Iluvalar & Shadow Wolf TJ's *Commander Mods*- Need Play Data

[Rman Virgil]

.

Just another day and I'll be able to go hands-on; get my feet wet with testing.

In the meanwhile I've been thinking about how to structure my 1st battery of tests such that they follow a scientific protocol and can yield valid, useful data.

I will share one such experiment structure later today for critique; is it sound or is it flawed and if flawed, how so.

Curious - has anyone followed a similar course? And if yeah, what was your structure & context?

Also - if you haven't yet done so, can you think of any such experiment set-ups that I (or others) may be able to pursue for useful data harvesting? (The more detailed, the better.)

Be sure to keep the following in mind:

....

What is in this test:

stat wise :

COST : 250 => 190

HP : 1000 => 580

conceptually wise

Python medium => Cobra HMG.

That should make the commander just slightly OP when using full heavy bodies, but should make it usable in lighter body situation.

...... I need commanders deployed as soon as you have the tech (or cobra ^^).

...

I need data to continue :

1) How much units are really in combat ?

2) What level do you commander reach ?

3) What kind of unit you plan to use ?

.

~ Thanks, Rman

.

[Rman Virgil]

.

Earlier today, Shadow Wolf TJC & I had a conversation about the testing which explains 2 different, equally valid, protocols and areas of valuable data gathering focus. The comments shed more light on how to proceed in practical terms that may be of use to others presently on the fence about lending a hand.

Here are the exchanges as they unfolded.

======>

I'm just worried about how overpowering the ability to provide the experience bonus to so many units at a time would be. A 5% bonus may not sound like much of a bonus for an individual unit, but it adds up pretty quickly when it's applied to, say, 20 units. A 10% bonus would add up even quicker, and it's not very difficult to level up to Green, or even Trained. See why I'm concerned?

I do now...and it's a good valid point.

Only testing can answer that ultimately ... and if it turns out to be the case then you dial back on the bonus %.... till it's balanced to not make everything else irrelevant or not viable.

As we continue with Iluvalar's Mod WIP I suspect we'll come to deal with this issue to some degree if we set-up experiments to specifically expose it.

.

------------------

We could sort of test the effects of how pitting a new commander leading 20 or 30 units against an enemy group of 20 to 30 units would turn out, though since commanders are currently limited to having 6 units assigned to it at a time, we'll need to use multiple commanders to test this out.

What I propose that we do is pit a group of 20 or so Cyborg Machinegunners, 20 or so Python Heavy Cannon Tracks, etc. against an opposing group of the same exact thing, only that one of these groups is being led by freshly-produced commanders (preferably light-bodied commanders if leading groups of Cyborgs or light vehicles, and heavy-bodied commanders if leading groups of heavy tanks) while the other has none. However, as soon as one Commander is destroyed, the player MUST unassign all his/her remaining combat units from all other commanders, in order to mimic the effect of losing a commander in battle, one that, while freshly produced, could lead up to 20 or so units.

I think that's a good test under the constraints of hard coded limits. (I wish this was externalized for modding because changing the source and compiling a new binary is not something I want to get into - at this stage anyway. Though down the road it may be unavoidable.)

I'll be following this base protocol when I begin my experimental testing set-ups for Iluvalar's Mod starting tomorrow.

BTW - in order to follow the one variable difference protocol between Commander led groups and opposing forces without, what unit will you substitute to maintain parity in unit numbers? In my experiment set-ups I was thinking of using a sensor unit.

...................

Perhaps we could create the 2 groups of 20-30 units based on budget constraints, and maybe research and infrastructure constraints as well (such as a lack of Factory Modules, or Medium or Heavy Bodies), like having enough power to produce 30 Cyborg Machinegunners for one test, or having enough power to produce 30 Python Medium Cannon Tracks for another test, etc.? Of course, for the group that's going to be led by commanders, we would devote part of that budget away from producing the group in order to produce just 1 of the commanders that would lead the group.

Due to the need to manually unassign all combat units from commanders once one of them is destroyed, I'd recommend slowing down the game speed to a pause while you unassign the units. This is if you're playing offline, in which you'd be using cheat mode to control both groups.

Oh, and we wouldn't necessarily need for the 2 groups to be equal in terms of numbers. In fact, we should probably be looking at the total cost of the units that survived as opposed to the total number of units that survived.

