Suggestion on the communication model

Matrix Games Forums

Forums Register Login Photo Gallery Member List Search Calendars FAQ

My Profile Inbox Address Book My Subscription My Forums Log Out

Suggestion on the communication model

View related threads: (in this forum | in all forums)

Logged in as: Guest

Users viewing this topic: none

Printable Version

All Forums >> [New Releases from Matrix Games] >> Command: Modern Operations series >> Suggestion on the communication model

Page: [1]

Message

<< Older Topic Newer Topic >>

Suggestion on the communication model - 2/18/2016 7:13:16 AM

acui

Posts: 7
Joined: 7/13/2013
Status: offline

Please excuse any errors as I am from China.

The CMANO is a really fun game.

But the God's eye view is not that fun. It made the development in the communications technology nearly useless. The ship of 50s can get the same threat infomations as a ship after 2000.

If we make the communication just as real as it in reality, an isolated unit(like a submerged sub in H3) will be controlled by the AI, and the AI controlled unit will make the game unreal and less fun.

From my experiences in the Combat Mission, I think they have got it solved. The idea is very simple: Let the player has the God's eye, but every units can only see what their sensors have detected and what they received from others(with some delays based on the comm method and the path from the source). When none is selected, the player can see all the units and contacts. After the player select an unit, the map will be changed to that unit's view and shows only what it knows. The player can still control all the units and is aware of the overall situation. But an unit will only react to what it knows. So a F-15 will not take evacuation maneuver to a SAM just in sight of a sailor. A ship will not turn on the radar and begin launching the SM-2 immediately when the white trails of vampires towards it are spotted by a P-3 pilot.

This method will also make us to play realistically more easily. We only see what an unit knows when it is selected, so we can order it to react based only on that information, thus make a more realistic move. And we can even make friendly fire possible under some circumstances.

In the future MP games. This method will make even more fun. We can lock players to sub-branches in the OOB. It works like the sides in the current game, but is more flexible.

The implementation will be hard but not impossible. I know nothing of the implement of sensors in the CMANO. To the best of my knowledge, the CMANO has a global sensor map for each side, it is merged from the contacts of all the activated sensors of that side, and each unit will react based on the global sensor map. This will not works with the new model because of the delay. I made a simple implement of the sensor map to be used with the new communication model. It has multiple sensor maps with different update interval. The first one is the real time one, it's updated real time. Then the 5s, 30s, 1min, 5min...1 day and 1 week ones. Each unit will then monitor different contacts in different maps based on their delays to the sensor's platform. It could deals with 100,000 units with multiple sensors easily on an i5-4690 using one thread. But this is without the complexity of the real sensor model and unit AI. It is only a test to see whether where is a way to implement a more realistic communication model in the scale of CMANO.

< Message edited by acui -- 2/18/2016 8:53:58 AM >

Post #: 1

RE: Suggestion on the communication model - 2/18/2016 10:22:15 AM

Dimitris

Posts: 13282
Joined: 7/31/2005
Status: offline

Hello and welcome! We're glad you like CMANO and thanks for taking the time to put this together.

First of all it is important to realize that CMANO's comms and sensory model is not as simple as you may think. Units may indeed share a common tactical picture but the "you cannot engage what you cannot see yourself" limitations are still there; in your ship engagement example the P-3 pilot who has visually spotted the vampires may flash an advance warning to the ship but the ship's radars still have to detect the targets in order to engage them (however, the advance warning will ensure a faster reaction by the ship's crew, just like in RL). So we are a lot closer to the Combat Mission example that you are describing than the simple Borg hivemind.

There are of course a number of exceptions to this, like inertial/GPS-guided munitions or the new "blind" engagement opportunities offered by CEC-capable weapons and these are also modelled.

