Commander Command Improvements -- What ? Details ?

Discuss the future of Warzone 2100 with us.
User avatar
whippersnapper
Regular
Regular
Posts: 1183
Joined: 21 Feb 2007, 15:46

Commander Command Improvements -- What ? Details ?

Post by whippersnapper »

.
This has come up often over the last 10 years. Again in the most recent "Wish List" thread.

I think it represents a major area of improved WZ game play so merits a dedicated thread.

I know there are quite a few of you out there with good ideas on this flowing directly from your specific frustrating experiences with Commanders.

First let me start with a couple quotes from the "Wish" thread.
Per wrote:The problem with the commander isn't that it is weak (I am not sure if it is), but that it is really annoying to use. Fixing the first problem without addressing the latter will just make the situation worse, not better.
Zarel wrote: I think it's both. I agree that we should do something about the latter. Ideas?
As much as I have considered Commanders over the years it is In Black Project's "Hardcore Campaign Mode Mode" that I have found a fine lab for analysis and possible remedies.... In "Hardcore" Commanders are absolutely essential to success therefore it is the perfect testbed to identify where improvements are needed to make Commanders more effective in Multi-Player, IMO.

I always like to begin by framing the situation in a generative manner:

Abstract summary: Commander Improvements to make Battle Groups Smarter by reducing baby-sitting and making more effective microing, & switch-tasking with non-combat management.

Some Key Areas:
  • - Commander Engagement Orders-Schema

    - Commander Re-supply Orders-Schema

    - Commander Repair Orders-Schema

    - Commander Retreat Orders-Schema

    - Commander Maneuver Orders-Schema

    - Commander Special Abilities

    - BetaWidget / LUA
I bear in mind that these areas are separated only for analysis convenience and that in reality they have to be smoothly integrated.

ALSO: A brief insight into where I'm going with this. IMHO, improving Commanders is the best way to make Asymmetric GPMs viable in WZ.

Now I must go through my chit and better sort it out before I can share on my end stuff that goes back to discussions I had in Pumpkin's own BBs with
several of WZ Creators input as well fans from the early days clear through RTS.net.

But please do share your thoughts because I believe anybody that has used commanders has insights of value..

Thanks, whipper :cool:

Addendum: I wasn't sure how many folks were aware of the following so I thought it best to put it forward here up-front....

Command turrets DO NOT, I repeat, DO NOT need to be of a "command" classification! In other words, it is entirely possible to have a Commander shooting laser beams, missiles, cannon shells, MG's, Flak, SAM's, rockets, BBs.... whatever !

You can have a Commander armed with Missiles instead of the usual Ruby Red Laser - lancer-strength missiles, for example, so such a Commander doesn't just lead armies, he can kick butt, too!

Anyway.... to use or not use becomes a balance issue vis a vis a Commander's native durability.............

.
Deus Siddis
Trained
Trained
Posts: 235
Joined: 18 Aug 2007, 06:58

Re: Commander Command Improvements -- What ? Details ?

Post by Deus Siddis »

Here's the best improvement you can have: REMOVE THEM.

They are a gimmick that can offer no possible value to the game, because AI is dumb, and thus AI commanders are dumb under any implementation. And it subtracts value because you lose control of the underlings, have to give up a platfrom to a command turret and have another layer of complexity to manage.

But worst of all, the commander gpm has for some reason poisoned mobile artillery. Artillery units have to be assigned to a sensor that acts as their commander, so that you can't control them individually or anywhere near precisely, and they get flanked and murdered by the enemy because you can't just tell them to retreat of get out of the way of reinforcements moving in to protect them.

However, for no reason at all, artillery structures don't have this limitation, they can fire on anything visible to any friendly unit or structure, making firebases the only sane option for field artillery. The problem is firebases require turtling and cannot go on away missions, and the weapon section of the design interface gets cluttered with artillery weapons you will never use because of their stupid assignment requirement.
Samowar
Trained
Trained
Posts: 42
Joined: 03 Jun 2009, 19:46

Re: Commander Command Improvements -- What ? Details ?

Post by Samowar »

Deus Siddis wrote:Artillery units have to be assigned to a sensor that acts as their commander, so that you can't control them individually or anywhere near precisely, and they get flanked and murdered by the enemy because you can't just tell them to retreat of get out of the way of reinforcements moving in to protect them.
Why, I just have my artillery grouped (via ctrl+number) and then assigned to a sensor. Now if I need my artillery to retreat, I just press the appropriate number key, they are selected, and I can tell them to go anywhere. Admitted, they are automatically unassigned from their sensor and stop their bombardement, so you can't bombard while retreating, but in general, it works fine that way.

However, a more general problem is that there isn't a clear division between "direct fire - internal sensors" and "artillery - external sensors". I think the sensor range of all direct-fire units and structures should be adapted to their fire range (I don't know about Railguns, but e.g. the Lancer Bunker needs a nearby sensor tower to fully utilise its range), on the other hand, indirect fire structures should only take in account the sensor informations from dedicated sensors or from their own internal sensors, but not from the internal sensors of battle units. A MG Viper's sensors are only built to tell the MG where to shoot - not to relay this information to a nearby artillery emplacement which thereupon starts bombarding the area.

Furthermore, artillery units should be able to switch between "gather information from all dedicated sensors and shoot anything that moves within their (and your) range" and "assign to a particular sensor and only shoot within its range". Furthermore, sensors should be able to switch between "auto-target anything in your range" (like sensor towers) and manual targetting (like it is at the moment).
User avatar
zoid
Trained
Trained
Posts: 125
Joined: 13 Jun 2009, 00:45

Re: Commander Command Improvements -- What ? Details ?

