Jump to content

MK 20 Rockeye (Is it done?)


Gentoo87

Recommended Posts

I went up to do some ground ordinance training, Mk82's 84's ZUNI Rockets did great, I scored a direct hit on 3 BTR's with  Rockeye. Watched from the F6 perspective watched them get surrounded in smoke but zero damage was registered, Put a MK 84 in the center of the cluster no damage, a few soldiers killed. Landed a direct hit with a MK 82 the BTR died. I don't know what to blame? Is a BTR tough enough to survive a bombs that close or a direct hit with a cluster? or is something wrong with the munitions?

Link to comment
Share on other sites

7 minutes ago, Gentoo87 said:

I went up to do some ground ordinance training, Mk82's 84's ZUNI Rockets did great, I scored a direct hit on 3 BTR's with  Rockeye. Watched from the F6 perspective watched them get surrounded in smoke but zero damage was registered, Put a MK 84 in the center of the cluster no damage, a few soldiers killed. Landed a direct hit with a MK 82 the BTR died. I don't know what to blame? Is a BTR tough enough to survive a bombs that close or a direct hit with a cluster? or is something wrong with the munitions?

 

It's a general DCS-ism; a combination of the damage model of units, and the fact that fragmentation isn't really modelled (it's usually approximated by increasing the explosive mass of munitions).

Modules I own: F-14A/B, Mi-24P, AV-8B N/A, AJS 37, F-5E-3, MiG-21bis, F-16CM, F/A-18C, Supercarrier, Mi-8MTV2, UH-1H, Mirage 2000C, FC3, MiG-15bis, Ka-50, A-10C (+ A-10C II), P-47D, P-51D, C-101, Yak-52, WWII Assets, CA, NS430, Hawk.

Terrains I own: South Atlantic, Syria, The Channel, SoH/PG, Marianas.

System:

GIGABYTE B650 AORUS ELITE AX, AMD Ryzen 5 7600, Corsair Vengeance DDR5-5200 32 GB, Western Digital Black SN850X 1 TB (DCS dedicated) & 2 TB NVMe SSDs, Corsair RM850X 850 W, NZXT H7 Flow, MSI G274CV.

Peripherals: VKB Gunfighter Mk.II w. MCG Pro, MFG Crosswind V3 Graphite, Logitech Extreme 3D Pro.

Link to comment
Share on other sites

39 minutes ago, Northstar98 said:

 

It's a general DCS-ism; a combination of the damage model of units, and the fact that fragmentation isn't really modelled (it's usually approximated by increasing the explosive mass of munitions).


Is this WIP or finished product/idea? Thank you for replying I appreciate your input.

Link to comment
Share on other sites

Actually, in this situation it's because DCS does not simulate all 247 HEAT submunitions that it should contain. The Rockeye doesn't rely on fragmentation IRL - rather, it contains so many bomblets that more than a few should score a direct hit. 

 

This is obviously unfeasible in DCS. Simulating a single bomb could, perhaps, be achieved without melting the CPU if it was optimized to hell and back, but when you consider that it's a 500lbs-class weapon, you'll find that something like the A-6 would be able to produce 6916 bomblets if it rippled them off all at once. 

 

Currently DCS handles this by reducing the number of bomblets, making the weapon practically useless. A statistical approach could work better - calculate the overall area covered by the drop pattern, pick 247 random points in that area, and if any of those coincides with a vehicle, spawn a bomblet over that point. However, I don't know if that's technically feasible.

  • Like 8
Link to comment
Share on other sites

2 minutes ago, Dragon1-1 said:

Actually, in this situation it's because DCS does not simulate all 247 HEAT submunitions that it should contain. The Rockeye doesn't rely on fragmentation IRL - rather, it contains so many bomblets that more than a few should score a direct hit. 

 

This is obviously unfeasible in DCS. Simulating a single bomb could, perhaps, be achieved without melting the CPU if it was optimized to hell and back, but when you consider that it's a 500lbs-class weapon, you'll find that something like the A-6 would be able to produce 6916 bomblets if it rippled them off all at once. 

 

Currently DCS handles this by reducing the number of bomblets, making the weapon practically useless. A statistical approach could work better - calculate the overall area covered by the drop pattern, pick 247 random points in that area, and if any of those coincides with a vehicle, spawn a bomblet over that point. However, I don't know if that's technically feasible.


So if I'm reading you right drop more than one in a target area. Regardless of target type.


Edited by Gentoo87
Added detail
Link to comment
Share on other sites

  • 2 weeks later...

I can't say I've had that experience. Using a ten degree CCIP pass from medium/low altitude, ideally dropping at least two bombs usually nets very favorable results against light and unarmored targets, and will even take out medium and heavy armor with a bit of luck. 

