Selrahc Wrote:So is there any reason you've skipped KoTE for now? In your tech plan you list it in before philosophy and way of the wicked rather than after.
Great question, and very observant. You remember my tech plan better than I do!
The reason is simple. Right now, all of my cities are building either monuments, military to protect new city sites, workers or settlers. I'm behind the ball in cities, and I need to concentrate on catching up in expansion and development. This means I cannot spare a city to pump out adepts.
KoTE therefore only gives me things which I simply won't use yet. Slavery on the other hand gives me the whip, which at the very least I expect to use to build the monument in Tongurstad, which has 3 excellent food tiles but no decent production in the inner ring.
When I wrote the tech plan, I didn't really know how the timing of techs and development would work out, and thought I might be ready for it now, but I'm not. In actual fact I'm thinking of getting Education and Code of Laws next and putting it off even longer. Haven't decided for sure though, depends how many turns they will take. I'm also considering only revolting to slavery when I can revolt to Aristocracy on the same turn, but again I need to check how many turns that will hold up slavery, and where Tongurstad in particular will be at by then if I do - if it's going to be 6 population with still no monument and working desert scrubs then that's not so good.
I'd say three cities at turn 70 is a little behind the curve, even for FFH2. Good to know that you're planning on fixing that up soon.
Good plan with heading for Slavery next; makes a lot of sense to get the whip online as soon as possible.
So have you completely given up on a Pyre Zombie-driven attack against WarriorKnight, or are you still considering it at some point in the future? Obviously it doesn't make much sense to go invading before you've got more cities founded and more of your empire developed at home, but eventually I imagine it'd be a good idea to start thinking about how to take him down a peg or two. He looks to be the clear leader in the game at this point (or at least one of two leaders), so striking while you have a decent advantage will be key to not eventually getting crushed by sheer numbers (whether economically or militarily or both).
Anyway, just a few thoughts. Expansion definitely seems to be the key for the near future.
So, the two workers roading their way through the many hills to Tongurstad made me think about a more general problem. You've got 'W' workers, which start next to a line of 'H' hills that you need to road. What's the best way to do it, and how many turns does it take? Based on this I bring you another of the more mathematical posts.
Of course you can substitute forest or jungle for hills in the discussion below. I had hills, so I'm talking about hills.
Well, with 1 worker it's easy, it takes 3H turns. It maxes out at 4 workers taking H+1 turns - basically each turn two workers road, then the other two use the road to move to the next hill, and you get one more hill roaded each turn. Any more than 4 workers and it doesn't get any faster. It's easy to describe why - there's no way to get a worker to the last hill in less than H turns, so H+1 is as good as you can do no matter how many workers you have. With 2 and 3 though it's quite tricky.
The obvious pattern for 2 workers is to simply move both onto a hill one turn, then road the next, taking 2H turns in total. However, this is suboptimal, and if there's one thing I don't like it's suboptimality Here's a pattern which takes 7H/4 (give or take some rounding) turns. The notation is w below for where the workers are positioned, H for a bare hill, 1 for a hill with a turn of road building, and r for a roaded hill:
Code:
H H H H H
w
w
1 H H H H
w w
r 1 H H H
w w
r r H H H
w w
r r H H H
w w
r r 1 1 H
w w
r r r r H
w w
r r r r H
w
w
So, it repeats itself every 7 steps, and roads 4 hills in the process, hence the 7/4.
Three workers is interesting. The best repeating pattern I can find is 4H/3. There's a lot of ways to get 3H/2, for instance:
Code:
r H H H H
w w
w
r r H H H
w w
w
r r 1 H H
w
w
w
r r r H H
w w
w
However, this feels very inefficient, and it is, because it's suboptimal. Here's the repeating part of a 4H/3 pattern:
Code:
r H H H H
w w
w
r r H H H
w w
w
r r H H H
w w
w
r r r 1 H
w w
w
r r r r H
w w
w
What made me take a while to find it was that just starting from the initial position it takes a lot of moves to actually get to the repeating part. Here's one sequence which gets there, but it might not be the best.
