Home   Help Search Login Register  

Author Topic: Forward Observer Pack V1.0  (Read 4096 times)

0 Members and 1 Guest are viewing this topic.

UNN

  • Guest
Forward Observer Pack V1.0
« on: 03 May 2005, 02:25:17 »
Hi,

I've re-packaged my Forward Observer Addon, fixed a few bugs, added support for some more third party addons. Increased the dispersion of impact points and added some minor functionality. While this does not contain all the updates I would have liked, it does fix a few bugs. I plan on re-writing the Observer to incorporate all the functions I would like and those requested. Formost will be adding better support for other weapons systems and expanding on the strategic decision making.

AI Forward Observer capable of calling in indirect fire over a predefined area. Observers have the same eyesight quality as Snipers, with the added advantage of using Binoculars more effectively than regular AI, in the role of static observation. This provides them with an enhanced detection ability of about 1000+ meters (depending on terrain and target type), while keeping within the confines of the game engine itself.




The core system files are now contained in two addons:

FOU & FCU System Addons

These addons are required by default.

If you dont have any of the follwing Mods\Addons then you can use the default OFP units and a Generic Mortar addon donated by GFX.

FOU Default OFP units

Defaul OFP units

Observer
Observer (Sniper)
Black Op (Observer)
Black Op (Day Observer)
Spetz Natz (Observer)

FCU 81mm Mortar By GFX (West,East & Res)

The FOU Pack currently supports the following MODS & Addons if you have them. If not, then download them now :)

US Army Rangers By Laser

US Delta Rangers By Laser
US Weapons Pack By Laser

Delta Observer
(S) Delta Observer
DES Delta Observer
(S) DES Delta Observer
DES 2 Delta Observer
(S) DES 2 Delta Observer
Ranger Observer
(S) Ranger Observer
DES Ranger Observer
(S) DES Ranger Observer
Ranger Ghillie Observer
DES Ranger Ghillie Observer
Delta Ghillie Observer
DES Delta Ghillie Observer

Deltas/Rangers By Ballistic Addon Studios

DES Delta Observer
DES Delta (S) Observer
DES Ranger Observer
DES Observer
Delta Observer
Delta (S) Observer
Ranger Observer
Desert Delta (S) Observer
Ranger Observer
Desert Ranger Observer
Ranger (S) Observer
Desert Ranger (S) Observer
Delta HD Observer
Desert HD Delta Observer
Delta HD (S) Observer
Desert HD Delta (S) Observer
Ranger HD Observer
Desert Ranger HD Observer
Ranger (S) HD Observer
Desert Ranger (S) HD Observer

Retextured Guerillas by Ivan/Edge Of The World

Observer
Terrorist Observer (Camo)
Observer (Black Op)
Observer (Cap)
Observer Marksman
Observer Sniper

Infantry By Finish Defence Forces Mod

Finnish Mortar Observer
Finnish Jaeger Observer (Sniper)
Finnish Mech Jaeger Observer (Sniper)
Finnish MP Observer (Sniper)
Finnish Special Jaeger Observer (Sniper)
Finnish Sissi Observer (Sniper)
Finnish Sniper Observer (Bush)
Finnish Sniper Observer (Winter Ghillie)
Finnish Sniper Observer (Leaf)
Finnish Sniper Observer (Fir)
Finnish Reservist Observer (Sniper)
Finnish Paratroop Recon Observer (Sniper)
Russian Observer (Sniper)
Russian OMON Observer (Operator)
Russian Spetsnaz Observer (Sniper)
Russian SOBR Observer (Operator)

FCU Finnish 81mm KRH 71Y Mortar
FCU Russian 82mm Mortar
FCU Finnish 81mm KRH 71Y Mortar (2)

Scout Sniper by Ghost in the Grass

Scout Sniper Observer

Invasion 1944 Mod

Wehrmacht Observer
Fallschirmjaeger Observer
82nd Observer
Waffen SS Observer (Sniper)

FCU Inv44 81mm Mortar
FCU Inv44 60mm Mortar

WW2 French Infantry (demo version)

(Hvy) Infantry Observer
(Lt) Infantry Observer
(Hvy) Overcoat Observer
(Lt) Overcoat Observer