 

image.png

Link to comment
Share on other sites

It appears to be rather sensitive to release parameters. It does work, just not nearly as reliably as it probably should. Best used in pairs or in ripples of four. As far as the Hornet goes, the "one pass and haul ass" rule will apply to most missions, anyway, and if you pickle off all 8 along a column, it should work OK.

  • Thanks 1
Link to comment
Share on other sites

You should dive more than 10 degrees.  The successful hits I've found are around 25 degrees at least with a speed of about 400-500 knots.  If you stay within that profile, your chances are good.  I remember seeing this in a video Wags made that mentions this profile.

 

As to being done?  I made a post about how the sound is very weak and not even sounds like the real thing.  More like a 'swish' than a 'boom' or 'multiple booms'.  I think either Nineline or Wags mentioned that they were looking into improving the explosion sound.  However, that was said a few years ago.  That may have been prioritized way down the list or just forgotten.

 

View at 2:55 for the profile.

 

 

Edit: Just noticed this was in regards to F-14.  It's still the same release parameters.  I've have Harrier, Hornet, and Tomcat.  They all release the same way and require the same profile.


Edited by Eagle7907
Last note on bottom
  • Like 1

Win 10, AMD FX9590/water cooled, 32GB RAM, 250GB SSD system, 1TB SSD (DCS installed), 2TB HD, Warthog HOTAS, MFG rudders, Track IR 5, LG Ultrawide, Logitech Speakers w/sub, Fans, Case, cell phone, wallet, keys.....printer

Link to comment
Share on other sites

On 2/10/2021 at 9:18 AM, Dragon1-1 said:

Actually, in this situation it's because DCS does not simulate all 247 HEAT submunitions that it should contain. The Rockeye doesn't rely on fragmentation IRL - rather, it contains so many bomblets that more than a few should score a direct hit. 

 

This is obviously unfeasible in DCS. Simulating a single bomb could, perhaps, be achieved without melting the CPU if it was optimized to hell and back, but when you consider that it's a 500lbs-class weapon, you'll find that something like the A-6 would be able to produce 6916 bomblets if it rippled them off all at once. 

 

Currently DCS handles this by reducing the number of bomblets, making the weapon practically useless. A statistical approach could work better - calculate the overall area covered by the drop pattern, pick 247 random points in that area, and if any of those coincides with a vehicle, spawn a bomblet over that point. However, I don't know if that's technically feasible.

This is a straight up fabrication, and you can check this yourself by looking at the weapon definition in the lua:

 

declare_bomb("ROCKEYE", _("ROCKEYE"), "rockeye", wsType_Bomb_Cluster, "bomb-cassette-2", {
fm =
{
    mass            = 222.000000,
    caliber         = 0.335000,
    cx_coeff        = {1.000000, 0.390000, 0.600000, 0.168000, 1.310000},
    L               = 2.340000,
    I               = 101.298600,
    Ma              = 0.197848,
    Mw              = 1.987409,
    wind_time       = 1000.000000,
    wind_sigma      = 100.000000,
},  

launcher = 
{
    cluster = cluster_desc("Bomb_Other", wsType_Bomb_Cluster, combine_cluster(MK118_DATA,
    {
        cluster = {
            count        = 247,
            effect_count = 20,

            wind_sigma  = 50,
            impulse_sigma = 2,
            moment_sigma = 0.0001,
        }
    }, 
    "cluster"
    )
    )
},

control = 
{
    default_delay       = 1.2,
    default_open_height = 457,
},

 

The rockeye is modeled as having 247 heat projectile submunitions.   The problem is the damage of the submunitions is very low, meaning they usually don't score enough hits (and thus enough damage) to actually kill the target they hit.

 

The mk 118 submunition weapon data:

 

warheads["MK118"] = -- Mk-20
{
mass = 0.59,
expl_mass = 0.25, 
other_factors = { 1.0, 1.0, 1.0 },
concrete_factors = { 1.0, 1.0, 1.0 },
concrete_obj_factor = 0.0,
obj_factors = { 1.0, 1.0 },
cumulative_factor= 10.0,
cumulative_thickness = 0.25
};

 

I'll admit I'm not 100% certain of how these values work, since I haven't done extensive testing, but I believe the low explosive mass of the submunitions is the cause of the problem.

They appear to have good armor pen damage bonus (cumulative_factor of 10) but the thickness of only 0.25 I believe means they do very little damage on anything with moderate armor.