Code:
H H H H H H H
w
w
w
1 H H H H H H
w w
w
r 1 H H H H H
w w w
r 1 H H H H H
w w w
r r 1 1 H H H
w w w
r r r r H H H
w w w
r r r r H H H
w w
w
r r r r r 1 H
w w
w
r r r r r r H
w w
w
If you count the move onto the hill before the start (which technically you should), it takes 9 moves to build the first 6 roads. So it's (4H/3 + 1) in total, which is still only as good as 3H/2 until the repeating pattern starts after the 6th hill. In practice then the 4/3 is maybe a bit of an illusion. But in practice you don't necessarily always start with the three workers arriving together. If one worker arrived earlier than the other two in the first place for instance you could move straight into the repeating pattern.
Overall, the speed increases, but the efficiency reduces with more workers. I just give the repeating pattern time here, because the "+1" for 3 and 4 workers is really just a function of the starting setup, which was a little arbitrary anyway:
Code:
# Time Worker-turns-per-hill
1 3H 3
2 7H/4 3.5
3 4H/3 4
4 H 4
Have I learned a useful general rule out of this? Maybe not. Although knowing the patterns might save a few worker turns. And it's good to know that all the workers arriving at once is bad for the 3 and 4 worker case can save you a worker turn or two as well, by getting a couple of them to do something else before they head off to start roading. My experimentation to find these patterns also taught me that in general it's good to send a worker off ahead of the pack to road the way for the others. The classic example of this is actually roading grass then a hill with two workers - don't road the grass first with both, instead send one straight onto the hill so the other worker never wastes a turn climbing it.
The related question of roading grassland with a large number of workers is also quite interesting. Another interesting one is roading hills, but with grassland nearby to use to get the workers ahead onto further hills. This case doesn't max out until about 15 workers or so (I haven't worked our exactly how many). Although normally you'd just road the grassland instead in this case so it's a even more theoretical than the others. I might discuss these later if people are interested.
Overall, do I spend too much time optimising stuff like this and missing the big picture? I have no doubt that I do. But I enjoy it
Lord Parkin Wrote:I'd say three cities at turn 70 is a little behind the curve, even for FFH2. Good to know that you're planning on fixing that up soon.
Good plan with heading for Slavery next; makes a lot of sense to get the whip online as soon as possible.
So have you completely given up on a Pyre Zombie-driven attack against WarriorKnight, or are you still considering it at some point in the future? Obviously it doesn't make much sense to go invading before you've got more cities founded and more of your empire developed at home, but eventually I imagine it'd be a good idea to start thinking about how to take him down a peg or two. He looks to be the clear leader in the game at this point (or at least one of two leaders), so striking while you have a decent advantage will be key to not eventually getting crushed by sheer numbers (whether economically or militarily or both).
Anyway, just a few thoughts. Expansion definitely seems to be the key for the near future.
Well, you could be right. That's no why he was so willing to sign an NAP to T110. But I don't really see myself as having a reasonable opportunity. I don't think pyre zombies are such an advantage that I could pull a victory from an otherwise trailing (if only marginally so) position that I'm in at the moment. When you add to that 20 turns of being vulnerable but unable to make progress or regain lost ground it looks like pretty miserable prospects. Given that I'd be destroying my relationship with my only neighbour by doing so it just doesn't seem worth the cost.
Yes there's a risk I'm allying myself with a runaway and dooming us all in the process. It wouldn't be the first time I've done that either would it? But I still think this is the right choice. My plan A at this stage for WarriorKnight is to out-race him to the Tower of Mastery, which I ought to be able to do unless he gets completely out of control.
I also actually think that, as good as pyre zombies are, I'm going to have my strongest civ-related advantage when I reach Strength of Will. That's the pinnacle tech that I'm developing towards. There'll be another local peak at Sorcery, and a series of improvements from a range of Planar-Gate and Ashen-Veil based acheivements, but Strength of Will is when I have the most chance of taking down an otherwise superior rival. If by then I'm simply defending my empire while building the Tower of Mastery all the better.
