[Idea to improve capturing (dunno if it would be possible)]

Quote

This could simulate the idea that maybe the walls and doors were damaged allowing invaders to enter through the breaches.

Note that attackers didn't necessarily need to damage the walls to capture the city. Surrounding the enemy and starving them out was a popular strategy. In some sieges, a traitor would open the gates for the attackers. Siege ladders and siege towers were used to put troops directly on top of the walls without needing to damage them. Huge siege ramps, made out of dirt, sloping up to the top of the wall, were another strategy.

[How to quickly locate code for menu/unit from within the game?]

What I might do is look at some text in the GUI that appears over the item, then run a recursive grep for that text in the source tree that I have checked out from svn. I find the following bash function useful.

function codegrep () {     
    find ${2:-.} -regex '.*\.cpp\|.*\.h\|.*\.hpp\|.*\.js' -exec grep "${1}" '{}' +
}

You can also look in xml files for menus and stuff (just add |.*\.xml to the regex), though that will take a lot longer. And if you know the ancestor directory, such as source/ for cpp or binaries/data/mods/public for javascript and xml, that will also speed it up a lot.

[It looks like the animals are climbing where they shouldn't be.]

[Quick experiment with shorter turn lengths]

Range queries can be heavily optimized.

• Do the target filtering in C++ not in JS.
• Use a quadtree instead of a fixed grid so that you can more quickly find the closest unit without having to iterate over all the units in a grid cell.
• Don't generate the entire list of units in range and then filter afterwards. Instead, filter as you iterate over the units, and return as soon as you have a matching one.

 

My own computer is too slow with the current turn lengths. A large battle becomes almost unplayable as it drops to 2 FPS and the camera stops scrolling properly.

["auto-sniping" thoughts and tests]

1 hour ago, Atrik said:

 

True, the test I did with NoobDude WAS emulating the sniping of 1 unit attacking 1 unit automatically. I can show you if interested. I am not sure it's necessary to have a feature doing perfect sniping, If units would attack individually the closest enemy (matching the criteria) to themselves, this would already naturally spread the unit focus.

I insisted this was a illustrative example, demonstrating that "sniping" wouldn't require any bots or scripts, but would look like a very basic amendment, if ever it was considered like a desirable feature.

There would still be significant overkill, especially with archers, because at any time there are only a handful of the "most forward" enemy units. Might be an especially large overkill with archers because all the archers stand in roughly the same spot so they would all have similar or the same "closest" units.

Perfect sniping wouldn't be 1 unit attacking 1 unit, but maybe 5 attacking 1, depending on the average turns to kill. It is better to kill 50% of the enemy army while leaving the rest untouched, than to damage the whole enemy army by 50%. So you do want to focus on a subset of units, while not focusing so much that overkill is excessive.

["auto-sniping" thoughts and tests]

21 minutes ago, Atrik said:

else if (Engine.HotkeyIsPressed("session.attackWeakUnits"))
   targetClasses = { "attack": ["Ranged", "Support"] };

2 Lines of code => Hotkey to do "Auto-sniping" available in vanilla 0ad to everyone.

That's only part of what sniping does. The other part is that you distribute your fire over a big portion of the enemy army, instead of wasting tons of damage on overkill of one unit at a time.

[Alpha 27 - Roman Fasces might be interpreted as a fascist symbol]

@hyperion As long as it is a faithful representation of an authentic Roman fasces, there is no problem. Again, the current minimap logo is much closer to this flag of Italian fascism than to any authentic Roman fasces artifact.

In the link you most recently posted (here) there are some apparently authentic Roman fasces, though we can't be quite sure because it is a modern reproduction rather than an ancient original. Anyway, those fasces could be distinguished from various Italian symbols by the shape of the blade. The similar Italian ones curve up around the animal head, where the authentic ones have a straighter axe head. They could also be distinguished by the size of the eagle head; the Roman ones have a larger eagle head than the Italian ones. By faithfully reproducing these differences it can be made clear that it is a Roman fasces and not an Italian one.

[Alpha 27 - Roman Fasces might be interpreted as a fascist symbol]

1 hour ago, hyperion said:

Fascist symbol:

https://en.wikipedia.org/wiki/File:Fascist_symbol.svg

Logo of the PNF:

https://commons.wikimedia.org/wiki/File:National_Fascist_Party_logo.jpg

 

They don't look similar but the coat of arms of St.Gallen does somewhat:

https://en.wikipedia.org/wiki/File:Coat_of_arms_of_canton_of_St._Gallen.svg

 

So for me as long as it's clear what is the source and the source is unproblematic the current logo seems fine. Not that I object to changing it either.