Post by zoid »

I already posted this in the wishlist topic, but it's worth reposting here.
The problem with Commanders is that you can't issue a command to an attached unit without it becoming unattached. A good solution might be to assign unattachment to Ctrl-click. Any unit Ctrl-clicked on would immediately become unnassigned. You would not have to give it an order to unnassign it. A normal click would select just the unit, not its Commander as well. Issuing an order to the normally selected unit would not unnassign it. Units could be micromanaged without removing them form the commander system.
My new system wuld not let you select a command group by clicking on any of its units. I don't see that as a problem, however, as you could still select Commanders via two hotkeys, the button on the hexagonal console, or by clicking on the actual Commander. :)
User avatar
whippersnapper
Regular
Regular
Posts: 1183
Joined: 21 Feb 2007, 15:46

Re: Commander Command Improvements -- What ? Details ?

Post by whippersnapper »

.

Sorry for not attending sooner - been deep in the mountains, off grid... Back in civilization for a few
so I'll catch-up here and yonder other threads.

---------------------->

Interesting. The PoV to get rid of 'em entirely and the suggestions to improve (which are good, imho).

I did a search of the BBs over the last few years and found some fine proposals by still active folk.
Wondering if I should take the liberty to aggregate them here as they are scattered all over the place
or leave 'em as is because maybe those folks have changed their minds or they may yet still choose to
express thier proposals here. Dunno yet what course to follow there....

On my end - I am adding to my proposal which takes into account using BetaWidget-LUA when it's fully implemented -
meaning it is a full new command scheme with a GUI widjet mock-up and accompaning design doc of the GPM, switch-tasking
and supporting modular A.I. relationships..This proposal is not something I'm hoping someone will do but rather I'll do it myself
thus putting it squarely in a realm other than "fulfill this wish for me" or some pipe dream with little to no possibility of materializing.
I'll hold off a few more days till I decide whether to aggregate those other suggestions posted hither and yon over the years by
insightful members of this board.

Regards, whipper :)

.
User avatar
whippersnapper
Regular
Regular
Posts: 1183
Joined: 21 Feb 2007, 15:46

Re: Commander Command Improvements -- What ? Details ?

Post by whippersnapper »

.

Widget Summary:

Did and posted first in 2004 before source liberation. Only one significant change since - the addition of a UAV schema...

I'll post the explanatory summary document l8r. I'm thinking of ripping-out the "Experience-Rank-Special Ability" relationships
from the doc otherwise I'd post it as is.

L8r, whipper :ninja:
.
Attachments
GCI v.07 GUI.jpg
User avatar
whippersnapper
Regular
Regular
Posts: 1183
Joined: 21 Feb 2007, 15:46

Re: Commander Command Improvements -- What ? Details ?

Post by whippersnapper »

.
whippersnapper wrote:.

Widget Summary:

Did and posted first in 2004 before source liberation.* Only one significant change since - the addition of a UAV schema...

I'll post the explanatory summary document l8r. I'm thinking of ripping-out the "Experience-Rank-Special Ability" relationships
from the doc otherwise I'd post it as is.

* BTW - the 2004 WZ Wiki I created still stands with the GCI 1st cut widgets & explanation from that time thanks to the Wikia Org. HERE

Here goes....

GCI GUI Design Doc and Supplemental notes v.07: New Command and Control GPMs through Commanders, UAVs and enhanced UI A.I. 2-tiered Modularity.


* Button #1: If checked Turns on GCI over-ride of that Commander's local level A.I. & implements GCI A.I. Preset set of
scripts.


* Button #2a: Pop-up list of available Field Commander Units to select from.


* Button #2b: Pop-up list of available Commanders for Reinforcement Duties. These are Commanders that are still at HQ and have not been fielded or deployed in a battle theater. These Commanders can also be assigned base perimeter defense duties with apt scripts designed to that end.


* Button #3: 'Go To Deployment' - Press to go to that Combat Group in the Game World. This button will also 'Blink Red' when the Command Group parses to the LOWEST Health Preset of your assigned 'Base-Line Unit' warning you, The General, that it's best to attend the group locally as soon as possible..


* Button #4: Pop-up list of all units linked to that commander. From this list you must assign-select your 'Base-line Unit' for the A.I. Script Trigger Thresholds (aka 'Health of Base-line-Unit')..


* Button #5: 'Mini-Map' - LOS area of that command group. Combat Groups have identity icons and can be maneuvered on the Mini-Map. Mini-Map maneuvers automatically overide local commander a.i. &/or GCI pre-sets. Once Mini-map maneuvers are completed the combat group defaults back to its last set - GCI a.i. or Local Commander UI a.i..


* Button #6a: 'A.I. Set Select' Press to launch drop-down #6b Screen. Screen #6b: You select A.I. Presets & Thresholds for that command group. After setting, close like standard window.


* Button #7: Press to launch re-sizable 'Known World Map' which includes all command groups and UAV recon penetrated fog-of-war.

For example.. You have 3 Combat Groups: #1, #2 and #3. You see those #s-icons on the MM (Mini-Map) or KWM (Known World Map) and can do the following: set-up Way-Points, Patrols, Rally-Points, coordinated offensive maneuvers from multiple vectors and at varying velocities. Basically, this is how 21st Century RL Military OPs are executed. It's called "C3" for "Communications, Command, and Control". IMO, it is thru this mechanic that you can make asymmetric tacs & strats viable in MP mode.

This would also have the side benefit of relieving some of the bogus switch-tasking strain of having to attend NON-combat tasks while
trying to control a major offensive. Or as I have said elsewhere -

"Non-combat switch-tasking is excessive if it degrades the battlefield maneuver possibilities of the game (including asymmetry) which for me is the #1 experience that all other RTS GPMs, features and GUI's should serve to enhance."

Lastly - if you have a Commander led Combat Group loaded onto a transport you should be able to navigate that transport directly from these maps. Also... you can close or roll-up like any standard window.


