• redone from scratch
• delete trader movement speed technology
• exclude ships from the health technology and change its costs
• make first technology

• redone from scratch
• exclude problematic Spartan women history string

• reduce spread, per @wraitii

Tnx. Add+10 pop is better.

Right now it has nearly double the population of the house (15 vs 8).
[EDIT] Besides, they count towards the city phase requirements.

• redone from scratch
• make cart apartment independent from cart house
• make cart house 20% smaller to increase the difference

Some more things need to be changed.

Yes, I'm aware of that; I think it's fine. The Seleucids have to choose, whereas the Macedonians get two infantry champions with the technology and some civs with specific structures have three (Athenians, Kushites, Mauryas). Besides, the Seleucids have both cavalry and chariots at the stable, so in the end they'll be spending just as much on unlock technologies as most factions.
Thanks for the heads-up, though; I'll update the summary.

Just to be clear, I meant 4, 3, 2 with the rank spread reductions, i.e.:

           j   ,  s   ,  a
basic   : 4    , 3    , 2
advanced: 3.2  , 2.4  , 1.6
elite   : 2.56 , 1.92 , 1.28
champion: 1.6  ,      , 0.8
hero    : 0.8  ,      , 0.4

Still fine?

Well, we'll see. They can always be made stronger or cheaper in A25.
Anyway, the idea of having separate structures for separate functions is to force players to make choices on what to prioritize, rather than just booming and building a structure that's best for everything.

Anyway, as @borg- reminded me repeatedly, having some civs able to train their champions at the barracks, others at (not too expensive) specific structures, and some only at the (still expensive) fortress is neither consistent nor balanced.

Defensive strongpoint + territory root.

Exactly! That's precisely the point. Their purpose is to defend your territory, there are other structures for training units. Fortresses have been made a bit more affordable too (D3324/rP24552).

Hero training area (for some civs)

I'd like to see heroes removed from them as well (it's actually more of a penalty compared to civs that don't), but that's something for a future patch.

I mean upgrading your archers to elite is much more important than for other units.

How quickly an unit gains experience is based on how quickly they kill their opponent, and since archers inflict less damage per second than javelineers, they promote more slowly.

I find missing arrows 100% annoying.

Well, I kind of like the unpredictability.

Well yeah, but also no. The point of spread is that ballista can't be used against units effectively. Not that archers can't be used against units effectively. Our archers are basically very very micro heavy

Isn't the idea ranged troops are ineffective vs single targets, but effective vs massed groups?

Basically I disagree with the spread increase here, and I think we should instead decrease in general, possibly make this somewhat a non-factor for these units.

So what numbers would you prefer? Basic javelineers 4, slingers 3, archers 2? Or even lower?

In D3381#150241, @Angen wrote:

just note:
someone updated ranges here: https://code.wildfiregames.com/rP24539
after https://code.wildfiregames.com/rP24217 was already commited

Yes, I'm aware of that. The purpose of D3297/rP24539 was not to rebalance ranges across the board, it was merely to remove some oddities (e.g. elite soldiers having more range than their champion counterparts).
More people did some testing recently and concluded the attack range of archers and structures is too high, hence this patch.

Obsolete because of rP24623.

Sorry, I don't follow why?

Recent feedback indicated people are concerned about champions, about ranged troops, and about archers. While elite archers become more accurate, basic archers remain unchanged, and champions (and javelineers) are a bit less effective, so I'd say overall it's an improvement.

Mostly my point is that this makes archer XP much more useful than any other XP, which doesn't really seem "working as designed"

What do you mean with XP?

I did a basic test against a dummy infantry unit, and I got at max range, for the same amount of time:

Thanks for testing. Basically javelineers are most effective, which is hardly surprising, given their much higher base damage. To me it just confirms increasing javelineer spread is an improvement.

Therefore, I think archer DPS is unpredictable

Isn't that precisely the point of spread? If we want predictability, we might as well give everything a spread of 0.

I would prefer to reduce Spread across the board, so that DPS is more predictable, and then debuff accordingly.

Probably the most important thing to address is attack ranges (D3381); the new way of calculating (which is definitely an improvement) has the consequence maps are relatively smaller and rushes harder.
Spread ought to be done next. I don't particularly care about the precise values, I was just explaining why I kept basic archers at 3 for now. However, I do think the proposed b:a:e:c:h ratio is better than the current situation, with the big jump of champions.
Finally, change base damage (in D3246 or elsewhere), simply because this is easiest to adjust.

