@Buttercup your post has many assertions that don't align with my experience with the game.
Yes, I build offensive units in the Ancient Age. Neither Warrior Code nor Iron working are dead-end techs, though Horseback Riding and later Chivalry could be considered dead-end. Isn't Horseback Riding required to exit the Ancient Age?
Neither Feudalism, Invention nor Gunpowder are dead-end techs in the Middle Ages.
Since acquiring land (and food) is key to achieving any of the victory conditions, expansion using both settlers and military is common across many play styles. It's considered a significant challenge to play "tall", with only five cities or one city. I don't try to "crush the AI" in every game. Most games I pursue Space, and I fight wars to gain land and valuable resources. About a third of my games I pursue Domination, so yes, I do plan to "crush the AI."
An element missing from your analysis about the shield cost of more advanced units is the ability to upgrade older units. Once I research Feudalism, I tend to spend gold to upgrade my swords to MDI. Yes, I am building new MDI but not a whole army's worth. Once I have acquire iron, I rarely build enough archers to upgrade them to longbows, but I could. If I choose to research Chivalry, I would upgrade my horsemen to knights. Neither I nor the AI have a surplus of shields in the Ancient and early Middle Ages. Making decisions about what to build (and the AI loves to build wonders!) is a key advantage that human players have.
Yes, after the first 50 to 100 turns, I look at the terrain that the game has given me to use -- one of the many random elements in the game -- and decide whether it will be fun to try to win this game. I have had fun on maps where iron was scarce, and I needed to plan a significant invasion to get it. I have had fun on maps with few rivers, so that I couldn't easily build the Hoover Dam, when I got that far. I have had fun on maps where I had both iron and rivers in my starting core. I have abandoned games where I wouldn't have fun trying to dig out of the hole that the RNG gods have put me in.
A nice post this, very good quality in detail. I'm not seeing the "many" aspects of my failed assertations, but I can understand the sentiment, because the game is so rich and deep with different mechanics and debatable points that it is very difficult to highlight them all in one post, one just hopes the other person gets the gist. And I do get your gist.
Regarding the "dead end techs" angle, obviously, some are techs you have to take anyway, but for brevity's sake I didn't want to list the nuance of every tech. However, your post insists I clarify, so I'll try my best to.
The Ancient Age tech tree is deliberately tailored so that it's very cumbersome to bee-line the actually useful things. If you want Horsemen, you have hack through first rank techs and then it dead-ends. Yes, you'll still need it to progress to the next tech age, but going that route at the start of the game is severely cumbersome because, even if you are the first Horsemen, you wont have the time or capacity to spam horsemen at that point, due to prioritising settlers, workers, and aiming to get culture up and all that kind of stuff.
As a result, Horsemen is usually a relatively late consideration, at which point the AI probably has them as well and they will probably have loads of spearmen and swordsmen as well, to which the horseman doesn't have that great an advantage. Horses also rely on you actually having a horse resource, and you may need to wait for this to be connected or, as you say, you may need to fight for it, both of which can be a major hassle.
This all excludes Unique Units, of course, which adds a different set of variables, thought he cost of the Iroquois bonus Horseman is, of course, another point that matches my proposition, in that costs so much in shields that you may not be doing that much conquest before you've got Knights anyway. I must admit, I've had good romps with the Iroquois UU, but I did benefit from lovely starts, a point not to be forgotten.
Iron Working is necessary, but it's not necessary very early on, because its main benefit to learn is that it's a path to Construction, one of the most crucial AA techs because it provides the much needed aqueducts. Other than that Iron Working only provides you with the knowledge of where Iron Deposits are, and if none are nearby or in really awkward places, you won't be getting to them for many turns anyway, there are probably more useful techs you could be learning in the meantime. Because even if you get a good early Iron Deposit, you're still not going to be overly bothered with massing an army of swordsmen when you are building settlers, workers, and are permanently at your allowed units limit.
These are both techs that get learned, but they are much more 'at some point' rather than prioritised unless you are specifically aiming for one specific kind of scenario, such as defeating the AI in record time on a small to medium sized Pangea or whatever. The two techs don't actually offer anything beyond 'you have a percentage chance to build some better units'. And if your percentages fail, you've wasted your time, because Horseriding leads nowhere and Iron Working only gets you halfway to Construction, something you'll desperately want after about 150 turns, but not really before that anyway, almost whatever your playstyle.
Warrior Code is an interesting one. It's certainly nice to have, but is it worth it rather than any other tech at that point if you don't start with it? It only leads to Horsemen (see previous) and to Monarchy, a very specific tech that is itself a totally dead-end optional tech. Yes, you get access to archers, which can be very useful, I've quickly overrun enemy AIs with Archers before, but they tend to be a bit rubbish against other archers or swordsmmen and only have a mild advantage over Warriors for most everything else. And yet they cost twice as many shields as a Warrior. So, for Barbarians, might as well stick to Warriors and learn something else, but for crushing an AI quickly on a small to medium Pangea, sure, it has its value, if you're extremely quick about it.
