It looks good to me. Only small grey area for me is that citizen cavalry is slightly faster as a result (x1.95 to x2.0) but at this point it's just nitpicking, I don't think it will have noticeable effect to balance.

I personally have nothing against this, I think a slight imbalance between traders will not change much there. Can ask @ValihrAnt just in case.

I have nothing against this patch, in the current state of the game it would do no harm, but i would like to note that if we are in a situation where champions are balanced and viable then suddenly persians would become (by far i think) the civ who would be able to spam them the fastest entering city phase and it could become a problem in that case.

I also think it might affect balance. Here it is not the balance between civs but the balance between units which is also important (to compare, right now champions aren't really viable compared to citizen soldier which is a problem). If cav become harder to hit it would buff them. Now i don't know if this patch actually noticeably changes balance or not, but to me it would be the kind of patch that could use some testing to be on the safe side of things.
If it does affect balance but we still want that patch then maybe cavalry would have to be nerfed in some other stats.

Yeah, that's it. It's been a long time, I kinda forgot about it

To be honest I think a lot of work is needed to balance the heroes while keeping their bonus diversified and somewhat accurate historically. I think there was once a patch by @Hannibal_Barca with many of the changes I proposed or agreed on.

I agree with the changes now.

I was thinking of the generic +20% attack heroes (Vercingetorix iirc has radius 60, affects soldiers and maybe siege), Hannibal also has radius 60, affects own and allied units. Compared to them, Scipio, Seleucus and Boudica feel like in a lower tier.

Oh ok i thought it got removed too. Then I think it's ok.

For me a 25% bonus feels kinda low but i could be wrong. As a comparison i wonder by how much celts civilizations build faster than others ?

I'm not sure about that change. I feel that the +20% damage heroes with local are better anyway so making the +1 armor local makes it feel pretty weak for me.
On a side note, antiochus is fine (compared to his counterpart, that one persian hero that gives attack to cavalry) because it allows to micro away the cavalry to save them before they die and heal them which helps them get promoted

I agree with that change

I agree with the changes, suggested some of them

If it is historically accurate I have nothing against it. It could help kushite a little bit.

Didn't mean to accept btw

I agree that the build time needs to be reduced, however not sure if it should be to 100. Carthaginian embassies have 150 build time, too

Tried to fix compatibility problem with tech but i didn't test it, also for some reasons the decimals .0 will not show up when the multiplier is an integer and this is not the unit card for some reasons

There was a space in french translation so I missed it
In more information tooltip :

Here's how it looks now :

AI talk is about the eventuality that more hard counters are added
But the proposal itself doesn't affect AI yes

Could also consider to delay reaction of the AI by a multiple of AI difficulty, with 0 for the very hard one

By the first 2 screenshots, i meant overall in this differential. I posted 5 in total.
If unit classes are not in yellow wouldn't it look weird if line is inbetween attack and armor tooltip, with nothing other than digits in white ? The first 2 screenshot show the result if i put it on the top, where i would say white font looks less weird but putting counters here is also not logically pertinent. In the third screenshot the line is after attack tooltip and in white(but with "against" that is unnecessary and removed now)
I didn't find the missing semi-colon.
For effectiveness of petra, true it would reduce it but it can be possible to make him train counters to his opponent units (AoM AI does it IIRC). Of course, that AI train units well doesn't mean it would use those units well. (also i think AI has a big room for improvement)
And now i just noticed i removed sprintf where i shouldn't

What do you prefer ? As done in first 2 screenshots, or in these :

Here is how it looks after moving the line down attack :

Also for the if(bonus.Classes) code, I put it there because before i saw it in DamageBonus.js code in simulation, but didn't understand why it was there, if there are Bonuses in the template there should always be at least a bonus, which should always contain Classes and Multiplier, so maybe it can just be removed, didn't try it.

Now it works like it should. Here are the screenshots (where i didn't bother putting my game in english). First one in vanilla, second one in Delenda Est where we can have multiple counters.

Unrelated with last comment but for the noisiness, perhaps counter could be added after the details of one AttackType, without adding new lines except if it goes out of bounds (which would most of the time).
Something like Melee attack: 3.0 Hack, 2.5 Pierce, Rate: 1.0 Seconds, Counters: 3.0 Cavalry
But the problem is that i don't see how to harmonize it aesthetically, it would certainly look weird if i want to put unit classes in small font yellow letters

Well, indeed it got triggered, now it's template.attack (attack in lowercase).
But after correcting this, it still doesn't work, after I try warn(uneval(template.attack)) I don't find any bonuses even if some units should have them

Well, I added my function name in gui\reference\draw.js, gui\session\selection_details.js and gui\session\selection_panels.js which is where my file search lead me, but it doesn't seem to add the tooltip still. I will search some more.
Currently my code adds a new line for each attack type, but also for each counter contained in that attack type, it would maybe help to regroup counters in a line for each counters that share the same multiplier, but it would still remain quite noisy if someone wanted to regroup all possibilities in a single unit.

Not sure I understood everything but there's what i modified.
I still (temporarily ?) modify the identity tooltip property, while i search for something better.

Indeed i think that ptol barrack cost rebalance was needed