The Roman version is problematic on all maps, the Carthaginian only on water maps and useless elsewhere, therefore it's an improvement to delete both.
And the patch is complete.

Elite archers are probably _much_ stronger than CS archers with this patch, and this is hard to read.

True, they are, but champions are less accurate, so overall I think it's fair.
Elite archer spread is lowered by 36%, but if basic archers were to have a spread of 2, then elite archers would have 1.28, i.e. 57% less than what they have now.
More importantly, when javelineers and archers are firing at each other, archers are clearly more accurate.

I guess the higher gravity makes trajectories more of an arc? Pls double check at close range then.

Gravity is hardly noticeable at point-blank but it is at longer distances; try a gravity of 0 and then of 1000. Anyway, that's beside the point: rP23144 intentionally raised gravity to approximately half the speed, but rP24458 raised speeds without adjusting gravity, which I noticed when tweaking the spreads, which is why I included it here.

Yes, I kept the basic archer spread unchanged, both as a point of reference for the rest and because @Feldfeld pointed out in a private message yesterday that archers might be too strong already, so I'd rather change spread by too little than by too much.
As for hunting, just move your archers closer.

Keep in mind you still need food.

• do the right thing

Good to know, thanks for checking!

At the request of @borg-.

A grep -r shows the "Worker" class is use in baseManager.js, headquarters.js, and startingStrategy.js and the "CitizenSoldier" class in attackPlan.js, baseManager.js, headquarters.js, and victoryManager.js.
I don't know what they're used for or how the AI works, though.

That's fine!
Also, rP24635 might be problematic for the AI as well.

Great, thanks!
Do you think you could also make the AI able to use the Seleucid choice (traditional army vs reformed army)?

This penalty makes no historical sense (bonus: D).

Well, Sparta (and Athens) had a smaller army than e.g. the Macedonians, Mauryas, Persians, or Seleucids, but that's because the latter were simply much larger.

For the record, I dislike the approach taken here, with these unlock technologies. I'd prefer simply raising the cost of the city phase or the stable.
I'm also worried whether the AI will actually make use of this. I've observed (without this patch) the AI researches phase advances and economic technologies, but it seems to ignore everything else. It does not even take the Seleucid champion choice, despite it being free.

Now we need a patch to remove infantry champion from the castles and all ok, tnx for you hard work @Nescio .

Yes, I'm working on it, but it's more complicated, due to different champion infantry file names.

Fine by me.

To do, figure out if the Worker or Citizen class needs to removed (especially for the AI).

The Persians already have a lot of cavalry things:

• four types of citizen cavalry and two types of champions (most other civs have two and one, respectively)
• a very cheap town-phase technology giving cavalry −10% training time
• a technology enabling training citizen cavalry at triremes
• a technology increasing giving champion cavalry spearmen +20% health

Therefore I think it's better they're required to research both those unlock technologies.

No, it's a mistake, hack technologies should cost food and metal, pierce technologies wood and metal, thank you for the notification.

Yes, I know this patch is about cavalry, that's why I wrote what I wrote:

• if you lower cavalry hack resistance, then infantry swordsmen are a bitter better at killing them
• if you raise cavalry pierce resistance, then infantry spearmen and pikemen are slightly worse at killing them

I guess a separate patch would be nice, this is rather unrelated and I'm not sure we should merge it for A24 (I'd rather ranged units be not too strong, and this still buffs them).

Here you go: D3374.

The resistance changes effectively mean infantry swordsmen become a bit more effective and spearmen and pikemen slightly less.
I'm neither in favour nor against this patch.

I don't think it's necessary to divide between cavalry and chariots, for me both are cavalry.

It's not strictly necessary, but it makes no difference for Britons and Mauryas, is not more effort right now, and saves work if an when chariots are truly separated from cavalry.
For the Persians it means they have one technology for unlocking the cavalry spearman and one for unlocking the chariot archer, forcing them to choose which one to prioritize. This seems fair to me, given that other civs only get one champion per technology as well, and the Persians have four types of citizen cavalry by default, whereas most other civs have two.

Copy-pasting now, are you? Be careful, I can't guarantee the summaries of my patches are free of errors.

