IED/VBIED discussion

[hellfish]

Figured we could start our own little discussion carried over from the Blog:

Part 1

Part 2

Thus far, I think the idea of having decoys is the most appealing. The Syrian player can buy various cars, trucks or little civilian groups and also have a selection of identical units with various proximity devices attached to them - from a couple sticks of dynamite on a terrorist to a large fertilizer bomb on a truck.

They are all restricted to green/conscript experience but maybe with higher fanatacism to simulate misunderstandings/miscommunications by civilians and the determination of terrorists. Of course huge penalties for killing the civilian versions.

I also like the idea of having those civilian groups and vehicles under control of the Syrian player. In Somalia (maybe to an extent in Lebanon and Iraq as well) civilians actively aided insurgents/terrorists. The Syrian player can send waves of civilian units toward the Americans and goad them into trying to fire at some of them or to distact the Americans from their actual mission. There don't even have to be VBIEDs in the scenario - the US player would still need to react to the civilians and the potential threat they may pose.

Of course this brings up the distasteful aspect of having to model civilians. I think there are ways to "fudge" that graphically, however - making car and truck windows opaque, hiding the faces of civilians under turbans or burquas or something similar.

I'm sure there are complications that we're not aware of but I think it's the best solution I've seen so far.

[Battlefront.com]

The key to doing a simulation is knowing when you have to abstract and then knowing where to draw the line. We know we can not simulate civilians in any meaningful way. It would take a year to do even a half assed job, so the ideal situation is practically way beyond our reach. What's the other extreme?

A presumption that all civilians are out of the combat zone all the time and anybody in it is 100% considered a combatant. This is easy to do and is in fact what pretty much every wargame since the beginning of time has done. But in today's world the line between civilian and military is blurred and therefore leaving out civilians altogether isn't advisable.

The primary reason for having civilians in the game is to act as "background noise" for unconventional combatant forces. IEDs, VIEDs, "spies", and various types of disguised fighters require such "background noise" in order to be useful. Otherwise they are just sore thumbs that the other side can ID and pick off long before they become a significant threat. This presents an impossible to solve dilemma for us; we can't simulate civilian activity from a production standpoint, yet we must simulate civilian activity to include a rather important part of the modern day battlefield. Quite a conundrum for us to be in, which is why you don't see little, if any treatment , of civilians in wargames

OK, so where does that leave us? Damned no matter what we do So understand that there is no possibility, and I mean none, of simulating all combat situations that have some relationship to civilians. Therefore, we have to go with a system that makes the most amount of sense from a simulation standpoint that won't kill our development schedule and won't open up the door to gamey tactics. The best way to do this is to come up with a single abstracted system and define it as rigidly as possible so the concept is as pure as it can be. Anything that falls outside of those boundaries has got to be left out.

What this means is no decoy vehicles. They are a gamey element just waiting to be twisted into a development nightmare for us. There is zero, and I mean zero, chances that we're going to go in this direction. The reasoning is actually quite straightforward and airtight...

Since we can not simulate civilian traffic and traffic behavior, any and all civilian vehicles seen by the US player can only mean one thing -> enemy combatants. Oh sure, they might be unarmed and cause some sort of penalty if you destroy them, but inherently they are being controlled by the enemy and not a neutral third party. So even if they are not directly a combatant, they are aiding the combatant force. This gives the US player a huge incentive to waste any civilian vehicle it sees. That means they aren't decoys as much as targets. This undermines their purpose.

OK, so perhaps we give these decoys a huge point disincentive if killed. Great, now the US player can't get rid of these things no matter what they do. Picture a decoy buzzing a US convoy. Think of the problems this is going to cause the TacAI! And yet if the US player wastes the decoy he suffers a huge penalty. Either way, the US player is at a gamey disadvantage and the Syrian player is laughing his ass off. That's a dead design concept even before we start thinking about all the AI programming that would be needed to pull off even a responsible use of decoys (and it is substantial). That means that the dead design concept has just been shot about 20 more times and then buried without ceremony

As for decoy crowd units... this is almost the same as decoy vehicles, but not exactly. These are combatants, in a sense, because they are indeed controlled by the enemy force. Human shields and what not. Yet the US force isn't really at liberty to waste them. Unfortunately, simulating clearing them off requires us to simulate something else. And that something else is rather difficult to put one's finger on. In most cases a bunch of civilians would actually stop the military operation completely, like what happened with the 101st Airborne in An Najaf. In other cases it would delay it to the point of running down the scenario's clock to zero. Not much point in playing a scenario where you sit there for a half an hour waiting to hear back from Brigade what you should do about the problem. So although I think decoy crowd units have more place in CM:SF than decoy vehicles, they are equally impractical to simulate and tending more towards stability ops vs. combat ops.