* Button #8: 'Go to General'.. no matter where your eyes are, by pressing this you will be taken back to wherever your Command Relay Structure (CRS) resides. (Which does NOT have to be where you have made your HQ). The CRS is essential to the GCI which enables you, The General, to act fully as a field general does. Lose the CRS & you are severely handicapped in your General Duties - basically the current status quo of the game.


* Button #9: 'Deploy UAVs'.... These are pure aireal recon-drones. A Commander gets 2 upon being built. These 2 UAVs constitute a new Commander Special Ability. These UAVs are no bigger than 25% of a Bug Body, carry no weaps, are at least twice as fast as a Bug body power plant out-put, recon both map territory and any enemy force strength on the ground or in the air, plus structs. This intel is stored in your Command Relay Structure and you need a CRS to use UAVs which can only be deployed through the GCIs 'Known World Map'. Double-clicking on the CRS will give you a Graphic display of recon-gathered intel, including specific enemy force strength.

UAVs carry their Commander's ID-Icon designation on the 'Known World Map' (or Mini-Map) and are way-pointed from there to also include entering Fog-of-War space. They cannot be built separately. They are very hard to kill but once lost, cannot be replaced. They return to their Commander upon Recon completion till manually assigned a new mission by you The General. They circle their Commander till re-deployed and this is when they are most vulnerable. They do not need re-arming pads to function.


Additional GCI Notes A:


* I.) Individual Commanders still have their own command screens. The main difference being that a set of custom A.I. scripts are player assigned in the GCI and that this provides for 2 levels of modularity that reflect decisions on a local or tactical level and those that are multi-vectored, coordinated, strategic-level unfoldments between many Commander led combat groups. Also remember - If checked the GCI option will over-ride Commander screen A.I. settings with the GCI presets.

* II.) Individual Combat Group assembly & linking are still done at the Commander level.

* III.) The GCI has a toggle for each command group. When checked the GCI A.I. script settings prevail. When NOT checked the Local Commander A.I. prevails, but all other functions of the GCI remain operational: Maps, 'Go To', re-setting a.i. scripts and threshold triggers, etc.

* IV.) To start the GCI consists of 2 Field Commander Queues and 1 Commander-led Reinforcement Queue Totaling 3 Queues. Opening up
more command queues can be tied to a success metric such as a Kills v Losses ratio.

* V.) The GCI will still recognize Combat Groups if loaded onto a transport. In fact you should be able to use the 'Known World Map' to navigate Transports loaded with Commander led Groups.

* VI.) The Player can Name his Combat Groups (&/or choose some symbol) and that is how they will show up on mini-maps, the macro-map aka 'Known World', as well as the GCI pop-up select.

* VII.) The GCI is linked to the Command Relay Structure. If the Command Relay Structure is destroyed you lose GCI access and must revert back to local macroing your Combat Groups. However, building another Command Relay makes available the GCI again.
GCI Supplemental B:


* VIII.) Left to right, 0-100%, A.I. Slider Threshold Triggers (Health of Base-line Unit) can only be set as unique increments of 10% (relative to each other). You cannot, for example, have 2 scripts set at 90%.


* IX.) @ 100% (Health of Base-line Unit) Commanders Local A.I. settings are in effect (unless you are manually controlling or have checked "ON" the GCI).

If you have checked on the GCI (for that Combat Group) @ the 90% threshold the GCI settings will prevail.

Illustrating from the GCI widget mock-up settings let's break it down: @ the 90% increment the 'Cover Vanguard' script would implement. @ the 80% increment the "Fire & Fade" script would implement. @ the 70% increment the 'Hit % Run' script would execute & so on.

Your lowest threshold setting will prevail until you, the General, actively re-address that Combat Group.

Again using the graphic example: when the combat group hits the 60% mark the "Retreat Out of Range" script will prevail no matter how far down the % health of the base-line unit (even if it's killed).

Also, at that point, there will be the visual cue of the 'Go To Deployment' button blinking red in the GCI for that group - indicating this state.

The exception to this would be if you set-up the 'Go To HQ' with an A.I. & % threshold. Using the Graphic example: The A.I. script 'Avoid the Enemy' is selected and the threshold set below 60% - let's say 50%. So if your "Base-Line Unit" health hits 50% or dies, this Combat Group will head to HQ (or a pre-assigned Rally Point) and avoid enemy engagements along the way.


* X.) You can set-up the GCI as a Template before the start of game. What you can select are the scripts and threshold trigger %s for each command queue. Once in game you can call up the GCI Template and fill in the rest once the GCI pre-reqs are achieved as defined above: Specific Commanders, base-line unit, etc..


* XI.) The GCI screen has scroll bars and can also be re-sized to suit the player's needs / tastes. Doing it as a transparent over-lay to the player's game world PoV would seem best.


* This GCI, and associated GPMs, can also be coupled intimately to re-vamped Experience - Rank Mechanics and Commander Special Abilities (I only touched upon one - UAVs; there can be others easily) which is itself covered extensively and seperately but is not essential to implementing the GCI on its own so I have not included any more of that here. There are also relationships linked to Pilot / Driver, Battle Mode that I have also removed from this doc.

* Constructive input is always welcome for I consider this an unfinished work, readily admit to fallibility and because unforeseen stuff always seems to come-up when you think you've covered all your bases in design and, of course, in the initial implementation phase (& throughout making it real, to be honest). However, permission or approval or anything associated with popularity is, frankly, NOT being sought out here (same goes for snark)..

* I guess I will next go ahead and aggregate the suggestions of others made on this topic over the years because I value them whether or not they fit into my PoV schema or not.

Regards, whipper :cool:
.
Last edited by whippersnapper on 25 Jun 2009, 15:36, edited 2 times in total.
User avatar
whippersnapper
Regular
Regular
Posts: 1183
Joined: 21 Feb 2007, 15:46

Re: Commander Command Improvements -- What ? Details ?

Post by whippersnapper »

whippersnapper wrote:.

