Matrix Games Forums

The following is a somewhat lengthy post - apologies in advance for this.

When giving orders to lower-level formations, we generally have a specific reason for a particular tactical/movement order (for instance, we would move units to a particular hex to control/inderdict movement along a route overwatched by that hex). However, this is not incorporated in the actual orders given and executed in-game; the units given this order will simply execute it (with some flexibility based on SOP) but will not incorporate any awareness of the intent behind the order. While this largely works for simple situations, in cases where orders are made obsolete by changing circumstances it does not.

One example based on my understanding of "mission orders" would be a unit given an order to defend a bridge in order to prevent an enemy force from advancing across the bridge. The unit would be given a movement order to the bridge in this case, with the objective ("in order to") statement indicating the reason for the movement/tactical task. However, if the enemy force or a portion of it crosses the bridge first and the unit encounters them before the bridge, by some interpretations the original movement order would be rendered obsolete. A unit commander that had properly understood the commander's intent would presumably attempt to block/disrupt the enemy advance at the point of contact, rather than continuing to push towards the original objective (although this could change depending on the expressed intent of the order, such as a directive to destroy the bridge as opposed to blocking movement along the road passing over it).

Returning to the context of FCCW, this framework is largely, if not entirely, absent from the game in its current state. Although units have parameters for tactical initiative and their behavior can be adjusted via SOP settings, they are incapable of autonomously generating new pathing orders in response to changing circumstances (in other words, although they may withdraw from a movement order upon encountering the enemy, they will not then re-path to continue to their objective or achieve the commander's intent). I think that it would be interesting, and would definitely be more realistic, to incorporate a higher-level AI into lower-level formations to enable some of this behavior.

While I do agree it would be great to have more options and parameters for units, I also believe there should be a limit to the AI-fication or autonomous behaviors of units when given some macro directives. That would counter some of the tactical features of the game.

However, if the enemy force or a portion of it crosses the bridge first and the unit encounters them before the bridge, by some interpretations the original movement order would be rendered obsolete. A unit commander that had properly understood the commander's intent would presumably attempt to block/disrupt the enemy advance at the point of contact, rather than continuing to push towards the original objective (although this could change depending on the expressed intent of the order, such as a directive to destroy the bridge as opposed to blocking movement along the road passing over it).

So for example, on this I would say it would be great having an order or status called "Defend", where the unit is given a hex to defend with a radius of 1-2 hexes; meaning it would go to that hex and withdraw under its SOP conditions to a hex within that radius and best LOS / cover towards the original "Defend" hex. That would be simpler to implement and still avoid an over-simplification of the game mechanics.

That said, right now you could go over the area with a Move Deliberate order, that will make it stop and deal with the enemy units on its path. That, combined with a good synchronization to the command cycles can go a long way.

Although units have parameters for tactical initiative and their behavior can be adjusted via SOP settings, they are incapable of autonomously generating new pathing orders in response to changing circumstances (in other words, although they may withdraw from a movement order upon encountering the enemy, they will not then re-path to continue to their objective or achieve the commander's intent

I would love more parameters and settings, but again I would be very careful to allow autonomously generating complex new orders. For pathing, currently the units will do a re-path to the original order after scooting due to SOP.

We would love to add more of this intent type logic to the game for both the player and the CO (Computer Opponent). As noted a bit by DIVM, this becomes a double-edged sword, and more "control" can lead to more frustrating outcomes and behaviors the player does not want. A player trying to set up or predict what the unit needs to do in any situation becomes complext and cumvbersome really quickly in this type of game. Some folks want to tell units what types of targets to shoot at, what ranges to use, or where to go when condition X happens, but those matrices of event combinations are exponentially large and almost impossible to code, and they will not work most of the time. My best example is "only shoot at tanks within 2000m". What is the enemy tank unit stops at 2500m and starts shooting at you? What is a recon vehicle drives by, or an IFV closes to 1000m? Now we need this ever-growing list of conditions and exceptions that ultimately enable the CO unit to do what it needs to do as configured.

Granted, there are areas that need improvement and could better capture the commander's intent. Having more mission-based orders and control measures would be a step in that direction. Being able to order "movement to contact up to phaseline Able at 08:30hrs" provides more control and, when coupled with SOPs, improves flow and control. These are features we hope to improve and add in the future to gain that overarching intent and keep the players out of the weeds with micromanaging every squad and vehicle.

Keep the ideas and comments flowing. This is an area of interest to the team.

I definitely see the potential pitfalls with adding these features - no argument there.

As far as suggestions go, an algorithm that considers some/all of the following when choosing a withdraw location would be a good starting point:
- Line of sight to the objective (if the task is to control a particular area): this can be quantified based on the LOS value from various hexes to the objective hex
- Cover/concealment: this is (I believe) already factored in when choosing a suitable withdraw location
- Effective weapon range of the unit: hexes meeting the above criteria with shorter ranges to the objective would be weighed higher

In terms of the added interface elements, a dialog or other UI element allowing the player to designate a particular area, select from basic mission tasks (secure, control, etc.), and potentially assign priorities would be useful. In this example, the "secure" task would mean that the unit would attempt to actually enter the objective hex, while a "control" task would use the above criteria to command either the objective or the approaches to the objective.

This is limited to defense so far - offense with dynamic orders adds complexity and would probably require more toggles than defensive orders.

I will say that having the ability to (at the least) not fire on recon units while engaging all other units would be a game-changer; a number of my scenario playthroughs have had units in cover engage a single recon unit and immediately get plastered by artillery. And (on a related note) it would be great to introduce a feature where small units (1/2 subunits) have a chance to not even spot the enemy engaging them if all subunits are destroyed within a short enough time. This would be a partial step towards addressing the above issue with recon units being almost immediately destroyed, but still able to pinpoint the engaging unit's location and call in artillery before actually being taken out.

Zumwalt_446 wrote: Wed Feb 04, 2026 2:35 am In terms of the added interface elements, a dialog or other UI element allowing the player to designate a particular area, select from basic mission tasks (secure, control, etc.), and potentially assign priorities would be useful. In this example, the "secure" task would mean that the unit would attempt to actually enter the objective hex, while a "control" task would use the above criteria to command either the objective or the approaches to the objective.

This is limited to defense so far - offense with dynamic orders adds complexity and would probably require more toggles than defensive orders.

I will say that having the ability to (at the least) not fire on recon units while engaging all other units would be a game-changer; a number of my scenario playthroughs have had units in cover engage a single recon unit and immediately get plastered by artillery. And (on a related note) it would be great to introduce a feature where small units (1/2 subunits) have a chance to not even spot the enemy engaging them if all subunits are destroyed within a short enough time. This would be a partial step towards addressing the above issue with recon units being almost immediately destroyed, but still able to pinpoint the engaging unit's location and call in artillery before actually being taken out.

I would support all of these suggestions!

Especially the one about not attacking Recon units; I made an specific feature request to choose the range of attack for different kind of units (so for example you wouldn't attack Reccon till they are next to you) or aThreshold parameter to not start attacking a unit below certain parameter (so you could have your big units hidden till the real threat shows up.

One important issue right now is that Reccon units are often used as a bait instead of actual reccon. And this, among human players, forces the other players to play that way. Because what's the point of spending double the amount of time going through that forest when you are going to be blown up as soon as you touch open field by a tank battalion 3km away? You might as well save time, bait them out, and start bombing them with artillery earlier.