Page MenuHomeWildfire Games

[gameplay] add visible garrison points to gates
ClosedPublic

Authored by Nescio on May 22 2020, 8:36 PM.

Details

Summary

Now placing units on walls no longer interferes with the opening and closing of gates (D1418/rP23710, rP23714, D2775/rP23731), it becomes possible to add visible garrison slots to gates as well.
Currently people can garrison up to five ranged infantry on long wall segments. This patch gives most gates slots directly above the doors. However, there is no parapet visible above the doors of the kush gate, which therefore gets three slots on either side:


Likewise, the rome_siege_wall_gate, which consists of two square towers with doors in between, gets eight slots due to symmetry.

[EDIT] Because different civs have different actors, their gates have different numbers of slots:

  • 4: brit, gaul, maur;
  • 5: sele;
  • 6: kush, pers;
  • 8: athen, cart, iber, mace, spart, and rome_siege_wall_gate;
  • 10: ptol, rome.

See also D2477, D2769, D2783.

Test Plan

Check for mistakes and omissions.

Event Timeline

Nescio created this revision.May 22 2020, 8:36 PM
Owners added a subscriber: Restricted Owners Package.May 22 2020, 8:36 PM

Build failure - The Moirai have given mortals hearts that can endure.

Link to build: https://jenkins.wildfiregames.com/job/docker-differential/2205/display/redirect

Any opinions on the number of visible garrison slots on gates? In principle they could accomodate more units.

Nescio edited the summary of this revision. (Show Details)Jun 3 2020, 12:44 PM
Nescio added a reviewer: Restricted Owners Package.
Nescio removed a subscriber: Restricted Owners Package.

I personally somewhat dislike that we rely so much on garrisoning for defence, so I would keep it limited, but I like how the effect looks, so it's good to have the possibility. The first concern is more related to balancing ofc.

Nescio planned changes to this revision.Jul 20 2020, 1:01 PM
Nescio updated this revision to Diff 12835.Jul 21 2020, 7:37 PM
  • rebased
Owners added a subscriber: Restricted Owners Package.Jul 21 2020, 7:37 PM

Successful build - Chance fights ever on the side of the prudent.

Link to build: https://jenkins.wildfiregames.com/job/docker-differential/2739/display/redirect

borg- added a subscriber: borg-.Aug 4 2020, 3:02 AM

I didn't get any bugs, but I'm not sure I tested it enough.

genava55 accepted this revision.Aug 4 2020, 10:01 AM
This revision is now accepted and ready to land.Aug 4 2020, 10:01 AM

Good initiative, in the past I found it was a lacking feature for the defense.
The only thing is the difference between the factions. Maybe giving 6 slots to everyone and 8 to the Romans could be better from a balance perspective. But that's a detail.

The only thing is the difference between the factions. Maybe giving 6 slots to everyone and 8 to the Romans could be better from a balance perspective. But that's a detail.

Before proposing this patch I experimented a bit and tried giving everyone six slots. The problem is that the various gate actors have different designs. The outer parts can't be used, due to overlap with wall towers. As mentioned earlier, if the gate is open above the doors (kush), then the number must be even, i.e. two, four, or six. Gates that are level (athen, cart, mace, spart) can support up to eight in a row, like long wall segments. If gates include towers (iber, ptol, sele) there is room for six (which also looks best for pers). For those that have a roof in the middle (brit, gaul, maur) four looks best, five works too, more doesn't. Then there is one with a platform in the middle (rome), which is deep enough for two rows of five, but not wide enough for one row of six. Giving each civ a different number is probably not very fair. Therefore I opted to give gates five by default (the same number long walls currently have), and make an exception only when the centre is open (kush; rome siege).

Ok. Thanks for the clarification. Still fine to me.

I like the aesthetics of this. @balancing (@Feldfeld, @borg-, @badosu, @ValihrAnt)?

borg- added a comment.Aug 6 2020, 11:57 AM

I like the aesthetics of this. @balancing (@Feldfeld, @borg-, @badosu, @ValihrAnt)?

I really like this.

Do the symmetric ones fill up one tower first then the other? Or are entities added evenly? If the former, perhaps do the latter. Also perhaps you can use more descriptive names for the turret points of the symmetric ones (e.g. Tower1Pos1 or something).

I'm not sure there are any balancing considerations at this point since stone walls are severely under used at the moment. Looks like a good improvement though.

Any opinions on the number of visible garrison slots on gates? In principle they could accomodate more units.

Personally I would say: "pack 'm full" ;)

Personally I would say: "pack 'm full" ;)

The gates of different civilizations have different designs, which means some have space for only 4 or 5 spots, others for 8 or 10, unlike long walls, which all are fundamentally equivalent, and can easily support 8 (see D2783). So the question is whether it's desirable for gates of different civilizations to have different numbers.

Do the symmetric ones fill up one tower first then the other? Or are entities added evenly? If the former, perhaps do the latter. Also perhaps you can use more descriptive names for the turret points of the symmetric ones (e.g. Tower1Pos1 or something).

No, there is no fixed loading order. It seems units occupy the nearest open spot. Neither the node name nor the position they're listed in the templates matters. That's also why I preferred the node names One, Two, etc., which are generic and applicable to basically anything, rather than First, Second, etc., which imply a fixed order.

I'd say the position in the template does matter actually, since the code finds the first free spot and occupies that.

