In D3868#167306, @wraitii wrote:

I don't really agree with this for Fruit (at least for the early-game berries). These are meant to be consumable, and a slow regeneration just introduces micro.

With normal usage those will be exhausted and disappear. Apply the patch and try it yourself. It's just a nice extra for non-competitive players who want to explore the possibilities of the game.
Players who want an infinite supply of food can build farms and players who want to maximize their food income should go for the corral. This patch doesn't change that.

Regarding Fish, I don't really mind, but again to avoid micro we would need to make sure that the fish stay on map once fully collected, and ideally that units don't start collecting the 'regrown' fish right away.

As for fish, I quite liked the sigmoid growth proposed in the ticket, but I believe that hasn't been implemented.

While the template_unit_ship_quinquereme.xml has five children, there are only three quinquereme actors. It's not that much more work to add dedicated artillery to all of them.
I quite like the idea. If I understand correctly, those subunits can fire independently, which means the ship itself could have an arrow attack instead (like the trireme) and get additional arrows from soldiers on board.

Should the unlock_champion_cavalry.json and unlock_champion_infantry.json be changed from "notciv" to "civ" then?

Thanks! That's one untrue thing less in game.

The Ptolemaic ship is not that supersized: it uses the Roman quinquereme actor. Perhaps all quinqueremes ought to have one or two pieces of artillery?
Are they included by default? Can they be destroyed separately? How does it affect the ship costs?

The issue is how to differentiate crossbowmen without making them more effective than archers, which would unbalance the game.

I think these little diversifications between chariots, camels and horses are unnecessary.

Camels are visibly taller than horses, both in reality and in game. Given that elephants have a higher vision range than cavalry and cavalry than infantry, it is reasonable for camels to have a vision range in between elephants and cavalry. And for the same reason I put chariots in between cavalry and infantry.

To be clear, I'm not necessarily opposed to giving crossbowmen a different range than archers. I was merely pointing out the “historically ...” argument doesn't hold in this case. And as usual, just because Age of Empires II does something doesn't mean 0 A.D. must do the same.
Crush damage is a possibility. However, slingers used to be problematic precisely because of their crush damage. Besides, people complained on the forums bolt-shooters had crush damage and stone-throwers pierce damage, blurring their functions. (This was changed in A24.) Nor do I really see why crossbow bolts should be able to raze structures. Historically (there we go again), ranged troops and artillery were used, by the besieged, to kill enemies within range of the city walls and discourage the attackers from approaching; and by the besiegers, to clear the walls of defenders, to be able to safely approach them.

Actually handheld crossbows could be used by the besieged and the besiegers alike, just like artillery. And like artillery, modern estimates of maximum and effective range vary greatly. Sure, a crossbow is a more powerful weapon than a longbow and large artillery more powerful than small artillery, however, more powerful weapons were generally used to shoot heavier projectiles, which had a greater impact (i.e. base damage in 0 A.D.), and heavier missiles have a shorter range than lighter ones, obviously.

It seems to me they mostly suffer from the comparison against rams (edit or elephants for that matter), which are extremely annoying to destroy, having 150HP more against Hack damage & infinitely more against pierce damage.

And rams and war elephant are a bit cheaper too, don't need to pack to move, and have more than 3× the damage per second of stone-throwers.

An increase of 5 resistance levels means a 41% damage reduction or, equivalently, a 69% health increase.
By the way, economic structures have a pierce resistance of only 20.

In tests I did, 50 basic rank archers without an upgrade can take down a stonethrower with just 9 shots.

50 vs 1? Keep in mind 50 archers are a lot more expensive than 1 piece of artillery and that the latter only costs 2 or 3 population.

• rebased because of D3690 / rP25154

Wood gathering significantly improved with D3607 / rP25000 and gather technologies became more effective in D3703 / rP25134.
I opted for 0.7 because female citizens are the basic gather unit and because it's nicely in between the other gather rates:

0.35 × √2 ≈ 0.5
0.5  × √2 ≈ 0.7
0.7  × √2 ≈ 1

All infantry has 80.

