Tanks - Fire-control system

[darkChozo 133]

Regarding FCS failure: laser rangefinders work by sending out a pulse of laser light and timing how long it takes for the reflection to come back. They fail when the reflection isn't strong enough to detect above background noise. Any sort of laser countermeasure works by either decreasing the strength of the reflection or introducing new noise to make the signal harder to detect.

 

The strength of the reflection depends on the material of the object being lased, as well as its angle to the laser. In terms of material, you want something as reflective (shiny) as possible, but you don't want it to be polished like a mirror, because you want light to scatter everywhere and not just in the angle of reflection. Things that absorb a lot of light (dark materials), and porous materials that "trap" light like nets or foliage, are bad. Note that rangefinders use IR light, and objects that are visible-reflective aren't necessarily IR-reflective, and vice versa.

 

The ideal angle for a reflection is perpendicular. This should be fairly intuitive; you bounce light off a surface that's flat and perpendicular to you, it bounces right back to you. You'll still get diffuse/scattered reflections if the object isn't perpendicular, but these are going to be weaker the further from perpendicular you are (unless it's retroreflective, because retroreflective stuff is magic).

 

These factors aren't necessarily stable. Small changes in conditions can make a huge different in reflectivity. This is what causes snow or still water to sparkle as you move your head. In effect, this is a "random" element; it's not actually random (mostly), it just might as well be.

 

So if you wanted to model this without any sort of optics simulation, you'd take the material of the lased object, its angle to the laser, and the distance between the laser and the lasee (lasers don't drop off much over range, but their reflections certainly do), and use that to calculate a chance of success.

 

It's actually surprisingly doable in the ARMA engine considering that the material of objects and terrain is already tracked to some degree. Not that I expect BI to implement something along these lines, it's just interesting to think about.

[dragon01 902]

Aside from failures to range trees and certain kinds of foliage and camo netting (or mirrors), modern rangefinders don't generally have any problems getting a reading. That laser is strong, enough to blind a people even with a brief pulse. This feature would be best implemented as a restriction to ranging certain kinds of objects.

[AveryTheKitty 2626]

Will the Blackfish (and other aircraft) receive FCS? It's very hard to accurately hit targets with the Blackfish, currently.

[xxgetbuck123 945]

2 hours ago, Nightmare515 said:

Will the Blackfish (and other aircraft) receive FCS? It's very hard to accurately hit targets with the Blackfish, currently.

 

In the Dev Branch they gave the Blackfish gunners CCIP so they can lead their shots properly. It's not a FCS but its a CCIP. 

[dragon01 902]

Blackfish needs locking the camera to terrain, not FCS. Automatic lead computation would help, but it'd be of little use if you can't keep the gun on target in first place. It's got CCIP, which would be all that's needed (well, that and perhaps horizontal gun traverse, to center it a little better) had we been able to lock the camera to terrain like it's possible for TGPs. 

[dragon01 902]

Blackfish got the camera! Hurray. :) Thanks, BIS. 

 

That said, after trying it out, I have to take back what I said about FCS - it is the best solution for Blackfish, after all. For example, I noticed that the max zoom level is useless at normal flight speeds, due to CCIP reticle being so far to the right it comes off the screen. Also, Blackfish still has the "magic rangefinder", as opposed to the proper one. Blackfish doesn't need radar - its gunnery should be similar to that of a tank.

 

EDIT: OK, I did some more testing. It turns out that the CCIP reticle on Blackfish is, for some reason, very inaccurate. That might have something to do with zeroing, since it seems that it's not affecting it. In 1000/200 (radius/altitude) loiter, you need to aim above the target to hit it, as if the gun was zeroed to the default 600m, despite the display showing otherwise. So yeah, FCS it has to be.

[msy 22]

AI gunner still doesn't relay on FCS, when I as commander order gunner to target a moving enemy, it can directly calculates the ballistic without any hesitate.

[Imperator[TFD] 444]

Not sure what changes this has made to the vanilla tank AI but I've now been shot down twice by Kuma's using their main cannon while trying to make a strafing run in my Wipeout.

 

Love it!

[CaptainAzimuth 714]

Hahahaha... I know some people who that's going to tick off. Lost a few Wipeouts on our server to .50cals already, so... this'll be extra fun. I think challenging AI are fun anyhow. 

[Strike_NOR 898]

1 hour ago, Imperator[TFD] said:

Not sure what changes this has made to the vanilla tank AI but I've now been shot down twice by Kuma's using their main cannon while trying to make a strafing run in my Wipeout.

 

Love it!

 

I love this fact! This now forces pilots to fly higher than the tanks can elevate their guns, which in turn poses a significant threat from Anti Aircraft fire. If you want to bring your jet into the mud, you better be hugging terrain and performing pop-up attacks on enemies. Give the enemies the least possible amount of time to train their guns on you before you zoom out of there again!

 

Excellent!

[Beagle 684]

8 hours ago, Strike_NOR said:

 

I love this fact! This now forces pilots to fly higher than the tanks can elevate their guns, which in turn poses a significant threat from Anti Aircraft fire. If you want to bring your jet into the mud, you better be hugging terrain and performing pop-up attacks on enemies. Give the enemies the least possible amount of time to train their guns on you before you zoom out of there again!

 

Excellent!

