Above Demigod

[walletta]

First of all, I am not planning to make the leap from Demigod to the next level. I am still a long way from mastery of DG. However, having learned about micro-management, pre-building, city-placement and all the higher arts that are necessary at DG level, I am left wondering: what further skills are needed at higher levels still? Is it just the same but with yet more precision or are there tweaks of a wholly different kind? Is the game merely less forgiving of mistakes?

I am assuming the latter and that the AI remains unutterably dumb but at Deity and up, dumb with even more advantages in research speed, bonus starting units, settlers etc

 

[justanick]

The step from DG to Deity is moderate. A bit more precision may be needed but i have seen people winning Deity albeit making mistakes that i considered huge. So forget Deity, Deity is easy. Sid however is a whole new level. Cost factor 4 instead of 6 at Deity means AI is 50% faster and it starts with 50% more settlers, too. Deity is something you can win on your own. For Sid you need AI the wear down eachother. AI will have such great amounts of military resources that you cannot beat it, you need to make AI beat AI and then mop up what remains. Diplomacy will be more crucial than at lower settings.

As for research it should be mentioned that reserach costs will not vary for AI. AI gets no (direct) advantage there. You get a malus and a huge one that is at Sid. DG to Deity is plus 16,67%. Deity to Sid is +50%.

 

[walletta]

The step from DG to Deity is moderate. A bit more precision may be needed but i have seen people winning Deity albeit making mistakes that i considered huge. So forget Deity, Deity is easy. Sid however is a whole new level. Cost factor 4 instead of 6 at Deity means AI is 50% faster and it starts with 50% more settlers, too. Deity is something you can win on your own. For Sid you need AI the wear down eachother. AI will have such great amounts of military resources that you cannot beat it, you need to make AI beat AI and then mop up what remains. Diplomacy will be more crucial than at lower settings.

As for research it should be mentioned that reserach costs will not vary for AI. AI gets no (direct) advantage there. You get a malus and a huge one that is at Sid. DG to Deity is plus 16,67%. Deity to Sid is +50%.

Yikes! Sid sounds very intimidating. Might give Deity a bash one (far off) day. 'Deity is easy' LOL. Like that one

 

[justanick]

Once you try Sid you have no choice but to admit that Deity is easy. At least relatively speaking.

 

[walletta]

Once you try Sid you have no choice but to admit that Deity is easy. At least relatively speaking.

Have you won at Sid, Justanick?

 

[justanick]

Technically the correct answer would be yes by accident, but realisticly speaking i did not try start a Sd game so far. Nor a Deity game. Nor i have technically finished a DG game. Need to find time.

 

[Lanzelot]

Deity is something you can win on your own. For Sid you need AI the wear down eachother. AI will have such great amounts of military resources that you cannot beat it, you need to make AI beat AI and then mop up what remains. Diplomacy will be more crucial than at lower settings.

I agree. And to add a few points: Sid also takes a bit of luck (if the AI simply decides to take you out early on, there is nothing that can be done about it) and a lot of time. For that reason my level of choice is Deity: still an interesting challenge, but doesn't require the insane amount of time.

If you survive the early phase and can then use the AI to your advantage, you have a chance of winning. On archipelago this is achieved more easily than on other maps, however, then you can't let the AI wear each other down, you need to apply different methods for beating the AI's military. E.g. SirPleb's "Funnel of Doom", the "bait trick" combined with "taking and gifting away until the entire AI army goes poof".

I've played three Sid games so far, but all unfinished, because the time ran out:

• COTM 43, a nice archipelago map favoring the human player. I submitted an unfinished game in 1625 AD, which is however won at that point: the AI is isolated and backwards and drowning in unit support, and all that remains is to spend a couple of days clearing island after island until you hit the domination level...
• COTM 55, a quite difficult map with lots of floodplain, but lacking a bit in production. Retired from this one in a lost position.
• COTM 79, another favorable archipelago map, where I had a 20K victory in the makings, but didn't manage to finish before the deadline ran out. (Still some AIs to beat down in order to prevent them from launching before my 20K city accumulates the necessary 20K.)

