This is a very good comparison, because Georgian and Briton Keeps are functionally identical in stats and upgrades, save for the Georgian ones having piercing arrows dealing full damage to every unit hit. They both have +2 attack, and both have Arrowslits, Masonry, and Architecture. This makes the comparison very natural, as the only advantage the Georgians have over the Britons is piercing arrows. And boy, what an advantage that is!
I tested both against the same clumped-up group of Mayan archers, who were allowed to fire back. These are the results for both:
They were only marginally slower than the Georgians, at about 1:40, but took enough damage for the Keep to be lit on fire, as they lack Architecture. I still think Georgian towers are hands down the best, as they have high attack easily dealt to multiple targets, are quite tanky, and even take less damage on hills, where they will probably be built a lot. So it’s hard to say for sure, but I still think Svan Towers should be nerfed, or at least changed. Maybe the +2 attack should be removed, but in exchange, towers become cheaper, like my version of the technology. That way, individual towers are weaker in terms of attack, but they become much easier to mass. 50% is probably too much, but 33% is probably fine.
+2 attack for castle is small imo. And as I was replying this, I just remembered it now overlaps with new Persians UT too.
UTs shouldn’t overlap. If it is a stronger version of a free civ bonus like Aztecs vs Burmese, only then it should be okay.
Fortified church as a trash defensive building is too strong in the late game. And it is more important for Georgians than Armenians. So maybe just make the UT to Tower only as the name says. Tower +3 attack for both primary and extra arrows. No piercing ability. Keep that for a later civ. Reduce the cost of UT.
As for Britons, repetitive bonus for same civ really gets under my skin. Saracens is gone, Burmese and Britons to go. Britons (Elite) Longbowman +1 range, Yeoman UT effect change to Longbowman can be trained from Range.
I realized that where Svan Towers differs from Yasama is that it probably helps garrisoned units.
Did you test that? Are secondary arrows affected?
If they are, try these tests.
Note: in your tests, Svan Towers would have killed more than Yasama at start, and then slowed down, so that’s probably the main reason Yasama towers took more damage, in addition to more damage as a percent of HP. Damage dealt by the targets is more useful in a real-world scenario than the time it took to kill targets, but it would be hard to measure in these tests without affecting results (unless you use a trigger to change HP, damage or armor).
So, compare these scenarios, with target being trapped melee units:
Georgian Keep with 4 Georgian Crossbowmen standing next to it.
Georgian Keep with 4 Georgian Crossbowmen garrisoned in it. (Should add 2 arrows; test damage of secondary arrows by garrisoning 5 crossbowmen, which should add 3 arrows if secondary does 15 damage and 2 arrows if secondary does 17 damage.)
Briton Keep with 4 Briton Arbalests standing next to it.
Briton Keep with 4 Briton Arbalests garrisoned in it.
Predicted order, from fastest to slowest: 2, 1, 3, 4, though the last two may depend on the targets. (With halberdiers as target, ungarrisoned arbalests kill faster because of +3 bonus and probably less overkill; with paladins as target, garrisoned arbalests kill faster.)
Bonus round: 125+50+70*4 = 454 resources of heavy scorpions, i.e. 3 heavy scorpions.
With a single tower, the result is predictable (or nearly so, with a slight uncertainty on exactly how much Stronghold buffs fire rate: like, is it from 2.05 to 1.55, or from exactly 2 to exactly 1.5?). If the targets were post-imperial Arbalests, then Briton towers did 13 damage, killed in 4 hits; Celtic towers do 10 damage, also kill in 4 hits, but fire faster. Adjust for missing Architecture (towers still take 1 damage per hit, but have 10% less HP if you want to calculate percent HP lost).
I think the Svan towers tech is probably okay. If anything they can make the collision box of the projectile slightly smaller. But it’s hard to say how much use you’re getting out of the passthrough ability, as you’re normally not going to be fighting arbs squished together in a box. A better test might be to see how many moving melee units they can kill (either attacking the tower or villagers in nearby eco). It’s hard to establish a baseline for lategame balance, but where the generic FU Keep is basically useless, I’m down to have another civ that actually has a reason to build towers past Feudal Age. Although I kind of wish they would buff generic towers, and then rework UTs/Civ Bonuses as necessary.
Zealotry wasn’t that exciting, but it wasn’t repetitive until devs added the +10 HP as a civ bonus, which was boring and not that impactful anyway. Aside from that the new UT seems kind of random. Don’t hate it, but it kind of looks like they’re trying to speedrun adding all possible aura effects. Which has implications on power/mechanics creep, and suggests to me (along with mule carts and warrior priests) that they’re not intentionally saving bonuses for future civs on which they might be more fitting.
How about make the Yeoman give +2 range instead of +2 attack, and let the Eupseong have a new effect?
The longer range meets the identity of the British.
To the Svan Towers, I’d like to make their projectile visually become scorpion bolt.
Nah, I don’t mind the current arrow. A Scorpion bolt might be too clunky coming out of a tower.
How about the tech upgrades towers to Tower Houses, towers that provide 5 pop space and have extra LoS? The devs can finish the Svan Tower model from the Editor and add a destruction animation. It would be a nice touch, though I’d have to think of a different Guard Tower look for my own Caucasus set.