 

To complicate things further, the way armor is modeled in DCS is some really confusing code that appears to compare the incident angle of the projectile to the center of the object being hit, then applying an armor value based on that angle.  Here's the relevant generic schemes from the code for example:

 

-- armour scheme
unarmed_hull_elevation = { {-90, 90, 1 }, }
unarmed_hull_azimuth = { {0, 180, 1 }, }
unarmed_turret_elevation = { {-90, 90, 1 }, }
unarmed_turret_azimuth = { {0, 180, 1 }, }
unarmed_armour_scheme = {
										hull_elevation = unarmed_hull_elevation,
										hull_azimuth = unarmed_hull_azimuth,
										turret_elevation = unarmed_turret_elevation,
										turret_azimuth = unarmed_turret_azimuth
									};

tank_hull_elevation = { {-90, -45, 0.1}, {-45,11,1}, {11,19,2.9}, {19,40,1}, {40,90,0.15}, }
tank_hull_azimuth = { {0,10,2.9}, {10,30,1}, {30,150,0.67}, {150,180,0.20}, }
tank_turret_elevation = { {-90,18,2.9}, {18,90,1}, }
tank_turret_azimuth = { {0,10,2.9}, {10,30,1}, {30,150,0.67}, {150,180,0.25}, }
tank_armour_scheme = {
									hull_elevation = tank_hull_elevation,
									hull_azimuth = tank_hull_azimuth,
									turret_elevation = tank_turret_elevation,
									turret_azimuth = tank_turret_azimuth
								};

IFV_hull_elevation = { {-90, 30, 1 }, { 30, 90, 0.6 }, }
IFV_hull_azimuth = { {0, 30, 1 }, { 30, 150, 0.6 }, { 150,180, 0.5 }, }
IFV_turret_elevation = { {-90,18, 1 }, { 18,90, 0.5 }, }
IFV_turret_azimuth = { {0,180, 1 }, }
IFV_armour_scheme = {
									hull_elevation = IFV_hull_elevation,
									hull_azimuth = IFV_hull_azimuth,
									turret_elevation = IFV_turret_elevation,
									turret_azimuth = IFV_turret_azimuth
								};

T55_hull_elevation = { {-90, -45, 0.2 }, {-45, 30, 0.8 }, { 30, 90, 0.4 }, }
T55_hull_azimuth = { {0, 30, 1.2 }, { 30, 150, 1 }, { 150,180, 0.56 }, }
T55_turret_elevation = { {-90,23, 1.0}, { 23,90, 0.2 }, }
T55_turret_azimuth = { {0,30,2.0}, {30,150,1.6}, {150,180,0.65}, }
T55_armour_scheme = {
									hull_elevation = T55_hull_elevation,
									hull_azimuth = T55_hull_azimuth,
									turret_elevation = T55_turret_elevation,
									turret_azimuth = T55_turret_azimuth
								};

So from reading this, my take would be that anything using an armor scheme (if the unit doesn't have an armor scheme defined, it uses a flat "armour_thickness" Value), should be penetrated by the mk-118 since they should only be hitting from above.

 

For instance, a HMMWV has thickness 0.005, whereas BMP1 is 0.2, leopard is 0.125, T-72 has 0.1

BUT each one has a different HP value as well:  HMMWV: 1.5, BMP: 4, Leopard: 32, T-72: 25

 

Now, I don't know exactly how explosive mass and HP correlate, but I do know i've seen T-72's shrug off a MK20, while trucks and such get deleted by it.
I think we can reasonably assume the explosive mass is close to a 1:1 correlation with HP damage (assuming a penetration), so it might require upwards of 100 submunition hits to kill a T-72 for instance, or 320+ for the Leopard (Granted, you can start a vehicle burning without totally removing all the HP and it will die, but that's another discussion)

Given the dispersion pattern of the MK-118's, probably about ~60% of them don't end up hitting a target, which is why I think we don't usually see them killing vehicles (especially higher HP/armor ones) well.

 

 

tl;dr:  MK-20 is accurately modeled down to the submunition, unit HP/armor causes issues, since the MK-118s don't do enough damage.

 

  • Like 2
  • Thanks 1
Link to comment
Share on other sites

That's certainly interesting. It doesn't look like it does, but I suppose effect_count parameter is responsible for that. I don't generally dig through the files, and I've already seen DCS cheat somewhat with much less intensive CBU-97 (although to be fair, that's also based on the visual model).

 

It seems that the damage model is to blame, then. Realistically, it should be like being showered with anti-tank grenades. As it stands, actual tanks appear to shrug it off completely, while BMPs usually die.

  • Like 1
Link to comment
Share on other sites

1 hour ago, Dragon1-1 said:

That's certainly interesting. It doesn't look like it does, but I suppose effect_count parameter is responsible for that. I don't generally dig through the files, and I've already seen DCS cheat somewhat with much less intensive CBU-97 (although to be fair, that's also based on the visual model).