Testing using budgetary constraints, instead of a 1 to 1 relationship between unit number and type, is certainly valid and valuable but also a bit more complex in scenario testing variants and subsequent analysis.

To get started with Iluvalar's Mod i'm gonna go with the K.I.S.S. 1 to 1 relationship between unit number and type. Later on I'll explore the budgetary constraint protocol. Glad you brought it to the forefront because those variant scenarios should be tested as they do represent real in-game context.

.

[Iluvalar]

You dont get it guys, I have those theorical data :

Code: Select all

level 1
units in battle: 4 (66%)
bonus: 22%
Total With mass effect: 15.68 units in a row
Equivalence with unit only: 5.12 units
Commander Value: 1.12 units
Suggested Balance for this level using Heavy Machinegun Cobra on Half-tracks as guide: 84$ 368hp
Suggested Balance for this level using Inferno Mantis on Half-tracks as guide: 135$ 473hp
Suggested Balance for this level using Medium Cannon Python on Half-tracks as guide: 241$ 944hp
level 2
units in battle: 5.33 (66%)
bonus: 50%
Total With mass effect: 25.08 units in a row
Equivalence with unit only: 6.6 units
Commander Value: 1.26 units
Suggested Balance for this level using Heavy Machinegun Cobra on Half-tracks as guide: 129$ 465hp
Suggested Balance for this level using Inferno Mantis on Half-tracks as guide: 194$ 584hp
Suggested Balance for this level using Medium Cannon Python on Half-tracks as guide: 329$ 1115hp
level 3
units in battle: 6.66 (66%)
bonus: 85%
Total With mass effect: 36.91 units in a row
Equivalence with unit only: 8.1 units
Commander Value: 1.44 units
Suggested Balance for this level using Heavy Machinegun Cobra on Half-tracks as guide: 190$ 582hp
Suggested Balance for this level using Inferno Mantis on Half-tracks as guide: 274$ 717hp
Suggested Balance for this level using Medium Cannon Python on Half-tracks as guide: 448$ 1320hp
level 4
units in battle: 8 (66%)
bonus: 128%
Total With mass effect: 51.4 units in a row
Equivalence with unit only: 9.65 units
Commander Value: 1.65 units
Suggested Balance for this level using Heavy Machinegun Cobra on Half-tracks as guide: 275$ 724hp
Suggested Balance for this level using Inferno Mantis on Half-tracks as guide: 385$ 879hp
Suggested Balance for this level using Medium Cannon Python on Half-tracks as guide: 615$ 1571hp

What i need is REAL GAME experience with commanders used quickly. To know with more precisions :

*What proportion of units are in the battle (there is always some away) ?
*What kind of units are the first viable ?
*What is the mean level a commander reach (when playing human players that know how to get rid of nuisance)

[Shadow Wolf TJC]

@Iluvalar: I've used Commanders on light bodies effectively, and I've used Commanders on heavy bodies effectively. It all depended on what budget constraints I had at the time.

Perhaps we could have 2 different versions of Command Turrets: 1 for lighter bodies, and another for heavier bodies? That way, players can more effectively use Commanders for a wider variety of situations. (By the way, I'm considering adding them to my mod for these reasons.)

[Rman Virgil]

You dont get it guys, I have those theorical data :