• unlock technology for chariots
• lowered cost to 600 food

Any update on this?
The gui scale is an option many could benefit from, but most people are unaware of. Being able to set it from within the game is important.
I still think it ought to be listed under “Graphics”, but since there is no room there (D3034 hasn't got much traction lately), it's better to have it under “General” than not at all.
While I doubt anyone would really need a value greater than 200%, I'm not against allowing giving people the choice of a scale up to 300%, in case someone wants to make very little use of a very high resolution screen (who knows).

In D2950#149594, @Freagarach wrote:

In D2950#147482, @Nescio wrote:

Their structures already have lower wood costs

Where do I find that? ;P

You're right, I was mistaken, I apologize. Now I wonder why I then thought it was true. I guess I misread or misremembered the celt_structures.json technology; the houses of Britons and Gauls are cheaper, but that's because they're smaller; Iberians and Mauryas have that too.
Nevertheless, while what I wrote was wrong (or at least inaccurate), I still stand by the main point: having a resource discount, a time reduction, or extra population each are very useful bonuses to have, especially in the early game. Having more than one such bonus makes a civ simply too strong. Therefore I'm glad this patch has been committed.

Yes I guess my point is that that's perhaps not a good idea.

It makes templates a bit more difficult to read but easier to maintain. We also use relative movement speeds.
Perhaps it's not perfect for resistances, but let's postpone that discussion. For here and now using op="add">-2 is consistent and easiest.

I'm not sure it's better to subtract 2 or just replace with a fixed value.

Other champion and hero templates use relative resistance values too.

Or to put it differently:

• bolt shooters, bolt towers, stone throwers, quinqueremes, and artillery towers now clearly outrange archers
• bolt shooters and bolt towers have lower spread, while the spread of stone throwers, quinqueremes, and artillery towers was increased
• bolt shooters and bolt towers inflict only piece damage
• stone throwers, quinqueremes, and artillery towers inflict only crush damage
• fire raisers and fireships inflict only fire damage
• bolt towers, stone throwers, quinqueremes, and artillery towers no longer have splash attacks (but bolt shooters keep their linear splash)
• bolt shooters, bolt towers, stone throwers, quinqueremes, and artillery towers have their base damage raised significantly
• these entities keep their high reload times (from rP23285) and their new damage-per-second values are still well below that of elephants or rams

As a consequence, bolt shooters and bolt towers are more effective vs units but worse vs structures, whereas stone throwers, quinqueremes, and artillery towers are more powerful vs structures but less effective vs units, thus further differentiating the two, giving each a more specialized function.

The champion cavalry has more armor than the champion infantry (don't ask me why)

That's odd; for comparison, citizen cavalry has less resistance than citizen infantry. However, I think that should be addressed in a different patch, focussing this one exclusively on the difference between melee and ranged troops.

Increasing attack reload time without adjusting attack damage meant entities (here artillery) are significantly weakened. In this case it was addressed by D2494/rP24608.

elephants don't collect resources, so I don't see a good reason to add food

No food is added, mercenary food cost is reduced, to 40%.
There are currently no elephant mercenaries anyway, so the point is mood. However, they're included in this patch because it's not unlikely they'll be introduced in a future patch (or mod) and having two more lines here is cheap.

I think the Range bonus is irrelevant, and agree with removing it.

Already done in D3297/rP24539.

Ranged units are supposed to deal damage, I think increasing the resistance makes limited sense. An expert bowman is no more protected against swords than a beginner (the same isn't _quite_ true of melee units).
[...]
Reducing ranged champion resistance seems fair to me (more consistent with CS).

See D3369.

Remember that spread is "normal deviation at 100 meters".

Yes, I know, @fatherbushido once explained to me on the forums how spread works.

I don't really have comments on their actual firepower, but spread needs to be reworked IMO. [...]

I fully agree spread values need to be changed, but I'm unsure what would be appropiate values.
Moreover, if a basic javelineer has double the spread of an archer, then a champion javelineer ought to have double the spread of a champion archer, in my opinion.

So something like (D) but with spread reworked.

Do you think it's better to do spread in a separate patch (and update this one afterwards) or do you prefer to include it here, making this one more complicated?