From that experience I can confirm that Sid is winnable, even without using the tricks (some would call them exploits...) that Spoonwood describes in his thread about his Sid HoF victories. (Most of that is probably banned under GOTM rules anyway?!)
But of course it's never guaranteed to be a win, even with the most favorable start position. Some "bad luck" can always happen at that level...

 

[walletta]

May I trouble you for a link to these, Lanzelot, or a search term?

'E.g. SirPleb's "Funnel of Doom", the "bait trick" combined with "taking and gifting away until the entire AI army goes poof".'

I have a friend who managed the single city challenge with every civ on Deity in its previous iteration (when repeated wood chopping on the same tile was allowed). It seemed to involve capturing vast hordes of workers and using them in strange ways. I find this sort of play amusing but unaesthetic. My chess is limited for the same reason (and others). I like things to be, errrr, 'realistic' (what?)

 

[tjs282]

May I trouble you for a link to these, Lanzelot, or a search term?

'E.g. SirPleb's "Funnel of Doom", the "bait trick" combined with "taking and gifting away until the entire AI army goes poof".

Found this for the FoD. I think you already know how 'bait cities' work, yes?

I suspect that the other trick might involve capturing cities then gifting them to a distant, maybe weakened third party, ideally someone who's not at war with your current victim, and/or who you intend to become your next target. Well, that is, they think you're gifting them the cities, but you know that it's more like a loan...

It's win-win for you: you don't have to bother garrisoning troops to prevent flips or quell resistance, but the gifted cities will most likely be 1-shield towns for the 'lucky' contestant, so will remain relatively poorly guarded for a very long time. As your current target loses more and more cities, he loses unit-support and his remaining units should start going 'poof'. At the same time, the third party gets saddled with a bunch of cities (plus improvements) which starting putting pressure on their Treasury as well, so with any luck their military units start going 'poof' as well -- and that's when you stop being nice to them, and start taking back all those cities you lent them...

(I remember we did something like this on a very small scale in ABLES-SG, gifting a few English cities to Ragnar -- who we'd already hammered -- then taking them back later, once Lizzie was on the ropes as well)

 

[walletta]

Found this for the FoD. I think you already know how 'bait cities' work, yes?

I suspect that the other trick might involve capturing cities then gifting them to a distant, maybe weakened third party, ideally someone who's not at war with your current victim, and/or who you intend to become your next target. Well, that is, they think you're gifting them the cities, but you know that it's more like a loan...

It's win-win for you: you don't have to bother garrisoning troops to prevent flips or quell resistance, but the gifted cities will most likely be 1-shield towns for the 'lucky' contestant, so will remain relatively poorly guarded for a very long time. As your current target loses more and more cities, he loses unit-support and his remaining units should start going 'poof'. At the same time, the third party gets saddled with a bunch of cities (plus improvements) which starting putting pressure on their Treasury as well, so with any luck their military units start going 'poof' as well -- and that's when you stop being nice to them, and start taking back all those cities you lent them...

(I remember we did something like this on a very small scale in ABLES-SG, gifting a few English cities to Ragnar -- who we'd already hammered -- then taking them back later, once Lizzie was on the ropes as well)

Ah, thanks. I read Sir Pleb's description of the 'funnel of doom' with great enjoyment. I am a big fan of counter-intuitive ways of exploiting the AI, not that I use them particularly myself but rather that I admire the humour and ingenuity they display. Yesterday, I found a thread in the Hall of Fame area in which there had been a serious discussion of the ethics of repeatedly selling the same map to the IA Apparently the AI has the memory of a goldfish and will keep buying the same map over and over again without ever stopping to think, 'Hang on! There's something fishy going on here!' I bet that over on the AI forum there's a thread called 'where's all my money gone?' with bemused posters joining in saying, 'yeah, that happened to me too. But never mind that. The human sap just sold me his world map for 5 gold! What a moron!'

The ruling on this point was that you can re-sell the map on the same turn but only if something has changed (say you have captured a town in the interim).

