Some thoughts about veterancy

[JoonasTo]

I am sure there is a lot here that has already been said before but perhaps this will be a bit more relevant now that we have had some more experience with this current system and seen it's advantages and shortcomings.

There are different veterancy systems in use in games with different outcomes. One needs to understand them to choose a suitable one for the aim of the game. So let's first look at a few different veterancy systems, examples and their features(-) and effects(+). Qualities important to the system but game specific, not inherent in parentheses().

No veterancy, Age of Empires
+ Does not entice to keeping veteran units alive, rather dedicate low health military units to raiding duties
+ No information for casters
+ No imbalance issues

Kill counter only, StarCraft:Brood War
- Visible kill counter
+ Incentive to keep units alive for bragging rights
+ Good for casters
+ No imbalance issues

Cosmetic Veterancy, StarCraft 2
- Information visible
+ Incentive to keep units alive for RP purposes, great for single player and casting
+ No imbalance issues

Combat effectiveness veterancy system based on cost value killed, Wargame: Red Dragon*
- Unit gains boost to all combat stats
(- Information visible to all)
+ Incentive for keeping units alive throughout multiple encounters
+ Cost effective units vet faster
+ Snowball mechanic
+ Buff to own unit effectiveness
+ Alters expectations of what the unit can kill

Full veterancy system based on cost value of damage dealt, C&C 3
- Unit gains boost to all combat stats and survivability
(- Visual identification of full veteran units on the battlefield)
+ Hard to predict veterancy levels
+ Incentive for keeping units alive throughout multiple encounters
+ High DPS units gain vet faster
+ Veteran units are a high priority target in battles
(+ Great visual information for viewers and players about veterancy)
+ Veteran units alter gameplay, in practice different unit than the base unit
+ Snowball mechanic
+ Both buff to own unit effectiveness and debuff to enemy unit effectiveness

Kill based HP veterancy system, Supreme Commander
- Unit gains boost to survivability
(- Veterancy information easily visible to the owner of the unit and observers)
+ Helps in keeping unit alive through multiple encounters
+ Easily predictable veterancy ranks
+ Leads to disregarding easier to kill lower level units as the game progresses in favour of tougher higher level ones, the famous don't feed the EXP
+ Stronger units gain veterancy easier
+ Snowball mechanic
+ Disproportionate veterancy gain
+ Debuff to enemy unit effectiveness
+ Alters expectations of what is needed to kill the unit

Point based HP veterancy system, FAF
- Unit gains boost to survivability
(- Veterancy information easily visible to the owner of the unit and observers)
+ Helps in keeping unit alive through multiple encounters
+ Easily predictable veterancy ranks (for the owner)
+ Leads to priorising easier to kill targets over tougher ones, the famours engineer massacre
+ Stronger units gain veterancy easier
+ Snowball mechanic
+ Debuff to enemy unit effectiveness
+ Alters expectations of what is needed to kill the unit

HP veterancy system based on percentage of cost value of damage killed, current FAF
- Unit gains boost to survivability
(- Veterancy information visible to the owner of the unit and observers)
+ Hard to predict veterancy levels
+ Helps in keeping unit alive through multiple encounters
+ Cost effective units vet faster
+ Leads to priorising high value, low combat effectiveness targets, aka buildings
+ Debuff to enemy unit effectiveness
+ Snowball mechanic
+ Alters expectations of what is needed to kill the unit

*There is also a morale damage veterancy gain system in Wargame: Red Dragon but that is beyond the scope of this discussion.


Now that we have looked at some systems and their features let's analyze our current system a little more in-depth. The most pronounced feature is that units that gain veterancy don't become better at doing their job, they become harder for the enemy to eliminate. This has two profound effects.First, they are easier to keep alive through multiple encounters which facilitates the snowball mechanic. Secondly, it alters enemy expectations of what is needed to eliminate the unit. Both of these can lead to frustration in players, especially in combination with each other.
The second defining feature is that units gain veterancy based on percentage of cost value of damage killed. This has a few effects. Targeting low threat, high value units becomes preferable, generally this means buildings. The other important to understand effect is the obscurity of the veterancy gain, this means the system appears very unpredictable, especially to the enemy but also to the owner. Again, a possible source of frustration if not handled correctly.