* I guess I will next go ahead and aggregate the suggestions of others made on this topic over the years because I value them whether or not they fit into my PoV schema or not.
.
.

Part One: Commander Related Posts over the years aggregated:

Zarel wrote:Ehh. Trucks, commanders, sensors, and repair turrets (i.e. all systems turrets) aren't even considered regular turrets. The template code for them is really messed up.
T5k wrote:I want to make factories assignable to more than 1 commander, and since i usually have 15+ commanders it is really kinda hard to get reinforcements to them, so i though of making dinamic factories - factory closest to the commander would supply units to it, and commander would have it's own "build" menu - so you would basically order what units you want to have in it's group and how many of that kind. When a unit gets destroyed, factory would automatically make it's replacement :)

even more micromanagement automated and more time to focus on strategy :D
fulano wrote:I just want to say first I really like the rebalance, except I never can find the lancer AT rocket in MP games.

I would just like to suggest a simple interface change. When units attached to commanders are too damaged they return to the nearest repair turret and get fixed and return to battle.

When a unit in a number group gets too damaged it goes away and never comes back unless you specifically go get it again.

Why not have these units automatically return to their number groups after they are repaired?
Zi-Chan wrote:
Don't know, if somebody already suggested that, but i want:

Unit Formations:

Sometimes you want Units with LongRange-Weapons behind Units which don't have that high Ranges. It doesn't make much Sense to have LongRange-Weapons to be vanward, because sometimes they don't fire, e.g. Arch Angel Missilie-Units have to drive away first to fire the Enemy. Or Units without any Weapons, e.g. Commanders or Sensors. They shouldn't be vanward, they just get destroyed.

Especially when you have a Commander-Unit controlling it's Troop, there is the Problem, that if you want to move the Commander to get away from Bombardment, the whole Troop moves too. :/ One Click should only highlight the Commander Unit alone and double click should highlight the Commander Unit together with the designed Troop.

Control: Highlight Units, rightclick one of them and choose the desired Formation. They deploy immediately. Doable with a Commander too, then you just right click on the Commander and do the same.
unmorph wrote:
........ And the ability for commander units to learn tactics such as "hit and run", "ambush", and so on. Commanders could communicate with each other for multi-unit tactics. Like a situation where one commander wants to withdraw for repairs and signals a reserve commander to replace it. Or coordinated multi-unit base attacks.
gene69 wrote:
thing i would like to have.........

-> better way to use commanders. i would like make then drive unit in specified order (machinegun/flamer first followed by artillery units) i would like a hit and run button. so scripted behaviour.
TheDaze wrote:
I would like to point out two ideas that somehow got lost in the sands of time.

1. Better LOS for commanders. My idea was quite simple: that the mk version of the commander turret would increase the sensor range. mkIV would then have the range of a Wide Bandwith Sensor turret, or at least close to that.

There have been other ideas, as shared LOS od assigned units, but I guess those would be much more difficult to achieve.
fisk0 wrote:I recently noticed one thing, which is certainly intended, but makes no sense to me - that mobile sensor turrets only work if you 'micro manage' it, clicking on every single unit or building it should target.

I would have assumed that when not given a specific target it should work like regular sensor, automatically assigning targets (if not set to 'hold fire'), just like any combat vehicles automatically fires upon enemies in range if they are not told to do otherwise.

They do not seem to work when assigned to Commanders either, if I have a vtol strike turret vehicle assigned to a commander, the vtols won't do anything, even when the commander assigns a target, you have to select the sensor unit, detach it and asign it a target manually to have the vtols attack.

I really think sensor turrets should work like sensor buildings when set to 'fire at will', so you could for example put one on a hill overlooking a ravine and targeting the units passing below, get artillery support and all that, without having to construct sensor building everywhere.
cruise wrote:
Not requiring the commander to have LoS to a target would be a great help - if it has been sensed or seen by any member of the assignees then the commander should be able to assign it as a target.

Maybe the higher level commanders could use other teams sensing data, so that they'll target units sensed by anything you control (with reason, obviously)?

That would be a good help...
Kayiaxo wrote:Hello,

As most here on the forums I think Commanders are underpowered. I don't think they should be damage dealers but rather hit takers or add another ability.

I thought that maybe the commander could also include the sensor turret as well since it fits his job. He has to find and appoint ennemies so the units assigned to him can target them. And then remove the sensor unit as seperate unit.

If that fails I just suggest to give him some armor/hp boost and giving him more mobility but reducing the damage he deals.


As for heroes I think they are a little underpowered for what they are. Heroes are hard to get and gamma campaign just rips your units appart. I've never managed to get a hero long time since the ennemy aims at him and wont leave till he dies.

Heroes are HEROES, they are godlike fighters who manages miracles. I suggest 2 things for heroes:

1. DEcrease their fire range but INcrease armor/hp/accuracy/speed more.

OR

2. Give them a special ability like: a 1-hit KO every for exemple 2000 damage dealt.

Giving them a % 1-hit KO on each hit would make him über with a machinegun, so I thought rather add up the damage and once he reaches the damage done, the next hit he destroys for sure the target.

Please give input.

NEXT: Part Two
.
User avatar
whippersnapper
Regular
Regular
Posts: 1183
Joined: 21 Feb 2007, 15:46

Re: Commander Command Improvements -- What ? Details ?

Post by whippersnapper »

.
Part Two: Commander Related Posts over the years aggregated:
Watermelon wrote:In a more realistic implementation of commander/officer/squad leader,the members of a squad/group/division should not simply copy the order of the commander,rather it should prefer the target,destination commander assigned over its own target,destination, currently it simply overrides the individual units' behaviors,that's why the commander now is nothing more than a laser target designator/focus fire assigner.

The commander needs improved protection against all forms of damage,unit preservation always plays an important role in a tactical strategy game,ignoring it will make the game into a 'mass more tanks' pitfall,that's exactly one of the problems of wz's mp.
Kacen wrote:I think giving the commander a cannon like they did in 1.12 would be cool.

