reported SCUD precision

[Red_Camarada]

It is normal to SCUDs miss the target by far? I am setting a "fire at point" task and the SCUDs are missing every shot by like 1NM, even in flat terrain.

[flashben]

I think that's pretty normal for something that doesn't have GPS corrections and only relies on INS guidance during rocket powered flight, afterwards it goes ballistic

[sLYFa]

AFAIK CEP for the R-17 is about 500m, so missing by 1nm is a little too much

[Fri13]

CEP = 50% of the shots should fall inside the radius, rest to go outside. But if CEP is 500 then it should be about 1000-1500 meters at maximum where you can have majority to fall in. 

[Northstar98]

Last time I checked the accuracy seemed fairly reasonable, but the problem is it was incredibly precise - with missiles landing in the craters left by previous missiles. I'll have to do some more testing.

EDIT: Yep, still doing it - the missiles are always significantly off of the target, but they all hit the same place - they're inaccurate, but they're incredibly precise (and shouldn't they also account for the wind?)

Edited January 6, 2022 by Northstar98
formatting

[VFA41_Lion]

ED keeps saying the ballistics are work in progress, so

[Harry.R]

Scuds aren't that accurate, for example in 2003 we were more worried about being hit by bits of one downed by Patriot than actually getting killed by a DH.

[Northstar98]

On 5/19/2021 at 9:22 PM, Harry.R said:

Scuds aren't that accurate, for example in 2003 we were more worried about being hit by bits of one downed by Patriot than actually getting killed by a DH.

True, though in DCS it's a little more convoluted because they're very inaccurate (and don't seem to be compensating for wind at all - which might be the source of the inaccuracy) but incredibly precise.

But they all miss by the exact same amount - with subsequent missiles landing consistently in the crater of the previous missile.

Edited January 6, 2022 by Northstar98
formatting

[Raven (Elysian Angel)]

I remember even news reports from back in '91 saying the military was not at all concerned about the SCUD's accuracy. Still, you probably wouldn't want one aimed at a city.

[Shadow KT]

As far as I know, nothing has changed recently in regards to the SCUD's FM. 

You still get an error, if your launcher is placed on incline/decline. This is a bug, which has been reported many moons ago. That's why you can get huge offsets. 

Another issues is, the missile does not compensate for wind. If you have wind, it will blow it away.

Third issues is that, although you will get an offset, because of the two above mentioned issues, the offset its self will be consistent. Meaning, although you will miss, you will always hit in practice, the same spot.

Edited May 21, 2021 by Shadow KT

[Northstar98]

On 5/20/2021 at 3:17 PM, Raven (Elysian Angel) said:

I remember even news reports from back in '91 saying the military was not at all concerned about the SCUD's accuracy. Still, you probably wouldn't want one aimed at a city.

Oh yeah, the accuracy is several hundred metres to a km, which isn't great for something with a conventional (albeit a fairly large conventional) warhead. My main gripe though is that while the accuracy I'm okay with (though it doesn't seem to account for the wind at all), the missile is extremely precise, they all land within a few metres of each other, I've seen subsequent missiles land in the crater of the previous missile it's that precise.

Just in case anybody is confused here's an illustration of what I mean:

I'd expect the R-17 to be more like the one in the top left (inaccurate and imprecise), with 50% of missiles fired to randomly fall within a 500m/1km radius around the target. Right now they're more like the top right bullseye (not accurate, but precise), only with even tighter dispersion.

On 5/21/2021 at 12:38 PM, Shadow KT said:

As far as I know, nothing has changed recently in regards to the SCUD's FM. 

You still get an error, if you launcher is places on kind of incline/decline. This is a bug, which has been reported many moons ago. That's why you can get huge offsets. 

Another issues is, the missile does not compensate for wind. If you have wind, it will blow it away.

Third issues is that, although you will get an offset, because of the two above mentioned issues, the offset its self will be consistent. Meaning, although you will miss, you will always hit in practice, the same spot.