Now that we understand this, we should decide if those are favourable to us and if not, how do we deal with them. Looking at the current FAF implementation it seems that priorisation of low threat, high value units is not preferable, building veterancy value is getting nerfed by 50%.
It also seems that cost effective units vetting faster has been deemed a problem and some less cost effective units(experimentals) have been buffed in comparison to only require 50% mass value for a veterancy. This brings in unwanted effect of experimentals vetting from killing each other to what the current solution is limiting veterancy gain from a single kill to only one level.
Part of the same problem is the veterancy gain for the commander, which doesn't take into account the upgrades on the commander. This means that a late game double-gun, double-nano commander gets the same vet from killing an Othuum as a T1 commander does. This is mitigated by the limited veterancy gain from higher level units for commanders. What this, along with both of the solutions for the two above problems does, however is make it even more obfuscated for the players on how much veterancy units gain.
This is a problem that doesn't seem to be recognised or not deemed meaningful in the current implementation. I say this because all the solutions to problems make the veterancy gain even more inconsistent and there are no good visual clues to tell how close to a veterancy or how veteran a unit is on a battlefield. Only the owner of the unit can take a look at the unit card to find the small veterancy star symbols and the even harder to read veterancy gauge. The only way other players can find out if a unit is a veteran(and how veteran) is to look at the maximum HP value of the unit and learn all the unit HP values(and the vetted up unit HP values.)

I think this either an oversight or a mistake by the current team. I refuse to believe it was a conscious aim to obfuscate this system and hide the workings of it from the players. So if we wish to stick with the current veterancy implementation here's what I propose instead:

► Let's include the value of commander upgrades in commander veterancy calculations and remove the different veterancy gain from different tech levels of units.
♦ Will require a lot of testing
♦ Your commander gets more effective, veterancy scales accordingly
♦ Possible problem of veterancy hunting before upgrades to make it more effective

► Let's remove buildings from the veterancy system
♦ Simple to understand
♦ Defenses can still give veterancy but can not get it
♦ Removes the AA emplacements vs gunships balance issue
♦ This will remove the problem of veterancy hunting from buildings
♦ Ahwassa will require balancing without reliance on veterancy due to this

► Let's remove different veterancy levels from units and standardise on 150%
♦ Easy to remember value
♦ No problematic Experimental war veterancy issues
♦ Roughly equivalent to old Commander vet value of 20 kills/points
♦ Actually a nice threshold value for LABs not vetting from engineers
♦ Will allow normal units to vet more often to promote awareness about the system

► Let's get better visual indicators for veterancy
♦ To minimize frustration of both the owner and the enemy for false expectations
♦ Make the veterancy gauge/insignias more prominent in the unit card instead of the kill counter
♦ Ultimately aim for a visual indicator visible to all players to signify a veteran unit, C&C 3 is a great example of how to do it right, there's an audio-visual que for gaining veterancy and the vetted units are presented differently


The above suggestions were written with the assumption that why we want veterancy is to reward micro. But if we look at rewarding micro, is the current system even what we really want? It gives us units that require less micro to survive through consequent encounters with the enemy. I am going to suggest there's a better system out there.
If we really want to reward micro would not a better reward be a unit that is more effective in it's role? We should rather be using a system which rewards with a unit with more effectiveness rather than survivability. Normally this is done by increasing accuracy and target acquisition plus secondary stats. Sometimes rate of fire or damage. Due to how FAF works, we are left with increasing unit damage. So by now most of you are probably going "trololol ml with 6k dps" but you'll have to remember the Monkeylord will have to kill 7½ other Monkeylords to get there(assuming 150%), with the same hitpoints it starts with. And it will still die to strategic bombers and Percivals just as easily.
So okay, maybe you weren't one of those people but rather "omg summit with 14k alpha." Yeah okay, that's more of a problem that we have to solve. Units with higher survivability benefit far more from this than from the survivability bonus system. This also means shields can become really powerful in certain situations. I can't give a solution to this without play testing.
Or perhaps you were the one who saw the biggest problem of all, ACUs. This will inevitably change the way ACUs work in relation to unit spam in the early game and it's going to be different. This is not bad thing in itself but the change it brings will be huge to some players(yours truly one of the most affected.) In my mind this is the biggest danger with the system change. It has the possibility to alienate a group of relatively high level players. I do honestly think it's a better system for the game in the long run though if we can get across this obstacle. It absolutely must not be a one-sided, "oh btw. we are implementing this," change. Rather it has to be done with the community involved on a bigger scale than anything before.


