Just had a huge war against Huns-Aztecs in Medieval. Was pretty scary, which is good. The Hunnic Hordes' ranged attacks can shoot uphill or across cliff, a very interesting design and I quite like it; you can't just standing on a cliff for defense and thought it is a safe place.

My only issue being the attack twice ability, it might be far too powerful - a swarm of Classical hordes can easily devastate an army of Medieval Swordsmen and Knights, as the double attack can deal firearm level damage.

Edit: In addition, for some reason, a Hunnic Horde can ranged attack my Quadriremes if next to them, but my Quadriremes cannot fight back. Seems to be an overlook at the naval-land battle's coding.

Ok, so here we are with plenty of new Victor feedback. I'd like to first say that HK is surely shaping up to be a fantastic game, and this OpenDev already shown a great deal of improvement over the Lucy build. I'll focus this report on the issues I found, but this by no way means HK isn't a good or fun game, just that I'm pointing to what can be made better before release. I'll also try to not discuss bugs like the various graphical glitches or the fact that many Early Modern Emblematic Quarters could be built in multiples per territory since by now I assume you're well aware of those. I'll try (key word here is try) to present suggestions to each issue, but of course, my knowledge of civics, technologies and cultures is limited by the scope of the opendevs revealed so far. So, without further ado, lets dive in:

- Land Armies

- Issue: While the unit upkeep was a good move, it is nowhere high enough to make a dent on your gold income. It is way to easy to field armies big enough to attack the whole world, especially in later eras, even if you don't hard focus on gold generation. Also, the army upkeep reduction mechanic, while ingenious, is counter intuitive and a bit confusing. Solution:  Increase the upkeep on units, especially on medieval onward, based on their industry costs (higher industry costs = higher upkeeps). Also, remove the army upkeep reduction mechanic and instead add a per-army upkeep cost. This way, the more armies you have the costlier they become (so you create incentives for grouping your armies, just like the upkeep reduction mechanic does), and you can scale it better in the long game - say (not final numbers are not final) each army costs 0 gold in Neolithic, 2 gold in Ancient, 4 gold in Classical, 8 gold in Medieval, 16 in Early Modern, so on and so forth. You add that value to the total unit upkeep costs of the units that comprise that army and you have the total upkeep cost for that army. That would be easier to grasp and probably easier to explain on the upkeep tooltip than the current upkeep reduction. 

- Issue: After the Victor combat damage adjustment (which is much better than the state of Lucy's), Free Riders became extremely strong, and the Huns and Mongols, while being rightfully nerfed from its Lucy state, still dominate the battlefield. A good point of comparison (using both RNGZero's wonderful combat spreadsheet and my own combat simulator) is the Russian Cossack. While it is strong, it does not dominate the Industrial meta because it is weaker, CS-wise, than all other EU in the era, and the fact that the last eras is dominated by Gunners means it does not perform counterattacks as often as a horde do. Solution: Hunnic and Mongol hordes should be made weaker, CS wise. By using the combat simulator, I think the sweet spot is to have the following changes: a) Reduce the Hunnic Horde from 22 to 19 CS (so it is 22 CS after the Hunnic LT, not 25), and the Mongolian Horde CS from 29 to 28 CS (The medieval meta has much stronger units than the classical meta, so there's no need to reduce Mongols as much) and b) give the Hordes Melee weakness like other Ranged options (so it suffers -5 CS when being attacked by a melee unit). These changes should make the hordes much more bearable in their respective eras and solidify them in the "glass cannon" role. They'll have to either overwhelm the opponent with numbers or employ hit-and-run tactics, or the enemy counterattack will decimate them.

- Issue: Ransacking was too quick and strong in Lucy, and in Victor you increased the time to ransack to around 5 to 7 turns. This is too long and makes ransacking usually a bad option. As several civics and cultures rely strongly on ransacking, you feel punished by trying a ransacking route because it is so slow in Victor. Solution:  Adjust the ransacking timers to make it somewhat intermediary between Lucy and Victor should be the sweet spot, especially if you can keep lone ransacking units and ransacking admin centers on the high side (seven turns feels good) but can reduce the ransacking time for full armies on unprotected districts and extractors (especially with high CS units) to 1-3 turns.