Exactly this. It's very inaccurate (realistically the CEP means it isn't very accurate at all, but unrealistic in that it doesn't compensate for wind, I haven't tested it on an incline and I thought it needed to be placed on level ground), but incredibly precise in that all the missiles land in essentially the exact same place.

Edited January 6, 2022 by Northstar98
formatting

[antiload]

This Scud missiles are still very inaccurate. Also nothing can shoot them down or attempts to shoot them down.  

 

Please ED can you look at the scuds as I would use them more if they had some tuning and got engaged by air defence. 

 

Thanks for all your work.  

[Northstar98]

On 7/11/2021 at 1:28 AM, antiload said:

This Scud missiles are still very inaccurate.

Yeah, they're incredibly inaccurate (much more than the 1km CEP often quoted - I don't think they're accounting for the wind or launch angle), but incredibly precise.

On 7/11/2021 at 1:28 AM, antiload said:

Also nothing can shoot them down or attempts to shoot them down.

So far I've had success with Patriot, I've seen it successfully intercept Scuds.

Edited January 6, 2022 by Northstar98
formatting

[antiload]

On 7/11/2021 at 2:03 AM, Northstar98 said:

 

Yeah, they're incredibly inaccurate (much more than the 1km CEP often quoted - I don't think they're accounting for the wind or launch angle), but incredibly precise.

 

 

So far I've had success with Patriot, I've seen it successfully intercept Scuds.

 

Hi Northstar98

You are correct, the test I did was using a Patriot site I incorrectly set up. Just to reiterate the scuds can be intercepted.

Thanks.

[Northstar98]

On 7/17/2021 at 7:04 PM, antiload said:

Hi Northstar98

You are correct, the test I did was using a Patriot site I incorrectly set up. Just to reiterate the scuds can be intercepted.

Thanks.

No worries.

Edited January 6, 2022 by Northstar98
formatting

[VFA41_Lion]

didn't want to start a new topic,

but there is no way this unit is behaving correctly in game. A group of scud units will always have their missiles impact the same location all together, and will always have it impact that same location every time the mission is run. Seems like it has a very narrow beam of CEP that it hits every time. See attached mission for demonstration and screenshot.

scuds precise.miz

[orbiter28]

Hello, same here, i don't want to start a new topic about the same subject.

I have just done a test with 2 scuds, the target of which is oriented due north at 100km for one, and the target is oriented due west 100km for the other.
I used the crimea of the caucasus map in order to have a very flat relief.

The accuracy of the scuds is of the order of 300m when there is a 45m difference in altitude between the scud truck and the point of impact.
And when there is almost no difference in altitude between the scud truck and the point of impact, we achieve an accuracy of around 100m.

It is concluded that if you want to have accurate shots with your scuds, you must be careful that the launchers have the same altitude as the point of impact.

scud accuracy test.miz

[jackmckay]

Yeah. Had same issues with SCUD. Wanted to integrate it into server just for immersion treat.

Found out that the target point offset is really uncool besides having manual hydraulic pumps on human foot power for raising and lowering that forky ramp. Besides still cool but too long launch sequence default trajectory goes into high earth orbit. Who knows, maybe it really crosses some winds of distortion on the way up and down in RL.

Anyway, would be cool if it has more random spread lookup around "actual" target point. The offset is too far and too precise in offset point after hitting with barge. Same spot all. Looks too weird. The only thing I found as solution is to find actual bearing offset propagation over distance - somehow, or try some scripts to fix it. i did, but this one doesn't support IC and in that mod it flies very smoothly and hits target like artillery which looks fine and moves eyes away from that problem. Yet, cool neat and mighty thing. Maybe is just OP?

[jackmckay]

On 4/7/2023 at 9:09 PM, orbiter28 said:

And when there is almost no difference in altitude between the scud truck and the point of impact, we achieve an accuracy of around 100m

it looks like you solved vertical offset? what about horizontal?