 

It seems that the damage model is to blame, then. Realistically, it should be like being showered with anti-tank grenades. As it stands, actual tanks appear to shrug it off completely, while BMPs usually die.

 

Might be due to the fact that HEAT effects aren't actually modelled at all. From what I can test, HE, HEAT, HESH, HEF are all approximated as just HE. There's also the DM of ground vehicles, which seldom go into much differentiation of the armour - as far as I can tell DCS doesn't model the difference between protection equivalents for kinetic and chemical munitions.

 

Spoiler

Here's the damage model for the T-72B3 obr. 2016 - it's split into a lot of zones (but misses some), the armour width values look like they could be RHAe for kinetic projectiles.

 

No differentiation between chemical and kinetic equivalent protection though, ERA certainly isn't modelled. Nothing internal and the armour and projectiles aren't volumetric.

 

I'm pretty sure all these do is act as a modifier for the HP system, so a hit in a thinner area would remove more HP.

 



GT.DM = {
	{ area_name = "TURRET_FRONT", 		armour = {width=0.400}},
	{ area_name = "TURRET_L",			armour = {width=0.250}},
	{ area_name = "TURRET_R",			armour = {width=0.250}},
	{ area_name = "TURRET_BACK",		armour = {width=0.100}},
	{ area_name = "TURRET_TOP", 		armour = {width=0.080}},
	{ area_name = "COMMANDER", 			armour = {width=0.040}},
	{ area_name = "COMMANDER_SCUTE",	armour = {width=0.800}}, -- nearly invulnerable to prevent penetration speculations
	{ area_name = "HULL_FRONT_01", 		armour = {width=0.400}},
	{ area_name = "HULL_FRONT_02", 		armour = {width=0.300}},
	{ area_name = "HULL_TOP",			armour = {width=0.080}},
	{ area_name = "HULL_BOTTOM",		armour = {width=0.080}},
	{ area_name = "HULL_BACK",			armour = {width=0.080}},
	{ area_name = "ENGINE", 			armour = {width=0.040}},
	{ area_name = "BULWARK_R",			armour = {width=0.200}},
	{ area_name = "BULWARK_L",			armour = {width=0.200}},
	{ area_name = "CHASSIS_R",			armour = {width=0.160}},
	{ area_name = "CHASSIS_L",			armour = {width=0.160}},
	{ area_name = "GUN",		 		armour = {width=0.500}},
}

 

 

The thickness of 0.25 for the Mk118 sounds interesting, it seems that the Mk118 should have ~200mm RHAe penetration, which should slice through engine decks and turrets that don't feature additional protection (such as ERA). Though not sure how DCS models penetration, if at all.


Edited by Northstar98

Modules I own: F-14A/B, Mi-24P, AV-8B N/A, AJS 37, F-5E-3, MiG-21bis, F-16CM, F/A-18C, Supercarrier, Mi-8MTV2, UH-1H, Mirage 2000C, FC3, MiG-15bis, Ka-50, A-10C (+ A-10C II), P-47D, P-51D, C-101, Yak-52, WWII Assets, CA, NS430, Hawk.

Terrains I own: South Atlantic, Syria, The Channel, SoH/PG, Marianas.

System:

GIGABYTE B650 AORUS ELITE AX, AMD Ryzen 5 7600, Corsair Vengeance DDR5-5200 32 GB, Western Digital Black SN850X 1 TB (DCS dedicated) & 2 TB NVMe SSDs, Corsair RM850X 850 W, NZXT H7 Flow, MSI G274CV.

Peripherals: VKB Gunfighter Mk.II w. MCG Pro, MFG Crosswind V3 Graphite, Logitech Extreme 3D Pro.

Link to comment
Share on other sites

5 hours ago, Northstar98 said:

 

Might be due to the fact that HEAT effects aren't actually modelled at all. From what I can test, HE, HEAT, HESH, HEF are all approximated as just HE. There's also the DM of ground vehicles, which seldom go into much differentiation of the armour - as far as I can tell DCS doesn't model the difference between protection equivalents for kinetic and chemical munitions.

 

  Reveal hidden contents

Here's the damage model for the T-72B3 obr. 2016 - it's split into a lot of zones (but misses some), the armour width values look like they could be RHAe for kinetic projectiles.

 

No differentiation between chemical and kinetic equivalent protection though, ERA certainly isn't modelled. Nothing internal and the armour and projectiles aren't volumetric.

 

I'm pretty sure all these do is act as a modifier for the HP system, so a hit in a thinner area would remove more HP.