- Issue: There's a lack of control in the reinforcement phase. You can't select who will reinforce (outside of keeping your armies distant from one another), and as the reinforcing drains the movement of all armies, this can become very painful, especially on the later parts of the game. Having more control on who will fight and who will not allows for more gameplay expression and more impactful tactical decisions. Solution: Add an option to select, during deployment phase, which armies will reinforce and which won't. Endless Legend has a similar system (with checkmarks to enable/disable reinforcing), so a system similar to EL here would be great.

This is a post in a series of connected posts about the Victor Opendev. You can find the posts discussing other topics below:

Economy: https://www.games2gether.com/amplitude-studios/humankind/forums/210-victor-opendev/threads/39499-feedback-economy-and-game-pace?page=3#post-315472

Naval & Air: https://www.games2gether.com/amplitude-studios/humankind/forums/210-victor-opendev/threads/39509-feedback-naval-gameplay?page=1#post-315474

Religion: https://www.games2gether.com/amplitude-studios/humankind/forums/210-victor-opendev/threads/39525-feedback-religion?page=2#post-315475

Diplomacy: https://www.games2gether.com/amplitude-studios/humankind/forums/210-victor-opendev/threads/39502-feedback-diplomacy?page=2#post-315476

Civics: https://www.games2gether.com/amplitude-studios/humankind/forums/210-victor-opendev/threads/39508-feedback-civics?page=1#post-315478

Cultures: https://www.games2gether.com/amplitude-studios/humankind/forums/210-victor-opendev/threads/39500-feedback-cultures?page=2#post-315479

Independent Peoples: https://www.games2gether.com/amplitude-studios/humankind/forums/210-victor-opendev/threads/39526-feedback-independent-people?page=1#post-315481

I completed several playthroughs on up to 'Empire' difficulty. Overall, I think the combat system offers a refreshing take on the genre, and has a lot of great things going for it (which I won't list here, but in my view the game designers should feel really good about their work). There are also important issues, which I hope can get looked at during the home stretch. I think these can all be addressed with reasonably minor tweaks to gameplay, and without redesigning important systems so close to release.

1) Chasing armies around is not fun, unrealistic, and interacts poorly with other game mechanics.

Problem: Right now, the AI will frequently have tiny armies running around. When confronted, they will retreat, wasting the turn of the chasing army and those within range of it. This is often frustrating, and not terribly realistic (why is everyone in the region chasing after the scout? I thought we were marching on Paris?). In addition, the frequent minor engagements and retreats drain the ai's war score far too quickly, sometimes leading to its total surrender without any real battle.

Suggestion: Expand the radius of zones of control (ZOC) around non-retreating armies, forts and cities by at least 1 hex (perhaps more from civics/techs/traits), and cause entering hostile units to initiate battle. The game naturally creates chokepoints across the map, so larger ZOCs would decisively limit the diffusion of AI units towards the player's territory, and encourage the AI to concentrate its forces to defend or challenge key battle sites. This would also allow defenders to more effectively capitalize on their home terrain, make city placement more meaningful, and add greater realism to military campaigns. This change could also create a better niche for the 'spy' units by making them immune to ZOC, and grant navies much-needed utility as blockaders and troop transports (circumventing strongpoints)

2) Capturing territory is too easy. Battle victories are too decisive, and defeats too costly. 

Problem: Defeating the main enemy force often irrecoverably leads to total victory, as few further obstacles stand in the way of the victor, particularly in the early-game. Fortifications pose a small extra hurdle but not once, across several playthroughs, have I needed to besiege a city for any duration of time. This is not only ahistorical (as Hannibal Barca would like to remind you), but also drives much of the snowballing in Humankind. Since a single military action can easily spell disaster and knock a player out of contention for the rest of the game, the field of serious contenders quickly narrows as warfare develops during the very volatile early game.