Problem is in that it is only a cannon so by T3 it is outdated and the Elite Commander is used. I think at that point something like a rail gun commander is needed.

But yes, I think commanders need more than a laser designator.
Deathguise wrote:
Yeah, the reason they do a miniscal amount of damage is because weapons are not allowed to do 0 damage, personally id rather see more durable commanders than heavily armed commanders - the reason being the first command turret has about 1000 body points the machinegun has 75, now if the command turret does as much damage as a machinegun you might as well build 10 commander units and use them as combat units altering T1 game balance.

The Elite Commander if im not mistaken had an EMP weapon and imo it was a good combination.
Watermelon wrote:
A commander or any 'super' unit need to have superiority or leadship,superiority makes unit either very deadly (kill foes in mere seconds or armed with WMD) or extremely durable(can take several punches before perish),while leadership grants nearby friendly units massive bonuses,though wz's commander has neither 'trait'.Also the uselessness of experience/rank makes commander/experienced units less desirable.

Players should be rewarded for keeping their units alive,but in warzone it's 'quantity over quality',mass more tanks will always make more sense than maintaining elite units.

I agree about the fragility of commander,any player will try to FF the enemy commander on the spot,so the commander needs more HP or maybe revivable,the advantages accumulated from destroying outnumbered enemies or keeping commander alive over time wont get lost as soon as your commander is killed.
NucNut wrote:What i'd like to see is a split between the designator/commander/artillery spotter units, as the commandwerer as is is a little to varied in this regard.

*Make basic commanders as a normal unit with a slightly different turret, eg: has several whip aerials ion it to represent improved comms, but is essentially the same as all the other members of his squad (maybe a little more accurate/armour/etc.) you see this with modern day command tanks

*Make scout/spotter commanders as a dedicated "assign fire support" unit, but you can assign base artillery or mobile artilllery to him, would have an increased sensor & designator range to represent this (Maybe deleting the command turrets projectile effect could represent this - its just a camera)

Most of all, i'd just like to see commanders as normal nits with a slight tweaking to their stats, because in RL, commanders are not just a designator, the have advanced communications gear and targetting equipment, but they are still armed. Having the commander as a multi-turreted unit, with a light weapon defending the commander while the designator paints a target would be ideal.

Commanders a great thing in WZ2100, but they should be made a little more realistic - bring on the command tank!
EvilGuru wrote:
Work is underway to make commanders more useful. One of these will be the concept of global experience. This means that if a unit is assigned to a commander and the commander has a higher experience level that itself it will take on the commanders experience level for as long as it remains assigned to the commander. Currently this only applies when one is receiving damage, but when I revamp the experience system (more damage + more accuracy for higher ranks).

There was also some talk of multiple attack vectors for units assigned to commanders. So if you have 5 bunker busters and 5 heavy cannon tanks assigned to a commander and leave them to it the BBs will attack any structures first and the tanks will go for any units first.

Thus you can assign specialist units to a commander and have them always attack 'the right thing' with all of the advantages of focused fire.

As for the Mk1 => Mk4 my plan is this. Currently the formula for commanders is 6 + cmdrLevel * 2. I plan to change this to: 6 + cmdrLevel * 2 * cmdrMk.

I would be interested in knowing what you think of these ideas.

Regards, Freddie.
ouch wrote: the biggest problem I have with commanders and vehicles with sensors, and sensor towers. Is that any unit assigned to them loses the ability to think for themselfs. What I mean by that is if an enemy comes in range they don't fire on it unless whatever structure/unit they are assigned to orders them too. This happens even if they are under fire and the unit/structure thier assigned to doesn't have a target in range to shoot at. at the very least they shoot stuff when the structure/unit they are assigned to doesn't have a target. Ideally (expecially commanders) the unit should also say "hey, I'm under attack! heres the guys location." and beable to act as a radar extension for the group and/or the structure/unit they are assigned to.

Perhaps commanders should permanently have the ability to see the other units radars as it's own. that way anything in range for that unit to fire upon would be in range for the commander to target.
Molotov wrote: I believe that the commanders should be more VTOL friendly. What I mean is, I should be able to assign a VTOL Strike sensor vehicle or a wide-spec sensor tank to a commander, and let the commander select targets in the sensor's range for it's VTOL aircraft to attack. Preferably, I'd like to see my Air Commander telling each VTOL to attack a target it is effective against, meaning if the commander's MAIN target is a base structure, he will realize this and send the assault cannon VTOLs (if the player was retarded enough to build any) against 'Borgs in the area.
.
kipman725 wrote: commander turet is one of the few used through T1, T2 and T3 as such the number of body points is not enough past T1 but having too many body points would unbalance T1. The simple solution would be to make the commander turret upgrades increase commander arrmor aswell as increasing the controll limit. The cannon commander was a good unit but dosn't rearly fit the game, instead I prupose that the commander be given EMP powers when emp cannons are reserched paralysing the designated unit as this fits in with the designator on the commander.
psychopompos wrote:this is all 2.0.10, maybe some fixed in newer betas
......................

the lack of communication in commander assigned units bug, where one assigned unit sees an enemy, but gets no firing support from other units that can see said enemy.proper shared LOS would fix this.

another is the *sit-scratching-own-arse-while-being-killed* bug, where a com-assigned unit just sits there, doing nowt, just cause the commander doesnt have eyes on with the target again, proper shared LOS would fix that

another , is that if your commander is damaged & retreating, the units behind it will get it killed, as they piss about like decapitated chickens, cause they are too dumb to get outa the way.healthy yields to retreating could help that! retreating unit collides with healthy, healthy told to shift, unless strict hold position. healthy unit collides with retreating, healthy evades. its a pain putting the effort into a commander in skirmish just to have it die pointlessly, but i suppose traffic rules could be part of pathing.