 




GT.DM = {
	{ area_name = "TURRET_FRONT", 		armour = {width=0.400}},
	{ area_name = "TURRET_L",			armour = {width=0.250}},
	{ area_name = "TURRET_R",			armour = {width=0.250}},
	{ area_name = "TURRET_BACK",		armour = {width=0.100}},
	{ area_name = "TURRET_TOP", 		armour = {width=0.080}},
	{ area_name = "COMMANDER", 			armour = {width=0.040}},
	{ area_name = "COMMANDER_SCUTE",	armour = {width=0.800}}, -- nearly invulnerable to prevent penetration speculations
	{ area_name = "HULL_FRONT_01", 		armour = {width=0.400}},
	{ area_name = "HULL_FRONT_02", 		armour = {width=0.300}},
	{ area_name = "HULL_TOP",			armour = {width=0.080}},
	{ area_name = "HULL_BOTTOM",		armour = {width=0.080}},
	{ area_name = "HULL_BACK",			armour = {width=0.080}},
	{ area_name = "ENGINE", 			armour = {width=0.040}},
	{ area_name = "BULWARK_R",			armour = {width=0.200}},
	{ area_name = "BULWARK_L",			armour = {width=0.200}},
	{ area_name = "CHASSIS_R",			armour = {width=0.160}},
	{ area_name = "CHASSIS_L",			armour = {width=0.160}},
	{ area_name = "GUN",		 		armour = {width=0.500}},
}

 

 

The thickness of 0.25 for the Mk118 sounds interesting, it seems that the Mk118 should have ~200mm RHAe penetration, which should slice through engine decks and turrets that don't feature additional protection (such as ERA). Though not sure how DCS models penetration, if at all.

 

So, DCS does model different kinds of warheads.

 

Looking at the warheads.lua you can see the different types of "generic" warheads that exist, but are all based around the same definition:

 

function simple_aa_warhead(power, caliber) -- By Saint
    local res = {};

	res.caliber = caliber
	res.mass = power; --old explosion damage effect
    res.expl_mass = power;
    res.other_factors = {1, 1, 1};
    res.obj_factors = {1, 1};
    res.concrete_factors = {1, 1, 1};
    res.cumulative_factor = 0;
    res.concrete_obj_factor = 0.0;
    res.cumulative_thickness = 0.0;
    
	calcPiercingMass(res)
    return res;
end

function enhanced_a2a_warhead(power, caliber) -- By Yoda
    local res = {};

	res.caliber = caliber
    res.expl_mass = 1.7*power;
	res.mass = res.expl_mass;
    res.other_factors = {1, 1, 1};
    res.obj_factors = {1, 1};
    res.concrete_factors = {1, 1, 1};
    res.cumulative_factor = 0;
    res.concrete_obj_factor = 0.0;
    res.cumulative_thickness = 0.0;
    
	calcPiercingMass(res)
    return res;
end

function directional_a2a_warhead(power) -- By Yoda
    local res = {};

    res.expl_mass = 3.5*power;
	res.mass = res.expl_mass;
    res.other_factors = {1, 1, 1};
    res.obj_factors = {1, 1};
    res.concrete_factors = {1, 1, 1};
    res.cumulative_factor = 0.0;
    res.concrete_obj_factor = 0.0;
    res.cumulative_thickness = 0.0;
    
	calcPiercingMass(res)
    return res;
end

function simple_warhead(power, caliber)
    local res = {};

	res.caliber = caliber
    res.expl_mass = power*explosivePercent; --new explosion damage effect (explosive + fragments)
	res.mass = res.expl_mass;
    res.other_factors = {1, 1, 1};
    res.obj_factors = {1, 1};
    res.concrete_factors = {1, 1, 1};
    res.cumulative_factor = 0;
    res.concrete_obj_factor = 0.0;
    res.cumulative_thickness = 0.0;
    
	calcPiercingMass(res)
    return res;
end


function cumulative_warhead(power, caliber)
    local res = {};              
    
	res.caliber = caliber;
    res.expl_mass = power*explosivePercent;
	res.mass = res.expl_mass;
    res.other_factors = {1, 1, 1};
    res.obj_factors = {1, 1};
    res.concrete_factors = {1, 1, 1};
    res.cumulative_factor = 3.0;
    res.concrete_obj_factor = 0.0;
    res.cumulative_thickness = 0.2;
    
	calcPiercingMass(res)
    return res;
end

function penetrating_warhead(power, caliber)
	local res = {};
	
