I've heard it said that Epic/Marathon make warring easier. Are there any aspects of the game it makes more difficult rather than just taking longer?

I'm pretty sure it makes technologies take longer to research.

Playing the game as a builder is harder at the slower speeds. Especially on marathon where unit costs are 2x and everything else 3x the AI comes at you with huge stacks. Because units cost less every hammer you sink into a wonder could effectively be 1.5 hammers into a unit so you have to be pickier about what you build.

Also, warring is only easier on epic/marathon if you're doing well. If you find yourself in a hole you can easily be overwhelmed since that giant SoD the AI has on your borders isn't going to become obsolete anytime soon and if you don't have a stack of your own you can't exactly whip one up as quickly as you could on normal speed.

slower speeds meed your in each era longer. What that translates to is a longer number of turns where your axeman can pose a threat for quick expansion, or a longer period of time where calvary and cannon are a force to be feared. And, as pointed out earlier, at marathon speed units are relatively quicker to build than buildings.

You need more workers on epic compared to normal & marathon since it takes longer for them in relative time to complete their work. For example: chopping a forest on normal takes 3 turns, 5 turns on epic. Irrigating a grassland tile takes 5 turns on normal, 8 turns on epic. Also, the fact that roads take an uneven number of time to complete (3 turns) is bad imo as it complicates micromanagement since you usually can't afford the luxury of grouping three workers together so many roads take 2 turns to complete even with 2 workers which is twice the time it takes on normal.
Of course, less effective workers could mean less advantage for the AI on difficulties where it starts with workers so you could argue that it's an advantage for you.

The main added difficulty on slower speeds, and specially Marathon, is that it is far harder to research your way out of trouble.

Screwed up your early economy by overexpanding? Currency is going to be a long, long time away.

The neighbors are starting to bully you and you haven't researched Feudalism yet? Give in or be run over.

Peace treaties still only last 10 turns, and researching a technology typically takes longer than that, so you can't rely on continous trading to keep your neighbors happy either.

Keeping up-to-date with military techs is far more important on Marathon than it is on faster speeds. It's either that or play a good diplomatic game.

If you are playing defensively (defending against AI SoDs, hiding in your cities with longbows, etc.) for any extended period of time, you may have bigger problems than what speed you are playing.

game speed really affects the overall strategy. it almost makes it a different game

because if buildings cost 3x as much and military only 2x, this will almost always lead to troops having a better cost/benifit ratio than city improvements. not just in raw hammers, imagine this scenario:

you can't expand as fast, because it takes longer to get to CoL and currency. but early wars of attrition are even more crippling to the ai. say you capture some cities, because you have the troops, even though you know you can't keep them. when you gift them back to the ai, basically all of the improvements will have been destroyed. so while everyone's military losses are multiplied twofold, in terms of hammers, losses inside cities are threefold.

basically, slower speeds involves maintaining large armies and marching them around in circles, while faster speeds let you steamroll and conquer from very early on. i play epic.

@gliese 581:
if you split up worker actions, like by individually moving and assigning tasks to workers, you won't lose any turns on roads. on a blank tile, worker A builds roads, worker B builds roads. next turn, A finishes roads, B moves and starts new roads. turn three, A moves and finishes roads on that turn. you can also start two workers on the same tile and assign them individually build road to destination commands, and they will leapfrog without losing turns. however, if they are grouped together, they will lose that turn. this is on a mac playing warlords, and it seems like a bug.

Also golden ages don't last as long (relatively) as on faster game speeds. for example: normal 8 turns --> epic 10 turns = factor 1.25 whereas most of the other modifiers are scaled by 1.5.
So GPs are better used for bulbing & settling.

So GPs are better used for bulbing & settling.
...relative to normal speed. But this does not mean that golden ages are less optimal on epic than your other options.

...relative to normal speed. But this does not mean that golden ages are less optimal on epic than your other options.

Does this mean that burning GPs for GAs on normal speed is the "most optimal" option and GP usage is best balanced for epic speed?
Or, if the CIV-Gods intended perfect balance on all game speeds - why is GA duration scaled by 1.25 (normal2epic) while the bulbed beakers are scaled by 1.5? What am I missing?

I have no hard math on the subject, I can only say that when I play marathon games as I usually do I almost never utilise the GA ability from my GPs. Maybe if I get one I can't do anything useful with, normally a great artist or great prophet.

Else I find bulbing far more efficient on Marathon, could be that you see those very long research times being cut efficiently with a bulb but I'm sure if I would calculate on it, a early settled GP would likely earn more in the long run with the added turns. But on that subject there's already plenty of threads describing this and advocating one or the other solution. And there's probably some gain in a Golden age as well, even on Marathon.

Also, the fact that roads take an uneven number of time to complete (3 turns) is bad imo as it complicates micromanagement since you usually can't afford the luxury of grouping three workers together so many roads take 2 turns to complete even with 2 workers which is twice the time it takes on normal.

