Proposal workshop: unit stacking ideas

[azum4roll]

And some people play with one model per unit.

 

[Tekamthi]

Do you mean the 3D model? That wouldn't be enough, I don't even look at it.

Aye don't forget the SV players... Not me for the most part but I do flip to it occasionally as map gets cluttered, and there are players that use SV as default. We should pick a format that allows for further development of this concept if it's successful -- ie it was suggested a UA might allow ranged unit to be combined into an army -- our icon format should allow for this as well.

I'm thinking take two icons overlay one on top of the other slightly off set so one is visible behind the other, so that warrior army icon visually looks like two warrior icons combined, etc. This way even player that does not participate in Congress or even read patch notes will get the cue that this is 2 of these units in 1

Tech/policy-gating the ability has the advantage of keeping the AI from forming them in the early game. This is an advantage because it means the logic for when to form them can be simpler. Also without it, the early game fighting will change, possibly significantly, compared to the current play.

Revisiting this topic, consider having them gated by unit level. Say at least one unit in the army-forming merge must be level x, x sufficiently high that it's difficult to achieve in ancient but easily available by late game, maybe x = level 4?

 

[usadefcon1]

The same way units are highlighted in red if they are an enemy could they be highlighted in gold if they are an army?

 

[Stalker0]

Changing blitz promotion for an increase to damage dealt and taken on offense would solve the problem, remove one more click, and reduce the insane xp gain from double attack...

Since this concept was about removing tedium…perhaps this is the better Approach. If we are still clicking twice a unit (or have to space bar when we don’t want the second attack) it might not save that much

 

[ma_kuh]

There's also Spear Throw (new version would be renamed, of course) as a single-click double attack. It would also justify a lower max-hp increase because the first hit doesn't cost health. Adds some complications with moving forward if you get a ranged kill though.

Is the code feasible for a "double melee" Spear Throw ability?

The reason I feel a double-hit is better than just a CS increase is that CS increases have varying value based on how much bonus CS you already have, and also affect defense, which makes a fair value for bonus HP or CS harder to calculate.

and reduce the insane xp gain from double attack...

There's already a solution for this, and it's making XP only proc on the first attack in a turn. Until we fix it, this argument holds no weight.

 

[pineappledan]

We've had the backend to disable XP on 2nd attacks in place for over a year now. We've just been waiting for the community to come to their senses

 

[CppMaster]

We've had the backend to disable XP on 2nd attacks in place for over a year now. We've just been waiting for the community to come to their senses

Isn't it applied to logistic promotion? Could be imba without it

 

[Rekk]

Isn't it applied to logistic promotion? Could be imba without it

What? Double attacks have given double XP since vanilla.

We've had the backend to disable XP on 2nd attacks in place for over a year now. We've just been waiting for the community to come to their senses

Literally just requires a flag to be set to true.

 

[CppMaster]

What? Double attacks have given double XP since vanilla.

Yeah, I know, but I thought it was changed in VP, because we had this discussion.

 

[pineappledan]

We did have the discussion, and the community voted not to remove double XP.
And then someone coded it in as a toggleable option anyways.
So if you've been playing a balance mod, like my tweaks mod as an example, there's a good chance you have been playing with no XP on 2 attacks for a long time

 

[ma_kuh]

Okay, check my assumptions and math on this one:

Spoiler An Army should be relatively twice as effective as two individual units :

Effectiveness can be seen as the ability for a unit to take an equally-powered opponent from 100 > 0 hp.

We have the damage formula, we'll call F, and the inverse (used for the weaker party) 1/F, or G.
Then we can look at the effective damage of each as F/G. For example, if the Army deals x2 the damage of an individual against it, then F = 2 and G = 1, and the effective damage is x2.

We also have the HP to contend with. If a unit has +50 HP, then damage relative to its health is reduced. For example, if an Army deals 30 damage to 100 hp, it takes out 30% of the target's health, but if that same target has 150 hp it only takes out 20%.
This can be abstracted as D / HP, and D / mHP (D is damage, HP is base HP, mHP is modified with the bonus hp for an army, 50 in the above example).
* 50% ballpark is for 100-hp units, later on this % goes down with the Fieldworks promos.

So we have two components:
(F / HP)
----------
(G / mHP)

or

F/G * mHP/HP

With this in mind then, our goal is for that formula to work out to 2, to satisfy the premise that the army should be twice as effective.

With a flat +30% CS and +50 HP, the basic formula works out to 2.04, which is close enough to me.

Spoiler Here are some other specific comparisons :

So what this means is the later in the game it goes, the less relative power converting to an Army will yield. This makes sense to me because as tech improves, armies become less about the number you can field and more about precision and quality.
Similarly, as a unit's promotions and experience improves, the less they will contribute to an equally-experienced army. Individual prowess can't shine through as much, or something. Again it makes sense to me.