 

You may have missed my earlier post with the pictures. The current minimap icon is a close match to the flag of Italian fascism (different from what you linked) based on details of the axe (the diamond-shaped spike, the broad head, the square boss joining the two).

The Canton of St. Gallen formed in modern times (1803) so its coat of arms may be modern as well. It would be better to base the art directly on an original Roman artifact.

[Alpha 27 - Roman Fasces might be interpreted as a fascist symbol]

I want to say I don't have an objection to using a fasces as a Roman symbol - I'm not one to censor history because it offends modern sensibilities. But, if there is to be a fasces, it had better be very clearly based on a real Roman fasces, and not based on the flag of the Italian fascists. For example, if it had two opposing long and skinny axe blades at the top (instead of a wide axe blade and a diamond-shaped spike), like the rusted Roman fasces above, that would be sufficient to identify it as a Roman one and not an Italian one.

[Alpha 27 - Roman Fasces might be interpreted as a fascist symbol]

I see there are minimap icons now that show on unexplored areas of the minimap. The Roman minimap icon is a fasces:

The fasces is an authentic symbol of ancient Rome, but extraordinary care is needed, because it is also a symbol of modern Italian fascism. Does this accurately depict an authentic Roman fasces and not a modern Italian fasces? It is necessary to find an authentic archaeological Roman fasces that matches the picture.

I have doubts. The style of axe is nearly identical to the axe on the flag of Italian fascism. The axe head projects out of the top of the bundle, and it has the same shape of curved blade on the right, the same type of diamond-shaped spike on the left, and a square center-piece.

Doing some google searches, I haven't been able to locate an authentic archaeological Roman fasces with those precise features on the axe. Therefore, unless such an example can be found, this minimap art is representative of specifically Italian fascism and needs to be changed.

For comparison, here are some authentic Roman fasces:

 

[Has Building capture been eliminated in A27]

26 minutes ago, Stan` said:

IIRC that breaks attack move which is why it wasn't used.

EDIT: I think @Freagarach created a ticket for the default behavior to be optional 

Breaks it how? I don't see any problem with attack-move when binding capture and attack-move to Ctrl.

 

Ah, it's not attack-move that's the problem, it's garrison. You can't garrison in a structure without full capture points if you bind capture and garrison to the same key. https://code.wildfiregames.com/D4908

[Has Building capture been eliminated in A27]

Or you can use this mod I just made to restore the old behavior where you press the attack key to attack a building, and otherwise capture it by default.

oldcapturebehavior.zip

[Has Building capture been eliminated in A27]

57 minutes ago, seeh said:

hmm. the c key is already occupied in my autoCiv configuration (a important key). i love the old style. maybe a config option , will be nice/possibillity?

You can put "capture = Ctrl" in your local.cfg under hotkey.session, and delete "attack = Ctrl" there. Then you can capture by pressing Ctrl instead of c.

[cannot deselect units from group once it's formed]

Disable the formation.

Or you could go to Options -> Game Session and turn the "Battalion-style formations" option off.

[Alpha 27 Pre-release/Release Candidate Build Testing]

5 minutes ago, real_tabasco_sauce said:

eles have been redesigned to be more effective fighters, not just a second ram. They still do well vs buildings, but their primary role is now vs infantry.

Looks like their crush damage was cut to a quarter of its original value, in exchange for splash damage? So they aren't like rams at all anymore.

Anyway, for fighting infantry, it's pointless to pit them against 75 spearmen because that's not the army composition players actually use. Try instead a realistic army, say 80 archers and 20 spearmen vs 80 archers and 3-4 elephants (similar total cost, though in fact the elephants are more costly because metal is more valuable than wood).

[Alpha 27 Pre-release/Release Candidate Build Testing]

20 minutes ago, Lion.Kanzen said:

It's not about combat counter...

It is a distractor unit that serves as a metashield or mattress to stop elephants.

 

You need a melee unit to keep the elephant busy.

But the logical thing to do is to stop the elephant with something so that it doesn't destroy your buildings.

 

While you kill it with counters.

 

 

In that case (if you block the elephant with spearmen) it still doesn't matter how long the spearmen take to kill it, because your ranged units are going to deal most of the damage to the elephant.

