Introducing the Official community mod for 0 A.D. Empires Ascendant

[chrstgtr]

1 hour ago, real_tabasco_sauce said:

not sure what you mean with this. The ships are now like any other unit, (archer, pikeman, etc) in that they don't have buildingAI. So there is no buildingAI to be had.

im sure there will plenty of strategies and micro involved.

There's some disconnect. Right now, ship AI behavior is buildingAI (I think, it is referenced in the templates and that is how I remember it). I think it makes sense for ship AI behavior to non-random buildingAI instead of random buildingAI. 

For ships, I think non-random buildingAI is superior to random buildingAI and unitAI. Non-random buildingAI allows for positioning micro and allows the rock-paper-scissors in wow's proposal to work. Random buildingAI will mess that up by shooting randomly, so the efficient build in Wow's model is just a mix mashed together. UnitAI is similar to non-random buildingAI except ships will chase the first ship that they come into contact with. This chase can lead to problems, including luring with rock ships with rock and/or paper ships. 

Because naval battles are so clunky, I think the AI behavior should be something that emphasizes ship positioning. Non-random buildingAI seems best positioned for that.  

[real_tabasco_sauce]

ok so basically unitAI is responsible for just about all the entities, so ships and siege towers have UnitAI. BuildingAI is responsible for the building arrows.

in @wowgetoffyourcellphone's work, they no longer have anything to do with buildingAI, and instead behave like infantry, cavalry etc.

19 minutes ago, chrstgtr said:

UnitAI is similar to non-random buildingAI except ships will chase the first ship that they come into contact with. This chase can lead to problems, including luring with rock ships with rock and/or paper ships. 

By default they will target the first enemy they see unless you give a specific target, just like regular units. I think the ship classes do an excellent job on emphasizing ship positioning: Keeping vulnerable ships protected, flanking with ram ships, massing arrow ships together, raiding fish with scout ships.

The main thing is that ship gameplay won't be a total snooze fest XD.

[chrstgtr]

58 minutes ago, MacWolf said:

noooooooooooooooooooo, siege towers will revert to being useless, we having more units being used lately, people are understanding how to kill siege towers, don't revert all random arrows, theres some good parts about it !!!

You can just have separate behavior for siege towers and buildings. 

[chrstgtr]

10 minutes ago, real_tabasco_sauce said:

ok so basically unitAI is responsible for just about all the entities, so ships and siege towers have UnitAI. BuildingAI is responsible for the building arrows.

in @wowgetoffyourcellphone's work, they no longer have anything to do with buildingAI, and instead behave like infantry, cavalry etc.

So the overwork rakes out BuildingAI and replaces it with UnitAI? 

10 minutes ago, real_tabasco_sauce said:

By default they will target the first enemy they see unless you give a specific target, just like regular units. I think the ship classes do an excellent job on emphasizing ship positioning: Keeping vulnerable ships protected, flanking with ram ships, massing arrow ships together, raiding fish with scout ships.

 

That's all fine and good (and I agree with most of what you say). But because UnitAI chases, your ships may move without you doing anything. That can be problematic with how clunky ships are (ships getting stuck behind other ships, ships getting lured, etc.). Non-random BuildingAI wouldn't have that problem because ships would only move when you tell them to, which also means that an enemy can't exploit it as easily. 

UnitAI and non-random BuildingAI aren't really that different from each other aside from the chase aspect. 

10 minutes ago, real_tabasco_sauce said:

The main thing is that ship gameplay won't be a total snooze fest XD.

Long overdue. 

Edited May 22, 2024 by chrstgtr

[real_tabasco_sauce]

21 minutes ago, chrstgtr said:

Non-random BuildingAI wouldn't have that problem because ships would only move when you tell them to, which also means that an enemy can't exploit it as easily. 

UnitAI and non-random BuildingAI aren't really that different from each other aside from the chase aspect. 

AFAIK, ships will currently try and aggro enemy ships. Yeah the only practical difference is that the Building AI is a stream of arrows (or a multi-arrow blast :D) that can be done while moving.

[chrstgtr]

46 minutes ago, real_tabasco_sauce said:

AFAIK, ships will currently try and aggro enemy ships. Yeah the only practical difference is that the Building AI is a stream of arrows (or a multi-arrow blast :D) that can be done while moving.

Apparently it's been so long since any of us have played a game with ships (and for good reason) that we are all forgetting how the AI works. 

[ShadowOfHassen]

5 minutes ago, chrstgtr said:

Apparently it's been so long since any of us have played a game with ships (and for good reason) that we are all forgetting how the AI works. 

A solid map for sea would help. Currently I think we have one that is needed for naval battles

[chrstgtr]

23 minutes ago, ShadowOfHassen said:

A solid map for sea would help. Currently I think we have one that is needed for naval battles

We have maps that are functional enough. The most popular land map is mainland because it is just wide open space for battles. That more or less exists for naval maps.

