbjax explained the issue already. |
Is this change a step one for a step two? If yes what will step two look like and why is it decoupled from this change? I am asking, because while I do agree that vehicle drone armor is odd to be as strong as a tank, I am not sold on the presented setup. They assume that Spy Drone is stealth. Vehicle drones are not. |
When you really consider just how imbalanced / broken the original values are, it is more than reasonable. The primary question I asked myself while contemplating the changes was whether a drone should take longer to kill than a King Raptor or Humvee as they do in 1.04, which is completely nonsensical and no longer the case with this change. Drones would ideally die twice as fast or even faster (and not have the ×2 hit points), but adjusting it just enough so that they die only slightly faster than the aforementioned units seems like a reasonable compromise and (if necessary) starting point. A drone will now die slightly faster than a King Raptor, as demonstrated in the shots to kill table below.
I was contemplating adding a new armour type as a second step because it would affect the Spy Drone as well. I was wary of including it in the same PR after what happened with #1412. A new armour type also has far greater design considerations.
Is that such an unreasonable assumption to make? Using Nevertheless, this was originally a test branch just to see how the standardised values would feel (which do feel much better). The more I thought about it, the more it made sense, and so opened a PR for visibility and discussion on the current setup. If you have a more reasonable suggestion and/or justification, then by all means. |
Thank you for detailed response. Before going more in detail, could you add a few more target units to the "shots to kill" table if you have the time? It would be useful to see the original setup for the following:
|
I think you very nicely proved two things Stubbjax:
Ofcourse it still needs testing, but I predict spamming them as pure cannon fodder isn't nearly as viable as before and it shouldn't be. Spy drones will still be viable in limited quantities for their intended purpose. One more thing, keep in mind the armor can still upgraded from the Strategy Center. |
Can you also show in first post how this change affects |
|
Drone Armor, Chain Guns and AP Bullets would cancel each other out then I expect. I found the above videos and numbers are a bit deceiving when observing the total amount of time and shots to kill a drone, because standard Quad Cannon has low fire rate, but can in practice increase its fire rate and or damage on scrap, veteran or other modifiers. And the China Gattling Cannon needs to spin up fully. In optimal circumstances, Quad Cannon and Gattling Tank will perform similarly.
Drones are not Airplanes. The way drones approach enemy targets is not comparable with a plane. Planes will approach at high velocity, unload missiles, and then turn around to reload at home. Naturally the exposure to enemy fire is minimal with such an attack. A drone does not suffer from a return-home-to-reload penalty, but it will typically not engage targets as swiftly as a plane. The drone hovers around a slower host vehicle, so it is bound to its limitations in some capacity. Consequently enemy forces can shoot on vehicle drones much longer than on air planes. Drones are not vehicles either. They are not trainable and will therefore get no perks. They cannot be shot at by tanks, but they can be shot at by Anti Air guns from much bigger distances than ground units. This means drones will, like air planes, draw fire when engaging ground targets. The drones come at different prices and serve different functions. It is very reasonable to say that a Scout Drone for 100 cost should not be able to tank a lot of damage from enemy forces. But a Hellfire Drone for 500 cost naturally must withstand much more enemy fire to be useful. If it dies too quickly in situations where it was originally expected to be useful, then it may become a risky investment with this patch and cause major frustration.
I do think Drone Armor setup needs more nuance to account for the different drone types. |
Hmm something does not add up with the numbers. 5 seconds to kill per drone would make it 15 seconds, but video shows 35 seconds. There is error somewhere in the numbers. |
I don't understand or agree with this sentiment at all. The data has been presented in the least deceiving way it can be. Default attributes - before any further upgrades, ranks, propaganda, battle plans, bonuses, laser targeting or salvage - should always serve as the base comparison. It is far easier to extrapolate circumstantial values and modifiers in a single direction. I also don't see why you've listed 2 as the Quad Cannon's 'optimal' fire rate when it fires every 4 frames like the Gattling Tank. The Quad Cannon values you've listed are much more misleading.