So that's what I think would benefit the veterancy system in both the present and in the future.

[moonbearonmeth]

►Let's include the value of commander upgrades in commander veterancy calculations and remove the different veterancy gain from different tech levels of units.

This doesn't seem as bad as I first thought but you are right, it would require a ton of testing.

Spoiler: show

I would also be interested on what impact this would have on the fabled 'tele-sacrifice GC' I've heard people say is an issue

► Let's remove buildings from the veterancy system
♦ This will remove the problem of veterancy hunting from buildings

Assuming the other suggestions you want get implemented I actually don't see this as a problem, it makes sense to me that you get rewarded for destroying something of value especially if it's left lightly/un-defended.

► Let's get better visual indicators for veterancy

I agree with this but not necessarily in the same way.If given the choice I would prefer a more integrated approach to the representation of veterancy much like how we already determine what upgrades an enemy ACU has. Unfortunately that would require a ton of work to make.

Due to how FAF works, we are left with increasing unit damage

I don't think we're 'left' with it. Things like turret turn rate and secondary 'token' AA guns could possibly be looked into.

Another point maybe worth looking into would be increasing the veterancy value of vetted units so that destroying them rewards more vet.

[Franck83]

► Let's remove buildings from the veterancy system
♦ This will remove the problem of veterancy hunting from buildings

Assuming the other suggestions you want get implemented I actually don't see this as a problem, it makes sense to me that you get rewarded for destroying something of value especially if it's left lightly/un-defended.

I thought a lot of this for my mod. You have an intermediate solution : check if the unit killed is military to boost it's XP. So civilian building can still give XP but at a lower amount. Don't forget that defense turrets should be rewarded as mobile military units.

In Alliance of Heroes, veterancy is pushed deep. Unit damages and accuracy will increase by vet when the unit fire military unit. Defense armor veterancy increase only when unit take damage. So this is much more realism, because the player way of play gives the veterancy path.

Then when you kill a veteran unit in AoH, you are better rewarded than killing a rookie unit. And when a veteran unit kill a rookie one, it gains less XP.

[ZLO_RD]

can t1 arty get less vet for buildings and more vet for things that it is supposed to be bad at?

[Cyborg16]

I was sceptical but mostly agree with this. I've long thought the way experimentals get HP boosts was lame.

One game you missed: Total Annihilation
- veterency improves accuracy
+ LRPG (long range arty) becomes super powerful (to the point of shooting down air units)
+ most units are unaffected

IMO veterency boosts should be somewhat subtle but still significant, so e.g.:
- accuracy boosts where appropriate, but not unreasonable ones
- regen rate boosts
- small range boosts (this would make tanks and some experimentals significantly more powerful without being OP, and be very useful for ACUs)

Removing the HP boosts would definitely require rebalancing, but I think would be worth it in the end. I wonder though if assisted repairs should be cheaper/faster (i.e. make it worth repairing an experimental instead of reclaiming and building another).

[FtXCommando]

I think people really underestimate the importance that hp boosts have in providing value for experimentals. If you take away the hp boost from veterancy, you are significantly nerfing Cybran while also doing great harm to most other factions barring UEF really. You are going to do a huge change to the risk/reward dynamic of utilizing experimentals when experimentals are already a mass donation in plenty of situations. This would run the risk of nearly making them irrelevant unless you change the reclaim value they leave.

Also

- small range boosts (this would make tanks and some experimentals significantly more powerful without being OP, and be very useful for ACUs)

No. ACUs getting range advantages against other ACUs will immediately devolve the game into gun ACU meta and there is no way any other upgrade path will ever be viable with such a change.

Removing the HP boosts would definitely require rebalancing, but I think would be worth it in the end.