Ships are just clunky to move, which isn't very fun. Naval battles also don't lend themselves to reinforcements which make the battles frustrating because it is either one short winner take all battle or it is a never ending battle where the player that wins the first fight becomes weak and cannot cross the remainder of the sea and land troops before fresh enemy reinforcements can arrive to wipe out the low health (and not repaired) victor from the first fight. 

Right now, ships are most fun when their primarily used as water-based siege that fight against land units/buildings.

Edited May 22, 2024 by chrstgtr

[Gurken Khan]

1 hour ago, real_tabasco_sauce said:

AFAIK, ships will currently try and aggro enemy ships.

From my experience ships will aggro on anything enemy, and have been for ~all versions; this includes chasing, so if you don't want your ships to move (because you're currently repairing them or not to get lured) you have to put them on standground.

[MacWolf]

1 hour ago, chrstgtr said:

You can just have separate behavior for siege towers and buildings. 

i hope he doesnt change siege towers atleast.

[wowgetoffyourcellphone]

2 hours ago, ShadowOfHassen said:

A solid map for sea would help. Currently I think we have one that is needed for naval battles

At least one good random map for naval play is sorely needed. Something good enough that it could be played semi-regularly in rated. There are some great skirmish maps, but they are mostly limited to 1v1 or 3 players. 

[wowgetoffyourcellphone]

2 hours ago, MacWolf said:

i hope he doesnt change siege towers atleast.

This kind of attitude prevents improvement. Siege Towers are pretty weird at the moment, but because some folks have managed to use that weirdness to their advantage we shouldn't improve them?

[wowgetoffyourcellphone]

4 hours ago, real_tabasco_sauce said:

The main thing is that ship gameplay won't be a total snooze fest XD.

We need to surrect a clone of @Stan` to improve the ship models.

1. Reduce their size by about 20%. This requires their animations to be re-exported. But would definitely help with pathfinding! 

2. Ramming Ship animations. Something akin to the battering ram actor moving back and forth, while the entity itself doesn't need to move. The whole affect only needing to occur in the animation and no need for new unit motion code.

3. The Roman and Carthaginian Quinquiremes are pretty ugly compared to the Ptolemaic one. Probably just need an all new model.

[ShadowOfHassen]

31 minutes ago, wowgetoffyourcellphone said:

At least one good random map for naval play is sorely needed. Something good enough that it could be played semi-regularly in rated. There are some great skirmish maps, but they are mostly limited to 1v1 or 3 players. 

My favorite sea map was the Caribbean from age of empires III. Everyone would start on a separate island and there'd be another island with treasure on it.

[Feldfeld]

13 hours ago, wowgetoffyourcellphone said:

What is different about our "engine" that makes non-random arrows not play well?

Currently 0 A.D. has quite high scale fights and I find it absurd that all the fortress garrison would focus a single unit to oblivion. All arrows follow the same path, it is ugly and hard to tell what is happening, is the fortress effectiveness reduced because of overkill, or missed arrows? I would also find it not nice having to keep valuable units out of a fight under fortress from fear of having them focused with all arrows by the enemy without drawbacks.

Not more than exactly one unit is damaged in a volley. It's weird that all defenders perfectly coordinate and are perfectly accurate (not as in "the unit will be hit", but the arrow follows exactly the intended path). There is no deviation, there is not a single other unit that will be hurt by a missed arrow.

Meanwhile, random arrows are spread into the enemy army, for me that is a behavior that makes more sense with a strong fortification in a large scale fight. Arguably, they were spread too perfectly, which is also unnatural in the other direction, but I still prefer it over the alternative.

And there are the concerns about rushes mentioned by @chrstgtr already.

[Stan`]

Can't we have a system were depending on buildings, up to max n arrows can be focused on a unit ? That number n would be the max between the value in the template and a ratio of the current number of arrows with the max number of arrows that can be shot from that building.

[chrstgtr]

11 minutes ago, Stan` said:

Can't we have a system were depending on buildings, up to max n arrows can be focused on a unit ? That number n would be the max between the value in the template and a ratio of the current number of arrows with the max number of arrows that can be shot from that building.

So random arrows but there can be manual targeting for some arrows? Sounds worth trying to me 

[Feldfeld]

13 minutes ago, Stan` said:

Can't we have a system were depending on buildings, up to max n arrows can be focused on a unit ? That number n would be the max between the value in the template and a ratio of the current number of arrows with the max number of arrows that can be shot from that building.