[jackmckay]

Was firing from Maykop to Sukhumi. Made a CC there and tried several times until this happened:

The bearing deflection on 220716m is 13.7kms east on 151 deg 219km measured:

This was the first salvo hit:

Direct hit:

 

Edited September 5, 2023 by jackmckay

[BIGNEWY]

We do have a task open for tweaking SCUD. If you have short tracks showing an issue please add them. 

thanks

[jackmckay]

41 minutes ago, BIGNEWY said:

If you have short tracks showing an issue please add them. 

thanks

I do have mission track, but its complex server setup. what I discovered that R17 has exact deflection offset. at one point it aims at other it hits. the difference in that setup from others is initial firing angle and single result. it looks like there's no wind influence on any other larger deflection from impact points. in conclusion: the deflection is exact angle offset.

Edited September 5, 2023 by jackmckay

[BIGNEWY]

to show an issue its best to use simple and to the point examples. This makes it easier for the dev team to use the debug version and not spend to much time replaying scenarios. 

thanks

[jackmckay]

Mod works without this code:

GT.verticalDeviationCompensationPeriod = 10.0
GT.maxVerticalDeviationAngle = math.rad(5)

and instead has:

GT.turbine = false;

It also changes missile declaration script:

-- Missile R-17 Scud B
local R17 = {
    category = CAT_MISSILES,
    name = "R_17",
    wsTypeOfWeapon  = {wsType_Weapon,wsType_Missile,wsType_SS_Missile,WSTYPE_PLACEHOLDER};
    Escort = 0,
    Head_Type = 5,
    sigma = {10, 10, 10},-- AMENDED THIS BUILD
    M = 2000.0,-- AMENDED THIS BUILD
    H_max = 230000.0,-- AMENDED THIS BUILD
    H_min = -1,
    Diam = 880.0,-- AMENDED THIS BUILD
    Cx_pil = 1,
    D_max = 595470.0,-- AMENDED THIS BUILD
    D_min = 12000.0,-- AMENDED THIS BUILD
    Head_Form = 1,
    Life_Time = 9200.0,-- AMENDED THIS BUILD
    Nr_max = 6.0,-- AMENDED THIS BUILD
    v_min = 370.0,-- AMENDED THIS BUILD
    v_mid = 1900.0,-- AMENDED THIS BUILD
    Mach_max = 5.00,-- AMENDED THIS BUILD
    t_b = 0.0,
    t_acc = 9.0,-- AMENDED THIS BUILD
    t_marsh = 172.0,-- AMENDED THIS BUILD
    Range_max = 595470.0,-- AMENDED THIS BUILD
    H_min_t = 0.0,
    Fi_start = 3.14152,
    Fi_rak = 3.14152,
    Fi_excort = 3.14152,
    Fi_search = 99.9,
    OmViz_max = 99.9,
    warhead         = 
    {
        mass                     = 985;-- AMENDED THIS BUILD 
        expl_mass                 = 975;-- AMENDED THIS BUILD 
        other_factors             = {1, 1, 1};
        obj_factors             = {1, 1};
        concrete_factors         = {1, 1, 1};
        cumulative_factor         = 0;
        concrete_obj_factor     = 0.5;
        cumulative_thickness     = 0.6;
        piercing_mass            = 987.0;
        caliber                    = 880,
    },
    exhaust = {1,1,1,1},
    X_back = -5.5,
    Y_back = 0.0,
    Z_back = 0.0,
    Reflection = 0.3967,
    KillDistance = 0.0,
    shape_table_data = 
    {
        {
            name     = "R_17";
            file       = "R-17";
            life       = 1;
            fire       = { 0, 1};
            username = "R_17";
            index = WSTYPE_PLACEHOLDER;
        },
    }
};
declare_weapon(R17)

Hope it helps for now.

[jackmckay]

the track is attached.

another test, fresh start. distance 220km/136mi.

this one has different aiming line after first correction.

So it looks like compensation worked after some shots, nr4. salvo x3.

nr5 miss but close.

then missile nr6 hit the target.

except this time correction vector changed and also changed initial orientation of scuds too. 137 vs old 124.

looks like its possible to correct manually.

on eds side, code could be changed and aiming deflection fixed with some spread finetuned to within actual scud-b results. just a number. 

SCUD Target Correction.trk

 

 

 

Edited September 6, 2023 by jackmckay