It's just as reasonable to say that a dedicated anti-air unit should be super effective against a small, relatively cheap, airborne utility drone. Nobody should be building Hellfire Drones against anti-air anyway - you'll typically see them relatively early in a game against Technicals, Troop Crawlers, sneaky Dragons and Chinook drops or to harass Dozers and Supply Trucks; situations where it makes sense and is a viable investment. It seems reasonable to say that this change does not really impact the utility of the Hellfire Drone. It's already an incredibly risky investment in 1.04 when any type of missile weapons are involved, which is somewhat lessened with this change in particular (5 infantry missiles vs 3). It may even technically be an overall improvement, as Hellfire Drones are often involved in taking out missile infantry, especially early on against a Chinook drop, Tunnel, or Tunnel Defenders dropped from a Technical. Anyway, if you think the armour should differ between the different drones, I'd love to see some numbers. |
I agree with the differentiation between drones xezon. The Scout Drone changes are a good starting point, but the Battle and Hellfire Drones should be a bit more durable in battle because they're designed for it. They shouldn't be more durable than actual battle units however. Taking down a swarm of Drones should't take longer than taking down a swarm of Comanches (with Counter measures). This doesn't make any sense. |
Should the Spy Drone match the Scout Drone? |
Ah. My own table had me confused. Base Quad has 1 shot per 4 frames so it will take 10.6 seconds to kill 1 drone. This is very long time, especially against Scout Drone. We agree on that. As for misleading numbers: It is wrong to assume that Quad will only ever take 10.6 seconds to kill a drone. There are far more states where Quad will take closer to 5.3 seconds than 10.6 seconds. That is, by gaining veterancy, driving over scrap, getting AP Bullets, getting modifiers such as Propaganda (Team Games).
To me that looks like it goes into the right direction. Again, Vehicle Drones are not vehicles, nor planes. It is not reasonable to treat them as such, because they are different units. One must check how they perform and see how strong they should be. This change highlights the abundant armor of the original drones and reduces strength very much. I take issue with that, because we make China stronger through many other avenues already, so USA will already have more trouble as is. Plus GLA is very good against USA in late game as is, so USA by no means has an easy time even with strong drones. Yes armor against missiles is increased a bit, but armor against bullets is decreased much more and the healing rate is decreased as well. I do actually like the redistribution of bullet armor to missile armor. The original missile armor is too little. |
Your typical infantry missile deals 40 A 200hp drone with A 200hp drone with So yeah, like I said at the bottom of the description, some unspecified damage types effectively deal half damage now. But we can rectify that by introducing a dedicated armour type. |
That was never an assumption. The assumption is that the base numbers / default state are enough to extrapolate to other circumstantial modifiers. Otherwise it would be tedious and time-consuming to list out every combination in every which way and only make things more complicated and harder to read.
So what do you suggest? |
I'm not sure what you're saying with this, are you saying their resistance vs anti-air weapons can't be compared to other air units or the host vehicles? |
It cannot be compared to both. It is not a tank, it is not a plane. It is a slave drone with very unique behavior. I agree with overall direction of this change, but would like to see less nerf. Nerf:
Buff:
With this change, Drones are overall worse. Reduced healing means all drones are likely to die quicker. Scout Drone dies significantly quicker against bullets. Battle Drone is no longer as good when upgraded. It has better resistance against bullets and missiles than Scout Drone. Hellfire Drone can tank still as much as before, but the armor capabilities have been redistributed: Bullet armor is halved, missile armor is doubled. This gives Hellfire drone more balanced armor attributes. It has better armor than Battle Drone. |
Using AirplaneArmor, the values would then be Scout Drone Health 160 That should come close to proposed setup. Correct me if wrong. |
I can't tell from the video as it's alternating between all three drones making it hard tell how fast they die, but my point is that AA units will have much less time to spend on killing drones from a distance, which could possibly mean you will spend money on drones and die before paying off, so we might see them get used less often if they become to easy to kill
We could start by testing your PR as well np, but I don't think it's optimal currently, at least maybe until after adjusting the rocket weapons |
…tank armour and airplane armour properties.