Liberation 1941-1943

'41E Observer Helmet
'41E Observer Sidecap
'41E Artillery Observer
'40 Observer
Observer Cap
Observer Sidecap
'41 Observer Helmet
'41 Observer Sidecap
'41 Artillery Observer

Marine Assault Pack v1.0

USMC Observer
Force Recon Observer
(D) USMC Observer
Russian Observer
(D) Russian Observer
Naval Infantry Spetsnaz Observer

SEB Nam Pack - NVA Units V1.00

AirCav Observer
Army Observer
Marine Observer
SF Observer
LRRP Observer
MFR Observer
NVA Observer
VC Observer

FCU M19 Mortar (Air Cav)
FCU M19 Mortar (Army)
FCU M19 Mortar (USMC)
FCU NVA 60mm Mortar
FCU VC 60mm Mortar

WW2EC Mod

1st Infantry Observer
2nd Rangers Observer
5th Rangers Observer
4th Infantry Observer
29th Infantry Observer
90th Infantry Observer

FCU 105mm M2A1 Howitzer

Updated documentation can be found here:

FOU Manual

The entire pack including documentation, example missions and all the FOU & FCU addons:

FOU Pack

You can use any of the supported Mortars with any Observer, to swap Mortar crews just call FCU_NewCrew from each mortars Init field with an infantry class name:

Code: [Select]
[<Object>,<String>] Call FCU_NewCrew
For example, to put some of Laser's Deltas in GFX's west mortar:

Code: [Select]
[This,"LSR_delta_spotter"] Call FCU_NewCrew
Do not install, those you do not have the corresponding addons for.

Individual FOU & FCU third party addons and example missions:

FOU LSR US Army Rangers
FOU BAS Deltas/Rangers
FOU Retextured Guerillas
FOU Finish Defence Forces
FOU Ghost in the Grass Sniper
FOU Invasion 1944 Mod
FOU WW2 French Infantry (demo version)
FOU Liberation 1941-1943(Example Missions requires the Invasion 1944 Mod)
FOU Marine Assault Pack v1.0
FOU SEB Nam
FOU WW2EC Mod(Example Missions requires the Invasion 1944 Mod)

If I have overlooked anyone or made any errors regarding credits to addon makers, please let me know.

I have update the addon and function names to reflect the OFPEC tags for the entire system, so it will mean any mission and scripts that use the old FO & FC prefixes, will need to be udated. My apologise for any incovenience to anyone who might be using the old versions, I can only say in future updates, there will be no more name changes.

Cheers
« Last Edit: 03 May 2005, 04:58:09 by UNN »

Offline penguinman

  • Contributing Member
  • **
  • Money is worthless, just paper, ink, and threads
Re:Forward Observer Pack V1.0
« Reply #1 on: 06 May 2005, 00:57:47 »

I would like to try it but could you be a little more specific on what one needs to download for just the normal  mortar pack.

and also
« Last Edit: 06 May 2005, 00:59:34 by penguinman »

Offline penguinman

  • Contributing Member
  • **
  • Money is worthless, just paper, ink, and threads
Re:Forward Observer Pack V1.0
« Reply #2 on: 06 May 2005, 01:07:13 »
Quote
Formost will be adding better support for other weapons systems and expanding on the strategic decision making.

Actualy, to me

i disagree, I dont think you should add suport for other systems, its got enough.

I think what you should work on  is getting the scripts advanced enough that you can use.

[FO,Mortargroup] exec"AImortars.sqs"

so you can just type that in an init field, place your mortar group, place your FO, and have him start automaticly tagging targets on his own and sending them to the mortars. ;)

I love the addon but it seems like your drawing away from the AI controled mortar part a little.

this system is probly the only AI mortar system so i think you should work on improving the AI scripts and that kninda thing first. :)
« Last Edit: 06 May 2005, 01:09:30 by penguinman »

Offline penguinman

  • Contributing Member
  • **
  • Money is worthless, just paper, ink, and threads
Re:Forward Observer Pack V1.0
« Reply #3 on: 06 May 2005, 01:22:32 »
never mind i got it.

ok i tested the missions


the only bug i found was,

when the mortars hit targets

the men firing the mortars say "got him" and "soldier is history" even though they are like a mile away and have no idea what they are hitting.

other than that its great.