Code: Select all

level 1
units in battle: 4 (66%)
bonus: 22%
Total With mass effect: 15.68 units in a row
Equivalence with unit only: 5.12 units
Commander Value: 1.12 units
Suggested Balance for this level using Heavy Machinegun Cobra on Half-tracks as guide: 84$ 368hp
Suggested Balance for this level using Inferno Mantis on Half-tracks as guide: 135$ 473hp
Suggested Balance for this level using Medium Cannon Python on Half-tracks as guide: 241$ 944hp
level 2
units in battle: 5.33 (66%)
bonus: 50%
Total With mass effect: 25.08 units in a row
Equivalence with unit only: 6.6 units
Commander Value: 1.26 units
Suggested Balance for this level using Heavy Machinegun Cobra on Half-tracks as guide: 129$ 465hp
Suggested Balance for this level using Inferno Mantis on Half-tracks as guide: 194$ 584hp
Suggested Balance for this level using Medium Cannon Python on Half-tracks as guide: 329$ 1115hp
level 3
units in battle: 6.66 (66%)
bonus: 85%
Total With mass effect: 36.91 units in a row
Equivalence with unit only: 8.1 units
Commander Value: 1.44 units
Suggested Balance for this level using Heavy Machinegun Cobra on Half-tracks as guide: 190$ 582hp
Suggested Balance for this level using Inferno Mantis on Half-tracks as guide: 274$ 717hp
Suggested Balance for this level using Medium Cannon Python on Half-tracks as guide: 448$ 1320hp
level 4
units in battle: 8 (66%)
bonus: 128%
Total With mass effect: 51.4 units in a row
Equivalence with unit only: 9.65 units
Commander Value: 1.65 units
Suggested Balance for this level using Heavy Machinegun Cobra on Half-tracks as guide: 275$ 724hp
Suggested Balance for this level using Inferno Mantis on Half-tracks as guide: 385$ 879hp
Suggested Balance for this level using Medium Cannon Python on Half-tracks as guide: 615$ 1571hp

What i need is REAL GAME experience with commanders used quickly. To know with more precisions :

*What proportion of units are in the battle (there is always some away) ?
*What kind of units are the first viable ?
*What is the mean level a commander reach (when playing human players that know how to get rid of nuisance)

Now you tell us these details. Can't say we didn't ask up front for the whole kit ' kaboodle. We thought you needed more this type data. Oh well.... I think we gots it now - I hope.

That said... with the way the game's rigged now so that its decided in the first few minutes... you really think this line alone is gonna have any impact on that?

Here's the dilemma - what player who otherwise knows how to win is gonna wanna use Commanders unless they have at least a fair chance of winning? I dare say very unlikely without that fair chance because what they are agreeing to do is play to loose. How many good players wanna do that? But I think I'm pointing-out what is obvious to many, it's just not being stated bluntly. So there ya go.... what's the solution to that dilemma? Agreeing to loose for the sake of data harvesting? I honestly dunno the answer.

@Iluvalar: I've used Commanders on light bodies effectively, and I've used Commanders on heavy bodies effectively. It all depended on what budget constraints I had at the time.

Perhaps we could have 2 different versions of Command Turrets: 1 for lighter bodies, and another for heavier bodies? That way, players can more effectively use Commanders for a wider variety of situations. (By the way, I'm considering adding them to my mod for these reasons.)

I think this speaks to my question above directly.

.

[Iluvalar]

Here's the dilemma - what player who otherwise knows how to win is gonna wanna use Commanders unless they have at least a fair chance of winning?

You got a real good point here . That's why here is the V2

Code: Select all

CommandTurret4,Level All,70,200,1000,0,1,1200,GNLCMD1.PIE,TRLCMD1.PIE,FXHBLas.PIE,FXBeam.PIE,FXSFlms.PIE,FXSFlms.PIE,FXSFlms.PIE,0,2304,2304,100,100,1,1,0,0,8,0,0,0,0,0,0,0,0,2500,0,YES,HEAT,COMMAND,HOMING-DIRECT,ALL ROUNDER,180,90,-30,YES,NO,0,0,YES,1,100,0,0,0
CommandTurret3,Level All,60,200,1000,0,1,700,GNLCMD1.PIE,TRLCMD1.PIE,FXHBLas.PIE,FXBeam.PIE,FXSFlms.PIE,FXSFlms.PIE,FXSFlms.PIE,0,2048,2048,100,100,1,1,0,0,7,0,0,0,0,0,0,0,0,2500,0,YES,HEAT,COMMAND,HOMING-DIRECT,ALL ROUNDER,180,90,-30,YES,NO,0,0,YES,1,100,0,0,0
CommandTurret2,Level All,50,200,1000,0,1,400,GNLCMD1.PIE,TRLCMD1.PIE,FXHBLas.PIE,FXBeam.PIE,FXSFlms.PIE,FXSFlms.PIE,FXSFlms.PIE,0,1792,1792,100,100,1,1,0,0,6,0,0,0,0,0,0,0,0,2500,0,YES,HEAT,COMMAND,HOMING-DIRECT,ALL ROUNDER,180,90,-30,YES,NO,0,0,YES,1,100,0,0,0
CommandTurret1,Level All,44,200,1000,0,1,216,GNLCMD1.PIE,TRLCMD1.PIE,FXHBLas.PIE,FXBeam.PIE,FXSFlms.PIE,FXSFlms.PIE,FXSFlms.PIE,0,1536,1536,100,100,1,1,0,0,4,0,0,0,0,0,0,0,0,2500,0,YES,HEAT,COMMAND,HOMING-DIRECT,ALL ROUNDER,180,90,-30,YES,NO,0,0,YES,1,100,0,0,0