I've added and applied the dedicated |
Great! Ty And what about INFANTRY_MISSILE? |
That is correct. A 200hp drone with I added missiles to the table ( |
Ok but infantry missiles used to take 3 shots, and now is 5, Why do we need this changed? |
To stop this kinda nonsense! CHEAP.mp4But really, it was more of a coincidental change of matching the armour with the Spy Drone's |
Fair Enough, i think it's reasonable to give a little buff to drones vs rockets and a little nerf vs quads/gats, because both are bad extremes, as to jets, I'm not sure, but i think that a full jet volly should be able to kill a drone, makes no sense why it needs two, this goes to Raptors, KRaptors and MiGs
I'm aware of that, but it needed addressing as to why should it be that way |
How little are we talking here? Where would you place them in relation to other vehicles / aircraft? It's tough because giving drones higher durability than your average fighter jet makes absolutely no sense; but at the same time, any values in the same range or lower will likely have a much greater impact. Here's an alternative with 100% modifiers for |
I do agree with Xezon that they can't be compared to jets and vehicles, they aren't controllable and usually just stay under enemy fire for longer unlike aircrafts where they just hit and run, and vehicles don't get targetted with AA weapons so can't be compared either, that's why i believe nerfing them too much might not be the best solution here That said, i do wanna test that last table and how it fares ingame as I don't have a clue on how much worse they are now |
…mmon damage multipliers from 120% to 100%.
This makes no sense to me. They absolutely can be compared. Drones are airborne units just like any other aircraft. Why should their autonomy and battle proximity have any bearing on their damage resistance? This is not an intuitive relationship at all. Players have no reason to expect a small utility drone to absorb more anti-air rounds than a dedicated fighter jet or attack helicopter. The fact that players are well-accustomed to this flawed design is what makes this so tricky to solve. Feel free to propose a better solution!
I have pushed a commit with the adjusted values. You can check out the prior commit if you want to test the previous values. I'm always happy to test! |
They can be compared because they are faction units. But Drones are not Airplanes, nor Ground Vehicles. They are Drones. They have a different name, behaviour, utility, control, spawn method, size, damage, armor. Using an Aurora Bomber is vastly different from using a Scout Drone. They are different kind of units. Therefore why should their armor be the same if they are different things? |
You are misrepresenting my point. All units obviously have a different name, behaviour, utility, movement, size, damage, health, etc. Autonomy and battle proximity are not attributes that invalidate the comparison, nor should they play any major role in a unit's damage resistance. |
Would you say that a slow melee combat unit can do with just as much damage resistence as an agile long range combat unit? And would you say drones and airplanes are the same in terms of combat range and agility? |
It makes sense from gameplay prospective, if you're thinking about it logically, Realistically and by comparing to units with different roles then of course you're going to have troubles accepting it
The solution is to adjust the values to make them not frustrating to kill while not eliminating their viability by being too weak and not worth the investment. That is still undecided because we don't know which values are optimal yet.
Will do, although I'm not sure how to get the files from the older commits |
What sane army would place their bowmen on the front-line? A better question to ask: Why would a player expect a utility drone to have more hp / armour than a fighter jet / attack helicopter?
It really doesn't, though. It's completely counterintuitive.
Which has been the intent of this change. It feels far more reasonable, with no eliminated viability, from my own tests.