 

9 hours ago, Imperator[TFD] said:

Not sure what changes this has made to the vanilla tank AI but I've now been shot down twice by Kuma's using their main cannon while trying to make a strafing run in my Wipeout.

 

Love it!

There is a reason why, in modern times, strafing attacks on armoured forces are not recommended anymore. A-10 began to rely on Mavericks for two reasons long time ago, Fire Control of Vehixles able to shoot at airborne targets reliable and increase of armour, leaving only a few spots to hit with GAU 8/A. The cannon just recently got usefull again only because of asimetric warfare circumstances. In DCS for example, using the gun on dispersed IFVs and MBTs is a no go for quite a while. BMPs are notorious to shoot you down with ATGM if you line up slow and low and nicely for the shot. But have no fear, JETS DLC wil bring you a NAVY Multi Role Fighter that might be better suited for the Job compare to a A-10C knockoff

[oukej 2911]

On 23. 3. 2017 at 11:10 AM, dragon01 said:

That said, after trying it out, I have to take back what I said about FCS - it is the best solution for Blackfish, after all.

We've tried that approach, but the FCS isn't currently compatible with stabilized (area tracking) turrets. CCIP (even if with it's own issues) seems still like a better option than nothing. But it's more of a test at this moment to see what u guys think about it.
Thanks a lot for the notes!

[dragon01 902]

Fair enough. Assuming it's unfeasible to make FCS work with area-tracking (if it isn't, that'd still be the best solution), would that be possible to add some sort of horizontal "gun elevation", using the same keys as gun elevations for artillery? This was my previous suggestion, and would make aiming with CCIP much more manageable. 

 

Also, I found that zeroing doesn't work on the Blackfish. Could you look into this, too? This is a real problem, since a 600m loiter radius is a little tight.

[iratus 71]

I've noticed a problem with trees and bushes: If the rangefinder lases a tree or bush, the returned distance is not that of the tree or bush but that of the ground behind the bush (or ---- in case of sky) as if the bush or tree would not exist. If said tree or bush is in front of a building, the rangefinding laser will ignore that (otherwise rangefindable) building too and still return the distance to the ground (or sky) behind the bush/tree.

 

A easy place to reproduce this is Hill 149 north of Abdera [Altis095206]. Lots of bushes and trees in front of houses.

Tested on build 1.69.141143

[dragon01 902]

It would actually be nice (and realistic) if, when ranging through a bush, we could switch between the distance to the bush and to whatever is behind it. This is a real feature of most tank rangefinders.

[Beagle 684]

6 hours ago, iratus said:

I've noticed a problem with trees and bushes: If the rangefinder lases a tree or bush, the returned distance is not that of the tree or bush but that of the ground behind the bush (or ---- in case of sky) as if the bush or tree would not exist. If said tree or bush is in front of a building, the rangefinding laser will ignore that (otherwise rangefindable) building too and still return the distance to the ground (or sky) behind the bush/tree.

 

A easy place to reproduce this is Hill 149 north of Abdera [Altis095206]. Lots of bushes and trees in front of houses.

Tested on build 1.69.141143

Green leaves are a RL problem for IR Laser Rangefinders, since green chlorofill is transparent to IR Light and the cell structure containing it disperses the light instead of reflecting it. There are more reason than just plain simple visual camoflage for choosing positions in foilage and putting fresh foilage on the tanks exterior.

[iratus 71]

8 hours ago, Beagle said:

Green leaves are a RL problem for IR Laser Rangefinders, since green chlorofill is transparent to IR Light and the cell structure containing it disperses the light instead of reflecting it. There are more reason than just plain simple visual camoflage for choosing positions in foilage and putting fresh foilage on the tanks exterior.

If it is a feature, not a bug, it's still buggy because if you lase a bush in front of a house the laser ignores not only said bush but also the house and returns the range of the terrain behind the house. The house itself is, if lased directly, rangefindable, if you lase the same wall next to the bush the range of the wall will be returned.

[Imperator[TFD] 444]

Not sure if this has been noticed but the driver positions of tanks no longer have any way of determining N/E/S/W or bearing.

[oukej 2911]

2 hours ago, Imperator[TFD] said:

Not sure if this has been noticed but the driver positions of tanks no longer have any way of determining N/E/S/W or bearing.

Check Slammer's driver. WIP :)

[Imperator[TFD] 444]

7 minutes ago, oukej said:

Check Slammer's driver. WIP :)

 

Very nice!  Is the plan to roll something similar out to the Kuma/T-100 prior to this hitting 1.7?

[R3vo 2654]

46 minutes ago, Imperator[TFD] said:

 

Very nice!  Is the plan to roll something similar out to the Kuma/T-100 prior to this hitting 1.7?

 

You can always see your bearing via the Custom Info GPS.

[AveryTheKitty 2626]

1 hour ago, Imperator[TFD] said:

 

Very nice!  Is the plan to roll something similar out to the Kuma/T-100 prior to this hitting 1.7?

What's different? I can't check atm.

[Zygzak191 52]

@Nightmare515 

Spoiler

 

[AveryTheKitty 2626]

51 minutes ago, Zygzak191 said:

@Nightmare515 

  Hide contents

 

 

Oh yeah I remember that. 

[en3x 209]

I like compass and turret icon (is it fully working?)

 

But the letter box view is still the reason why 90% of people will

go 3rd person.