you cant even move other units in as shields, as the projectiles(rocket, bullet or shell) will simply travel through the shield units.
did you know scourge missiles & vtol bunker buster can go through walls without hitting them, still striking their target?

that said, i find commanders with 8/10 along with 8/10 unassigned expendable units (viper HVC & viper BBR) quite effective for smashing bases especially with arty support turned on.and the fact they act as repair destinations is pretty bloody useful in an assault.
elio wrote:Hi all

You all know the feature of assigning a factory to a commander. My idea: You have a group with a commander unit and when one of the members die, the assigned factory produces automatically a new tank (at best the same type which died). I propose a additional button at the commander menu (right click on a commander)

What do you think?

elio
Verminus wrote:For those of you who have English as a second language, I will try to explain carefully what we are talking about:
  • We would like to be able to assign a factory to a commander, then tell the commander which units we would like in the command group.
  • The commander would then order the factory to make the units.
  • The factory would make the units and they would join the commander as normal.
  • When a unit in the command group is destroyed the commander then orders the factory to make a replacement.
edit : Example;
  • You assign a factory and tell the commander you want 6 lancer cobra half-tracks, 4 heavy cannon python tracks and 2 Heavy MG python half-tracks.
  • The factory then builds the units and they join the commander
  • 2 of the lancers and a heavy MG are destroyed
  • The commander orders the factory to build 2 lancers and 1 heavy MG to replace the missing units
Hope that helps explain it for you. ;D
Troman wrote:I think this is an interesting idea.

I'd like to summarize what I think would be ideal (IMHO): factories are not assigned to commanders anymore. You select type and quantity of units you want to have in commander's group and when the group is not full some routine makes factories (those ones best suited for the job, without defining what 'best suited' means right now) start building units.

But sometimes you might want to quickly build some other unit, therefore it would be nice to have a way (just a button in factory menu for example) to exclude a factory from set of factories that produce units for commanders, this way it is always available for any other needs.
EvilGuru wrote:While I like the idea I see it as mostly a use-case for commander scripts (?topic=1233.0 ). A script could easily perform such an operation along with so much more.

Currently, I think the biggest changes to commanders needs to be to change the relationship between factories and commanders. At the moment it is many-to-one, in that many factories can service a single commander. I would like for it to be many-to-many; so a single factory could be assigned to multiple commanders. When a unit is build the unit will go to the commander with the fewest units.

Regards, Freddie.
Verminus wrote: .....................

Oh and while we're talking about repairs, could something be done about the units that do actually find their way to to a repair yard, (miraculous as it is), but end up just sitting there not getting repaired? This is especially annoying when it happens to units assigned to a commander and somewhat counteracts one of the main reasons for using commanders in the first place. (That damaged units go get repaired and then rejoin the group without you having to micro-manage.)
psychopompos wrote:they are called commander cause they relay orders, keep your units at the front you want them at rather then a specific repair station's way-point marker. they also make your units focus fire on one hostile at a time with out you micro managing a group that breaks up & go... god knows where, as they retreat with damage.

if your keeping groups (possibly specialised in some way) together, and using strategies & tactics rather then brain-dead swarming, commanders are much more useful then the standard group commands. an ability i feel almost lost without in C&C or total annihilation, although in wz2100 more emphasis is placed on unit survivability.

if anything, giving them a wide area guard, that allowed some level of pursuit. but calculate survivability threshold probabilities based on the area of the map its in, which then cause it to return to where it was if the risk is getting too high.

in the campaigns, the commanders i build in cam1 (and alot of my ranked units that i assign), are the same ones killing nexus at the end.
meaning they are noticeably stronger then unranked units. which is useful

thats allot of unit exp, let alone resources down the drain.......
NEXT: ??? Maybe a Part Three......
Last edited by whippersnapper on 25 Jun 2009, 05:06, edited 2 times in total.
User avatar
whippersnapper
Regular
Regular
Posts: 1183
Joined: 21 Feb 2007, 15:46

Re: Commander Command Improvements -- What ? Details ?

Post by whippersnapper »

.

Part Three: Commander Related Posts over the years aggregated:

kage wrote:building on your idea, perhaps units under a commander will retarget to an active structure or unit once the commander "notices" that it is not a threat. or maybe veterancy could play a role in retargeting away from these, or both (any middle rank unit would target something else if available, and if a commander is at least in one of the middle ranks, it'll apply that intelligence to units under its command.

that brings up another thought: commander smart targeting -- if you tell a commander to explicitly attack something, it will perform the following steps:
  • if it's a structure or stationary unit (which is thus easier to hit, especially with the new collision detection system), then the commander will order just as many subordinates as are needed to destroy the structure in one volley, while retraining the commander laser on another target (thus repeat step one with the new target)
  • if it's a moving unit keep the aiming laser trained on target, do a very rough calculation of how many units would need to fire simultaneously to score enough direct hits to destroy it at once (this may be more units than are available, in which case, just use the rest), and if there are any spare units after this calculation, have them hold fire. after the 'active' units fire a volley, if the target remains, have the spare units all fire at once (switching between fire groups each volley). if the target doesn't remain, then there are some units that are immediately available to fire at the next target.
with the above, commanders could substantially increase efficiency -- specifically in the kills/minute metric, and, if global ammo is implemented, units under a commander would have vastly better ammo efficiency.

RBL-4NiK8r wrote:Well for the most part Commanders were useless do to the fact they die to easy, yes that could be fixed, but when you kill one all units that were hooked to that Commander become useless and basicly sit there till you go round them back up.

Now they were nice for targeting long range fixed a mobile artty that you had, or use them for CB fire, hell I recall one game where I took out all enemy CB towers with them and then let my artty go to town. In the SP game they were a nice thing to have at times, but once you learn most of the hot key commands you can do just as good without them, and in MP games they are just basicly cannon fodder. Also in MP games they dont last long enough to get high enough rank to make them worth usings.