Why? What is so bad about HP boosts? I mean, it would be a lot easier to predict when the HP boosts will happen on an experimental if we got rid of the current total trash vet system and went back to the point system but I GuESs ThAt SyStEm WaS toO CoMpLeX FoR FaF PlAyErS.

[TheKoopa]

I think people really underestimate the importance that hp boosts have in providing value for experimentals. If you take away the hp boost from veterancy, you are significantly nerfing Cybran while also doing great harm to most other factions barring UEF really. You are going to do a huge change to the risk/reward dynamic of utilizing experimentals when experimentals are already a mass donation in plenty of situations. This would run the risk of nearly making them irrelevant unless you change the reclaim value they leave.

Also

- small range boosts (this would make tanks and some experimentals significantly more powerful without being OP, and be very useful for ACUs)

No. ACUs getting range advantages against other ACUs will immediately devolve the game into gun ACU meta and there is no way any other upgrade path will ever be viable with such a change.

Removing the HP boosts would definitely require rebalancing, but I think would be worth it in the end.

Why? What is so bad about HP boosts? I mean, it would be a lot easier to predict when the HP boosts will happen on an experimental if we got rid of the current total trash vet system and went back to the point system but I GuESs ThAt SyStEm WaS toO CoMpLeX FoR FaF PlAyErS.

Is that not the meta already?

[Cyborg16]

Why? What is so bad about HP boosts? I mean, it would be a lot easier to predict when the HP boosts will happen on an experimental if we got rid of the current total trash vet system and went back to the point system but I GuESs ThAt SyStEm WaS toO CoMpLeX FoR FaF PlAyErS.

I personally hate the old system of CTRL-K THOSE T1 NOW - DON'T FEED THE ML!!!

Experimental rushes have always been a case of win-or-make-big-mass-donation. But lets not pretend raids with other units aren't also potential mass donations. Give the ML an HP boost to start with and it's still very viable.

No. ACUs getting range advantages against other ACUs will immediately devolve the game into gun ACU meta and there is no way any other upgrade path will ever be viable with such a change.

SMALL range boosts. The Aeon ACU already gets more range than others and it doesn't prevent T2 builds (other things might like spam or lack of choke-points on the map, but ACU range isn't the deciding factor).

[FtXCommando]

Why? What is so bad about HP boosts? I mean, it would be a lot easier to predict when the HP boosts will happen on an experimental if we got rid of the current total trash vet system and went back to the point system but I GuESs ThAt SyStEm WaS toO CoMpLeX FoR FaF PlAyErS.

I personally hate the old system of CTRL-K THOSE T1 NOW - DON'T FEED THE ML!!!

Experimental rushes have always been a case of win-or-make-big-mass-donation. But lets not pretend raids with other units aren't also potential mass donations. Give the ML an HP boost to start with and it's still very viable.

No. ACUs getting range advantages against other ACUs will immediately devolve the game into gun ACU meta and there is no way any other upgrade path will ever be viable with such a change.

SMALL range boosts. The Aeon ACU already gets more range than others and it doesn't prevent T2 builds (other things might like spam or lack of choke-points on the map, but ACU range isn't the deciding factor).

If you give even 1 range per vet level you end up with a 40 range aeon ACU facing off against 26 range harbs and 32 range bricks. Sound like an advantage that can be given up? No, it doesn't. Therefore T2 ACU will see no play barring UEF and Phim ACUs that can take both gun and t2.

My problem is that now you are doing a change to deal with an unnecessary change. Why do we need to give ML extra hp boosts in order to recover from the unnecessary change of getting rid of vet HP boosts? Why not just keep the normal HP boosts? Personally, I hate the system where you kill a 4000 hp structure and you get 5 vet. That makes even less sense than vetting off of t1 engineers in my mind. It's just changing the game for the sake of changing it.

[Cyborg16]

According to the unit DB, Brick and Harb ranges are 35 and 28. If Aeon upgraded gun range were reduced to say 32, max range with vet would be 37 with a bonus of 1-per-vet. Alternatively the range boosts could be reduced to maybe 2/3, or if fractional ranges are not allowed, not boost at every vet level (I'm assuming range might not be the only boost from vet).

I'm not entirely sure this is a good idea, but think it could work. But mostly I'm just saying that FtXCommando is over dramatising and has poor reasoning