Bottom line is we must maintain this simple, and straight forward, abstraction:

Absolutely no units shown to both players that are non-combatant. Instead, the Syrians will have control of units that remain hidden even after being "spotted" by US units. They only become visible when the US units figure out the identity of the hidden unit (this part is simulated behind the scenes, obviously). The Syrian player can do things to influence his hidden units from becoming visible, like not rushing them around, limiting their exposure to US unit's, keeping them distant, and avoiding putting them in spots where civilians obviously would not be. This is not a sure way to keep them hidden, but it at least offers a pretty good chance of it. And when the cloak of invisibility is removed, all US units within eyeshot will have a heightened sense of interest in their new found discovery

So there you go... an abstraction concept that can be explained in a few sentences. It's straight forward, practical, and we so far haven't found any gamey tactics that can arise from it. Therefore, we'll be extremely reluctant to budge from this design

Steve

[hellfish]

Well... poop. So much for that.

[Homo ferricus]

Originally posted by Battlefront.com:

So there you go... an abstraction concept that can be explained in a few sentences. It's straight forward, practical, and we so far haven't found any gamey tactics that can arise from it. Therefore, we'll be extremely reluctant to budge from this design

Steve

i love this idea, this is exactly how i wanted it to be from the start! A system that doesn't allow "gamey" tactics, but still gives the Syrians another asset that can make up for their lack of Fire Support, and it still makes sense. And let's be honest, from a technical standpoint, modelling non-abstracted civilians as units and modelling their AI would be almost impossible, not to mention all their animations and how US forces would react to them.

just one question, in the map editor (when making a scenario) can you edit and adjust which areas a civilian likely would or wouldn't be, thus setting values for how much suspicuois activity in a certain area it takes to arouse the interests of nearby American soldiers?

[akd]

But what about MBRPGs?

An Iranian take on suicide VBIEDs.

[aka_tom_w]

Thanks for the post

fytinghellfish

From the blog:

"

I think will be very frustrating to players.

<BFC Voice 2>

Which has been my point... you can't make VIEDs and IEDs less frustrating for the player and still have them be realistic The whole reason why these things are such a problem is that they are SO difficult to deal with. Any feature that allows the US player to understand a) that the scenario contains a high probability of having such a unit and that allows him to react unrealistically to it, undermines the very thing that makes VIEDs and IEDs so deadly.

If VIEDs and IEDs aren't frustrating, then we haven't simulated them correctly. Sorry guys, there is no way around this fundamental conundrum.

"

That voice sounds like Steve.

-tom w

[Battlefront.com]

fytinghellfish,

Well... poop. So much for that

I knew it, I knew it! You're not just a gamey cheat'n bastich playing DropTeam, but you're also a pending gamey cheat'n bastich playing CM:SF. I've got your number, mister!

Steve

[Sixxkiller]

I for one am glad that civilians wont be simulated, but I can see where people would want them in. But I think it would make this too realistic. Besides, most people here wouldnt cry if they offed civilians, especially in the game.

[hellfish]

Originally posted by Battlefront.com:

fytinghellfish,

</font><blockquote>quote:</font><hr />Well... poop. So much for that

I knew it, I knew it! You're not just a gamey cheat'n bastich playing DropTeam, but you're also a pending gamey cheat'n bastich playing CM:SF. I've got your number, mister!

Steve </font>

[Battlefront.com]

Sixxkiller,

Besides, most people here wouldnt cry if they offed civilians, especially in the game.

Correct, they would more likely laugh There were heated discussions about showing blood in early CM games and the argument was made by the supporters that it would bring the emotional impact of death and destruction home to people. Right... like GTA, Quake, and the like make people sit down and contemplate what death is and why we are so prone to bringing it on so deliberately. Obviously didn't buy into the flawed logic.

Fishy, thems fight'n werds! I'll see you and your Paladin on the battlefield soon, and we'll see if you remember who's your daddy

Steve

[hellfish]

Bring. It. On.

[pad152]

Think of the problems this is going to cause the TacAI! And yet if the US player wastes the decoy he suffers a huge penalty. Either way, the US player is at a gamey disadvantage and the Syrian player is laughing his ass off.

Isn't the real issue Rules of Engagement? Either shoot everything or shoot back only if it shoots first. Iraq, US is at the disadvantage and Iran is laughing thier ass off.

CMSF; Allied force - only shoot things that have been or can be identified and try to limit damage to infastructure! Emeny - direct enguagement by regular forces, and face ambush and pop-up attacks by irregulaer forces?