Yes, I recognise the bait thing now (must make more use of that one) and I have used the gifting cities trick. In a recent game, I gave a crappy little tundra town in the far north to Rome (I think) and then watched an enormous German army cross my land to take it, precipitating a long drawn out global conflict in which I remained a bemused spectator . That was the one I ditched after incompetently missing out on the Hoover Dam.

Another crafty ruse, which I have never used, is the connect-disconnect trick, in which you disconnect your own supply of iron so as to build horsemen before reconnecting and upgrading them to Knights or whatever. Now that one was thought up by an evil genius. Planting and destroying forest over and over again on the same turn is no longer possible but that's another funny one.

By the way, how many cities did Sir Pleb have in that game? There seem to be hundreds!

 

[Lord Plebian]

By the way, how many cities did Sir Pleb have in that game? There seem to be hundreds!

Sir Pleb was going for a domination victory for the hall of fame, if I recall. I believe it was one of the first tries with Sid on this site. Honestly, I'd recommend reading the thread if you have the time. It provides a very analytical (and entertaining) experience with the difficulty level and some of the things you'll face, aside from some tactics used for score purposes.
Thread is here: http://forums.civfanatics.com/showthread.php?t=81424

 

[Lord AJ]

This is off the topic but I couldn't keep from asking. Is Lord Plebian the same person as SirPleb? The similarity in the names is just too much to be coincidence.

 

[walletta]

Sir Pleb was going for a domination victory for the hall of fame, if I recall. I believe it was one of the first tries with Sid on this site. Honestly, I'd recommend reading the thread if you have the time. It provides a very analytical (and entertaining) experience with the difficulty level and some of the things you'll face, aside from some tactics used for score purposes.
Thread is here: http://forums.civfanatics.com/showthread.php?t=81424

Thanks, I will. I loved the funnel of doom

This is off the topic but I couldn't keep from asking. Is Lord Plebian the same person as SirPleb? The similarity in the names is just too much to be coincidence.

I too am similarly suspicious.

 

[Lord AJ]

Sir Pleb was going for a domination victory for the hall of fame, if I recall. I believe it was one of the first tries with Sid on this site. Honestly, I'd recommend reading the thread if you have the time. It provides a very analytical (and entertaining) experience with the difficulty level and some of the things you'll face, aside from some tactics used for score purposes.
Thread is here: http://forums.civfanatics.com/showthread.php?t=81424

Wow. That's one heck of a thread. There's just SOOOO much to learn from there. Completely different strategies than what I've ever read in any strategy article (although slightly similar to Spoonwood's, who took inspiration from here) . This is a great thread. I'll read more before I start the COTM 115.

 

[Lord Plebian]

This is off the topic but I couldn't keep from asking. Is Lord Plebian the same person as SirPleb? The similarity in the names is just too much to be coincidence.

Nah. I'm nowhere near as good at Civ. It's just a coincidence of screen-names. I probably should have picked something different, but I already use this some other places.

 

[Lord AJ]

Nah. I'm nowhere near as good at Civ. It's just a coincidence of screen-names. I probably should have picked something different, but I already use this some other places.

Hmm. I see.
"What's in a name " said Shakespeare.😉
Anyways nice to have you here.

 

[Lanzelot]

The main point of the "gift until poof" trick is to destroy the entire (huge...) army of a Sid AI without actually having to fight against it. With unit upkeep this can't be done: if the AI has 200 units and disbands one every turn, the game will be over by the time the army is defeated...

It works best on Archipelago and if the AI doesn't have horses. What you do is: land a decent Knight/Cav force on the AI's island. It needs to be just strong enough to capture one city. In the interturn a huge stack of the AIs foot soldiers will arrive next to the city you just captured. Move your entire force out of that city towards the next target city. Then gift the first city to a distant AI, which is currently at peace with your victim and does not have a RoP with your victim. His huge stack is now stuck and needs another turn to get back onto the road network and still another turn to get anywhere else. During that time your Knight/Cav stack has captured the next city. (Should be easier now than the first, because a large part of the AIs defenders are still trying to return from their little trip to the first city...)
Move your stack towards the third city, and right when the AIs stack arrives next to the second city, you gift it to your "good friend". Rinse and repeat. It helps, if you have a Knight/Musket/Cav Army for that operation, so that your stack will not get attacked while moving from one city to the next... Or if the territory is right and you can use mountains for cover.