[Alpha 27 Pre-release/Release Candidate Build Testing]

In what world does spearmen vs elephants matter? Elephants are used against buildings, and are killed by skirmishers, slingers, or archers.

[Disable unit overlapping]

@Yekaterina It had the problem I mentioned: units traveling any distance would tend to narrow into a single file line. Also it had serious performance issues.

[Disable unit overlapping]

This is an issue with unit motion, and unit motion is really freaking hard. I made some attempts back in like 2017 to fix unit motion at the time, without full success.

After coming back to the game after years of absence, I'm actually quite pleased with how a26 unit motion works. It's true that units can be packed more than before, but unit used to have the opposite problem where they'd travel single-file everywhere with excess spacing.

Anyway, there's room for improvement in helping units to not bunch up so much, but it's not so simple as just enabling or disabling a setting. You'd have to implement a different algorithm (and then make sure it's performant enough, and doesn't interfere with woodcutting/gathering).

[Resume playing from a replay?]

I want to play a replay for a few minutes and then switch over so it's no longer a replay and I control the units.

My question is, has anyone worked out a method for doing this?

I'm trying to work out a method myself. Perhaps it could be done by playing the replay, then doing "save game" partway through, and then parsing simulation.dat for the saved game and changing whatever variables mark it as a replay.

[Alpha 27 Pre-release/Release Candidate Build Testing]

The Harmya (Palace) is where Mauryas train heroes and maiden guard champions.

[What's the point of siege walls converting to enemy?]

4 hours ago, Atrik said:

Romans can build walls inside your base, and since you can't captured them, you need to ram it all, even if it isn't defended by any garrison then?
As romans you could build walls anywhere and keep control over gates until they're ramed down?
How OP would that be? I'll say this looks annoying but...

How would being able to lock or unlock a gate (without being able to see it) be OP? Siege walls are an age 3 building so you can expect players will have rams by then.

The main use I can think of for building siege walls in enemy territory would be to build little squares to protect catapults and bolt shooters against melee cavalry raids. If you're thinking of "nuisance walls" that prevent the enemy from walking around his base, you kind of have to already defeat his army before you could build those. More of a troll strategy than a serious one, since if you've defeated his army you could instead just ram his cc.

[What's the point of siege walls converting to enemy?]

Also, garrisoned wall turrets do not shoot arrows and do not have any LOS.

Permitting siege walls to be garrisoned, and then relying on that to prevent territory decay, sucks for several reasons.

• Once they convert to Gaia or the enemy, you can't recapture them. You have to continuously occupy them or lose ownership over them forever. That's just... weird.
• New players won't be aware that they need to immediately garrison their siege walls after building them, so their first reaction is still going to be confusion that their new building is now the enemy's.
• Under current settings, the garrison can't shoot or see. That's not very nice. Better not to have any garrison than to have one that's locked in the coal cellar.
• It makes siege walls harder to make use of, as it costs you some soldiers. Who actually uses siege walls anyway? Siege walls need a buff to make them worth using.

Better to instead continue to not permit garrison, and just disable territory decay entirely for Roman siege walls.

[What's the point of siege walls converting to enemy?]

On 12/05/2023 at 1:20 PM, Player of 0AD said:

If I'm not mistaken they can be garrisoned by ranged infantry, but this is still no great solution

 

18 hours ago, chrstgtr said:

Everything you say is irrelevant--garrisoned walls already don't decay/get captured.

 

13 hours ago, chrstgtr said:

Force players to garrision them like they do with camps.

12 hours ago, real_tabasco_sauce said:

Unfortunately, I failed to double the siege wall garrison space like I did with all the other walls. Some people will probably not like that inconsistency, and it means it will remain a chore to keep the walls from being captured.

Fortunately the siege walls are fairly easy to destroy if I remember correctly.

Let's get one thing straight. In the current state of the game (a26 and svn), you cannot garrison walls. I mentioned this in the OP. The only thing you can garrison is stone wall turrets, which still leaves the wall to decay, and not even the turrets can be garrisoned for siege walls.

[What's the point of siege walls converting to enemy?]

@chrstgtr Well, it depends on what makes more sense for gameplay. If Roman siege walls are intended to be built in enemy territory, then those walls shouldn't decay when they are built there. But normal walls could decay.

Building walls and trenches in enemy territory was a major aspect of siege - protecting your troops from the archers in the enemy fortress, while allowing your troops to get closer to the fortress. Actually I'd also like for walls to block low-arcing ranged attacks, depending on the height of the wall.