[birdstrike]

Steve,

when you say there won't be civilian vehicle traffic, will there still be things like abandoned/wrecked civilian vehicles in the streets (I'm thinking about cover here )

And will there be a way for a player ingame to determine that there is "abstract civilian activity" or could the enemy just pop up anywhere?

[Battlefront.com]

Pad,

Isn't the real issue Rules of Engagement?

True, however picture a situation in the game where you are the US player and you see a taxi 8 blocks away the very first time it comes into LOS. In real life there might be a stream of traffic or at least no reason to suspect that taxi of anything. Yet in the game you KNOW that it is a combatant in that it is controlled by the combatant player (i.e. it is not independent).

Therefore, the US player unrealistically has his attention focused on something which has not posed a threat and a strong incentive to blast it before the game can simulate that it is a threat. To counter this urge to whack something "prematurely" we have to put a huge penalty on its destruction. This is for sure going to be used by the Syrian player to his advantage. Unrealistically so, because the decoys are supposed to simulate innocent civilians so in real life the Syrian commander wouldn't be directing their action.

The net result is that the US player will become extremely frustrated with the game itself because he knows he's being screwed with unrealistically and yet he can't do anything proactive because the game will likely remove any hope of him winning the scenario.

When we come up with an abstracted system the first thing we try to do is figure out how well it approximates the part of combat it is supposed to represent. If we can kick big holes in it early on, it probably isn't the right way to go. The reason is that it is highly likely that there are far more subtle holes that the players will find an exploit after the game is already coded. Therefore, if we see potential problems now we need to be looking for some other way to go. We have another way, the other way is simple and straight forward, and so far we haven't found any significant holes in it. Shortcomings, sure, but all abstractions have shortcomings by their very nature. Heck, even our painfully detailed features have shortcomings. That's just par for the course when trying to model the real world in a sim

birdstrike,

when you say there won't be civilian vehicle traffic, will there still be things like abandoned/wrecked civilian vehicles in the streets (I'm thinking about cover here )

I hope we can do this because, as you say, cars are an important form of cover.

And will there be a way for a player ingame to determine that there is "abstract civilian activity" or could the enemy just pop up anywhere?

No, it will be scenario wide based on terrain. During the Briefing the player will know what that level is and he'll have to abstract his thinking accordingly. A taxi out in the middle of the desert is going to be spotted quite easily regardless, a group of 4 men hanging around 20m away from active combat isn't going to look good either, etc. So while in theory the enemy can pop up just about anywhere, often times they will pop up far sooner than the Sryian player would like

Of course, just like everything else, the Syrian player will have no idea when his hidden units are being looked at by the US player's units. Therefore, he has no idea when the enemy will "catch on" until he starts taking fire. This should make the Syrian player quite cautious about how he uses his specialized units.

Steve

[birdstrike]

Okay, thanks for the answers.

[hoolaman]

I quite like the "spy" abstraction. In game terms, a "spy" could be either used for spotting or used as a trigger point for an IED. In this way the concept could be abstracted to the point that a civilian spy could be treated like a TRP. It could be simply a point that provides spotting information, and the ability to detonate a mine in its line of sight. It needn't be depicted as a fully animated fat little syrian man to either the syrian player or the US player, to have the same game effect.

In a similar way, I don't think vehicle IEDs should be simulated or at least depicted graphically, for a couple of reasons. First, during the conventional phase of the conflict, rules of engagement would very likely preclude civilian vehicles from getting within 100m of blue forces. I believe this was the case during the initial invasion in Iraq. There should be no penalty tactically or points-wise to the player for following these sensible rules of engagement.

Second, it would be far more disbelief inducing to have a Toyota appear from nowhere than to have an abstracted "spy" that can detonate an IED. If they are all abstracted into one system, there will be the same net result as a VBIED.

Maybe the system could work like this. A "civilian spy" can be an abstracted unit that acts as either

1. A slightly mobile spotter for intel gathering

2. A FO to detonate IEDs

3. A fast mobile suicide unit.

In all three examples the graphical depiction could be abstracted, like a TRP or a chess pawn or something. In all three examples there is a chance the unit will be identified and very quickly destroyed.

The Syrian player could have hordes of these little pawns and make very good use of them in a realistic manner.