• mercenaries cost 60% less food than their citizen counterparts
• their wood, stone, and metal costs are replaced with just metal: infantry 60, cavalry 80, elephants 120

As I wrote earlier, getting a combination of two or more resources is easier, so the total cost ought to be somewhat higher than if they would only cost metal.

Britons and Gauls no longer get additional population from ordinary structures (D2950/rP24623) so the current bonus is less problematic and can stay as it is. The Gauls received a number of new things in A24 so they are not as similar as the Britons as they used to. The Britons deserve a little something as well, but that's something for a future patch.

I hope the resource Gatherer rate to 0 thing works?

It did last time I checked (which was at the end of last year).

I'm not sure if collecting treasures is still supposed to be possible? Depending, see inline.

Yes, they're supposed to be still able to collect treasures, hence the large number of lines.
I guess D3303 might make it easier, but I don't know for sure.

I guess my point is mostly that reducing the cost but increasing build time is essentially playing on the 'same thing', i.e. worker time, but in both directions at once.

That's another way of phrasing what I meant.

This makes it harder to understand what the bonus does. I would favour a straight small wood reduction over a complicated bonus like we have now.

Feel free to replace the time increase with a health reduction.

Have to say this is still a pretty weird bonus. Sundiata has a point that they probably should have both faster build time, lower cost and lower HP.

The sun-dried mud bricks need time for the mud to dry in the sun, though.
More importantly, the reason Britons, Gauls, and Ptolemies are by far the most popular civs in multiplayer is fundamentally simple: they have an economic bonus at game start. Being able to develop more quickly early on is a clear advantage. Having both a cost reduction and a time discount (or population bonus) makes things a lot worse, the modifications compound and create a snowball effect. Therefore it's necessary to give civs only one such bonus (hence D2950), and ensure it has some kind of penalty.

To be honest I'm not entirely sure "dried mud" is actually specific enough that it should be a civ bonus at all. Further, Kushites seem like they should benefit too?

In the initial version of D3329 I proposed extending this civ bonus to cart, kush, and pers as well (for comparison, brit and gaul share a structure bonus, as do athen, mace, spart), but @borg- recommended against.

The problem with vision technologies is that entities with already high vision range benefit much more than those with smaller vision ranges (the area is pi times radius square).

I might actually vote to disable this tech ATM otherwise.

That would be my preferred solution too.

• changed cost as discussed with @borg- (sorry for the haggling)

180 food and 120 metal?

Previously: 1.1^3=1.331, i.e. a 33% increase.
Patch: 1.15^3=1.521, i.e. a 52% increase.
I'm fine with increasing the trade gain technologies from +10% to +15%, but please remove the +25% movement speed bonus to compensate, and keep the trade_commercial_treaty.json unchanged.

Fine by me.

Was waiting for this to be finished before committing anything :)

Could you commit the icon first, though, like you did with the carnyx-trumpeter?

Here is my reasoning:

• Reducing enemy attack damage by 10% is equivalent to raising your resistances by one level.
• Champion infantry costs 80 food, 60 food, 80 metal and the swordsman has 47% more damage per second and two more resistance levels (equivalent to +23% health) than the proposed carnyx-trumpeter.
• Infantry heroes (e.g. Brennus) cost 200 food, 150 wood, 200 metal, have a lot more health, resistance, and attack damage, and much more powerful auras.

I called 250 food excessive because it's over three times the champion infantry's and even more than the hero's. I think we can agree the cost of this proposed unit ought to be somewhere in between ordinary champion infantry and infantry heroes.
So how about one of the following?

• 100 food, 200 metal
• 150 food, 150 metal
• 100 food, 100 wood, 100 metal

Or just make them cost 50% more than ordinary champion infantry, i.e.:

• 120 food, 90 wood, 120 metal.

What do you think?

