Delnar_Ersike
Prince
New units won't necessarily arrive at "the end of the carpet", but it will certainly take a while to bring their forces to bear compared to if you only had one or two units: you would need to shuffle your armies around and/or go the long way around to attempt a surround just to get a new unit to a position from which it can deal damage. This is also one of the reasons why being able to move after attacking is so powerful: it lets you essentially attack with multiple units from a single tile, as if you weren't limited by 1UPT. If you just suicide your frontline units to bring in your reserves from the back, the opponent can just use the insta-heal promotion to nullify almost all of your potential gains. Unit promotions are valuable, yes, but because the XP requirement for new promotions becomes incredibly high after the first few levels, it's generally better to keep units at around level 4 (to pick up Blitz, March/Air Repair, +1 range, and Cover III for blockers) to maximize promotion benefits while still keeping XP requirements low enough that insta-heals are offered often; this is also why Alhambra is better than Brandenburg Gate, because Alhambra's promotion is free while the promotion you'd get from Brandenburg Gate's XP bonus increases the unit's level, making future promotions (and insta-heals) significantly harder to get.
I did not say the falloff was linear (it's probably more logistical, being slow when you have 6-7 units but really picking up pace around 16 or so non-air units all attempting to fight on a single, ca. 5-tile front), but it's still there.
On the other side of the fence, single units on their own are much more powerful in Civ5 than in Civ4 because they do not die instantly in combat and they block a lot more movement in hex-based grid than in a square-grid with diagonals (two units in Civ5 block the same amount of movements as three units in Civ4), with a third bonus for ranged units, since they do not take damage when attacking and for the most part do not need to be adjacent either. I cannot understate the importance of the first bonus (units do not die instantly from combat), because even if Civ5's unit production costs are roughly 2-3 times that of Civ4's unit production costs (while building costs are roughly the same), players can usually keep units alive for a lot longer than 3-4 combats, especially given how combat strength scales (in Civ4, the first melee upgrade to Warriors, Axemen, had twice their strength, while Civ5's Warrior upgrades only double their original strength at Pikemen). As a result, peaceful Civ5 empires can usually make do with producing a few units for defense and keeping them around for a long time (especially if they survive combat), while even peaceful empires in Civ4 had to cycle and build new units constantly to replace the ones they would lose to barbarians and the occasional war.
The primary problem with ranged attacks is that unlike all other types of combat in any of the Civ games, they give the attacker the advantage. In Civ4 combat, a stack's defender was always the one who had the most combat strength against the attacking unit, so the defender had the advantage in most cases (exceptions being attackers with collateral damage when it was stack-of-doom vs. stack-of-doom, where weakening all units in the other stack gave the attacker the advantage in attrition). In Civ5's melee combat, if two units face off against each other on even terrain, both units will suffer the same damage no matter who attacks first. However, in Civ5's ranged combat, if two archers face off against each other, the player who fires with their archer first wins, because the other player only gets the chance to fire back once their archer is damaged and has lower combat strength as a result (exception is Japan).
This is also why, mind you, players did not complain about most air combat in Civ5 or in previous Civs: even though bombers could technically attack from afar in all Civ games just like ranged units in Civ5, the fact that they could be intercepted and could still take damage from their ranged attack meant that sometimes the defender did have the advantage. It's also why you'll notice the same sort of complaints pop up the instant air units can no longer be intercepted, as is the case with Civ5's Stealth Bombers (Civ4's Stealth Bombers came much earlier but only had 50% evasion) and Atomic Bombs (which can only really be defended against with Nuke Bunkers, which unlock an entire era after Atomic Bombs). Let's not even begin to talk about the effects of not being able to intercept paradrops... *cough*XCOM*cough*
Heck, even Fall from Heaven had this problem with mages who could produce fireballs (which essentially acted as a ranged attack), which is why a combination of very low combat strength on mages, the ability to gain fire resistance, and the quick availability of assassin units were required to even hope to balance things out.
The only way to keep ranged attacks against units but not allow them to be too powerful is to make sure ranged attacks could suffer immediate counterattacks from targets: cities must fire back at ranged units immediately when being fired upon (if they are allowed to fire, which is actually another thing I dislike about Civ5's combat, but that's another topic), enemy ranged units could automatically fire back upon being fired upon, melee units being fired upon from range 1 could initiate an automatic counterattack after the shots land, etc.
I absolutely love ZoC (Zone of Control) and am so thankful they did not remove it from Civ5. ZoC's most important role is that it makes mobility useful even during melees where neither side wants to retreat: units with higher mobility can move a tile back (out of ZoC), make a few flanking maneuvers, and move back into ZoC to attack, while units without such mobility have to make do with attacking from their current position. Even with 1UPT, your mobile units will usually be the ones you'll be attempting to flank with or at least move around a lot with, so you're more able to get them into positions where they aren't limited by units occupying the tiles behind them. Reducing the effects of ZoC on melee units significantly punishes cavalry, who already have negatives from not getting terrain bonuses.