[yuvuphys]

How about something like this. There are two levels of spotting for VBIEDs, "Possible Threat" and "Threat". When a vehicle is defined as a threat, the US player knows it is a unit controlled by his opponent, it can be engaged with no fear of negative consequences. If the vehicle is defined as a "Possible Threat", it could be one of two things. It could be a unit controlled by the Syrian player that is not yet positively identified as a threat. The US player could take protective measures and fire warning shots if he so chooses. However, the vehicle could also be a civilian vehicle *not* controlled by the Syrian player, but randomly created based upon the amount of civilian traffic in the area. All of the normal traffic one would encounter does not need to be simulated; only an occasional civilian vehicle that approaches too close to the US forces. (This is less work than simulating all civilian traffic; but I'm still not sure if it'd be too much work. It could be as simple as a car driving down a road, and taking a turn, and no longer being a threat...the car would have to respond to warning shots somehow; but usually by stopping and backing away.) These vehicles would never become identified as a definite "threat", and would stay not perform any actions which would classify it as a threat to US forces under their SOPs.

So, here's the beauty of this. The US player can target "possible threats" if he so chooses; if the vehicle happens to be a Syrian controlled vehicles, he'll suffer no ill effects, job well done. But, if it happens to be a civilian vehicle, he suffers the consequences. One could allow the US player to give instructions to his troops on the engagement of these "possible threats"; how to react, to fire warning shots, adn generally how agressive to be in contronting them. Since the Syrian player can't control these possible threats (and can't even see them!), he can't use them in a gamey manner; but it provides the US player with the same difficulties a real US invasion force might encounter. Again, it's still some work, but less work than simulating all traffic, and maybe that's enough?

[Battlefront.com]

All these ideas have already been thought of and rejected. They blur the line and that is bad, bad, bad for so many reasons. I've already listed the bulk of them above and in the Blog.

There will be no visual depiction of anything unless it is a threat. It eliminates gamey behavior, the need for independent AI behavior, unrealistic influence on US units (besides gamey behavior), and unrealistic depiction of civilians. The latter is the thing I don't think you guys quite understand. If you blur the line then you get the worst of both worlds.

As I've said before, it is impossible to come up with an abstracted system that represents all non-combatant and unconventional situations without spending a couple of years making a virtual city sim. So we're going to focus on a system that probably does 80% what we need, with 5% chance of abuse, and minimal strain on our overstretched development schedule. Introducing a lot of gamey potential and massive increases in development needs to eek out another 5% functionality is simply not the way to go.

Steve

[dan/california]

The other way to think about a lot of this is the fact that in many high intensity engagements the civilians who can leave do, and the rest crawl into the depths of their basement and pray real hard that the house doesn't get dropped on top of them. It is after the invading side takes theoretical control that the nightmare starts.

This game is about establishing that control and if you are willing to make the point with 50 caliber warning fire most civilians will keep as much distance as they can as fast as they can.

People who want to keep living just do not cozy up to armored columns that are operating with active combat ROE.

[Dirtweasle]

The discussion on the blog is very interesting. For what it's worth I like the decision you've made.

[Dirtweasle]

Hey, what about Special Ops in civilian attire?

I assume the "cloak of invisibilty" *could* be used to some effect with highly trained special operators infiltrating Syrian lines and mizing with the invisible civilan background noise. Yes?

[C'Rogers]

When this game was first being discussed I thought 'well, it needs to have civilains'. The more I think about it I think, even if all of the extensive AI problems could be solved, it is better off without them.

Mainly because I don't think the player's role in the game has anything to do with decisions regarding who and who is not a threat.

While the player is not defined into a specific role, he is something of a god company commander (battalion or whatever). His job is to provide orders and the overall strategy. A lot of the posts here about US players engaging potential threats would require an intense level of micromanagement.

Just picture a turn with civilians (or even harder, real time). The player must go to each squad, and scan the crowd. Is anyone edging to close? Well we better fire a couple warning shots just to be safe.

On top of that the time between someone going from normal civilian to insurengent in disguise is a few seconds. With vehicles there might be a brief period of time as it closes on the US players units but otherwise a person will go from non-hostile to hostile in an instant.

Unless of course the game would suddenly become CM: psychological anaylsis. Did that guy just twitch? What about that one over there, it looked like he sweating pretty heavily as he prayed. Hmmm, that little kid looks a lot more serious than he should.

I think the current idea is the best to go with, even if the AI problems weren't so huge.

My 2 cents on how everything should be (because what wouldn't be a post without throwing out personal opinions).

Insurgent units (armed but plain clothed), for lack of a better term, would have very very high stealth ratings in heavily populated areas until they have started firing (and maybe regain them after they have stopped firing and retreated).

IEDs. Current system seems the best. Placed, with different types, US troops have a chance to detect before hand based on numerous factors.

