MRG's Random Game of Randomness

[Lanzelot]

I meant to do this when I discovered writing, but I forgot. Korea asking for it during the IBT reminded me.

Korea will trade Writing for The Wheel, Iron Working, and 10 gold. America will do the same thing but with 20 gold. I trade with America.

Korea now will only trade Masonry and 10 gold for Writing. I take it.

Sorry, I forgot to mention it, as for me it is already "second nature", but: never ever trade Writing around! Especially not on the higher difficulty levels. I thought this is general knowledge.

This definitely kills any remote hope we might have had of achieving the full Republic slingshot, and may even put the "CoL slingshot" at risk, because an AI that knows Writing may now pop Philosophy from a hut any time and it's "good-bye our free tech"... (Or an AI may start researching Philosophy now and be faster than us with the big discount they get...)
We may still be lucky, but this mistake could also ruin the game already at this early point. Should we hope for the best, and carry on, or should we restart?


Worker management: never move two workers on a tile that doesn't have a road yet. (Ok, that is a general rule, there may be exceptions, if there is a really good reason.)
This early in the game we cannot afford wasting precious worker moves like that. This is a twofold waste:

• First of all, both workers spend a turn, where they can't work
• And then, a road needs 3 worker-turns, but you are doing it in 2x2 = 4.

So in total, this road has cost us 6 worker turns, instead of the usual 4. That's 50% more than it should have!
Much better is to move both workers to different tiles, let them complete the roads alone, and once they can move on the roads and start their job the same turn, you can combine them for mines or irrigations.

Building scouts is absolutely not worth the trouble with the amount of barbarians active in the world. We need to send curraghs instead to explore.

Completely agree.


Regarding next town sites: we have now finished our first ring, but all the next spots are so far away, that the settlers will need forever to get there. And when they get there, the new town will not produce much, because of corruption. I suggest we stop settler production for a short while and produce more workers. We urgently need more improved tiles for our productive towns, and we need roads to the next prospective town sites, so that our settlers can get there faster (e.g. in 3 turns instead of 9).

And also, we might be lucky and the AI settles that land for us... Let them walk their settlers for 20 turns, while we build more useful stuff (like archers) and then take these towns, once we have roads into that area...

A note on Karakorum:

KK produces settler and will produce another settler.

In the current save I see KK at size 3, building a settler which will complete a turn after KK grows to size 4, which means KK will drop down to size 2!!!
Why are you hurting our capital, our most productive city, that badly??
A settler should be finished at size 6 (optimum is in the interturn, when KK would grow to size 7). The settler will not be ready much later compared to building it in a smaller capital, but in the meantime you can build lot's of other usefull stuff (like archers...) Think about it that way: at size 2, KK produces 4 shields. Meaning the settler will require about 7 turns to be produced. At size 5-6. KK can already make 10 shields = 3 turns for a settler!
Plus all the extra commerce/science output! (Size 2: 6 gold, size 5: 12 gold!)

We should switch the current settler to archer (or even barracks?) and start the next settler only after KK has reached size 5. (Modulo that we should pause settler production altogether for a short while now.)


Not having horses is indeed a bummer. We need to prepare for an archer rush, that's going to be tough...

 

[Lanzelot]

PS: wait a moment. You traded Writing away and then started research on Code of Laws??
There is no chance we complete that and Philosophy, before the AI has Philo! (Especially not with a size 2 capital.)
And there are already 7 turns of research (165 beakers) accumulated in CoL, so switching to Philo now would waste all that.

(BTW: why do we need 10% lux at the moment? CoL would finish a turn faster, if we increase science to 100%)

Ok, I think, we need a team decision: continue (and hope we are very very lucky) or restart?

 

[MrRandomGuy]

Sorry, I forgot to mention it, as for me it is already "second nature", but: never ever trade Writing around! Especially not on the higher difficulty levels. I thought this is general knowledge.

I didn't know/remember that one. Gentle reminder that I hadn't played this for several years and am just coming back. There's a good chance I'll make more mistakes so please bear with me.

This definitely kills any remote hope we might have had of achieving the full Republic slingshot, and may even put the "CoL slingshot" at risk, because an AI that knows Writing may now pop Philosophy from a hut any time and it's "good-bye our free tech"... (Or an AI may start researching Philosophy now and be faster than us with the big discount they get...)
We may still be lucky, but this mistake could also ruin the game already at this early point. Should we hope for the best, and carry on, or should we restart?

So, here was my mindset. In another solo game I went for a Monarchy government because I want to win the game by conquest without having to worry about WW. Most of the civs had researched writing. To my surprise when I did code of laws then philosophy I was the first one to research it and managed to get construction for free. This mistake is one where I should have known better because every game is different.