4nE
Deus Siddis wrote:I think the main things that needs to be fixed with commanders also needs to be fixed with forward observers and these really need to be fixed before any balancing or additional features for commanders can be added. They be:

1) Assigning combat and artillery vehicles to them should not be necessary for those underling vehicles to benefit from them. If you drive a sensor-turreted forward observer into sight range of an enemy that is within firing range of stationary artillery emplacements (but beyond their natural sight range) they will fire on it even though you did not (and I think in fact cannot) assign them to that observer. Why shouldn't mobile artillery be able to use this same simple system? And why shouldn't commanders do this as well, automatically increase the effectiveness of fighting units within their range without them being assigned to that particular commander?

2) You need to be able to command underling units separately, just as you normally can, even though they are assigned to a commander or observer. Otherwise you have units marching into enemy fire, traffic jams, just ridiculous chaos that is totally unnecessary because you are not allowed to manage any of it.

Until these are fixed, mobile artillery will be assigned to forward observers only because the game forces you to for them to be useful as artillery at all and commanders will continue to go unused because they are pointlessly and massively detrimental to commanding (oh, such bitter irony) your forces effectively.

This was one of only two real problems I had with WZ back in the day, if it could be fixed that would make the game much more intuitive and fun, as well as eliminating all of those "#$*&%!" moments where you can do nothing to stop your artillery column as it takes a shortcut into a tank battalion while trying to keep up with their forward observer.
RBL-4NiK8r wrote:Well Daze most of the things you can do with the Commander you can do on your own, like bringing units up to the front of the battle, you either build factorys closer, or click on the number outside of the factory and move it to a better spot, and I did both thru out a game. As for repair factorys I used them sometimes, but this had a lot to do with the map or how much power I had, most of the times I used them to recycle units not repair them if you can make more, well thats for MP games SP games are different, but you can also adjust the way point for Repair factorys to so they come up near the battle. It could be just me, but I could micromanage my units on my own then I could with a Commander.

If I was doing a T3 game all my trucks would be under 2 or 3 hot keys depending on what all I was doing at the time, sometimes I would pull 1 truck off from a group to do single set of tasks. Then I would use 2 or 3 for my combat units, say my main force was HC Mantis Hovers, but I had a few Bunker Busters mixed in, so I would ground them up by them selfs, or with VTOL's that I might be using to take out Sensor towers or Hardpoints. Its been awhile but on the map Teamwars I could have a full base with 5 Factorys, 5 Research Centers, and 5 Power Gens up and making tanks around 4 mins and 30 seconds. In T1 no base I could have 50+ Twin MG Half tracks at the 5 min mark, and found any holes in your defense before that so I know where I can sneak my main force into your base or extra oil wells. I may not win with them, but that was not always my goal, most of the time it was to slow you down and distract you from something like me getting Lancers at 8-9 mins.

Anyways the thing I am getting at is that the means are in the game to micromanage your army by your self, and do a damn good job at it, granted it might take awhile to figure out all the hot keys and some of the tricks to make you faster, but thats how we won MP games in the old days.


4nE
RBL-4NiK8r wrote:Daze dont get me wrong I think Commanders are a very big part of the SP game, but when your in the heat of battle in a MP game you dont have time to deal with them. Now IF they were made a lot stronger then they are now, then maybe they could work, and also fixing some of the problems with them like leaving an ant trail of new units if its kills, I think all the units should go to the spot of the commanders death or at lest the next vehicle close to the dead commander.

I also think it should be given more abilitys make it like a Truck so you can build with it, and do basicly the same as you could with any other vehicle in the game. Like it could have Radar or some kind of weapons system to but only for short range.

In SupCom they have a Commander type unit for all sides, but its higher in the tech tree, and a real bitch to kill, granted its not the same as the one in WZ or have the same abilitys, but having something thats able to take a lot of hits would be a nice touch. Say when a Commander comes into play it could not be one hit by ANY T1 tech, and by the time its had all of its upgrades it cant be one hit by anything in T3. Just some ideas.


4nE
fryBASS wrote:Yeah I usually just end up giving up on commanders because it ends up being so much easier to control your units normally. Again, the part about the commander being in front, it's just horrible.

Due diligence done for now here. Back to working on "War School"

Regards, whipper :cool:
.
User avatar
whippersnapper
Regular
Regular
Posts: 1183
Joined: 21 Feb 2007, 15:46

Re: Commander Command Improvements -- What ? Details ?

Post by whippersnapper »

.

The GCI does not disassemble and reassemble individual Commander A.I. as we know it for reasons covered in the design doc.. ALL of the suggestions over the years that I've aggregated in 3 Parts here are piecemeal attempts to do just that. Out of that collection of posts I left out my attempts to deal with individual Commander A.I. I will begin dealing with that now because there are GCI postulations that also deeply impact HOW individual Commander A.I. is handled. The most obvious new mechanic with this impact is the introduction of UAVs. UAVs can have the most positive impact on reassembling Commander A.I. along first tier modularity that flows seamlessly into the GCI's second-tier.

First we must briefly parse current individual Commander A.I. and then touch upon a new framework in pseudo-code (all the while keeping in mind just what UAVs represent in Recon of Intel and it's subsequent "Fusion & Distribution" as "Actionable, S.M.A.R.T, Maneuver" without baby-sitting or hyper-microing. This also impacts HOW Fog-of-War is handled.)

I.) WZ's current Command Interface supports 17 defined "Behaviors".

Only 4 can Interact simultaneously on the field at any given time.

Just 2 are incrementals to action (Retreat at Heavy or Medium Damage).

And 3 are Fiat Commands: Go for Repairs, Go to HQ, Go to Transport.

So however many A.I. Presets make for an elegant GUI (< 17 or 17+), something that play-testing must reveal, there are some other points to explore.