What value would you favour then?

What would be even better is to allow players to set custom starting values for each resource.

Actually I wonder why resource amounts are given names at all, just the number should be informative enough. The number of players and population cap settings have no names either.

https://en.wikipedia.org/wiki/Deathmatch
https://en.wiktionary.org/wiki/deathmatch
https://www.lexico.com/definition/death_match

In D3770#164280, @wraitii wrote:

I suppose we should also rename "Low" to something else someday.

Because it's not really low? Perhaps rename Low to Fairly Low (300) and High to Fairly High (1000) then? And maybe insert two more levels, Low (200) and High (2000).
I'm also in favour of making None (0) the default, but that's probably more controversial.

In D3770#164282, @Freagarach wrote:

The idea of Deathmatch is that there should be close to no gathering required.

The phrase “Deathmatch” is a bit of a misnomer anyway. How about renaming it to Enormous (50000)?

The idea is nice.
The attacker's surroundings shouldn't be revealed, nor the terrain it's standing on. I'm not even sure the attacker itself ought to be revealed. Perhaps you could try indicating only the position with something ugly (e.g. a magenta cylinder)?
Also add an option to opt out (or better: opt in).

You could use mul_round.

Yeah, I know. That's not what I meant, though. If resource costs and loot are made relative too, they ought to be multiplied by the same factor as build time and health are (e.g. medium segments have 4× the values of the shared wall parent). The base cost number is 6 (same as time) and by keeping the loot at 20% of the cost (as is the case for all structures) the base loot would be 1.2. Now 1.2 mul_round 4 gives 5, that's not the problem; the base number 1.2 is: loot must always be an integer.
Initially I wanted to use multiplications for costs and loot (and even wall piece length) too, however, when writing this patch, I noticed it created more problems than it solved, hence why I didn't include it here.

Yes, the idea is to make civic centres less dangerous by default and raiding somewhat easier.
Not every patch has to be committed, of course.

Yeah, but we already have special cost cases for maur and siege walls too, so it would be 4 cleaner files and 11 exceptions.
Also loot ought to be defined were costs are defined, and loot can't be multiplied easily in this case, because walls are rather cheap. Right now loot is about 20% of costs, the same ratio as other structurs have, but to apply the same multiplications of loot as those of costs (which are multiples of 6), loot would either have to become 17% (multiples of 1) or 33% (multiplies of 2).

Yeah, I agree. See how it's done in the right selection panel: if you select both units and structures, then the panel simultaneously displays (un)locking gates, (un)packing artillery, trainable units, and buildable structures:

The cost determines how expensive it is to repair gates and gates have a wood cost because upgrading long wall pieces to gates costs wood.

@Stan, it certainly looks nice! However, the left selection panel has a width of 5 icons. Medium wall segments have 4 slots, long wall pieces 8, gates up to 10 (the siege wall gate could have 16 if the lower platform level were used), and the unused double tower of the Mauryas 16. And one never knows how many slots might want to add to their structures (42?).
Moreover, what if multiple entities with slots are selected simultaneously (e.g. double-click on a wall piece)?

@Stan, if and when you have time and motivation, could you make I, II, III versions of the arrow.png, population.png, spear.png, and sword.png icons? And maybe _01.png, _02.png, and _03.png images with a transparent background too, so I don't have to bother you every time I want numbered versions of an existing icon.

I think it's fair, but two possible problems:

Hence the test plan:

Play a couple of games with the patch.

Eventually I'd like all structures to have exactly the same resistances (I'm thinking of 0 crush, 20 hack, 40 pierce), which means you only have to look at health to determine how strong something is, making comparing structures much easier; I believe @wraitii wants that too. However, currently structures have very different resistance values; to keep changes reviewable and understandable, I'm doing it in steps. Hence D3600, D3756, D3757; more will follow.
Why not change the hack resistance of defensive structures to 20 right here and now? Currently siege walls have a hack resistance of 15, city walls of 25. A difference of 10 levels is huge: it's equivalent to about a factor 2.868× in health. Giving both the same resistance without adjusting health would greatly distort the balance between them; and if health is altered, palisade health ought to be reconsidered too, thus things would quickly become quite complicated.
The idea is to keep this patch small and simple.