Once you capture the AI's last city, their entire army (which may still be 100-200 units) goes "poof". Then you turn around and re-take all those cities from your "friend". Most should still be empty or protected by at most one spear.
I have seen people conquering an entire empire protected by hundreds of Pikes and Medieval Infantry with just a dozen Knights and a Knight Army using this "technique". Our team has applied it (on a smaller scale) in the Demigod SGOTM 14 (see the team thread of "Team Ivan") and I've also tried it in the Deity COTM 61 (using the Berserk in galleys as my "fast units").

 

[walletta]

The main point of the "gift until poof" trick is to destroy the entire (huge...) army of a Sid AI without actually having to fight against it. With unit upkeep this can't be done: if the AI has 200 units and disbands one every turn, the game will be over by the time the army is defeated...

It works best on Archipelago and if the AI doesn't have horses. What you do is: land a decent Knight/Cav force on the AI's island. It needs to be just strong enough to capture one city. In the interturn a huge stack of the AIs foot soldiers will arrive next to the city you just captured. Move your entire force out of that city towards the next target city. Then gift the first city to a distant AI, which is currently at peace with your victim and does not have a RoP with your victim. His huge stack is now stuck and needs another turn to get back onto the road network and still another turn to get anywhere else. During that time your Knight/Cav stack has captured the next city. (Should be easier now than the first, because a large part of the AIs defenders are still trying to return from their little trip to the first city...)
Move your stack towards the third city, and right when the AIs stack arrives next to the second city, you gift it to your "good friend". Rinse and repeat. It helps, if you have a Knight/Musket/Cav Army for that operation, so that your stack will not get attacked while moving from one city to the next... Or if the territory is right and you can use mountains for cover.

Once you capture the AI's last city, their entire army (which may still be 100-200 units) goes "poof". Then you turn around and re-take all those cities from your "friend". Most should still be empty or protected by at most one spear.
I have seen people conquering an entire empire protected by hundreds of Pikes and Medieval Infantry with just a dozen Knights and a Knight Army using this "technique". Our team has applied it (on a smaller scale) in the Demigod SGOTM 14 (see the team thread of "Team Ivan") and I've also tried it in the Deity COTM 61 (using the Berserk in galleys as my "fast units").

Brilliantly subversive.

 

[Spoonwood]

A while back in some of my 20k games/attempts I learned that on an archipelago map where you have your home island for yourself, the AI will land their units right next to your capital if it lies on or near the coast and stands undefended. By which I mean, there don't exist units exactly in the city (at least not without them stealing your plans). I mean, it makes sense in a way, since if you destroy an enemy's capital, their spaceship disappears. So, to turn that your advantage, you leave only one square open next to your capital and block the others. Preferably the weakest production square. Have enough artillery type and offensive units to destroy any invaders immediately. And then declare on the AI. And in this way you fairly safely generate a leader... though it can take some time.

If you really get lucky with it, the AI lands next to your capital at just about the right time when you have enough to kill any invaders instantly. They only do that when they plan to declare war. So, then you can pay gpt for techs/gold and then ask them to "leave or declare" and you get techs/gold for free and war happiness.

After a quick test, it looks like it also can work on a pangea map if you have the right type of pangea. Here's an example of them landing in the right spot, though I don't have enough military to handle things, since I used an already lost game for a test:

 

[Spoonwood]

I've checked my six published 20k Sid saves to help give anyone interested how long it might take to spawn a leader using the "let them come at your capital" tactic. Note that most of those games did NOT emphasize military early on. I believe I preferred to put in a library in those games BEFORE putting out say trebuchets or catapults. I also had a city which did nothing but build wonders, so I might have gotten that first MGL earlier in those games. They were 80% water maps.

The dates of the first MGL were 210 AD, 130 AD, 590 AD or 770 AD, 640 AD, 260 AD (I had no ivory in that game... it looks like I spawned an MGL using knights), and 920 AD. All of which strike me as early enough for a more conquest oriented victory.