Imo that would be an improvement but I still also would like to see deviation with arrows being able to hurt a different unit than the target, not sure if that's possible without hurting performance

[Stan`]

1 hour ago, chrstgtr said:

So random arrows but there can be manual targeting for some arrows? Sounds worth trying to me 

Yep. Either a ratio, or a max of two or something.

1 hour ago, Feldfeld said:

Imo that would be an improvement but I still also would like to see deviation with arrows being able to hurt a different unit than the target, not sure if that's possible without hurting performance

I think arrows can still miss and hit something else, which is probably the reason of the lags @maroder noticed when there are many enemies.

[real_tabasco_sauce]

9 hours ago, Feldfeld said:

Currently 0 A.D. has quite high scale fights and I find it absurd that all the fortress garrison would focus a single unit to oblivion

So is it a realism argument?

9 hours ago, Feldfeld said:

there is not a single other unit that will be hurt by a missed arrow

Missed arrows damage nearby units.

9 hours ago, Feldfeld said:

Imo that would be an improvement but I still also would like to see deviation with arrows being able to hurt a different unit than the target, not sure if that's possible without hurting performance

To be fair, the accuracy pretty effectively accomplishes this, especially for forts. Also, the natural movement of units often spreads damage over a handful of units, as the "closest unit" changes often.

9 hours ago, Feldfeld said:

I would also find it not nice having to keep valuable units out of a fight under fortress from fear of having them focused with all arrows by the enemy without drawbacks.

If the player notices this (not easy, to be fair) keeping that unit in motion will dramatically improve its survival and even decrease the effectiveness of the building arrows. So there is ideally a nice back and forth should a player use the manual targeting like this.

9 hours ago, Stan` said:

Can't we have a system were depending on buildings, up to max n arrows can be focused on a unit ? That number n would be the max between the value in the template and a ratio of the current number of arrows with the max number of arrows that can be shot from that building.

I suppose a max number of attackers value could be used sort of like something @wowgetoffyourcellphone proposed a while ago. IMO this approach is heavy-handed.

One idea I brough up on overkill discussions was this: add a new range query for determining the closest unit with a flexible degree of rounding. In these large scale battles mentioned by @Feldfeld, what is currently "the closest unit" could be 2 to 5 units, or maybe more.

@Stan` is there already some sort of tiebreaking mechanism when two targets are exactly the same distance?

 

[ShadowOfHassen]

What if instead of a building focusing on only one person, it focuses on a few like 2 or 3?

[real_tabasco_sauce]

1 hour ago, ShadowOfHassen said:

What if instead of a building focusing on only one person, it focuses on a few like 2 or 3?

I was thinking something like this would be an elegant way to achieve this effect:

2 hours ago, real_tabasco_sauce said:

One idea I brough up on overkill discussions was this: add a new range query for determining the closest unit with a flexible degree of rounding. In these large scale battles mentioned by @Feldfeld, what is currently "the closest unit" could be 2 to 5 units, or maybe more.

[chrstgtr]

2 hours ago, real_tabasco_sauce said:

I was thinking something like this would be an elegant way to achieve this effect:

4 hours ago, ShadowOfHassen said:

What if instead of a building focusing on only one person, it focuses on a few like 2 or 3?

I don't think this addresses any of the underlying concerns with buildings' strength in early or late game. It is the same system just watered down and it will have the same effect as decreasing arrow damage, which has already proved undesirable. 

[Feldfeld]

13 hours ago, real_tabasco_sauce said:

So is it a realism argument?

Also valid as a gameplay argument. I'd like my fortress to damage multiple units at a time and I even find it strange some people don't want that.

21 hours ago, Stan` said:

I think arrows can still miss and hit something else, which is probably the reason of the lags @maroder noticed when there are many enemies.

 

13 hours ago, real_tabasco_sauce said:

Missed arrows damage nearby units.

 

13 hours ago, real_tabasco_sauce said:

To be fair, the accuracy pretty effectively accomplishes this, especially for forts. Also, the natural movement of units often spreads damage over a handful of units, as the "closest unit" changes often.

That may be possible in theory but in practice I didn't really feel that much. In AoE2 even towers can kill multiple units at a time.

13 hours ago, real_tabasco_sauce said:

If the player notices this (not easy, to be fair) keeping that unit in motion will dramatically improve its survival and even decrease the effectiveness of the building arrows. So there is ideally a nice back and forth should a player use the manual targeting like this.

That could be an interesting thing, but if every arrow goes for that unit then the player probably won't have the time to react before the unit dies. Also, having distinguishable arrow volleys would help with that (not saying that's not the case right now, I didn't play recently enough to remember)

[Stan`]

16 hours ago, real_tabasco_sauce said:

@Stan` is there already some sort of tiebreaking mechanism when two targets are exactly the same distance?

I'm not sure but IIRC units have preferences with regards to their target e.g. the Elephant class.