I expect they will only control tmg viper, that they will only fight with 50% of their maximum unit, that the commander research worth a 50% bonus (just that research). And on top they are now balanced to be fair as soon as the level 2.

With all that put together, I'm pretty sure i'm being optimistic and the balance will favor a bit too much the commanders. they will fall either in the top or the god tier. But not so much so the players that play without will die like flies. The data i'm seeking are the same and should be experimentable with that mod.

thx to test it. I repeat the information lacking :
1- The mean level of commanders in action
2- The proportion of the units it is reasonable to assume in combat (also on a mean basis)
3- What kind of unit we can attach to them (the heavier the better, but what we really have at that point ?)

[Rman Virgil]

.

Sweet. That sets-up an ideal challenge under real game conditions that has more appeal (good with Cs vs. good without Cs... or average with Cs vs average without Cs...some parity between skill levels, in other words) And if OPed gets born-out in the data, dialing back stats is just as valid to attain balance as dialing up but without the psychological sacrifice element.

===============>

Aside: Hello Kitty pink is good for a chuckle or 2 but prolonged exposure is like spending way too much time in a carnival fun house hall of distorting mirrors.

L8r...

.

.

[Rman Virgil]

.

Ok.. got the word out to my LAN club buds to see if I can enlist them to help with some testing here. They have always come thru in the past with my own WZ mod/maps which was a real boon otherwise I would have been up chits creek without a paddle when it came to getting valuable MP data for my WZ work & experiments.

In the meanwhile - since I'm not into being idle with this effort, I'm gonna incorporate this mod into one of my own map-mods. While it may not be suitable for the hard data purposes required here, it will at least give me an indication if we're on a fruitful track. The player start position is being assaulted within 15 seconds and the offensives escalate such that at the 2:45 min mark the base is being attacked from 4 different ground vectors & 1 air vector continuously. Trying to deploy stock Commanders under these conditions was self-assured annihilation (there was another strat-tac that allowed for survival but commanders was not it). Incorporating this mod should make a notable difference in using Commanders to survive this early, relentless, massive & muti-vectored assault, I'm thinking. Should be intense fun finding out.

.

[Shadow Wolf TJC]

I just quickly put together this commander mod for testing.

Two_Commanders_Mod.wz

(124.03 KiB) Downloaded 221 times

Basically, this provides 2 versions of Commanders: a comparatively fragile version that is designed to be mounted onto lightweight vehicles, and a more durable version that is too heavy to be mounted on anything lighter than a heavy vehicle. Here are the stat changes:

Code: Select all

CommandTurret1,Level All,50,100,500,0,1,300,GNLCMD1.PIE,TRLCMD1.PIE,FXHBLas.PIE,FXBeam.PIE,FXSFlms.PIE,FXSFlms.PIE,FXSFlms.PIE,0,1536,1536,200,200,1,1,0,0,1,0,0,0,0,0,0,0,0,2500,0,YES,HEAT,COMMAND,HOMING-DIRECT,ALL ROUNDER,180,90,-30,YES,NO,0,0,YES,1,100,0,0,0
CommandTurret2,Level All,100,200,500,0,1,550,GNLCMD1.PIE,TRLCMD1.PIE,FXHBLas.PIE,FXBeam.PIE,FXSFlms.PIE,FXSFlms.PIE,FXSFlms.PIE,0,1792,1792,200,200,1,1,0,0,1,0,0,0,0,0,0,0,0,2500,0,YES,HEAT,COMMAND,HOMING-DIRECT,ALL ROUNDER,180,90,-30,YES,NO,0,0,YES,1,100,0,0,0
CommandTurret3,Level All,150,300,500,0,1,890,GNLCMD1.PIE,TRLCMD1.PIE,FXHBLas.PIE,FXBeam.PIE,FXSFlms.PIE,FXSFlms.PIE,FXSFlms.PIE,0,2048,2048,200,200,1,1,0,0,1,0,0,0,0,0,0,0,0,2500,0,YES,HEAT,COMMAND,HOMING-DIRECT,ALL ROUNDER,180,90,-30,YES,NO,0,0,YES,1,100,0,0,0
CommandTurret4,Level All,200,400,500,0,1,1320,GNLCMD1.PIE,TRLCMD1.PIE,FXHBLas.PIE,FXBeam.PIE,FXSFlms.PIE,FXSFlms.PIE,FXSFlms.PIE,0,2304,2304,200,200,1,1,0,0,1,0,0,0,0,0,0,0,0,2500,0,YES,HEAT,COMMAND,HOMING-DIRECT,ALL ROUNDER,180,90,-30,YES,NO,0,0,YES,1,100,0,0,0
CommandTurret1Hvy,Level All,125,250,10000,0,1,800,GNLCMD1.PIE,TRLCMD1.PIE,FXHBLas.PIE,FXBeam.PIE,FXSFlms.PIE,FXSFlms.PIE,FXSFlms.PIE,0,1792,1792,200,200,1,1,0,0,1,0,0,0,0,0,0,0,0,2500,0,YES,HEAT,COMMAND,HOMING-DIRECT,ALL ROUNDER,180,90,-30,YES,NO,0,0,YES,1,100,0,0,0
CommandTurret2Hvy,Level All,250,500,10000,0,1,1575,GNLCMD1.PIE,TRLCMD1.PIE,FXHBLas.PIE,FXBeam.PIE,FXSFlms.PIE,FXSFlms.PIE,FXSFlms.PIE,0,2048,2048,200,200,1,1,0,0,1,0,0,0,0,0,0,0,0,2500,0,YES,HEAT,COMMAND,HOMING-DIRECT,ALL ROUNDER,180,90,-30,YES,NO,0,0,YES,1,100,0,0,0
CommandTurret3Hvy,Level All,375,750,10000,0,1,2620,GNLCMD1.PIE,TRLCMD1.PIE,FXHBLas.PIE,FXBeam.PIE,FXSFlms.PIE,FXSFlms.PIE,FXSFlms.PIE,0,2304,2304,200,200,1,1,0,0,1,0,0,0,0,0,0,0,0,2500,0,YES,HEAT,COMMAND,HOMING-DIRECT,ALL ROUNDER,180,90,-30,YES,NO,0,0,YES,1,100,0,0,0
CommandTurret4Hvy,Level All,500,1000,10000,0,1,3935,GNLCMD1.PIE,TRLCMD1.PIE,FXHBLas.PIE,FXBeam.PIE,FXSFlms.PIE,FXSFlms.PIE,FXSFlms.PIE,0,2560,2560,200,200,1,1,0,0,1,0,0,0,0,0,0,0,0,2500,0,YES,HEAT,COMMAND,HOMING-DIRECT,ALL ROUNDER,180,90,-30,YES,NO,0,0,YES,1,100,0,0,0

For calculating the HP of the heavier turrets, I used a Python Half-Tracks' HP as the base for the T1 version, then added an additional 150 HP to the T2 version, 300 HP to the T3 version, and finally, 450 HP to the T4 version so that it would equate to the base HP of a Wyvern Half-Tracks. I then multiplied the T2 version by 1.9, the T3 version by 2.8, and the T4 version by 3.7 to simulate an increase in HP provided by Composite Alloy upgrades (which are required to unlock the more advanced versions btw).

For calculating the HP of the lighter turrets, I took the HP of the corresponding tier of heavier turret, and divided it by 3 (since the lighter bodies typically had 1/3rd of the heavier bodies' HP). Finally, since the HP of turrets seemed to matter most in T1, I had to add, for the lighter turrets, an additional 100 HP to the T1 version, 75 HP to the T2 version, 50 HP to the T3 version, and 25 HP to the T4 version, while for the heavier turrets, I had to add an additional 200 HP to the T1 version, 150 HP to the T2 version, 100 HP to the T3 version, and 50 HP to the T4 version.