UNN

  • Guest
Re:Forward Observer Pack V1.0
« Reply #4 on: 06 May 2005, 03:11:15 »
Hi,

Quote
I think what you should work on  is getting the scripts advanced enough that you can use.

[FO,Mortargroup] exec"AImortars.sqs"

It will if I get to add the improvements I want, but I do understand your point. I want to be able to assign the observer to all sorts of different addons\groups, in pretty much the way you mentioned. The other weapon systems in their simplest form, will be Mortars and Artillery in different groups.

Although it is possible to call fire from independent mortar groups now, but it's not exactly an ideal solution and you would loose the radio dialog between the mortars. I can knock up an example missions if you're interested?

Quote
I love the addon but it seems like your drawing away from the AI controlled mortar part a little.

Yeah, I do want to expand on the strategic side of things. But that should make the Observer more decisive when picking targets, giving his decisions some tactical weight.

Quote
this system is probly the only AI mortar system so i think you should work on improving the AI scripts and that kninda thing first

The bulk of my time will be spent on updating how I gather and process target information. The trouble is, unlike the features I mentioned. It might not be that obvious, but I'm hoping target selection will feel a bit slicker.

Quote
the only bug i found was,

when the mortars hit targets

the men firing the mortars say "got him" and "soldier is history" even though they are like a mile away and have no idea what they are hitting.

That's just default OFP behaviour, short of turning the radio off for everyone, I don't think I can stop the group radio.

Cheers

Offline h-

  • OFPEC Site
  • Administrator
  • *****
  • Formerly HateR_Kint
    • OFPEC
Re:Forward Observer Pack V1.0
« Reply #5 on: 06 May 2005, 13:48:07 »
You know UNN, you could always delete/re-create the grenade upon firing...
This way you would get rid of the AI saying those chats...
Project MCAR   ---   Northern Fronts   ---   Emitter 3Ditor
INFORMATIVE THREAD TITLES PLEASE. "PLEASE HELP" IS NOT ONE..
Chuck Norris can divide by zero.

Offline penguinman

  • Contributing Member
  • **
  • Money is worthless, just paper, ink, and threads
Re:Forward Observer Pack V1.0
« Reply #6 on: 06 May 2005, 23:33:01 »
Quote
Quote:I love the addon but it seems like your drawing away from the AI controlled mortar part a little.

Yeah, I do want to expand on the strategic side of things. But that should make the Observer more decisive when picking targets, giving his decisions some tactical weight.

well, what I ment was,

insted of working on stuff for the player as an FO, where the player selects targets on a map, work on stuff like what hater said, and also like making the first salvo fall short and then the AI FO says ajust 50 meters NW. Because the mortars are too deadly IMO, even though you increased dispersion they are still to acurate.

but anyways, about the player/AI FO stuff.
 To be frank, sombody looking for a player controled Artillery mission is going to use the COC system, im not trying to be mean but its the truth. COC however unlike yours does not have the ability for an AI unit to call in mortars.

I hope i cleared it up

thanks

UNN

  • Guest
Re:Forward Observer Pack V1.0
« Reply #7 on: 07 May 2005, 11:54:27 »
Quote
You know UNN, you could always delete/re-create the grenade upon firing...
This way you would get rid of the AI saying those chats...

Yeah, but I do like the simplicity (thinking of MP) of just being able to fire the weapon like a regular OFP weapon. I calculate the entire trajectory so it's easy enough to simulate rounds like many of the artillery scripts here. But it's only an issue when you and the Observer are part of the Mortars group. Can't remember now if you still hear kill confirmation when your not the group commander, so I'm going to test this out first as a way to avoid unnecessary dialog.

Quote
insted of working on stuff for the player as an FO, where the player selects targets on a map,

None of this is for the player, as the Player still has a far better spotting range than the AI. Player support was added as an extra, took about a day to implement as opposed to months for the AI :)

The progression was. I did my indirect fire scripts when there was no alternative, once I finished them I thought hmm I need to get the AI to use them. Now I'm thinking how to improve the way the AI picks the most important target (ATM it's just based on the detected groups KnowsAbout total and class type). Also how does the AI react to incoming fire, in most cases they start to run if they have a waypoint, or stand around in formation waiting for the next round if stationary. When in reality I think they would all scatter, hit the dirt and look for cover?