You can checkout individual commits. |
Because due its decreased agility and combat range it will naturally receive more damage than an airplane. You can compare this with Crusader Tank vs Tomahawk Tank. Crusader is at the front. Takes damage. Tomahawk stays back, can shoot far away. Crusader Tank has more armor than Tomahawk. Reversing or equalling the armors on both these tanks would not be justified, because their attack ranges differ. |
So your argument is that close-combat units should generally have more health? |
I predict that, when we're going to test Stubbjax's changes that even at those lower resistance levels, that mass HellfireVees will still totally dominate mass Quads/Gats, the difference however is that the Quads/Gats will atleast be able to trade a good number of drones for it and earn some XP. |
This logic is correct, but it doesn't prove Drones need a mega ton of armor just because of the combat range or lack of agility. The thing is that these Drones only have a very few counters in the field and the majority can't even target the Drones at all. The units that are left (Gats/Quads) can't even take them out efficiently either! This is outrageous broken design tbh. |
Gattlings, Quads, Rocket men, Tow-Missile Humvees are core faction forces. Currently it does take a lot of firing power, but is the nerf of -58% bullet armor justified? If yes, why by this much exactly? And why all drones equally as much? I do agree with reduction in armor, but not with how much as presented by the first iteration of this change. It is just too much for a first step. We have questioned other changes for much much less impact than this one. |
The answers to this question should exist in the original description. Future iterations and further discussion / contemplation has resulted in a more thorough understanding of just how broken the values of 1.04 are, and why the values make more sense. It feels very unreasonable to give drones more hp / armour than a King Raptor or Humvee. Either way, it serves as a good starting point for testing.
All drones were originally changed equally as much because all drones in 1.04 have the same hp / armour. Regardless, this is the latest iteration after your suggestion to distribute different health based on the unit cost / value. |
Percentage before/after is irrelevant really, it's whatever values are needed to make it more balanced. If that proves to be alot then it only shows how broken it was before. Stubbjax already brought this up; if you want to fight anti-air units, for what reason would you mass invest in airborne units? If you want an edge in tank vs tank combat, or vs light vehicles (vees vs techs) it makes total sense. But empty Hellvees shouldn't dominate Gats/Quads like they currently do. Anyways yes, I believe it makes sense that Hellfire Drones are more durable than Scout Drones in battle. I think the earlier proposed 160, 200, 240 values are a good start. If that's too little, lets crank it a bit. That's not what I predict however #1463 (comment). |
Here my feedback on setup. Green ok, red not so ok. Ok:
Not so ok:
|
So your layout of good vs bad indicates that it is impossible to align the values without each drone having its own armour type. Is this what you would prefer? There is no way to keep those health values while satisfying the other points otherwise. The acceptability of the Battle Drone's Quad shots vs Gattling shots can't be different because the damage modifier of Also, the Spy Drone is actually identical in size to the Scout Drone. The preferred height might give the illusion that they are different. |
I began splitting this change into smaller ones. This way we can hopefully reach good setup more effortlessly. |
I think proceeding incrementally will just complicate things further as each change won't be a direct comparison to 1.04 but also to all preceding changes. It really needs to be tested and compared / contemplated as a composite change. Fixing the Battle Drone armour upgrade inconsistency does make sense as a standalone change, but it has little to no bearing on the complexity of this change. Besides that, the entire contents of the healing change are already superseded by this PR and any future health changes, and will possibly be replaced with a |
It means we can focus at one thing at a time and lock in a particular change step. |
i use GitHub Desktop, don't know if that's possible there, will have to try also i've been been busy the past two weeks, i might not be available for testing as much for the time being, but will try and catch up with the latest changes |
Battle, Hellfire and Scout Drones provide USA with a huge advantage, as they distract enemy anti-air units and act as an additional decoy and damage sponge that significantly and disproportionately extends the longevity of USA vehicles. They are fairly cheap and build incredibly fast, which exacerbates the problem even further, and are easily and readily replaced. They are also incredibly frustrating to play against, as a player's anti-air units will prioritise the drones over vehicles. It is clear that drones have a far greater impact on the game than they have any right to, and are part of an incredibly flawed design that should really be addressed.