The Combat Mission model is not without its own flaws. For example an individual troop squad may not be aware of an enemy unit far away, but the all-seeing player is, and can direct that squad towards the enemy unit until they establish contact themselves and engage. This of course is also true for CMANO but in the latter case the omnipresence of fast communications (incl. completely hands-off datalinks) from the 1950s onwards makes this a realistic RL tactic instead of a "Borg cheat".

One of the things we are currently working on is a realistic depiction of comms disruption that may lead to units being completely cut off from their side's "comms cloud", even to the point that (in addition to not sharing any contact info) they are completely outside the player's control and are even oblivious to what their own side forces are doing. Their behavior then is dictated by their existing doctrine for handling such a situation (Press on with the mission? Abort and RTB? Head towards the nearest reported friendly? Stay put until comms are re-established?). This will enable us for the first time to faithfully model situations like the 1982 Bekaa valley shootout where the Syrian air force was annihilated as a result of radar and comms jamming (among other factors).

You mention a test implementation. Is this a software prototype? What details can you share?

Thanks.

_____________________________

Facebook
Twitter

(in reply to acui)

Post #: 2

RE: Suggestion on the communication model - 2/18/2016 10:32:18 AM

quertice

Posts: 248
Joined: 9/9/2007
From: Italy
Status: offline

quote:

ORIGINAL: Sunburn

One of the things we are currently working on is a realistic depiction of comms disruption that may lead to units being completely cut off from their side's "comms cloud", even to the point that (in addition to not sharing any contact info) they are completely outside the player's control and are even oblivious to what their own side forces are doing. Their behavior then is dictated by their existing doctrine for handling such a situation (Press on with the mission? Abort and RTB? Head towards the nearest reported friendly? Stay put until comms are re-established?). This will enable us for the first time to faithfully model situations like the 1982 Bekaa valley shootout where the Syrian air force was annihilated as a result of radar and comms jamming (among other factors).

I'm really glad to hear this

_____________________________

(in reply to Dimitris)

Post #: 3

RE: Suggestion on the communication model - 2/18/2016 12:22:34 PM

acui

Posts: 7
Joined: 7/13/2013
Status: offline

Maybe I did not explain it clearly.

The current problem is that the AI acts as if it can see all the contacts. Maybe the P-3 pilot is not a good example. Another example is a sub has detected enemy fleet. Without breaking the radio silence, other friend ships will have not chance of knowing where the enemy is. But in the current game with the default AI behaviors the friend ships will attack the enemy immediately(if they are in range). With a realistic communication model this will not happen. You will have the sub break the EMCON to let friend ships know the enemy location and make an effective attack. Or let the sub launch a radio beacon and have friend ships get the target with some delay.

I think the flaws in the Combat Mission like model is not unbearable. The player can choose to play realistically or not.

And to my knowledge, even in the U.S. there are a lot of platforms(for example, the A-10As) do not have datalinks until 80s-90s. And a submerged sub does not have any real time communication ability. Even for the unit with datalinks, sometimes they cannot share target info without delay. For example, a U.S.Army's M1A2 tank can not ask an nearby naval super bug to help him even when the F/A-18 has some unused A-G ammunitions and has to jettison them before landing.

I am against the idea to have the AI control the uncontactable units. Because in reality, these units are controlled by human. If we let the AI take the command, they will behave very unrealistically like the subs in H3.

I wrote some C++ code to test the complexity of the model. It's a very simple one. The map is just an flat 2000km x 2000km 2D. And every units have a sensor with 100km range. They will detect an target if and only if it's in range. Every units will not change course or speed in the duration of simulation. The units is random organized in a tree of 10 levels, each unit can only direct-communicate with its children or parent. and the communication delay between two level is 10s. If you like, I can share the source code with you.

(in reply to Dimitris)

Post #: 4

RE: Suggestion on the communication model - 2/18/2016 2:31:30 PM

mikmykWS

Posts: 11524
Joined: 3/22/2005
Status: offline

Hi Acui

We definitely understand and think its great you've put time and effort into this. Our comms model is more complicated though than you think and unfortunately we're not going to release the code for people to pick through just yet. This is mostly due to the fact that as D mentioned we're still adding things.