But I do plan on adding a Forward Command HQ, more for the Observer as dedicated recon. But it should benefit the Mortar Observer to, if you choose to combine them.

Quote
making the first salvo fall short and then the AI FO says ajust 50 meters NW. Because the mortars are too deadly IMO

I calculate the trajectory about three times before actually firing, a spotting round could easily be inserted at any point. The one thing the put me off from the start was, it plays havoc with predicting the paths of AI targets.

You can get into a loop where the AI react to the spotting round, causing them to move beyond the 50m radius. So the Observer recalculates the next impact point, but the AI has calmed down by then and returns to it's original route, so you never actual get to fire a full salvo.

I know in WW2 the most effective way to use mortars was, firing every available tube at an unsuspecting enemy, moving in the open. The idea was to hit them with as many rounds as possible before they could react. I'll have to go back over the training docs I have, to see when spotting rounds are used.

But yeah, they are to destructive for the amount of AI you can have in OFP at any one time. I think the SEB Nam mortars are about as powerful as I would want in my own missions. Anything bigger than the 81mm mortar would result in a very short game. Might be part of the reason why COC developed Infantry on Demand, to keep the meat grinder ticking over :)

Quote
To be frank, sombody looking for a player controled Artillery mission is going to use the COC system, im not trying to be mean but its the truth. COC however unlike yours does not have the ability for an AI unit to call in mortars.

No worries, I would rather constructive discussion than a pat on the back. Things like spotting rounds will only really benefit the player as an Observer, giving him an idea of the ETA e.t.c But they will help players being targeted by an AI Observer, at least it will give you some warning.

Cheers

Offline penguinman

  • Contributing Member
  • **
  • Money is worthless, just paper, ink, and threads
Re:Forward Observer Pack V1.0
« Reply #8 on: 13 May 2005, 04:58:39 »
right

but now your artillery has the accuracy of most laser guided ordinance. For AI FOs usualy the first round wipes out the entire squad of the person the FO is targeting.

so I think some not spotting rounds but the first slavo being off and the FO quickly correcting would be good

UNN

  • Guest
Re:Forward Observer Pack V1.0
« Reply #9 on: 13 May 2005, 09:58:22 »
Quote
For AI FOs usually the first round wipes out the entire squad of the person the FO is targeting.

It's easy enough to increase the dispersion and decrease the killing power of each round. Part of the problem is OFP's terrain, it does not offer much cover. There is no variation between the most lethal (Tree bursts) to the most ineffective (hitting water).

I really want to spend some time on the AI targets to, I can trigger an event that can be used for targets. From an ear candy point of view, I can get the squad leader to shout "Incomming", but I can get a stationary AI squad to take evasive action before the round has hit the dirt.

With the advent of laser range finders, GPS and hand held computers, I think modern artillery could be as accurate? But I do need options to degrade the accuracy depending on the time period, but still allow for the accuracy of TRP's (Target Reference Points) for prepared positions.

Quote
so I think some not spotting rounds but the first slavo being off and the FO quickly correcting would be good

AI spotting rounds will not be to hard, I sample trajectories until I find one that lands with 10 meters of the target. I will add a user defined value (say defaults to 100m divided by 50m intervals), so the trajectory lands within any of those intervals it will trigger a (fake) spotting round, until it hits the 10m target radius. So a value of 100m would probably trigger two spotting rounds, one at a 100m and the other at 50m.

I still have the problem of not being able to rearm mortars from a script, of the many times I ran the Observer & Mortars example missions, I ran out of ammo after finishing off just half of the squads attacking the position. So I might skip spotting rounds when targetting sqauds moving at speed?

Thanks again for the feedback. I always try and add as many user defined options as possible, I just need pointing in the right direction sometimes.