The funny thing about melee-front, ranged-back, and cavalry flanks is that it has little historical basis, even if you consider them aesthetically pleasing (I'm ambivalent): the closest thing I can think of is infantry-front, artillery-back, and light cavalry flanks during the 17th-19th century, but you must remember than "infantry" were still ranged combatants with bayonet-equipped muskets (and later rifles) and that the reason cavalry were relegated to the flanks is because firearms became so powerful that even the strongest cavalry armor became no match for bullets. Artillery were vulnerable to cavalry because they equipped to support combatants from afar via suppressing fire that took a while to reload, not to snipe horses or hold their own in a melee; it is the only ranged military unit I can think of that had that relation with cavalry. Before firearms came about, archers and infantry were usually mixed at the flanks, while heavy cavalry would be front and center. Unlike traditional game logic (cavalry good vs. archers), archers were often incredibly powerful against cavalry because riding on horseback made soldiers much easier-to-hit targets (if an archer shot your horse dead, chances are you'd be incapacitated for the rest of the battle, especially you ended up getting trampled or breaking a few limbs), and cavalry would often have to equip so much armor to make up for this that they would lose their increased mobility in the process (they'd still have incredible stopping power). This is why Mongolian horse archers often dismounted before facing off against Chinese foot archers. Even after firearms and pike squares, musketmen and pikemen would be mixed (look up "pike and shot", the famous successor to the pike-and-crossbows formation) to most effectively exploit firearms' stopping power while simultaneously defending from cavalry charges by a disciplined pike square (the Civ5 equivalent would be if musketeers had range 1 or 2 instead of being melee units and could occupy the same tile as pikemen). The only saving grace of cavalry was that firearms took forever to reload and were quite inaccurate, which is why they really started disappearing from primary battlefield roles around the times of rifles and breech-loaded firearms.
Even if you think about early 20th century battles, tanks and armored units were the ones who blazed forward, with infantry covering their advance from behind and the flanks.
What I meant was that 1UPT's problems arise from its hard limit, so if that limit were increased to 2 or 3, we would just be pushing its problems to become relevant at later stages (eg. the production falloff would ramp up at around 32 units instead of 16) with the added negative of making mini stacks-of-doom (since players are better off stacking two units in one tile if the tile defender is always the best defender), instead of eliminating the problems altogether. If a system were created where the limits were soft, ie. the player could place stack as many units on a single tile as they wanted but spreading armies out would be a better idea than forming a stack of doom, you would solve most of 1UPT's problems in a more simpler (more elegant?) fashion than if you tried fixing it with an array of dynamic production costs, combat strength modifiers, application of 1UPT to air units (air units will always be incredibly powerful if they can stack when nobody else can), and other arbitrary rules (eg. only one unit from a stack can attack per turn except in certain circumstances, the tile's defender is not necessarily the best defender but sometimes is, etc.).
Besides making collateral damage more prevalent, another "soft limit" that leans more towards Civ5's 1UPT than Civ4's stacking would be attrition: if there are two or more units of the same type occupying a tile at the end of a turn, all units on the tile take damage depending on the tile and the number of units stacked. It's still a soft limit because ultimately the player decides whether attrition damage is worth it (plus it would make the March promotion even more useful) and they can move units around without having to worry about traffic jams (they'd just need to make sure no units were stacked at the end of the turn or else those units would suffer attrition damage), but stacking units would be punished even if the opponent does not attack. The biggest problem would probably be designing an attrition system that is intuitive, powerful enough to incentivize players to spread out armies, but not too powerful as to make stacking never viable.
Adjacency bonuses can be a lot more frustrating to work against than stacks of 2-4 units (if you also have stacks of 2-4 units) because you would need to rely on breakthroughs in enemy fronts a lot more than in current Civ5; plus it would make units that can take out enemy units behind their frontline (ie. ranged units) even more powerful than they already are. While I honestly haven't given it much thought, I can easily imagine trench warfare situations arising even in the Classical Era should adjacency bonuses be made important. Though I can definitely see it as a possibly interesting way of mimicking the rise of trench warfare (prevalent and strong adjacency bonuses would mean that killing a unit and moving in your own unit into its tile would instantly surround your newly moved unit and would be easily taken the turn afterwards) and the importance of infiltration (eg. tanks ignore opponents' adjacency bonuses, thus they would be the only units that can really cause breakthroughs even though you would still need to fill in the gaps behind your tanks with infantry), having it take place throughout the entire game would lead to stalemates between players who line up their forces against one another to maximize adjacency bonuses.
Thing is, I don't think the ideology system was meant to be flexible: it was more designed to be an extension of the policy system with a few extra perks (eg. ideological pressure, tenet levels, a requirement system that is not as set-in-stone as the ones for policies). Switching ideologies has so many penalties that it honestly feels more like the justification for ideological pressure (which I honestly think is an afterthought and something Firaxis shoehorned into the culture game just because they thought it would be cool) than a viable gameplay strategy. The policy system, for better or for worse, is definitely not meant to be flexible: it is more meant to let you double down on your plans (eg. Tradition doubles down on tall empire building, Liberty doubles down on wide empire building, Piety doubles down on religion, Aesthetics doubles down on trying to get a cultural victory) than to always assist in your plans even when those plans change (eg. switching to Pacifism when you realize you've started on your own continent, then switching off of Pacifism when people start building Galleons to transport units across oceans).