The other issue you may want to consider is that because this is the entertainment version of the game we have to make some compromises with reality to keep the game fun. If we implemented things 100% as they are in real life the game wouldn't be all that fun to most people. We rely on most people to provide most of the return for our effort so have to code to their spec

Pro version of the sim is a totally different ball game though

Anyways thanks for your time and effort. As we get closer to implementing the comms things would it be ok to ask you to take a look?

Mike

_____________________________

(in reply to acui)

Post #: 5

RE: Suggestion on the communication model - 2/18/2016 7:10:38 PM

FTBSS

Posts: 201
Joined: 8/25/2014
Status: offline

Actually submerged subs do have some comms capabilities I don't think I can offer details with how this works but I have first hand knowledge of it.

(in reply to mikmykWS)

Post #: 6

RE: Suggestion on the communication model - 2/18/2016 8:01:21 PM

SuaveWatermelon

Posts: 65
Joined: 1/3/2014
Status: offline

@Mike

Are you guys considering adding some of the pro features as options in later builds of the game?

While it wouldn't be fun for most people, I for one would appreciate having the option of playing realistically with submarines etc.

I don't know much about coding or game development so I have no idea how this would be done which is why I ask; but I can't see any reason that
players could not have the option to play CMANO as a game and as a sim.

(in reply to FTBSS)

Post #: 7

RE: Suggestion on the communication model - 2/18/2016 8:27:07 PM

mikmykWS

Posts: 11524
Joined: 3/22/2005
Status: offline

quote:

ORIGINAL: SuaveWatermelon

@Mike

Are you guys considering adding some of the pro features as options in later builds of the game?

While it wouldn't be fun for most people, I for one would appreciate having the option of playing realistically with submarines etc.

I don't know much about coding or game development so I have no idea how this would be done which is why I ask; but I can't see any reason that
players could not have the option to play CMANO as a game and as a sim.

No pro features will end up in the civilian version.

That being said things we feel should be in both will land in both and aren't considered pro. Does that make sense?

As far as sub comms you may be able to do with it with a coming feature but there will be no change to the basic model because most players like to drive their subs.

Mike

_____________________________

(in reply to SuaveWatermelon)

Post #: 8

RE: Suggestion on the communication model - 2/18/2016 8:28:57 PM

quertice

Posts: 248
Joined: 9/9/2007
From: Italy
Status: offline

cancelled post (writed before Mike's post)

< Message edited by qwetry -- 2/18/2016 9:32:58 PM >

_____________________________

(in reply to SuaveWatermelon)

Post #: 9

RE: Suggestion on the communication model - 2/18/2016 11:53:36 PM

Dysta

Posts: 1909
Joined: 8/8/2015
Status: offline

Oh wow, just I wake up and already have a long wall of comments. First of all, welcome.

I also do wonder about the comm system in game, how it gather information of your own units in and out. I haven't actually see a strategy game that will not accurately show up your units in far distance, not even from Harpoon or Fleet Command -- their location are always exact, as well as sending and receiving command instantaneously.

Another feature like CEC also make me thinking, what if the unit can add/modify the comm/data link(s) for specific scenario, so they could communicate with the other units with different set of comm system, as well as the missile guidance. The typical Link-16 from USAF and Naval fighters also proven its effectiveness along with alliance jets that have it installed.

And jamming with OECM and DECM are more likely a lottery than simulating the signature scrambling of how it was modeled, but I think something like a cloud or radius view of jamming, and showing its bandwidth and intensity are seems like a possibility out there. Although I am still content of the current system it is.

Wait for few more months for next update, and maybe we will see the difference.