So the non-blitz proposal I'd go with is:
Armies cost double supply and maintenance, and gain a unique promotion that grants +30% CS and +50 HP.

 

[pineappledan]

The above analysis ignores the complications of it being harder to bring 2 attacks to bear on a 2 move unit, and the loss of flank bonuses relative to 2 unstacked units.

addition of an army mechanic would nullify the main argument against implementing XP on 1st attack only. The reason that ultimately swayed players not to want it was that, while it was a warranted nerf to logistics (ranged), it would be an even bigger nerf to blitz (melee). If blitz is being given to units that have a large amount of bonus XP, and effectively pool the XP from hits from 2 units into 1, then that is a good reason not to also give them XP on attack.

 

[Rick Drayson]

All the problems of late game wars (every tile filled, congestion around choke points, units can be replaced faster than cities can replace, hours of repetitive gameplay for a single war. etc etc)
Can be fixed by unit tile stacking.

1. People hated stacks of doom. No idea why considering that in reality that is what wars were like until the Napoleonic era. Distributed forces were not a thing. Decisive battle doctrine. Lose on the battlefield, retreat behind city walls or capitulate. Wham bam wars over.
2. Fix stacks of doom with a dynamic stack limit per tile based upon a supply concept. A tile with a road and a river next to a city can supply more units than a desert tile. So a single unit only on a remote desert tile vs. say 6 within a major city. The stacks don't have to be huge in order to hugely improve the aspect of war especially late game.
3. Stacks will fight as a single unit.

This is the last feature VP needs to become the greatest strategy game of all time.

 

[Tekamthi]

Can be fixed by unit tile stacking

I tend to agree, however when this discussion got started, we forgot that deviation from 1UPT was long ago declared off limits for congress... Hence the turn towards armies as a pseudo-stacking alternative that may be allowable

I have a request on GitHub that, if implemented, may allow us to do some stacking via modmod, similar to your suggestion. That's probably best hope for something like that. Ultimately, and though I've made my own stacking suggestions, I think stacking won't work well for AI, even if it seems easy to us. On the other hand the armies stuff will potentially be AI manageable with only small tweaks.

Only other alternative we have yet to explore is whether combat units could be flagged as cargo for one another (the database structure seems to exist for this), and thus have a sort of stacking without stacking. However AI will be clueless there too and the concept seems a little wonky to begin with.

 

[ruskyandrei]

we forgot that deviation from 1UPT was long ago declared off limits for congress

I think it's worth discussing still, ultimately if this is the best way forward why not consider it ?

It is possible (though not easy by any means) and I think there are very good arguments for why some form of limited stacking would make the game considerably better.

 

[Anarcomu]

I am personally against deviations from 1UPT.
It makes the map harder to read, doesn't reduce the amount of clicking, changes nothing to battlefronts tedium (there is just more units fighting at the same time).
The army system have the advantage (without double attack) to considerably reduce the amount of click on the map. And even with double attack, it is still reduce micro since it is double click against mostly the same enemy.

 

[Heinz_Guderian]

I am personally against deviations from 1UPT.
It makes the map harder to read, doesn't reduce the amount of clicking, changes nothing to battlefronts tedium (there is just more units fighting at the same time).
The army system have the advantage (without double attack) to considerably reduce the amount of click on the map. And even with double attack, it is still reduce micro since it is double click against mostly the same enemy.

+1

 

[Tekamthi]

I think it's worth discussing still, ultimately if this is the best way forward why not consider it ?

By all means! The purpose of this thread was originally intended to theorycraft, or rule out, a possible proposal, and to be a proposal there are rules.

But to be a discussion thread there are not -- if we want to move forward on breaking the 1UPT paradigm in any kind of dynamic way, we'd need some additional database features implemented by devs -- as it stands, stacking can be enabled very flatly. I've made one request to sort of support this but so far no takers on implementation.

The existing AI only stacks by accident even if it's enabled, it's not part of its analysis, and outside of our main branch DLL devs it's almost impossible to affect change to AI. This will be the major challenge of any stacking project. It may be the kind of challenge that requires the whole project to be forked -- forks of VP have been successfully done, but it's a major commitment, probably beyond the capacities for most that visit here.

 

[Stalker0]

I think it's worth discussing still, ultimately if this is the best way forward why not consider it ?

Because we aren't going to do it...its that simple. At some point the head of the project said "we aren't going in that direction" .... so we don't.

There are so many things we actually could implement and change, why waste time and effort on things that we won't?

 

[epicivfreak]

I think it's worth discussing still, ultimately if this is the best way forward why not consider it ?

It is possible (though not easy by any means) and I think there are very good arguments for why some form of limited stacking would make the game considerably better.

One Unit Per Type, Per Tile, has worked extremely well in the past, despite Vox Populi's reservations to that.