To lessen the severity of the respective issue(s), the following changes have been applied:
TankArmor
withAirplaneArmor
With the above changes, the three drones are now standardised with the Spy Drone, which already uses
AirplaneArmor
with 200 hit points. This provides a nice reference, as players are already familiar with the Spy Drone and its attributes. As much as these changes may make sense, it's important to note that drones are used incredibly often, and so a great deal of care and consideration must be taken when altering them. In an ideal world, these drones would have minimal armour and die in half the time they take to create, but that would have greater implications and is beyond the scope of this patch.The key differences between the two armour types are displayed in the table below.
The above
TankArmor
properties - particularly the low25% SMALL_ARMS
and10% GATTLING
modifiers - are the primary offenders here, and the reason drones are so hard for Quad Cannons, Gattling Tanks and Avengers to shoot down.In 1.04, a Quad Cannon deals 5 damage per round with its anti-air weapon.
TankArmor
's 25% modifier forSMALL_ARMS
damage reduces this down to a pitiful 1.25 damage per round against drones. With 100 hit points, this means a Quad Cannon must fire 80 rounds to take down a single drone. At one round every 100ms (~133ms) - roughly 7.5 rounds per second - a drone takes ~10.5 seconds to destroy. This is an outrageously long time, especially when a Humvee takes 48 rounds to destroy and a King Raptor takes just 40 rounds (half of a drone). Quite ludicrous!Gattling Tanks are a bit trickier to compare due to the barrel spin logic, so let's assume a fully spun up state. A Gattling Tank deals 12 damage per round with its anti-air weapon. Due to the odd fact that Gattling Tanks deal
SMALL_ARMS
damage with their anti-air weapons, this damage is also multiplied byTankArmor
's 25% modifier to 3 damage per round against drones. With 100 hit points, a drone takes 34 rounds to kill. At one round every 100ms (~133ms), a drone takes ~4.4s for a Gattling Tank to destroy. A King Raptor takes just 17 rounds to destroy, for reference.It is also important to note that an Ambulance's healing aura can provide drones with an additional 5 hit points per second, which is equivalent to 4 Quad Cannon Rounds and ~1.67 Gattling Tank rounds. This effectively doubles the time it takes for a Quad Cannon to destroy a drone to ~21.7s. (It takes the same time to destroy a Cold Fusion Reactor.) This bonus will be effectively halved after the change, as drones will take twice as long to heal up to 200 hit points.
With these crazy numbers in mind, let's compare the results of the change. The number of shots to destroy a drone are as follows:
The Quad Cannon and Gattling Tank's prospects seem much more reasonable against the
AirplaneArmor
, and will destroy a drone (200hp) slightly faster than a King Raptor (240hp) when both use the same armour.An unfortunate consequence of the
AirplaneArmor
change is that it (strangely) protects strongly againstJET_MISSILES
. This is an existing problem in 1.04, as a King Raptor cannot take down a simple Spy Drone in one volley (6 × 125 × 0.25 = 187.5). WhetherAirplaneArmor
's damage multiplier forJET_MISSILES
should be slightly increased to offset this anomaly is another question and ideally separate from this change as it would impact most aircraft.There is also the issue of some unspecified damage types effectively doing half damage after the change (due to the ×2 hit points), though other damage types are typically powerful enough and never really a problem in the original game. Regardless, a subsequent follow-up alteration would likely be to create a dedicated armour class for drones and apply additional damage for the respective damage types to keep them consistent with 1.04 or adjust hp where it makes sense.
1.04:
A dedicated anti-air unit struggles to take down three pitiful $100 drones.
DRONE_OLD.mp4
Patch:
The three drones are dispatched in a much more reasonable time frame.
DRONE_NEW.mp4
With the proposed changes, drones no longer stall anti-air units for ridiculously long periods of time, and their reduced durability aligns much more logically with their prices, build speed and function. Ultimately, USA will effectively have to pay more to maintain their drones.