	res.caliber = caliber;
	res.expl_mass = power*explosivePercent;
	res.mass = res.expl_mass;
	res.other_factors = {1, 1, 1};
	res.obj_factors = {1, 1};
	res.concrete_factors = {5, 1, 10};
	res.cumulative_factor = 0.0;
	res.concrete_obj_factor = 5.0;
	res.cumulative_thickness = 0.0;
	
	calcPiercingMass(res)
	return res;
end

function antiship_penetrating_warhead(power, caliber)
    local res = {};
    
	res.caliber = caliber;
    res.expl_mass = power*explosivePercent;
	res.mass = res.expl_mass;
    res.other_factors = {1, 1, 1};
    res.obj_factors = {2, 1};
    res.concrete_factors = {1, 1, 1};
    res.cumulative_factor = 2.0;
    res.concrete_obj_factor = 2.0;
    res.cumulative_thickness = 0.0;
    
    calcPiercingMass(res)
	return res;
end

function HE_penetrating_warhead(power,caliber)
	local res = {};
	
	res.caliber = caliber;
    res.expl_mass = power;
	res.mass = res.expl_mass;
    res.other_factors = { 0.5, 0.5, 0.5 };
    res.obj_factors = {1, 1};
    res.concrete_factors = {1, 1, 1};
    res.concrete_obj_factor = 2.0;
	res.cumulative_factor = 0.0;    
    res.cumulative_thickness = 0.0;
    
    calcPiercingMass(res)
	return res;
end

 

 

From this, it appears that the MK-118 definition does make it a HEAT warhead, given the cumulative thickness number.

Looking at other HEAT warheads, they appear to also have an above 0.0 cumulative_thickness value, so it appears that's the armor pen value.

Honestly, people don't give the devs enough credit sometimes, because things like the CBU-97 are modeled with insane detail (there are probably ~200 lines of code that define the aerodynamics and the model animations for the dispenser casing pieces, skeets, and parachute submunition dispensers, and that's not even considering the definitions of the explosives and the seeker heads on the skeets)

  • Like 1
Link to comment
Share on other sites

12 hours ago, Hextopia said:

This is a straight up fabrication, and you can check this yourself by looking at the weapon definition in the lua:

 


declare_bomb("ROCKEYE", _("ROCKEYE"), "rockeye", wsType_Bomb_Cluster, "bomb-cassette-2", {
fm =
{
    mass            = 222.000000,
    caliber         = 0.335000,
    cx_coeff        = {1.000000, 0.390000, 0.600000, 0.168000, 1.310000},
    L               = 2.340000,
    I               = 101.298600,
    Ma              = 0.197848,
    Mw              = 1.987409,
    wind_time       = 1000.000000,
    wind_sigma      = 100.000000,
},  

launcher = 
{
    cluster = cluster_desc("Bomb_Other", wsType_Bomb_Cluster, combine_cluster(MK118_DATA,
    {
        cluster = {
            count        = 247,
            effect_count = 20,

            wind_sigma  = 50,
            impulse_sigma = 2,
            moment_sigma = 0.0001,
        }
    }, 
    "cluster"
    )
    )
},

control = 
{
    default_delay       = 1.2,
    default_open_height = 457,
},

 

The rockeye is modeled as having 247 heat projectile submunitions.   The problem is the damage of the submunitions is very low, meaning they usually don't score enough hits (and thus enough damage) to actually kill the target they hit.

 

The mk 118 submunition weapon data:

 


warheads["MK118"] = -- Mk-20
{
mass = 0.59,
expl_mass = 0.25, 
other_factors = { 1.0, 1.0, 1.0 },
concrete_factors = { 1.0, 1.0, 1.0 },
concrete_obj_factor = 0.0,
obj_factors = { 1.0, 1.0 },
cumulative_factor= 10.0,
cumulative_thickness = 0.25
};

 

I'll admit I'm not 100% certain of how these values work, since I haven't done extensive testing, but I believe the low explosive mass of the submunitions is the cause of the problem.

They appear to have good armor pen damage bonus (cumulative_factor of 10) but the thickness of only 0.25 I believe means they do very little damage on anything with moderate armor.