But, again, Warrior Code doesn't offer anything other than the improved unit, it doesn't particularly lead anywhere, hence, generally speaking, it's quite the dead-end tech, but not quite as much as the previous two.
While Feudalism, Invention and Gunpowder are not dead-end, they mainly rely on you punting for resources again. Longbowmen are the odd-unit-out, and, as we all know, these were a later addition to the game, another mad attempt to paper over the cracks of flawed combat system of the original design. While Longbowmen are an offensive unit, they cost quite a bit and offer no defensive threat and will be attacked no matter how many you pile up, and they'll die easy, and they are furiously slow, none of which bodes well for most styles of play beyond remorseless spamming and invading under a government type with zero war weariness, a fine playstyle, but a playstyle that would probably be all over by the middle of the Middle Ages anyway.
I feel somewhat confident in assessing that anyone who's made it to the Middle Ages and is only reliant upon Longbowmen by that point and has a mind to play for war is either going to be quitting or waiting for Riflemen or possibly Rubber. Or grinding it out by vast numbers, of course.
And as for Medieval Infantry, they're not much to write home about either. Yes, MIs can carve through Spearmen quite easily, but then so can Archers and Swordsmen. The real advantages in the MeA come from Knights and Cavalry, both of which require two resources and dead-end techs. And the advantage isn't in the numbers, the advantage is in the movements per turn and the retreat chance.
To which the point of the initial post was that, for example, if you have 6 cities spamming MIs and the AI has 6 cities spamming Swordsmen, you're not really gaining much by being an entire tech page ahead of them, and, in fact, could be at a disadvantage, because you're paying far more for barely any change in combat prowess and, exploits aside, the AI Swordsmen will probably win in the long run because they're making them at a far greater pace than you can produce MIs. And, similarly, Pikemen have barely any noticeable advantage over Spearmen, they die almost as easily regardless of terrain and etc.
It's unlikely that the human player will be in this situation because the human player knows this and therefore plays to not get into that situation. Or just quits, of course.
And my point really counts when I read your concluding paragraph where you say, yes, you do indeed quit if the circumstances don't pan out to something whereby you have access to everything anyway by some means, when, in reality, wouldn't it also be fun to play a Swiss game and have appalling terrain but just have the fun of watching the enemy fall like mayflies upon your mountain-top entrenched Pikemen as you calmly try to ignore the surrounding world and do your own thing. Alas, no, having a rocky start with not much but snow to wander about in is an almost automatic quit, why? because we know already that those Pikemen will easily succumb to those relentless AIs unless we have 4 on each mountain all fortified in Barricades, something the game's economy doesn't allow, assuming you still want to win any of the VCs.
Actually my "one-and-only-one type of playstyle" is 20k with no wars. (While I play a lot of different games, this is my favorite.) I have hydro plants because I build Hoover Dam, not because I have a lot of rivers. I trade for iron, coal, and oil if I don't have them, and I'll usually be missing at least one, because my empire is so small.
Yes, my favourite is 20k as well, so I'm surprised you aren't in more agreement.
You can't build Hoover's Dam without a River, so I'm not sure what your point is about the HD is here. Also, just because the HD puts a Hydro in all your cities, I'm not sure if any have any effect if that city isn't near a river anyway. I thought I tested this once and found it only affected cities with a river nearby and for all the other cities it was a useless folly. I could be wrong on this point. Also, my 20ks are usually long over before HD even comes into relevance, so there could be a difficulty level misunderstanding between us here.
Yes, you can trade for resources, but this requires the AI to be at the same tech level as you, so in those circumstances you are not in an advantageous position to the AI anyway for the notion of "being more advanced isn't that relevant to the combat" to be relevant, which was the point you were initially addressing.
But, yes, I can see and understand how threads and discussions can get confused via side-tracks and point-to-point deviations, so I don't hold it against you, I think you're a great poster and a great Civ3'er, and, as I said earlier in the post, with so many things to talk about with this game it is difficult to get it all out in one post, but if we could try to stay on track and try to get the most relevant details into each post it would save a lot of potentially enraging communication as the piece-meals turn into different topics.
The topic being: to what extent would your choice of playstyle be improved if you knew a Pikeman fortified on a Mountain would be much more of a fearsome rock that could probably take out at least 4 units with an attack value up to 5 before it looked like dying instead of the random 50/50 chance of it dying first time against a Swordsman that we currently endure?