II.) Now let's use pseudo-code:

* Check settings:

- (1.) What retreat level

- (2.) What aggression level (return, hold, ffa)

- (3.) What movement can I perform (hold,guard,pursue)

* Check current objectives:

- (1.) Am I moving somewhere

- (2.) Am I moving to attack

- (3.) Am I targeting/attacking

* Check Current Visibility

- (1.) What enemies are in sight

- (2.) What enemies are in range

- (3.) What enemies can I put in range with a small movement. Take moveto{} as its default zone, it can 'move' say 8 units away to engage.

* Check current locality

- (1.) Are the enemy buildings in sight

- (2.) Are they in firing range given my orders

- (3.) can I put them in firing range

- (4.) do I have permission to fire

* Check Status

- (1.) am I under fire

~ (a) How do I respond within my settings

+ (i) Complete current action

+ (ii) Disengage building and respond

+ (iii) retreat to point A / safe zone

* (2.) Whats my armor

~ (a) head to repair facility

~ (b) whistle up a repair unit

~ (c) head to a repair unit


* That's not precise code-wise but you get the drift.

* NOW THIS is KEY

* Once a UNIT is assigned to the commander, you cut it's processing over head considerably. Slave its 'thoughts' to that of the Commander, and share its tactical data. This means the Commander knows what its units can see and it's less likely to allow 'stragglers' to be picked-off. The slave units move to where the Commander is, using its pathfinder routine (with some flex built in) and target what the commander says.

* This would shift you more toward a parallel type of AI rather than multi-thread. ie. one larger routine doing tasks for many children, over each child having its own routine.

* Each status check for the Commander has a different weighting value which determines their sequence. Adjusting the AI's 'personality' would also reflect in that sequence. IE a reckless 'human wave' type ai would set his ai to do or die and this would start ignoring the amour check.

* Extra status checks would need to be employed. Things to keep his tactical-group together, ie slow down all units to allow the slowest units to keep up. Telling AA units to spread, or keeping heavier indirect fire units to the rear of a formation.

* Tricky, but quite possible.... no ?

* This also leads us to "Priority Threat Assessment"... but we'll save that for l8r.

We'll return to this by & by, whipper :cool:
.
andron
Trained
Trained
Posts: 69
Joined: 25 Jun 2009, 14:21

Re: Commander Command Improvements -- What ? Details ?

Post by andron »

hmm i dont like retreating commanders that take all units with them from the fight.

I would suggest change commanders to non combat units who cannot be attacked.
Per
Warzone 2100 Team Member
Warzone 2100 Team Member
Posts: 3780
Joined: 03 Aug 2006, 19:39

Re: Commander Command Improvements -- What ? Details ?

Post by Per »

What we really need is some implementation of swarm behaviours, in which the swarm becomes something smarter than the sum of its parts. The commander can function as the brain of a swarm, to which you can allocate various types of units and it will make appropriate use of them.
User avatar
whippersnapper
Regular
Regular
Posts: 1183
Joined: 21 Feb 2007, 15:46

Re: Commander Command Improvements -- What ? Details ?

Post by whippersnapper »

andron wrote:hmm i dont like retreating commanders that take all units with them from the fight.

I would suggest change commanders to non combat units who cannot be attacked.


What I have been in the habit of doing is assigning 2-4 Borg Repair Mechanics to each Commander led Combat group so that units
are repaired "in the field, continuously" and never retreat to Repair Facilities, thus side-stepping the short-comings raised.

I do not not think making anything that cannot be attacked is viable in WZ. Even the land itself should be subject to attack, IMO, as in deformable terrain - but that's another subject....
Per wrote:What we really need is some implementation of swarm behaviours, in which the swarm becomes something smarter than the sum of its parts. The commander can function as the brain of a swarm, to which you can allocate various types of units and it will make appropriate use of them.
That is an avenue rich in possibilities - not the least being the "training" aspects.

But before I comment any more I'd like to offer a couple references on "Swarm-Based A.I." for the uninitiated:

Swarm AI: A General-Purpose Swarm Intelligence Technique HTML version of the .PDF d/l HERE In 3 short pages this doc covers the concepts clear as day.

Using Particle Swarm Optimization for Offline Training in a Racing Game ..... this covers the application of the PSO algorithm in a game quite nicely in 2 pages.

We'll come back to this. It merits closer examination, all the way around. Brings to mind Peter Molineux's game A.I. work introduced in the first "Black & White" game back in 2001 and further evolved in "B & W 2". I'm thinking that on-the-fly, in-game, "Reinforcement Learning" would be better than the "Off-line Training" of the PSO algorithm.... but I need to cogitate on it some more in either case.

Regards, whipper :cool:

NEXT: I will share the results of a survey I conducted in the WZ community back in 2004 on the subject of applying Reinforcement Learning A.I. Techniques to Commander A.I. Far from being strictly scientific the survey nonetheless yielded some interesting results that informed my approach to the GCI in MP AND individual Commander A.I. in MP as well. Again these results reflected, starkly, the differences that must be kept in mind between the 3 Modes of game play - Campaign, Skirmish & MP - and their distinct audience tastes-appeals.
.
.
andron
Trained
Trained
Posts: 69
Joined: 25 Jun 2009, 14:21

Re: Commander Command Improvements -- What ? Details ?

Post by andron »

whippersnapper wrote: What I have been in the habit of doing is assigning 2-4 Borg Repair Mechanics to each Commander led Combat group so that units
are repaired "in the field, continuously" and never retreat to Repair Facilities, thus side-stepping the short-comings raised.

.
the problem with every solutions ist that:
The Computer normally focus its fire on one unit, and when he focus on the commander it will either quickly die because the repair bots are not quick enough or when the repair bots are extreme quick he will catch all the fire, or he will be able to retreat quick enough, in that case it not only need repair skill but also unearthly healt points on the commander turrent then it will work that you will mostly never loose a commander.
Post Reply