In D3600#160854, @Nescio wrote:

Yeah, I prefer 0 too, however, that makes foundations about 10% weaker; I think it should be done together changing the entity crush resistance of structures from 1 to 0 while compensating their health, i.e. something for a future patch.

The idea was to unify the resistances of all structures next. However, there are simply too many different values in too many files, so I've decided to do it in smaller steps.
[EDIT] The next steps are D3756 (ought to be uncontroversial) and D3757. More will follow later.

What kind of “in-depth testing” do you have in mind? There are many variables (map size, distance, directness of route, number of traders involved, enemies in between, etc.).
Anyway, displaying trade gain in the interface (D3692) would still be nice.

In D3599#162861, @Edwarf wrote:

In this case will they attack only units or buildings aswell, ?
Also, spreading the arrow over all ennemy target can be very strong because there is less overkill (that's why defensive building are very strong)

That's simply not true: defensive structures are very strong because they can adsorb a lot of damage, not because they have BuildingAI.
BuildingAI is generally a disadvantage: when facing a group of enemies, BuildingAI results in shooting one arrow at the first target within range, the second arrrow at the second target, the next at the next, and so forth. In contrast, UnitAI means the entity focuses on a single target and continues shooting at it until the target is destroyed (or out of reach and out of sight, or changes ownership, or the attacker gets a different order), before moving on to another target. With each target killed the enemy becomes less dangerous, whereas with BuildingAI, all targets are damaged a bit, but none are quickly killed, thus the enemy continues to inflict full damage for much longer.

In D3599#163294, @chrstgtr wrote:

If we can't do option 2

It's doable (and not difficult); the question is whether it's desirable.

Thanks for the clarification, now I understand the misunderstanding!

The idea of this patch is to make merchant ships simply the naval counterpart of land traders (in contrast to D3693, which makes them more effective). The movement speed increase (effectively a trade income bonus) is removed to make comparison easier.

• rebased (wall tower)

In D3680#163291, @chrstgtr wrote:

I personally like keeping the attack bonus at 25%.

That's what this patch does.

In D3680#163305, @ValihrAnt wrote:

It makes it very easy to afford this upgrade.

That's the idea: “still very expensive, yet also more affordable.”

In D3735#163365, @ValihrAnt wrote:

Why withhold cavalry upgrades from them though?

The purpose is to differentiate camels from cavalry. A camel is not a horse. Moreover, axemen don't benefit from swordsmen technologies either, nor are crossbowmen affected by archer technologies. Furthermore, if camels would remain cavalry, those technologies would practically cancel the movement speed penalty (0.9×1.1=0.99) and amplify the health increase (1.1×1.1=1.21>1.2).

(For the record, D3150 / rP25123 was what caused the conflict; had that one be committed a day later or this one a day earlier, everything would have been fine.)

Thanks! You missed one or two things, though. See D3752 (and verify I didn't make mistakes).

Something really went wrong here, only part of D3706 was implemented: compare the version that was accepted with the one that was committed. As a consequence this change does not remove the attack of wall towers, quite the opposite.

The recently added cavalry crossbowmen templates are missing (see D2886 / rP25122).

Can't you temporarily use the pack.png icon for now? It may not be perfect, but having a single button for two functionally different things is a rather problematic

Yes, that would certainly help.

It does make walls a bit harder to use, though.
And what if an entity has both (internal) GarrisonHolder and (external) TurretHolder?

And another thing: I can no longer choose and click to eject individual units in the left selection panel:

True. My point is visible slots are something specific to the entities (typically structures) that have them and the proper way would be to define what enities (typically units) can go into those slots in the files where the slots are defined, as used to be the case (with the outpost template specifying Infantry and the wall parent template Ranged+Cavalry). This commit effectively defines it in an arbitrary selection of units.