The gates of different civilizations have different designs, which means some have space for only 4 or 5 spots, others for 8 or 10, unlike long walls, which all are fundamentally equivalent, and can easily support 8 (see D2783). So the question is whether it's desirable for gates of different civilizations to have different numbers.

Which boils down to balancing again I guess (although it has been said before that walls are not made often). For me aesthetics is more important but I am aware that I am not the only one working on this project.

I'd say the position in the template does matter actually, since the code finds the first free spot and occupies that.

If that's the case then I failed to notice it in game; to me it seemed units merely occupied the nearest free spot; I guess I was mistaken then.
Anyway, the nodes are listed from centre outwards (also in D2783), so it shouldn't really matter in practice.

Nescio removed a reviewer: Restricted Owners Package.Aug 15 2020, 9:04 PM

Anyway, the nodes are listed from centre outwards (also in D2783), so it shouldn't really matter in practice.

I just tested this, but the nodes are not occupied from the centre outwards, rather the other way around. (This is because the template is read in the specified order, but stored in reverse.)
They do load, however, symmetrical, which is nice.

This comment was removed by Freagarach.
Nescio updated this revision to Diff 13276.Aug 24 2020, 1:20 PM
  • rebased

I just tested this, but the nodes are not occupied from the centre outwards, rather the other way around. (This is because the template is read in the specified order, but stored in reverse.)
They do load, however, symmetrical, which is nice.

Even better!

I'm not sure there are any balancing considerations at this point since stone walls are severely under used at the moment.

@Nescio It looks like this also allows for unequal amounts of entities on walls? Perhaps the only thing to take into account is that the amount of entities allowed does not exceed the amount allowed on a piece of long wall (after D2783).

Nescio added a comment.Sep 3 2020, 2:41 PM

It looks like this also allows for unequal amounts of entities on walls?

What do you mean?

Perhaps the only thing to take into account is that the amount of entities allowed does not exceed the amount allowed on a piece of long wall (after D2783).

That's not a problem, if an entity is upgraded to one with fewer garrison capacity (or turret slots), then upon finishing the upgrade units beyond the new maximum are automatically ungarrisoned.

What do you mean?

I mean that since there is no balancing consideration you can just increase the amount of slots.

That's not a problem, if an entity is upgraded to one with fewer garrison capacity (or turret slots), then upon finishing the upgrade units beyond the new maximum are automatically ungarrisoned.

It is not a problem in that sense, but rather that people won't just build gates all over their wall instead of keeping also pieces of walls.

Nescio added a comment.Sep 3 2020, 2:51 PM

Now you've lost me.
Upgrading all long wall segments is certainly possible, yet it's expensive and gates have less health, thus unadvisable, but if people want, they can do it.

I mean that, IMHO, it would be weird if a wall has less slots than a gate. Also because people may just convert all of their long wall segments to gates because there fit more entities on a gate so have a greater stopping power to anything other than siege. Of course people _can_ still do that, but there would be less reason to do so.

Nescio added a comment.Sep 3 2020, 3:15 PM

If gates are clearly larger (e.g. those in Millennium A.D., or gatehouses in reality), then I don't see why they can't have more garrison slots. Anyway, this patch gives most gates five slots, the same number as long wall segments currently have, and if gates must have strictly fewer, then I suppose D2783 ought to be committed first.
Regardless, each soldier can't gather resources, build, or fight elsewhere while garrisoned, and they contribute to the population limit, unlike towers.

strictly

Just my opinion ^^ Nothing strict here. (For the record, I meant <=.)

this patch gives most gates five slots

So the question is whether it's desirable for gates of different civilizations to have different numbers.

Which boils down to balancing again I guess (although it has been said before that walls are not made often). For me aesthetics is more important but I am aware that I am not the only one working on this project.

Ergo, I think you should add more spots wherever possible and if there are no objections we can finally get this in ;)

Nescio added a comment.Sep 3 2020, 3:30 PM

You want gates of different civs to have more slots than those of others? Fine by me.

You want gates of different civs to have more slots than those of others? Fine by me.

Well, I thought we discussed that earlier ^^

Nescio updated this revision to Diff 13379.Sep 3 2020, 4:27 PM
Nescio edited the summary of this revision. (Show Details)

Different numbers, as requested by @Freagarach:

  • brit, gaul, maur: 4
  • sele: 5
  • kush, pers: 6
  • athen, cart, iber, mace, spart: 8
  • ptol, rome: 10
  • rome siege wall gate: 8 (could be doubled to 16 if the lower level is also used)
Freagarach accepted this revision.Sep 3 2020, 5:44 PM

(could be doubled to 16 if the lower level is also used)

IIRC That would need the floor to be lifted a bit? Either way, eight is fine for now.

Changes look very nice in-game.

Freagarach requested changes to this revision.Sep 3 2020, 5:57 PM

Oh, almost forgot, but you need to add the wall-aura.

This revision now requires changes to proceed.Sep 3 2020, 5:57 PM
Nescio updated this revision to Diff 13381.Sep 3 2020, 6:09 PM
  • now with aura
Freagarach accepted this revision.Sep 3 2020, 6:31 PM

Thanks for the patch :)

This revision is now accepted and ready to land.Sep 3 2020, 6:31 PM
This revision was automatically updated to reflect the committed changes.