It's shaping up to be fairly likely for there to be war between the Elohim and Malakim, due to their leading positions. Let's just hope there's no clear winner in that battle.
PS to all: I've put code tags where needed in the previous post, so if the diagrams and table all looked shoddy when you read them before, they're fixed now.
I'm curious, what makes you think you can outrace Warriorknight to the Tower of Mastery? He's already ahead of you, demographics-wise, and you don't have economic traits. It's true that attacking him form a position of weakness is a bad idea. Apart form Pyre Zombies and Spectre spam, what do your military plans look for in the future? Do you hope to have someone else pop his worldspell before attacking?
Caustic Soda Wrote:I'm curious, what makes you think you can outrace Warriorknight to the Tower of Mastery? He's already ahead of you, demographics-wise, and you don't have economic traits. It's true that attacking him form a position of weakness is a bad idea.
The same way I used to consistently be the first person on my train to the top of the steps at the station, despite most other people on the train being fitter. It's the oldest trick in the book: The best way to win a race is to be the only one running.
I'm going to be focusing my game on the arcane techs and racing to strength of will regardless. I'll be building the Tower of Necromancy ASAP to support my many, many undead units (spectres, wraiths, skeletons, pyre zombies at the least), and I plan to have all the mana I need for that without a single mana node. I'll be building the Tower of Alteration as a very high priority item to boost all the support spells I'll be using. I'd be crazy not to want to build the Tower of Divination, and my best guess is that I'll take Malevolent Designs with it (assuming I take Strength of Will with the Infrenal Grimoire). So it's only the Tower of Elements that's out of the way and not especially useful. I'll get that last as part of a run to victory.
The Elohim on the other hand are inclined strongly towards following the religious path. Ice mana may pull him a little towards the arcane line, but that remains to be seen. Corlindale can cast snowfall, that might be enough for him. On the whole though I don't expect the towers are in his sights particularly.
He's ahead at the moment, sure, but it's not that big a lead, and it's certainly not over yet. Also, he may retain his economic lead, but with my plans to prioritise sorcery, bulb my way through Arcane Lore and pick up Strength of Will with the Infernal Grimoire, I expect I will get to the only expensive tech I actually need well before I or anyone else is prepared to build the last tower. So it's organising the mana, and physically building the towers that takes time. A stronger economy isn't needed for either of those, just mana nodes (and therefore land) and one or two good production cities.
Caustic Soda Wrote:Apart form Pyre Zombies and Spectre spam, what do your military plans look for in the future?
The core of my offensive military in particular will be spectre-summoning mages like you say. Supporting them though will be all manner of things, basically anything that can keep up movement-wise. Which notably doesn't include pyre zombies, as they cannot be hasted. But does include all the support spells I can get my hands on (from other mages, adepts, ritualists and Succubi), a fair number of some sort of living (i.e. haste-able) unit to defend the mages (haven't decided what yet), and I'm not sure what else.
Permanent undead like pyre zombies will mostly play a defensive role, ready to smash down invading stacks, with help from the defender's home ground advantage (except against the Lanun anyway) to gain the initiative (although the pyre zombies don't always even need it).
I have a long post semi-prepared with a lot more detail about how my magical units will be organised, which I plan to give you when I get around to researching Knowledge of the Ether.
Caustic Soda Wrote:Do you hope to have someone else pop his worldspell before attacking?
I meant to say this in my previous post: Oh goodness me yes! Someone else popping that thing, while I'm still at peace with him, would be an absolute godsend. How to convince someone to do it, particularly without having to get involved myself, now there's a question. We'll just have to wait and see I guess.
@Tower of Mastery: That makes sense. I would still be paranoid about others winning beforehand, but it's not like there are many faster options that ToM anyways.
What I really mean is that, as your map looks now, you seem to have the Elohim to your north, ocean to your E and S, and desert to your W, which will limit your expansion prospects. This is mitigated by beelining for arcane techs like you'd do anyways, but it will still put a damper on your economic potential. Eh, I suppose you'll cross that bridge when you get there.