 

To complicate things further, the way armor is modeled in DCS is some really confusing code that appears to compare the incident angle of the projectile to the center of the object being hit, then applying an armor value based on that angle.  Here's the relevant generic schemes from the code for example:

 


-- armour scheme
unarmed_hull_elevation = { {-90, 90, 1 }, }
unarmed_hull_azimuth = { {0, 180, 1 }, }
unarmed_turret_elevation = { {-90, 90, 1 }, }
unarmed_turret_azimuth = { {0, 180, 1 }, }
unarmed_armour_scheme = {
										hull_elevation = unarmed_hull_elevation,
										hull_azimuth = unarmed_hull_azimuth,
										turret_elevation = unarmed_turret_elevation,
										turret_azimuth = unarmed_turret_azimuth
									};

tank_hull_elevation = { {-90, -45, 0.1}, {-45,11,1}, {11,19,2.9}, {19,40,1}, {40,90,0.15}, }
tank_hull_azimuth = { {0,10,2.9}, {10,30,1}, {30,150,0.67}, {150,180,0.20}, }
tank_turret_elevation = { {-90,18,2.9}, {18,90,1}, }
tank_turret_azimuth = { {0,10,2.9}, {10,30,1}, {30,150,0.67}, {150,180,0.25}, }
tank_armour_scheme = {
									hull_elevation = tank_hull_elevation,
									hull_azimuth = tank_hull_azimuth,
									turret_elevation = tank_turret_elevation,
									turret_azimuth = tank_turret_azimuth
								};

IFV_hull_elevation = { {-90, 30, 1 }, { 30, 90, 0.6 }, }
IFV_hull_azimuth = { {0, 30, 1 }, { 30, 150, 0.6 }, { 150,180, 0.5 }, }
IFV_turret_elevation = { {-90,18, 1 }, { 18,90, 0.5 }, }
IFV_turret_azimuth = { {0,180, 1 }, }
IFV_armour_scheme = {
									hull_elevation = IFV_hull_elevation,
									hull_azimuth = IFV_hull_azimuth,
									turret_elevation = IFV_turret_elevation,
									turret_azimuth = IFV_turret_azimuth
								};

T55_hull_elevation = { {-90, -45, 0.2 }, {-45, 30, 0.8 }, { 30, 90, 0.4 }, }
T55_hull_azimuth = { {0, 30, 1.2 }, { 30, 150, 1 }, { 150,180, 0.56 }, }
T55_turret_elevation = { {-90,23, 1.0}, { 23,90, 0.2 }, }
T55_turret_azimuth = { {0,30,2.0}, {30,150,1.6}, {150,180,0.65}, }
T55_armour_scheme = {
									hull_elevation = T55_hull_elevation,
									hull_azimuth = T55_hull_azimuth,
									turret_elevation = T55_turret_elevation,
									turret_azimuth = T55_turret_azimuth
								};

So from reading this, my take would be that anything using an armor scheme (if the unit doesn't have an armor scheme defined, it uses a flat "armour_thickness" Value), should be penetrated by the mk-118 since they should only be hitting from above.

 

For instance, a HMMWV has thickness 0.005, whereas BMP1 is 0.2, leopard is 0.125, T-72 has 0.1

BUT each one has a different HP value as well:  HMMWV: 1.5, BMP: 4, Leopard: 32, T-72: 25

 

Now, I don't know exactly how explosive mass and HP correlate, but I do know i've seen T-72's shrug off a MK20, while trucks and such get deleted by it.
I think we can reasonably assume the explosive mass is close to a 1:1 correlation with HP damage (assuming a penetration), so it might require upwards of 100 submunition hits to kill a T-72 for instance, or 320+ for the Leopard (Granted, you can start a vehicle burning without totally removing all the HP and it will die, but that's another discussion)

Given the dispersion pattern of the MK-118's, probably about ~60% of them don't end up hitting a target, which is why I think we don't usually see them killing vehicles (especially higher HP/armor ones) well.

 

 

tl;dr:  MK-20 is accurately modeled down to the submunition, unit HP/armor causes issues, since the MK-118s don't do enough damage.

 

That's the real issue here, the ground units in DCS are just HP pools like in an RPG game and it could be either dead, or alive with some states in between in some cases but always based on that HP pool. For instance (and I'm just making out these numbers), and SA-6 search radar will have it's radar inoperative when the HP pool down to 30%, not because you hit the radar antenna. You can hit the tracks of the vehicle and get it down to 30% and the radar will stop working even if you haven't hit it.

 

This could work more or less acceptable, but the problem is with small warhead like the bomblets in the Mk 20 because the damage attached in DCS for this warheads is so low, that you will need many hits to a vehicle to completely kill it whereas with a proper damage model, you would have destroyed their tracks rendering it combat inoperative, killed the crew, the turret, hit munitions depot, etc. It is the same issue we have with the Harpoons, RB-04 and RB-15 work because Heatblur put a very high damage value for them in order to offset this. In real life, anti-ship missiles won't sink a ship, but they will render it combat ineffective but not in DCS where you have to sink them.

 

The whole RPG approach to the damage modelling is just wrong.