_____________________________

(in reply to quertice)

Post #: 10

RE: Suggestion on the communication model - 2/19/2016 12:06:34 AM

Marder

Posts: 242
Joined: 10/25/2013
Status: offline

I think i have a good idea.
Lets take the example with the submerged sub (A) which detects an enemy ship (X) by sonar but can not communicate with a friedly unit (B) to share:

Simply let (X) disappear from the map when (B) is selected.
Make it visible again if (A) is selected.

It does not eliminate the "problem" completely because the player can steer (B) manually towards (X) until in (B)s sensor range but it would add a lot more immersion.

What do you think?

(in reply to quertice)

Post #: 11

RE: Suggestion on the communication model - 2/19/2016 12:20:09 AM

Dysta

Posts: 1909
Joined: 8/8/2015
Status: offline

quote:

ORIGINAL: Marder2075

What do you think?

Yeah, exactly point out my question that way.

The thing is, for submarine the VLF and ULF are intensive bandwidths and hard to detect where they come from by target, and could reach quite of depth and range too. The only downside is the data transfer is extremely slow. A 5-digit command or code that usually used in 80s or older still need quite a long time to send and/or receive it.

< Message edited by Dysta -- 2/19/2016 1:22:07 AM >

_____________________________

(in reply to Marder)

Post #: 12

RE: Suggestion on the communication model - 2/19/2016 6:55:38 AM

acui

Posts: 7
Joined: 7/13/2013
Status: offline

I am glad to know that the comms model will get some improvement. And I will be really happy if I could be of any help.

(in reply to mikmykWS)

Post #: 13

RE: Suggestion on the communication model - 2/19/2016 7:45:40 AM

acui

Posts: 7
Joined: 7/13/2013
Status: offline

quote:

ORIGINAL: Dysta

I haven't actually see a strategy game that will not accurately show up your units in far distance, not even from Harpoon or Fleet Command -- their location are always exact, as well as sending and receiving command instantaneously.

In the Combat Mission Shock Force with the most realistic settings, the player can see an select all his/her units. But when a unit is selected, it will only show contacts it can see. The other contacts seen by other friend units will be sent to that unit with some delay. So when you select a M1A2 from Platoon A, the other tanks in the same platoon will be at their right position thanks to the data link. But other unseen vehicles and personals will not be at their current location. Depending on their position at the OOB and the communication methods they have, they will be displayed at the location of sometimes ago or will not show up at all. This affects a lot to the gameplay. Two units in the same platoon will react to the enemy spotted by each other more quickly. So It's a good practice to use units from same platoon together (just as in real life).

And in game Point of Attack 2, A more realistic C3I model is used. A lost-contact unit will not only disappeared from the map. It will be uncontrollable to the player and will be taken by the AI. But because the AI can not handle all the situations, this will lead to a unrealistic result most times.

(in reply to Dysta)

Post #: 14

RE: Suggestion on the communication model - 2/19/2016 8:08:16 AM

acui

Posts: 7
Joined: 7/13/2013
Status: offline

The problem is more complicate here. Some platforms were meant to be used as a whole. This includes the ASW helos and frigates, the AWACSs and fighters, ships in the same formation. They do shared contacts in real life, so it will be unrealistic if they can only see and attack what they have seen.

(in reply to Marder)

Post #: 15

RE: Suggestion on the communication model - 2/19/2016 10:36:45 AM

Dysta

Posts: 1909
Joined: 8/8/2015
Status: offline

quote:

ORIGINAL: acui
Combat Mission Shock Force

And

Point of Attack 2

I missed both of those.

_____________________________

(in reply to acui)

Post #: 16

RE: Suggestion on the communication model - 2/19/2016 1:05:19 PM

mikmykWS

Posts: 11524
Joined: 3/22/2005
Status: offline

If you don't really understand how the comms model works in command there isn't much value in going on about it. Well other than to be annoying:)

Do us a favor and post what you'd like and some save games to show issues.

This would be a smarter approach.

Mike

_____________________________

(in reply to Dysta)

Post #: 17