@Army composition: Have you considered how to deal with enemy assassins, and if so is that part of your arcane writeup for when you get KotE? Do you know if hawks can see assassins? I don't have vanilla FFH installed so I can't check, but IIRC you can only get Guardsman on melee units, and since your PZ can't be hasted they won't help on the offensive.
Caustic Soda Wrote:What I really mean is that, as your map looks now, you seem to have the Elohim to your north, ocean to your E and S, and desert to your W, which will limit your expansion prospects. This is mitigated by beelining for arcane techs like you'd do anyways, but it will still put a damper on your economic potential. Eh, I suppose you'll cross that bridge when you get there.
My broad plan for further expansion beyond the area I have to myself - which is still pretty large for the most part anyway, is conquest. Possibly of land a fair distance away. Hard to say at this stage of the game. It's also quite possible that all I need is sailing to get to land that isn't that far away to the southeast. With all the jungle, there's certainly room for more land down there.
Caustic Soda Wrote:@Army composition: Have you considered how to deal with enemy assassins, and if so is that part of your arcane writeup for when you get KotE? Do you know if hawks can see assassins? I don't have vanilla FFH installed so I can't check, but IIRC you can only get Guardsman on melee units, and since your PZ can't be hasted they won't help on the offensive.
Assassins aren't invisible, hidden, or anything like that. They just have the Marksman promotion, which lets them pick their target (kind of) when they attack. Shadows are invisible, but assassins aren't. Hawks can, according to the manual, see invisible units, as can floating eyes - which I may end up using exclusively in place of hawks, and as can Revelers, which I plan to spread around my empire as well as take along on any foreign expeditions. So not much risk of something sneaking up on me.
As for protecting mages from assassins, I could use Royal Guard, but that requires an expensive and otherwise mostly worthless tech. In actual fact I plan to do something much more effective: Stay out of range, or kill them before they get the chance.
As I said in my post on mobility and initiative earlier, regardless of whether you're the aggressor or defender in the war, being the one to actually make the attack in a big stack-vs-stack fight, in other words fighting the battle during your turn, is crucial. Assassins are only a small part of that. A stack that is exposed to assassin attack is a stack which is exposed to magic - in particular the likes of snowfall, tsunami, maelstrom and (depending on what's in my stack) destroy undead, as well as debuffs like wither, dispell magic and rust. A stack that's exposed to magic is a stack that's exposed to a complete wipeout with minimal opposition losses. Assassins or not, I'm not putting my stacks where they can suffer a complete wipeout.
So instead, I'll stay out of range, while still attacking with 4-movement, 2-turn-lasting summons. I'll make sure I know I'm out of range with floating eyes. If they bring the stack to me, in particular trying to chase me down with potentially faster assassins, then they have to put their stack in range of me. In that case, it's their stack of assassins that's exposed to my magic, and I should be able to kill it easily.
I need to watch out for good maneuvering on the opposition's part of course, but the gist of it is that I won't leave them exposed to assassins in the first place.
The stack defense I have planned is primarily for protection from super-fast moving units like cavalry. Cavalry can have up to about 6 movement, and I can't outrun that so I need to be protected from it. Otherwise though I plan to keep my mages out of range of everything really. Assassins have base movement of 2, which gets them up to a maximum of 4 with haste and mobility. I plan to have haste and mobility myself for 3 movement. So they're potentially faster, but can't chase me down without leaving themselves exposed.
Do you have horses? If so, would it make sense to use Horsemen or Chariots to supplement your mages? They should be plenty fast and reasonably strong. Horse Archers would be a bit out of your way, but then again so would most other units to go with your mages, with the possible exception of the priests of whatever religion you go with.
Here's a pattern which takes 7H/4 (give or take some rounding) turns. The notation is w below for where the workers are positioned, H for a bare hill, 1 for a hill with a turn of road building, and r for a roaded hill: