"Relative spotting" and information sharing

[Veitikka]

We are considering if the next engine should have a "relative spotting" system, meaning that every unit maintains its own pool of information that contains relations to all non-friendly units. This includes if the target unit has been spotted or not, the level of identification and so on.

I'd like to hear some ideas from you regarding what the new system should take into account, when it comes to sharing the spotting information that an individual unit has stored. What should be the rules between individual units, and also between the platoons, companies, and even battalions when communicating via radios and other equipment? Our plan is to make the system support electronic warfare, spy drones, and other contemporary assets, so they need some thought.

This is a good opportunity for influencing the features Armored Brigade might have in the future.

[exsonic01]

I'm really glad to hear AB is considering this feature in future version. I already shared some idea and suggestions before, and I will share more if I can bring more. Also I really wish to read more and better brilliant idea and suggestions from others...

One thing I think important is the balance between realism and gaming fun factor or simplicity. As a realism wargame fan, I really like the info-sharing and new spotting mechanics, and this is one of the reason why I like CM series. However, at the same time, it is also true that such features will increase difficulty & learning curve, as well as increase complexity and this difficulty of coding (and thus future modding / modulation possibility). And I'm pretty sure you and your team members would know about this factor. Though it wouldn't be easy, I wish you found the best solution and find the optimal balance. I think some sort of simplified modeling of info-sharing would be good... I guess? But this will going to need a lot of tests. I think CM series found their own balance, I think AB can do the same.

[altipueri]

I agree with exsonic - don't make it so realistic that only experts can play. Or at least give easily changed options to make it a simpler experience for some of us.

[22sec]

I’m really giving this some thought before I answer. I love the question and what I think it means!

[exsonic01]

How about introduce difficulty setting concept? Or, just give an option to players to control to manage those info-sharing time cost. Then, depending on options, introduce a multiplier factor for info-sharing time, and command delay time too. And if we introduce info-sharing time, I think command delay time can be included in the info-sharing time, isn't it?

Introduce the relative spotting and info-share via datalink as hardcore realistic as possible you can design with given game engine, then set it as "hardcore" difficulty. Or "realistic" info-sharing drop-down option. Then:
"hard" difficulty or "slow" info-sharing speed option with 75% multiplier factor for info-sharing and command delay time.
"Medium" difficulty or "medium" info-sharing speed option with 50% multiplier factor for info-sharing and command delay time.
"easy" difficulty or "fast" info-sharing speed option with 25% multiplier factor for info-sharing and command delay time.
"very easy" difficulty", or "very fast" info-sharing speed option, and set 0% multiplier factor for info-sharing and command delay time. 0% means no time cost for info-sharing or command delay, just like current AB or any other RTS game.

Or, you could separate options for command delay time, and info-sharing time, differently. You could do the same for EW effect. Induce the most realistic effect to "hardcore" or "realistic" difficulty or setting, and induce almost non-existence level of effect for "very easy" difficulty or setting.

Even now, there is a mod to reduce or remove command delay of this game. If option is given, then any players will welcome and enjoy the setting they want.

CM series' difficulty setting also controls info-sharing feature, but there, difficulty setting does not related with time cost as far as remember.

[Lowlaner2012]

This I gonna take some thinking, I'll get back to you 😎

Thanks

[thewood1]

I think a couple things need to be considered on relative spotting:

1) Comms sophistication and timeframe. Not all comms systems are created equal and the maturity/reliability of various systems over the relatively expansive timeframe is very important

2) FoW is very important in setting up a sighting transmission process. That includes misID and generic IDs.

Can it function without those?...yes. But it is one of those things that really makes it more effective.

[22sec]

This is an important, but complicated discussion. I think the big thing that would need to be addressed is how do my AI controlled units act if their radio comms are jammed. The current programming doesn’t allow for much unit independence. If I a have two platoons defending a hill, and they start taking losses but I as the player can’t issue orders to them they would die on that hill. I know it’s always a question of programming, so how do you program some additional conditions that can be set for units to be able to act more independently when they are cut off from receiving my orders? I don’t think it’s worth the effort to limit what the player can see. Even if you implement a system where I as the player has to click on a unit to see the enemy units it sees to me all your doing is slowing down gameplay. The key is implementing the effects of electronic warfare, which needs to be included in what I think y’all are considering.

Of course there are nations that have networked force trackers which change things. It’s been a while since I’ve looked at things like BlueForce Tracker, but if I remember correctly the commander can issue orders. One would need to do the research on those systems susceptibility to jamming.

[exsonic01]

ORIGINAL: 22sec
This is an important, but complicated discussion. I think the big thing that would need to be addressed is how do my AI controlled units act if their radio comms are jammed. The current programming doesn’t allow for much unit independence. If I a have two platoons defending a hill, and they start taking losses but I as the player can’t issue orders to them they would die on that hill. I know it’s always a question of programming, so how do you program some additional conditions that can be set for units to be able to act more independently when they are cut off from receiving my orders? I don’t think it’s worth the effort to limit what the player can see. Even if you implement a system where I as the player has to click on a unit to see the enemy units it sees to me all your doing is slowing down gameplay. The key is implementing the effects of electronic warfare, which needs to be included in what I think y’all are considering.
Of course there are nations that have networked force trackers which change things. It’s been a while since I’ve looked at things like BlueForce Tracker, but if I remember correctly the commander can issue orders. One would need to do the research on those systems susceptibility to jamming.

That is very good point about AI. I remember there was a discussion about self-protective instinct in the thread about order delay time cost to turn the direction. I think this issue shares the similar concern about AI = How much the AI for unit (platoon ~ battalion) will perform proper tactical maneuver when it is cut from player's command.

Some sort of automated routines would be great to be added, which would help AI's independent behavior as tactically sound enough.

Maybe, some sort of additional SOP for battalion / company / platoon / squad behaviors might help those. Aggressive / defensive / opportunistic / conservative and etc... RTS games like C&C Red Alert and Supreme Commander has an option to set AI behavior in general when making skirmish game against AI. Something similar might be helpful in AB too.

This issue can be linked with recon AI, as recons always operate forward boundary and behind the first contact line, searching for possible enemy HQ or other HVT. Typical recon operations are not totally autonomous, but it is true that recon (and light infantry / SF) operations has some sort of degree of freedom.

I suggested some sort of spatial analysis algorithm before. I think this would helps AI to behave properly. But I also agree with 22sec, there should be something to consider about AI's behavior.

[varangy]

My recommendation:

Keeping ABs simplistic, easy to learn (and I think also to implement) approach and building over what we already have:

- each formation has its own info sharing with some base speed (example 60 seconds)
- if they don't have a radio (infantry squads for example), this is extended (so slower, for example by an extra 50%)

- each formation in contact with any HQ shares and receives info
- if there is 0% contact, that unit will not be able to do it
- info has to be "uploaded" and also "downloaded"
- speed of up and downloading is base speed - base speed * HQ contact percentage / 2
- base speed is higher than internal communication
- for a certain enemy unit external comms cannot start until internal comm info sharing has been finished (I am not sure if this is a good idea though. If the game keeps track of a lead unit, it can start once lead unit knows about it)
- once data has been uploaded to an HQ any other unit can start to download from any other HQ (so no comms between HQs, its a shared pool)


fantastic info graphics:
For example look at the graphics below. One member of the infantry squad just spotted an enemy. How is information shared after this?
(internel comms base speed: 30 seconds; external comms base speed: 90 seconds)
- infantry squad internal com takes 45 seconds because not every member has a radio (+50% delay)
- because one member has radio, time to upload data is 58 seconds (HQ contact is 72%, 90 - 90 x 0.72 / 2 = 58)
- once this is finished, T-62 and BMP-2 platoon starts to download this info
- for the T-62 platoon it takes 45 seconds as they have 100% contact (90 - 90 x 1.00 / 2 = 45)
- for the BMP platoon it takes 63 seconds as they have 60% contact (90 - 90 x 0.60 / 2 = 63)

I wouldn't make generic IDs and the "I think there is somebody" thing like in the Battlefront games.

I also highly suggest to "forget" contacts if they go into hiding!

[zacklaws]

In my day from the mid 70's to the late 90's all information was sent straight away up the chain of command, ie section, to platoon, to Coy HQ, to BN HQ and then Regimental HQ. But, it could also be sent direct too from the lowest level to BN HQ depending on what radio net you was using and other formations on the same net would also hear the information and react to it. Everything was sent from sighting reports to contact reports, shellreps, etc. So soon as it happened, the higher formations knew. The only delay would be if the message had to be encypted and then decrypted, but a contact report was just sent in clear, with the pre warning message of "Callsign, Contact, Wait Out", That would clear the airwaves of non essential chat, Higher HQ would know without any delay that a specific unit was in contact and that a Contact report would be soon coming, In the manner of, When, Where, What is it, What is it doing and What you doing about it.

It is a bit pointless if a lower level formation keeps information to itself as then the higher formations do not have a clue what is going on to manoeuvre other units to counter a threat, support an attack or "stand units too" to face an approaching threat etc.

[lancer]

Hi,

If you are going to implement some form of info sharing that involves info being transmitted over communication nets you will also need to model electronic warfare as the two are like intertwined siamese twins.

The other thing that may be required, in order to reduce the inevitable player confusion, is a means of visually showing communication nets on the map. An overlay of some kind that allows players to understand the process would, I'd suggest, be necessary.

Without some form of visualisation of the information flow you are likely going to spend a lot of time answering questions about a key game mechanic that is going to be a 'black box' for most.

Cheers,
Lancer

[varangy]

ORIGINAL: zacklaws
soon as it happened, the higher formations knew.

It is a bit pointless if a lower level formation keeps information to itself

This is a tactical level game. When you hear something on the radio you will not be able to magically see the enemy immediately.

You are a platoon commmander. You hear from another platoon that some kind of enemy is somewhere. You have to think, look around, look on the map, etc. Also the radio can be distorted depending on the conditions, you don't hear, you have to ask back for clarification, you have to find a location with better reception etc.

My solution simulates this. There is no point in making a system to your idea of instantaneous communications as it is just what we have now, so there would be nothing to change.

[varangy]

double delete please

[22sec]

ORIGINAL: varangy

ORIGINAL: zacklaws
soon as it happened, the higher formations knew.

It is a bit pointless if a lower level formation keeps information to itself

This is a tactical level game. When you hear something on the radio you will not be able to magically see the enemy immediately.

You are a platoon commmander. You hear from another platoon that some kind of enemy is somewhere. You have to think, look around, look on the map, etc. Also the radio can be distorted depending on the conditions, you don't hear, you have to ask back for clarification, you have to find a location with better reception etc.

My solution simulates this. There is no point in making a system to your idea of instantaneous communications as it is just what we have now, so there would be nothing to change.

How would what you propose actually effect what the player does? You have two platoons sharing info, but in game you still have to control those units so there is really no sharing info between units, it’s only what you as the player sees.

[thewood1]

This is as much about individual spotting as about comms. You might have another unit radio you that an enemy tank is by the rock. But it will still take x-seconds/minutes for the the receiving unit to orient and spot the enemy unit. And that information loop or OODA loop is variable based on comms network, experience, training, environment, etc. Up until the point of the spotting cycle is complete, an unknown designator is used. I also think you should be able to order a unit to fire on an unknown designator, but with decreased accuracy, ROF, and other ramifications.

The only real question is how do you treat the enemy unit at the player level. Like in Combat Mission, you only see what the unit sees when clicking that unit. But when you click on HQ, you now see all known and unknown spottings. That is the big question.

I also think you need a tool, like a spotting list, that allows you to quickly see who is spotting what and the age of the spotting report.

[varangy]

ORIGINAL: 22sec

ORIGINAL: varangy

ORIGINAL: zacklaws
soon as it happened, the higher formations knew.

It is a bit pointless if a lower level formation keeps information to itself

This is a tactical level game. When you hear something on the radio you will not be able to magically see the enemy immediately.

You are a platoon commmander. You hear from another platoon that some kind of enemy is somewhere. You have to think, look around, look on the map, etc. Also the radio can be distorted depending on the conditions, you don't hear, you have to ask back for clarification, you have to find a location with better reception etc.

My solution simulates this. There is no point in making a system to your idea of instantaneous communications as it is just what we have now, so there would be nothing to change.

How would what you propose actually effect what the player does?

Its not about the player as it can see all the contacts. Its about which of your unit can shoot at which enemy target, because its about changing the spotting system.

[Veitikka]

First and foremost, this topic is about who can shoot who. I think the player should have the access to global spotting information at all times, and also unit pathfinding should use global passability information for its side, as it currently does, because otherwise the system can get unnecessarily complex, without making the game any better.

[jirik]

Hi,

I'd recommend to implement the behaviour as a simple abstraction not going to tracking exact C2 network.

Generally I'd dived the units into two groups - in-formation and out-of-formation. There will be a separate (moddable) delay value for each of the group and in-formation will have a smaller value than out-of-formation.

When a unit spots a new unit then all units in the same formation will receive the information with the shorter delay. The delay will not be exactly same for all units but will be in a random value of a normal distribution around the configured delay.

The same will be executed for out-of-formation units just with the larger value.

This will
- simulate the immediate communication vs command chain delay
- prevent robotic reaction among to units to simulate comms problems, difficulty in target acquisition
- increase the motivation to use larger formations

Dedicated recon units can receive a special modifier to reduce out-of-formation delay to simulate a dedicated comms channel between recon and command post.

EW would be also abstracted. I'd say that EW will be in form of point bought during formation selection.
When the enemy has EW points then the soppting dissemination delay will be calculated in the same way but the distribution will not be symmetric normal distribution but skewed toward larger delays. The more EW enemy points the larger the skewness and higher difficulty in dissemniation spotting information.

[thewood1]

I think that is pretty close to how Combat Mission does command and comms, as it effects what they have called borg spotting.