Offline penguinman

  • Contributing Member
  • **
  • Money is worthless, just paper, ink, and threads
Re:Forward Observer Pack V1.0
« Reply #10 on: 13 May 2005, 23:59:15 »
I think you should make a shrapnel script that will allow you to have  a smaller explosion with still the same ammount of possible dammage but also the men may be protected by trees and the like.


sa8gecko

  • Guest
Re:Forward Observer Pack V1.0
« Reply #11 on: 19 May 2005, 07:54:17 »
I was wondering if the grenade flying path could be improved a little.
That is, currently mortars fire (for what I've been able to see) using the
lower angle, instead of the higher one, like cannons instead of howitzers.
That is, if the same range can be attained using 25 and 65 degrees, currently
your mortars seems to use 25. I've not tested them on reverse slopes,
but in a test mission in Malden (the AI observer and mortars one) sometimes
the grenades fall 'short', hitting the wrong slope (and killing friendlies).
I know that raising the angle of the parabola will raise the flying time and
so the grenade can't live enough to do any damage. Anyway it's possible to
substitute the grenade with a brand new one while in flight (after perhaps
ten seconds). Unfortunately the air in OFP seems very thick, that is for a
real gun the maximum range achieved firing on a target on the same level
is reached with an elevation of about 40 deg (not 45, for air friction), while
in OFP is about 30 degrees. This can be overcome making the grenade
(to be substituted) a missile. With this system a grenade fired at 300 m/s
muzzle velocity can reach 8000+ meters, while currently max range
for a similar muzzle velocity is about 2700 meters. But I've found this way not
completely satisfying.
Anyway, the important thing is this: you can use a program like NLREG to
generate the equation you need to calculate right elevation for your mortars
to fire. You need a lot of data, taken for a better precision at distances of
100 meters one from the other, and with different differences of height
between the target and the mortars. Basicly you could use the equation
General Barron refers to in his indirect firing script, even if this is more
complicated than needed. It's very precise, tough, once you have the
corrects parameters, if the target and mortars are at the same level (ASL).
This equation needs to be corrected for different heights ASL. The easiest
thing is to introduce another simple linear equation that will raise or lower
the firing distance in relation to the height displacement: raising it if the target
lies above the mortars, lowering it if it's below them. Or go the more complex
way and make a new equation. This should be not to difficult, keeping in mind
that the equation proposed by the guy referred by General Barron is even
unnecessarily complicated (some parameter can be eliminated).
Using this equation should lower the time needed to calculate the trajectory,
that at current stage is calculated in an iterative way (if I'm not mistaken).

Offline Dinger

  • Contributing Member
  • **
  • where's the ultra-theoretical mega-scripting forum
Re:Forward Observer Pack V1.0
« Reply #12 on: 19 May 2005, 10:40:12 »
The current OFP projectile system takes direction and angle of attack into account, with weird results. So substituting a shell 20 seconds in and applying the same vector will present different ballistic characteristics (as the new projectile will be horizontal to terrain, while the previous one at a high angle).

NLREG is fun to play with, and I've gotten some good results (using a variety of interesting data collection routines), but the problem with a polynomial fit is its inherent instability: at various points it can just give garbage without warning.
Dinger/Cfit

UNN

  • Guest
Re:Forward Observer Pack V1.0
« Reply #13 on: 19 May 2005, 16:01:12 »
Quote
I think you should make a shrapnel script that will allow you to have  a smaller explosion with still the same ammount of possible dammage but also the men may be protected by trees and the like.

Yeah, that could have many possibilities. I've been messing around with the Russian mine pack. They use what appears to be a scattering of bullets, to simulate shrapnel. Perhaps combined with making the direct hit radius smaller, it could be a nice balance. As the secondary (shrapnel) blast might be absorbed more effectively by the terrain?

Quote
I was wondering if the grenade flying path could be improved a little.
That is, currently mortars fire (for what I've been able to see) using the
lower angle, instead of the higher one, like cannons instead of howitzers.

I know, in real life I think it's considered dangerous to fire at such low angles, when relying on gravity to impact the round. The FDF mortars do use angles of 45 degrees +. But as you touched on, they need a muzzle velocity of 100m/s and the range is halved as a result. GFX's mortar (the one I can tinker with) could be made to cover both depending on the current target range.

Quote
I've not tested them on reverse slopes,
but in a test mission in Malden (the AI observer and mortars one) sometimes
the grenades fall 'short', hitting the wrong slope (and killing friendlies).