Now I'm confused. The current situation for ranged soldiers is:

        damage , spread ; resistance
basic   : 1    , 3      ; 1
advanced: 1    , 2.7    ; 2
elite   : 1    , 2.43   ; 3
champion: 2    , 1      ; 5
hero    : 4    , 0.5    ; 8

Now let's list a few competing suggestions:
(A) Initial patch, give advanced and elite +20% attack damage each, without spread improvements; leave champions and heroes unchanged:

        damage , spread ; resistance
basic   : 1    , 3      ; 1
advanced: 1.2  , 3      ; 2
elite   : 1.44 , 3      ; 3
champion: 2    , 1      ; 5
hero    : 4    , 0.5    ; 8

(B) Give advanced and elite +20% attack damage each, without spread improvements, but remove their resistance increases:

        damage , spread ; resistance
basic   : 1    , 3      ; 1
advanced: 1.2  , 3      ; 1
elite   : 1.44 , 3      ; 1
champion: 2    , 1      ; 3
hero    : 4    , 0.5    ; 6

(C) Give advanced and elite +10% attack damage and −20% spread each, weaken champions and heroes accordingly:

        damage , spread ; resistance
basic   : 1    , 3      ; 1
advanced: 1.1  , 2.4    ; 2
elite   : 1.21 , 1.92   ; 3
champion: 1.5  , 1      ; 5
hero    : 2    , 0.6    ; 8

(D) Give advanced and elite +15% attack damage and −25% spread each, remove resistance increases, and weaken champions and heroes accordingly:

        damage , spread ; resistance
basic   : 1    , 3      ; 1
advanced: 1.15 , 2.25   ; 1
elite   : 1.32 , 1.6875 ; 1
champion: 1.75 , 1      ; 3
hero    : 2.66 , 0.4    ; 6

(E) Something else.
I'd highly appreciate feedback from @ValihrAnt, @wraitii, and others as well.

I'm not sure I like the removal of the reveal shore, to me it seems a fun addition. <- Maybe another team member disagrees with me here.
I dislike the removal of the colony from the metropolis tech also, because that just seems to be a good part of what its supposed to do.

Those two things could be split off.

@Freagarach, if you have time, could you review this? It's small and doesn't affect gameplay balance (and includes map files).

Removing the wood cost is fine by me, but 250 food seems rather excessive. How about 100 food and 150 metal instead? Or perhaps 100 food and 200 metal? (How much is the aura worth to you?)

@borg-, how much should this unit cost?

It's pending on whether it's acceptable to lower ranged champion and hero attack damage; see earlier post.

D3287/rP24604 has been committed today, from which heroes still benefit, confirming for me the need to delete this hero-only technology.

I could also suggest to have I II III on tech icons if they feel identical?

Please do that for the gather technologies!

I can agree to keep two techs for commercial gain, but I see no reason to keep tech for international gain, it seems to me repeated.

It's fundamentally different. If you're only trading with yourself, it won't help you at all (which is my it is so cheap), but if you have allies, both you and them benefit. It may be a bit more difficult to understand, but it encourages cooperation, therefore I'd really like it to stay.

Purge 0 lines?

D3342 (I like consistency.)

• let bolt-shooters keep a linear splash, because the removal was deemed too controversial
• give fireship and fireraiser only fire damage, instead of a combination of hack, pierce, and crush, per @Freagarach
• raise fireship minimum range because of its footprint size

Back to this patch. Civilization bonuses hidden in templates:

• cart market and dock
• cart merchant ship
• pers trader
• iber tower
• kush, pers, rome ram
• rome bolt-shooter and stone-thower
• spart female citizen

and without D2950 a number of population bonuses. Anything else?
The first three are particularly important, since trade gain and international bonus increases won't show up in the structure tree or cursor pop-ups. As for the rest, I think it's an improvement highlighting them by inclusion in the civ pages, but I'm unsure it's strictly necessary. Opinions?

To summarize, @badosu, @borg-, @wraitii, and I are in favour of this patch, but I believe @Angen, @Freagarach, @Stan are against.
While I strongly believe removing linear splash from bolt-shooters is a real improvement, if that's the only thing blocking this, how about leaving it out then? I'd rather see most of it committed than nothing at all; artillery is simply not very effective right now.
Also, D2493?

What's wrong with having two trade gain technologies? The storehouse technologies are fine, aren't they?
If some market has to go, I'd say deprecate trade_convoys_speed. I don't really care about trade_convoys_armor either.
And if you want to simply have fewer similar technologies, have a look at the healer technologies at the temple.

Let's rephrase the descriptions then.
The commit message is quite short, the patch summary gives a more detailed reasoning.

I dislike the removal of the colony from the metropolis tech also, because that just seems to be a good part of what its supposed to do.

Is it? To me it seemed an oversight. The technology was introduced in rP12753, the military colony in rP14411.