VIEDS or suicide bombers. Treating these like the current (CMx1) setup for mine fields seems reasonable. Syrian player has access to a suicide bomber that doesn't even need to be represented by a unit. At the start an area is placed where the bomber operates in (obviously larger for vehicles). Players entering that area have a chance to be attacked. Experience, busyness, rules of engagement (if they are in place) and surronding population detemines the chance that the attacker will hit US units before they have a chance to stop him.

[Battlefront.com]

Dirtweasel,

I assume the "cloak of invisibilty" *could* be used to some effect with highly trained special operators infiltrating Syrian lines and mizing with the invisible civilan background noise. Yes?

Yes

C'Rogers,

A lot of the posts here about US players engaging potential threats would require an intense level of micromanagement.

Yup, or a ton of AI. Neither one is good, but for different reasons.

Steve

[hoolaman]

Steve, am I right if I understand that a VBIED will appear out of nowhere as a newly spotted unit that will be graphically shown as a civilian car/truck flying towards you, and known to be a threat.

That wouldn't be too bad, especially if there are four or five 3d models that could be used as static parked cars as well. That would give a much better feel of a living (if deserted) urban area.

OTOH, I personally wouldn't complain if VBIEDs were left out entirely. They are distasteful for one, but also because they seem to open a can of worms by way of accurate simulation. Also, would Syria use suicide attackers like this? Does Syria have martyrs waiting in line to suicide bomb US forces. Hezbollahs campaign seemed very non-suicidal, and I wonder wether a Syria would act the same.

Besides, I believe simulating only static IEDs gives the same net effect as I said above. Wether on a truck or buried in garbage, the bomb has a finite chance of successfully exploding, detection or neutralisation.

Just assuming the ROE involve wasting anything moving within 50m should not draw too many complaints from the player.

[aka_tom_w]

From the blog:

The problem here is we aren't simulating the real life traffic patterns and real life driver decisions. The player, in either proposed design, controls the vehicle and can therefore unrealistically exploit things to his advantage in theory. With your example, let's say that the vehicle is visible and the TacAI is programmed to avoid a civilian vehicle instead of shooting it. Well, I'm the Syrian player... I have a Toyota with a family of 4 in it... I know this and I want the US player to lose some points... so I continually buzz a bunch of Strykers to get the TacAI swerving them all over the place. The US player then either has his vehicles being annoyed and distracted, or he blows the crap out of the car with the family in it. Either way the Syrian player has benefited unrealistically from the situation. And since the penalty for wiping out civilians would have to be extremely harsh in order to prevent the US player from whacking civilian stuff whenever he saw them "just in case", a gamey Syrian player could probably get the US player to lose the game simply by buzzing him with a non-combatant vehicle.

Now, contrast that with my system...

(the Romulan Clocking Device System?)

There are no non-combatants, only combatants. If a Toyota Taxi (personnel carrier) is in the game, it's up to no good whether the US player sees it or not. Buzzing a bunch of Strykers will get it ID'd for sure and wiped out. No penalty for the US player, no benefit to the Syrian player. In fact, the Syrian player is now short a transport and has a minus score added to his total. So there is no incentive for gamey behavior. If the vehicle is a VIED then getting it close to Strykers is exactly what he wants to do, but not to annoy them like I wrote above. It's to blow them up! If the VIED gets into range then the Syrian player has succeeded and the US player appropriately penalized. If the VIED gets ID'd too soon and is shot up to all Hell, then the US player gets the points and the Syrian minus points.

The big problem with the other suggestion is that it requires the Syrian player to move around vehicles that are decoys. Otherwise the US player will know, without any doubts, that the Syrians have a vehicle that needs to be eliminated. The invisible unit concept requires no such action on the part of the Syrian player, no special action on the part of the US player, it requires no special coding of the OpAI to handle decoy runs, and does not introduce a concept that is different than IEDs and Spies.

My feeling is that your system is not bad, but when the work and potential flaws are corrected for it won't be substantially better than the system I proposed. It is possible, because of gamey tactics, to be less realistic even though it will take a lot more work. Since we don't have the time to mess around with this stuff, we have to go simple all else being equal. If my idea was the one with gamey potential, then we might have to go more complex, but the opposite appears to be the case. So the idea that I've tossed out is the one that offers a better return.

I am afraid I don't get it :confused:

Is the game abstracting a chance that somehow, (and it appears to be out of control of EITHER the US or the Syrian player) that a "cloaked" VBIED will "sneak up" on a US vehicle or position and just blow up?

Neither player having any real in game control over what gets spotted and what does not?

What does the Syrian player need to know or do to decrease the chance of getting spotted?

What does the US player need to know or do to increase the chance of spotting the cloaked threat?

confused

-tom w