Uh-oh. Due to Civ IV's anti-micromanagement bent, I have always assumed that, unlike in Civ III, workers assigned to a task that does not need an integer multiple of their worker turns just end up with only a part of the team having to work on the last turn of the job. And heck, I always thought that what I saw was corroborating that. Are you saying I've been wrong, or are you just saying that the mentioned job "objectively" takes two turns?

EDIT: Ah never mind, I just read Zombiestomp's reply.

Bostock

@gliese 581:
if you split up worker actions, like by individually moving and assigning tasks to workers, you won't lose any turns on roads. on a blank tile, worker A builds roads, worker B builds roads. next turn, A finishes roads, B moves and starts new roads. turn three, A moves and finishes roads on that turn. you can also start two workers on the same tile and assign them individually build road to destination commands, and they will leapfrog without losing turns. however, if they are grouped together, they will lose that turn. this is on a mac playing warlords, and it seems like a bug.

Yes I always assign each worker manually and don't group them and I always cancel their orders after doing part of a work on their way to my target tile if possible so that I minimize loss of efficiency.
What I mean is that I'll frequently move two workers together although I still give them individual orders (if I can do it without losing work). If 2 workers are set to building roads over open land then every other road will take 2 turns to complete, that extra turn compared to normal can mean a military unit moves 1 step less for that turn which is what I was referring to.
This is all small stuff, chopping takes about 11% longer to complete than on normal, farming grasslands and plains about 7% longer. But it can add up for a large empire so to get the same work done you would need an extra worker or two on epic compared to normal/marathon because of this.

Does this mean that burning GPs for GAs on normal speed is the "most optimal" option and GP usage is best balanced for epic speed?
No, it means that golden ages on Epic can still be a good idea even though the statement I quoted would imply it is not.

I would agree that the 'penalty' of slower speed is essentially a slower ability to react to changing circumstances. If you are the aggressor and carrying the battle to the AI, it is not a big problem.

Other differences are the effects that traits have. For example if I remember correctly you only get 2 turns of anarchy on marathon and 1 turn of anarchy on both epic and normal in the early game. This makes spiritual realtively weaker on slower speeds and more powerful on normal.

Lots of faulty math here. We really need an article that summarizes the differences.

Epic:
Faster unit travel.
Shorter relative golden age length.
Less impact from automated worker stupidity.
More opportunities to micro-manage civics (if spiritual).
More opportunities to micro-manage diplomacy.

Marathon:
Much faster unit travel.
Even shorter golden age length.
Cheaper units.
Workers cost less, making serfdom less useful (if not spiritual).
Anarchy is relatively shorter, making spiritual less useful.
Even less impact from automated worker stupidity.
Even more opportunities to micro-manage civics (if spiritual).
Even more opportunities to micro-manage diplomacy.

More opportunities to micro-manage civics (if spiritual).


True, I didn't consider this.

the thing about marathon is it allows you to use units before they go obsolete. i hate playing a normal speed game on a standard or large map because by the time your maces reach the enemy, they already have rifles to defend their cities with.

marathon, i have discovered is a lot more fun than other speed once u hit the AD's. there is simply so much going on, and so much more time for more things to happen. losing a wonder is that much more painful, but leveraging a tech advantage is so much more satisfying, because it will take the other civs 30 turns to research it. likewise catching up a tech deficit is so much more a of challenge.

but i guess it is more favourable to the warmonger than the builder. and if your a builder and play marathon, your gonna get bored stiff.

its just preference i guess.

Here are some other differences:

In my observation, goody hut results do not scale with game speed. So a money or maps result on quick is so much better for you than on marathon.

Units with more movement are more valuable on quick speed than on marathon.

The number of turns before hammer decay starts is the same, regardless of game speed. However longer build times at marathon/epic means you are more likely to actually hit decay if you are shifting things around in the queue.

Diplomatic bonuses and penalties fade as a per turn chance which (if I understand correctly) does not scale with the game speed. So you gain much more value for doing "nice" things on faster game speed than on Marathon.

I have noticed that its far easier to explore on slower speeds as well, another function of the movement factor. You can get a LOT of info about your surroundings with a Warrior on Marathon, and thus have a better idea of where to settle.

On Normal, I can see why Hunting is actually an advantage over techs that start with a Warrior. I was wondering why some were choosing Hunting as a "favorite starting tech" in the Traits/Techs thead, but after playing some Normal and some Marathon recently, I can see why now. With a warrior on Normal, if you dont go the "right" way early, you could miss some very strong settling locations.

If you are playing defensively (defending against AI SoDs, hiding in your cities with longbows, etc.) for any extended period of time, you may have bigger problems than what speed you are playing.

Just an example, it applies to any military tech really.

@Bleys: Definitely on Marathon you will have more information about your surroundings, and it's slightly easier to pick a good spot for your second and third cities.

Also IIRC, the chance for a random event doesn't change with gamespeed, but of course the AI gets roughly the same number of events as you do, so it kind of evens out.