You never know what mods want to do, though:

What do you have in mind?

Why is <Turretable/> necessary? And shouldn't all units have it (cf. <Garrisonable>)?

One can still specify the entities that may occupy specific slots.

Good to know! I actually think it would be better to explicitly set it to Ranged Infantry on each wall slot, and Infantry on the outpost.

What does <Turretable/> do? Allow units to be placed on turret slots? Shouldn't crossbowmen have it too then? Or perhaps all infantry (see outpost)?
It used to be possible to specify what classes could occupy a specific slot, e.g. allowing a structure to have one slot for slingers and four others for archers. Is that still true?
And does this mean units on visible slots no longer contribute to recovering capture points? If so, it's a gameplay change.

Personally I would prefer animals not giving any loot at all, keeping experience exclusively for soldiers.

Nice catch!

To do: add a third level.

And still needs an icon, proper description, and interesting name.

As for this patch specifically, I would prefer it to be focused on a subset of technologies. Or if it must apply to every technology, then how about a time reduction instead of a resource discount?

In D3675#160589, @chrstgtr wrote:

Athens intellectual center-->cheaper techs

Yes, Athens was a centre of literary activity and many philosophers moved there. That doesn't mean it was technologically more advanced. Although it was not a rural backwater such as Sparta, even at its peak Athens was a provincial town compared to Babylon. Within the Greek world, the best doctors came from Kos (where Hippocrates lived) and advances in engineering (including artillery) came from the Greek city states in Southern Italy (the region where the Pythagoreans were active; not a coincidence) and Sicily. Syracusae was especially important, arguably the largest and most powerful city state in the Greek world (Athens' Sicilian expedition was a great failure) and the great rival of Carthage for centuries (Rome was a late entry); Syracusae attracted talent from everywhere (even Plato stayed there for a few years) and was also the first Greek state to have quadriremes, quinqueremes, and hexaremes. Polyremes may actually have been invented in Sidon, Tyre, or Carthage; those also produce high-quality goods such as glass, dyes, and textiles far superior to anything the contemporary Greeks could produce.
Virtually all surviving texts from the classical period (5th and 4th C BC) are from Athens and the Athenian dialect (Attic) was prestigious in the Hellenistic and later periods and formed the basis of Hellenistic and Modern Greek. As a consequence historiography is largely Atheno-centric and it is especially important to remain critical to avoid unjust stereotyping.

In D3675#160591, @ValihrAnt wrote:

Athens was the home to many philosophers and had a pretty decent education system.

Philosophers, true; “education system” is rather an exaggeration, though. It was mostly rich young men following and listening to gurus to learn how to win debates (an useful skill for those with political ambitions in the Athenian democracy).

The lower experience requirement affects all ranks, not just advanced. And the lower training time is essentially an economic bonus (as you already pointed out yourself).
Why exactly did you chose these modifications? And did you consider alternatives (e.g. higher health)?

The Greek word ὁπλίτης (hoplitēs) means ‘heavy-armed foot soldier’ (i.e. melee infantry in 0 A.D. terminology) and could be and actually was used to describe Macedonian phalangites, Persian kardakes, or Roman legionaries as well. Moreover, the traditional hoplite wasn't uniquely Greek either, Carthaginians, Etruscans, and various Italic peoples fought in the same way. Furthermore, only the wealthy fought as hoplites, since a full panoply (arms) was rather expensive.

Don't you want to wait for Vulcan?

• revert ranged cavalry unit movement clean up

Maybe increase their <Promotion/RequiredXp>?

Idea started by @wowgetoffyourcellphone at https://wildfiregames.com/forum/topic/36934-civ-differentiation-athens/page/2/?tab=comments#comment-421616

Was it? It's not in the linked forum post.
See also https://code.wildfiregames.com/D3688#160875

• Archer, spotted by @Freagarach
• attack spread
• movement speed

Length wasn't an issue for the snap-to-edges option:

Entity names:

• Specific, then generic names
• Generic, then specific names
• Only specific names
• Only generic names