In any case, my point was more along the lines of how Civ4's civics have a lot fewer total possibilities than Civ5's ideologies and policies, but Civ4's civics have vastly more different, possibly viable gameplay possibilities in practice than Civ5's ideologies and policies. Newer players get punished a lot more from unnecessarily switching from Decentralization to Mercantilism in Civ4 that they do from delaying finishing their Tradition policies in Civ5, but can also lose out on a lot more by not switching to Mercantilism when it's appropriate versus taking Liberty when Tradition is more appropriate. It's why the Civ4's civics and SMAC's social engineering are more intimidating than Civ5's policies. BTW, if you aren't familiar with SMAC's social engineering system, I highly recommend you read up on it, even if you dislike SMAC's sci-fi setting.
Just because people create more streams and Let's Plays of a game doesn't necessarily mean they like that game better than another, similar game: it's why you'll see more streams and Let's Plays of Nightmare-level Dragon Age Inquisition than of Nightmare-level Dragon Age Origins (YouTube returns 2x the results, despite the fact that Origins has been out for 5.5 years while Inquisition has been out for only half a year), even if most veterans agree that Nightmare-level Origins is strategically deeper (and probably more fun to play) than Nightmare-level Inquisition. Origins might be the better game, but Inquisition is still good enough to be enjoyable, and people expect to generate more new viewers by playing a newer game with better visuals than an older game whose action is also slightly slower (returning viewers are returning for the person, not the game). I can think of similar cases with Deus Ex vs. Deus Ex: Human Revolution, or Divinity 2 vs. Divinity Original Sin, or Masters of Orion vs. Endless Space, or HoMM3 vs. Might and Magic Heroes 6, etc.Only the conclusion I have to disagree withI don't think it's a completely logical step from figuring out Civ5 and then thus preferring Civ4. Strictly speaking if that were true, we wouldn't be seeing all these Deity players on Civ5, streaming and recording their matches even after BE came out, and a lot of them being proficient in Civ4 as well.
Civ5 mixes in tactics to what was originally a (predominantly) strategy game; this is not just apparent in 1UPT, how units don't die instantaneously, and ranged attacks, but also how the randomization range of combat is capped to effectively +/-5 damage at most. While this is good at giving more immediacy to combat and making the player feel like they are more in control, it has too many negative effects on the strategic experience (the production stuff, the unit composition imbalance, etc.) to be worth it, especially once stacking air units come into play. However, the role of tactics does make for a better viewing experience, and the payoff for taking a city from executing a quick surround on it is a lot more apparent (and possibly more enjoyable) than for taking a city solely due to you deciding to drop your science funding to 0-10% for 2-3 turns so you'd have enough gold to upgrade your axemen into macemen before attempting to take the city. As a result, Civ5 will probably be more enjoyable to spectate than Civ4.
That guide was written for singleplayer Civ4; a lot of its unit composition recommendations will not work in multiplayer, where the other players will anticipate your Trebuchet + Maceman stack and will build Horse Archers (which always attack siege units first, even if the defending stack has a Pikeman) + Crossbows (which counter Macemen mainly because of its +50% vs. melee bonus) to counter. In fact, from my experience with Civ4 multiplayer, stacks of doom are a lot rarer than in singleplayer, primarily because of earlygame permawar (everyone is pretty much at war with everyone else during the first 20% of Civ4 FFA's in my experience) and how people can simply do so much more damage by having their opponent run after four Guerilla II Axemen pillaging in four different directions than by bunching them all up in one tile.
Let's not bring singleplayer into Civ5's 1UPT dispute. We both know that the AI is terrible at handling 1UPT (it's actually one of the biggest things I keep improving, or at least trying to improve, in my mod), even more terrible than handling stacks of doom in Civ4; I'm also afraid to say that based on my Civ5 AI efforts, an AI that is OK/decent at handling 1UPT would probably prolong game turns a lot (think 10 seconds -> 2 minutes per AI player, and each CS is a different AI player), though this could probably be remedied if the game's pathfinder code was not 100% serial (ie. it could make use of tasks running in parallel).
What does the immense return from investing into units when you have a small number of them have to do with the fact that building an army is a huge investment, supposedly with a lot of payoffs? Building a large army definitely is a huge investment, but the payoff may not be worth it: the power difference between 8 battleships and 12 battleships is significantly greater than the power difference between 12 battleships and 16 battleships (simply because you can bring 12 battleships to bear on a single front almost as easily as 8 battleships, but things get traffic-jammy with 16 battleships), even though the production differences are identical. This is a big part of the problem: because of how hard individual units die and how difficult it is to bring your reserves' power to bear (you'd just be waiting around with them until one of your frontline units died, at which point you'd move a unit in only to see that the enemy unit you almost killed just insta-healed, effectively negating most of the gains made by sacrificing the frontline unit), large armies can only be used for wars of attrition. Even then, the player is simply better off constructing a moderate-sized army that can bring all its forces to bear and building a few science/growth buildings on the side than constructing a gigantic army whose units are constantly stuck in a traffic jam and getting left behind in science/growth.