Suggestion : Until gunpowder becomes widespread, fortifications should require overwhelming force to overcome (and I mean overwhelming), or force a prolonged, costly siege. Together with zones of control, this should limit early-game volatility by creating a substantial defender's advantage in their core regions. This would incidentally make plundering more important, as a way to bring a well-entrenched defender to the negotiating table. This would need to be balanced somewhat delicately to avoid the entire map from turning into the Cadia fortress-world. To ensure that players carefully decide where and how to fortify their assets, the game could for instance make fortifications expensive to build and maintain, or limit their availability based on infrastructure, resources, population, tech, etc. As a principle, outskirts and contested territories should be reasonably vulnerable to plunder and even invasion, but a nation's core, well-developed regions should not be targets of opportunity.

3) The war score system is immersion breaking, and snowbally

Problem: Skirmishes and inaction drain warscore, and can lead to total surrender. Wars usually end in total defeat or total victory, with little room in between, and vassalization or territory gains are the only negotiable terms of a peace deal. Generally, war outcomes can often seem absurd and drastic, given the participants' objectives and their relative positions.

Suggestion: The effect of battles on warscore could account for their scale, their decisiveness, and the remaining strength of the players. There should be large differences between the loss of a single brigade (Charge of the Light Brigade), the annihilation of a nation's whole military (Azincourt), and the defeat of one among uncountably many armies (Russia during operation Barbarossa). Rather than the original aggressor, the last player to refuse a peace deal should lose warscore each turn; AI aggressors should seek a white peace as soon as their initial assault is repelled. 

Additional, era-specific items could be added to peace talks to make the outcome of wars more interesting, and move away from the current all-or-nothing state of warfare in Victor. For instance: forced conversion, supply mercenaries (provide troop free of maintenance), force treaty (e.g. open borders), prevent military buildup (create grievance if they exceed X total unit strength), and so on. 

4) Conquests bear insufficient costs, and facilitate snowballing

Problem: Players are able to add conquered cities to their empire at no cost, and snowball out of control. More generally, Players aggressively expanding need to be more seriously penalized in other areas of gameplay (stability, culture, trade, religion, or some other system)

Suggestion : Strongly throttle the stability and productivity of newly-conquered areas and make them a net drain on the player for a time. Scale the duration and severity of penalties based  on differences in culture, religion and ethos between the player and their newly-conquered areas.

5) Unit upkeep is negligible.

Problem: At no point during any of my playthroughs did I ever need to seriously contemplate unit upkeep. This not only breaks immersion, but takes away one of the more interesting choices in 4X games: balancing out military vs. civilian investments. 

Suggestion: Very simply, count living military units towards the cost of population growth in their city of origin. In other words, cities spawning a unit would still lose one pop; however, so long as it exists, the unit would still be counted as a citizen of that city (which, it is!) for the purpose of determining the food cost of growth, just as if the city had not shrunk by one. To prevent exploits, disbanding units within friendly territory would return them to their home city, rather than the nearest one. A couple edge cases would need to be resolved, like what happens when units lose their home city, and what to do with nomads, but they shouldn't be hard to address. Overall I think this small change would feel great from a historical/realism standpoint, while creating very interesting dilemmas for the player.

Additional suggestion: Without being overly onerous to implement, I do think that adding growth penalties to military units would carve an interesting design space to uplift underused mechanics, like mercenaries, and allow new concepts to emerge. One idea I would really like: traits, techs or civics could grant players the ability to quickly spawn weak but numerous levy units, allowing for rapid temporary mobilization. These conscripts could save the nation in a pinch, but their immense burden on city growth would naturally make them unsuited for long campaigns (mileage could vary based on civics and civ traits. A 'Revolutionary France' type of culture just might make an invincible Grande Armee out of these angry peasants)  

0) To end on a more abstract note

Like in the Endless games, player progress in Humankind is more exponential than it is linear (the graphs presented after each game show this beautifully). As a result, once the early-game ends, flat costs get quickly overtaken by the economy and no longer provide interesting choices. The number of forum posts about Gold being too strong is I think a good illustration of that. It also makes balancing difficult, since linear cost adjustments in this exponential context can crush empires lagging behind, yet not be felt by those farthest ahead. For this reason, I am intrigued by the idea that some important tradeoffs could involve growth penalties, flattening the exponential growth of a resource to secure another benefit (like military units, to use my example).