 

  • Like 2

i7 12700KF | MSI Z690 A-PRO | Corsair Vengeance 2x16 gb @ 3200 Mhz | RTX 3070 Ti FE | Acer XB271HU 1440P 144HZ | Virpil T-50 CM throttle | Virpil WarBRD Base + MongoosT-50 CM2 Grip | MFG Crosswind | TrackIR 5 | HP Reverb G2

Bf 109 K-4 | Fw 190 A-8 | Spitfire LF Mk. IX | P-51D | Fw 190 D-9 | P-47D | Mosquito FB VI | F/A 18C | F-14 A/B | F-16C | MiG-15bis | MiG-21bis | M-2000C | A-10C | AJS-37 Viggen | UH-1H | Ka-50 | Mi-24P | C-101 | Flaming Cliffs 3

Persian Gulf | Nevada | Normandy | The Channel | Syria

Link to comment
Share on other sites

4 hours ago, Hextopia said:

Honestly, people don't give the devs enough credit sometimes, because things like the CBU-97 are modeled with insane detail (there are probably ~200 lines of code that define the aerodynamics and the model animations for the dispenser casing pieces, skeets, and parachute submunition dispensers, and that's not even considering the definitions of the explosives and the seeker heads on the skeets)

This might be because they look at in-sim results, like I do. From what I've seen of CBU-97, it splits apart into several identical bundles of four submunitions, that descend under their parachutes until they hit the ground, at which point the skeets shoot up in to the air and home onto the nearest armor. This behavior makes them far more susceptible to wind than they should be, and the "fixed bundles" effect looks really cheap. Even if ED made provisions in code for very detailed workings of the bomb, what we see does not reflect how it works IRL.

 

Also, there's one more thing. Does DCS really simulate the ~2000 bomblets that four-Roceye ripple can produce? Because one particularly sneaky optimization technique is to limit the number of projectiles, and delete those over the limit. The code might be set up for the proper number of bomblets, but we need to know if they're all spawned and not culled due to some hardcoded limit. 

Link to comment
Share on other sites

59 minutes ago, Dragon1-1 said:

This might be because they look at in-sim results, like I do. From what I've seen of CBU-97, it splits apart into several identical bundles of four submunitions, that descend under their parachutes until they hit the ground, at which point the skeets shoot up in to the air and home onto the nearest armor. This behavior makes them far more susceptible to wind than they should be, and the "fixed bundles" effect looks really cheap. Even if ED made provisions in code for very detailed workings of the bomb, what we see does not reflect how it works IRL.

 

Also, there's one more thing. Does DCS really simulate the ~2000 bomblets that four-Roceye ripple can produce? Because one particularly sneaky optimization technique is to limit the number of projectiles, and delete those over the limit. The code might be set up for the proper number of bomblets, but we need to know if they're all spawned and not culled due to some hardcoded limit. 

An easy way to check would just be to run a slow-mo track file and watch them release.  Most likely they're all simulated, since watching it happen can turn your computer into a slideshow momentarily.

 

As for the CBU-97, I don't think you're really paying close attention to it then, since it visually looks almost exactly like the real things do in the textron videos.  Also, the submunitions absolutely are susceptible to winds, which is why in high wind conditions you'd use a lower HOF, or offset your aimpoint appropriately.

 

Here's a decent video of them in DCS: 

 

And here they are in real life:

 

If you watch them in slowmo, you'll see that they're quite well modeled.


Edited by Hextopia
  • Like 1
Link to comment
Share on other sites

25 minutes ago, Hextopia said:

Here's a decent video of them in DCS: 

That's very much not what I saw when dropping them from the A-10. And I definitely did follow them, up close, all the way until impact (partially because I was wondering why they miss so much) They struck the ground, bounced up, and then started doing their thing. Even if they did work right on the Viper when this video was made, they didn't seem to be working so well when I flew the Hog.

Link to comment
Share on other sites

17 hours ago, Dragon1-1 said:

That's very much not what I saw when dropping them from the A-10. And I definitely did follow them, up close, all the way until impact (partially because I was wondering why they miss so much) They struck the ground, bounced up, and then started doing their thing. Even if they did work right on the Viper when this video was made, they didn't seem to be working so well when I flew the Hog.

Hmm.  That sounds like a bug, because I've only ever seen them work like the video I linked, and that's for the A-10 and the Viper.

Are you using any strange settings in the stores page when you get ready to drop them, or are you just using a normal CCRP drop without editing the stores page?


Edited by Hextopia
Link to comment
Share on other sites

Normal CCRP on a TGP target. In the A-10A, it was CCIP. As it happens, it doesn't really matter what you set in the SMS with the real bomb, as it doesn't affect the bomblets or the skeets, just the canister. This was a few months ago, but I don't remember any changelog saying they fixed it.

Link to comment
Share on other sites

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...