I did notice this once in a while, when testing the latest version. I'm almost 75% sure it's down to the delay I use, to allow the mortars to adjust for a new target. I'm using an arbitrary time to let the mortars get into position. That was before we worked out how to use game logics as an indication of the barrels current position :)

Quote
Anyway, the important thing is this: you can use a program like NLREG to
generate the equation you need to calculate right elevation for your mortars
to fire. You need a lot of data, taken for a better precision at distances of
100 meters one from the other, and with different differences of height
between the target and the mortars.

NLREG's inability (or at least the equation it produces) to accommodate differences in altitude between source and target is a pain. Plus I'm not sure it could return the time of flight? As this is vital for the Observer to do his stuff. The polynomial equations I used to compensate for variations in addon design, could possible compensate for the differences, but without a suitable amount of test data it also returns some weird results.

I must admit, apart from potentially being a CPU hog, I do like the flexibility and accuracy the iterative process provides.

Quote
This can be overcome making the grenade
(to be substituted) a missile. With this system a grenade fired at 300 m/s
muzzle velocity can reach 8000+ meters, while currently max range
for a similar muzzle velocity is about 2700 meters.

Never considered making it a missile. I do want to do more with GFX's mortar, mainly lavishing it with pointless detail like custom anims and a proper crew. Realistic operating and setup procedures e.t.c

Quote
But I've found this way not completely satisfying.

What is it, that lets it down?

Quote
The current OFP projectile system takes direction and angle of attack into account, with weird results. So substituting a shell 20 seconds in and applying the same vector will present different ballistic characteristics (as the new projectile will be horizontal to terrain, while the previous one at a high angle).

Yeah, it's the same titling problem everyone has with turret MG's. Fingers crossed, the next version of OFP (AA...or whatever) will give us that option to affect angle of attack.

sa8gecko

  • Guest
Re:Forward Observer Pack V1.0
« Reply #14 on: 19 May 2005, 18:05:37 »
Quote
The current OFP projectile system takes direction and angle of attack into account, with weird results. So substituting a shell 20 seconds in and applying the same vector will present different ballistic characteristics (as the new projectile will be horizontal to terrain, while the previous one at a high angle).
Are you sure ? In the recoilless jeep Hater_Kint made, the grenade is substituted
to the bullet when the trigger is pressed, while switching to the ranging gun
the fired bullet is left to go by itself. The substituted grenade and the bullet
have the same initspeed and they SEEM to have the same ballistic.
Quote
NLREG is fun to play with, and I've gotten some good results (using a variety of interesting data collection routines), but the problem with a polynomial fit is its inherent instability: at various points it can just give garbage without warning.
Not if the function is continuous and differentiable in the range you are going
to use. Just plot the function, and you should instantly recognize those
instability points.
Quote
What is it, that lets it down?
I couldn't set a friction coefficient so that the missile (with thrust =0, so that
I could use the internal OFP gravity) would slow down its horizontal speed
fast enough for my tastes. The horizontal speed slows down, but not fast
enough (as respect to the grenade). Maybe giving the missile a negative
thrust may work. I doubt it. Anyway the range is largely augmented and the
ballistic at short ranges resembles the grenade one. At long ranges I couldn't
say cause the trajectory it's still ballistic, but the grenade can't travel that far. I could make Hawk 2S19 hit targets at 7000+ meters away, if they
lied on the same plane as the artillery.
Quote
NLREG's inability (or at least the equation it produces) to accommodate differences in altitude between source and target is a pain.
This is not NLREG fault. If the function is the correct one, it will give you
the correct results. In the function proposed by the guy in his artillery
tutorial, there's no space for a Z variable. You must introduce it for the
function to give you the correct results with height differences. The
simpler way seem (to me) to have a function that corrects the distance
(and so the elevation) relating to the height difference between target and
shooter, as I explained above. It would not work perfectly, but it should be
acceptable. The better result will be obtained including the z variable in the
main function.
Quote
Plus I'm not sure it could return the time of flight?
Again, wrote a function that does just that. How are you at calculus?
Mine is rusty...
Quote
Fingers crossed, the next version of OFP (AA...or whatever) will give us that option to affect angle of attack.
I hope so, that was one of the things more requested about having a revision.
This will make possible, even if BIS should have not done so for 'Armed Assault',
to make good looking multiple turret vehicles and this way one can make
tow humvees without resorting to MCAR. And even MCAR will be a lot simplier.
I hope we'll get a demo soon.