I have two primary points of feedback; one is regarding the combat damage calculations and the other is regarding the interactions between defensive structures, rivers, and roads. Stay tuned for screens!

Firstly with the combat calculations. While I think it's much better than it was in the Lucy OpenDev, it's still in need of a significant rebalance. The main issue is that the combat damage caps out far too quickly, especially when attacking with ranged units. Take a look at this image:

That's a difference of 12 combat strength between the javelineer and the super buffed up immortal, yet the immortal can still be damaged up to a quarter of his health. There's nothing innately wrong with this; in fact, I'm glad that some civilisations' only ranged option (being the archer) can still do decent damage in the classical age, so that's great (it could maybe be debuffed just a bit, but it's otherwise fine I think). However, the real problems show when you compare that example to these:

Now we have the attacking javelineers in a MUCH better position for taking on these immortals. And yet look at the difference in damage - it's almost non-existed. A difference of two strength leads to 9-25 damage, and a difference of three strength leads to 7-25 damage. Anything below that, to my knowledge, whether it's 4 strength below or 50 strength below, will always lead to the defender taking anywhere between 5-25 damage.

So the problem is clear now; in a game where combat is supposed to be rewarded with good positioning and tactics, combatants are often NOT rewarded with good positioning and tactics. I actually kind of like the combat pace how it is right now so I'm not suggesting a total mechanical rebalance, but the gradient of damage dealt/taken in relation to power needs to be tweaked. Perhaps the range of damage at this extreme is too high - a difference of 20 is extremely random (25 is 400% higher than 5 - very random indeed). On the other hand, I don't think archers should be made obsolete as soon as anyone enters the classical era, and the system works much better for melee combatants, so I don't anything too drastic needs to happen. It's food for thought; I just wanted to point out there is a problem when a difference of two power, especially for ranged units, functions almost identically to a difference of 30 power.

On to the next issue: defensive structures and rivers/roads. I had this game where I picked a great place to defend from any invasion from the south and west; it was right next to a river. The problem is that as soon as I researched the wheel, roads started snaking all over the place and completely negated almost 100% of the defensiveness I gained from the river. Example:

I've outlined in yellow all the hexes that should have been affected by defensive rivers but were negated by roads, and the few that retained their defensive bonuses in blue. The path in red is the main perpetrator - it doesn't even connect to my own settlements, but to my vassal's Hunnic Ordu. The Greeks were able to march right up to my city and suffer from little to no river penalties in the ensuing sortie. It's such a shame, because we have this amazing system where you can choose to settle cities in places that will provide great defensive gameplay, but it's just being negated too easily at the moment by roads and road placement. It gets even worse when you consider clarity on the map:

Notice how you can barely see the roads! You'd assume this place is extremely fortified and will be difficult to overcome by any attacker due to the large number of rivers, but in reality that's just not the case. I think we have a huge problem of clarity and gameplay systems that interfere with other great systems here. I offer two solutions to this: Either make roads have no effect in strategic battles, or allow players to at least partially define the placement of roads in their territories, perhaps through drawing them directly or being able to designate a few hexes through which no roads should be allowed to pass. Unlike with making the walls in sieges more noticeable on the map, I don't think making roads more noticeable on the map is the solution here; there is a fundamental problem with how roads are formed too randomly that makes the strategic placement of cities - one of the most fun aspects of the game, in my opinion - much less impactful. Rivers contribute hugely to the gameplay, both on the main map and strategic combat map, and I don't think that roads should interfere with them to this degree.

I agree too with the above points and I'd keep the river crossing CS penalty on roads however I'd be okay if the movement speed bonus stayed the same, bridges are hard to fight on but easy to cross. I also appreciate Roryn bringing to light how little of a difference combat strength now makes compared to Lucy. Before it was too harsh of a penalty for having even 3 lower combat strength. I disagree with him thinking archers are fine though, archers are certainly too good, shelling fortifications and well defended enemies like artillery with no respect to positioning since dealing minimum damage is still so effective. The only thing you worry about is shielding them and cycling them out occasionally so they don't die to other archers.