If all this is too much to take in at once, here's the formula that I used to calculate turret HP.

HP = ( s x ( 150 + ( 50 x t ) ) x c ) + ( ( ( s + 1 ) / 2 ) x ( 125 - ( 25 x t ) ) )

s is the size of the body meant to mount the turret: 1 for a light body, and 3 for a heavy body.

t is the tier of turret, meant to increase along with the HP provided by more advanced bodies in higher tiers.

c is the HP multiplier provided by the Composite Alloy upgrade required to research the turret: 1 for no alloys, 1.9 for Composite Alloys Mk3, 2.8 for Dense Composite Alloys Mk3, and 3.7 for Superdense Composite Alloys Mk3.

[Rman Virgil]

.

Cool. I'm really curious about this approach, Shadow Wolf TJC. I'll go ahead & update the other thread.

.

[Iluvalar]

I then multiplied the T2 version by 1.9, the T3 version by 2.8, and the T4 version by 3.7 to simulate an increase in HP provided by Composite Alloy upgrades (which are required to unlock the more advanced versions btw).

why do you simulate it ? It's not applying ?

I'll look further your mod when i'll have the time. But the last time i checked, I figured that a commander fit better on heavier bodies. One of your commander will end up being better than the other in any condition.

[Shadow Wolf TJC]

@Iluvalar: Are you sure about that? The heavier command turret may have much more HP than the lighter one, but its weight was set to 10000 to balance it out.

Also, here's another version of my experimental Commander Mod to test out:

Two_Commanders_Mod_v2.wz

(151.1 KiB) Downloaded 221 times

This one is much like the other one, though the research prerequisites were changed up a bit. The Command Turret is now researchable before the Command Relay Center (which I'm uncertain as to whether or not it'll be kept as a requirement for producing Commanders), and the Research Module now requires Command Turret I as a prerequisite instead of Command Relay Center.

[Rman Virgil]

@Iluvalar: Are you sure about that? The heavier command turret may have much more HP than the lighter one, but its weight was set to 10000 to balance it out.

Also, here's another version of my experimental Commander Mod to test out:

Two_Commanders_Mod_v2.wz

This one is much like the other one, though the research prerequisites were changed up a bit. The Command Turret is now researchable before the Command Relay Center (which I'm uncertain as to whether or not it'll be kept as a requirement for producing Commanders), and the Research Module now requires Command Turret I as a prerequisite instead of Command Relay Center.

Hmm.. Could you explain a bit your reasoning & goal for changing up the research pre-requisites order?

Thanks.
.

[Iluvalar]

@Iluvalar: Are you sure about that? The heavier command turret may have much more HP than the lighter one, but its weight was set to 10000 to balance it out.

Yes, I'm pretty sure of that. The commanders have very intense constraints on HP as I said earlier.

They need to have an higher hp/efficicency than their units at low level (to avoid prematurated death)
And still have a lower hp/efficicency than their units at high level (to be targeted before they reach the god tier)

Your light body commander mounted on an heavy body will increase his hp/efficicency. Therefore it will reach higher level for a pretty small fee

[Shadow Wolf TJC]

@Iluvalar: Are you sure about that? The heavier command turret may have much more HP than the lighter one, but its weight was set to 10000 to balance it out.

Also, here's another version of my experimental Commander Mod to test out:

Two_Commanders_Mod_v2.wz

This one is much like the other one, though the research prerequisites were changed up a bit. The Command Turret is now researchable before the Command Relay Center (which I'm uncertain as to whether or not it'll be kept as a requirement for producing Commanders), and the Research Module now requires Command Turret I as a prerequisite instead of Command Relay Center.

Hmm.. Could you explain a bit your reasoning & goal for changing up the research pre-requisites order?

Thanks.
.

It's not really much of an improvement. In fact, since Command Turrets and Command Relay Centers require the same amount of power to research, it probably wouldn't make any difference if you're not planning on using Commanders anyways. However, if Command Relay Centers aren't required in order to produce them, then players would be able to get Commanders out sooner, though if they are still required, then at least players could design them sooner, before they could research Command Relay Centers.

Edit: Oh, and Iluvalar, I had to decrease the price of the Command Turrets significantly in order to make them more accessible. I've seen fortresses that are cheaper than the Command Turret IV for crying out loud!