Worker management: never move two workers on a tile that doesn't have a road yet. (Ok, that is a general rule, there may be exceptions, if there is a really good reason.)
This early in the game we cannot afford wasting precious worker moves like that. This is a twofold waste:

• First of all, both workers spend a turn, where they can't work
• And then, a road needs 3 worker-turns, but you are doing it in 2x2 = 4.

So in total, this road has cost us 6 worker turns, instead of the usual 4. That's 50% more than it should have!
Much better is to move both workers to different tiles, let them complete the roads alone, and once they can move on the roads and start their job the same turn, you can combine them for mines or irrigations.

Does that include clearing a forest? I thought it would be faster to have both clear a forest and then build a road to the other town. The workers in Alma is mistake. I should have sent one of them south to start another road. I see that now.

Regarding next town sites: we have now finished our first ring, but all the next spots are so far away, that the settlers will need forever to get there. And when they get there, the new town will not produce much, because of corruption. I suggest we stop settler production for a short while and produce more workers. We urgently need more improved tiles for our productive towns, and we need roads to the next prospective town sites, so that our settlers can get there faster (e.g. in 3 turns instead of 9).

We might be able to make a few more settlers to make southern cities with less corruption. Other than that I do agree we need more workers.

And also, we might be lucky and the AI settles that land for us... Let them walk their settlers for 20 turns, while we build more useful stuff (like archers) and then take these towns, once we have roads into that area...

I like that idea. We just have to hope they make towns in good spots not bad ones.

A note on Karakorum:

In the current save I see KK at size 3, building a settler which will complete a turn after KK grows to size 4, which means KK will drop down to size 2!!!
Why are you hurting our capital, our most productive city, that badly??
A settler should be finished at size 6 (optimum is in the interturn, when KK would grow to size 7). The settler will not be ready much later compared to building it in a smaller capital, but in the meantime you can build lot's of other usefull stuff (like archers...) Think about it that way: at size 2, KK produces 4 shields. Meaning the settler will require about 7 turns to be produced. At size 5-6. KK can already make 10 shields = 3 turns for a settler!
Plus all the extra commerce/science output! (Size 2: 6 gold, size 5: 12 gold!)

We should switch the current settler to archer (or even barracks?) and start the next settler only after KK has reached size 5. (Modulo that we should pause settler production altogether for a short while now.)

I didn't realize it was hurting our capital. I thought it was already set to be a settler factory.

Not having horses is indeed a bummer. We need to prepare for an archer rush, that's going to be tough...

Or iron either! America has access to iron for sure. Not sure about Korea.

Ok, I think, we need a team decision: continue (and hope we are very very lucky) or restart?

I need to learn from my mistakes. If you think the best way is to replay my turns I can do that. I don't think a full restart is necessary.

 

[Lanzelot]

So, here was my mindset. In another solo game I went for a Monarchy government because I want to win the game by conquest without having to worry about WW. Most of the civs had researched writing. To my surprise when I did code of laws then philosophy I was the first one to research it and managed to get construction for free.

But that was not on Demigod, I assume?

Does that include clearing a forest?

I would say so. Clearing that forest was not that urgent, the town still has other 2-food tiles it can use. (Edit: also, if the timber would fall a bit later, it might be more useful: e.g. finish that curragh without shield overrun, and then use the timber for something else, perhaps even a second curagh?!)
At the moment, roads are our most important tool for fast research. (Plus they speed up the settlement of new towns. It is not important, when a settler is produced, but when it reaches its destination! Sometimes you can build other things before the settler, and the settler still founds the new town at the same time, because meanwhile some important roads have finished and also because of what I already said earlier, i.e. even when the settler-production is started later, it may complete at the same point in time, because later the town is bigger and has more production, so needs less turns for the settler. Or in other words: the "extra" population that is available due to not building a settler, helps producing the "later settler" faster, e.g. i 3-4 turns instead of 7. The 3 "saved" turns can be used for building an extra archer, which would otherwise not come into existence...)
So anyway, when in doubt, build roads...

I didn't realize it was hurting our capital. I thought it was already set to be a settler factory.

A settler factory needs 5 food per turn, but we only have 4 (meaning growth cannot keep pace with production), so KK always needs a bit time to grow back. See tjs282's turn-log, where he always builds a warrior, scout or archer in between two settlers. (Except for once, I just realize... Need to give him a good talking to... )

I need to learn from my mistakes. If you think the best way is to replay my turns I can do that. I don't think a full restart is necessary.

I usually play without reloads, what's done is done. (Old GOTM habit.)
But if the main goal of this game is to learn a bit & get back into shape, I don't mind if you replay the turns without the Writing trade.
Also, after Writing is finished, we should pause a moment and make a team decision, whether we

• want to play it safe, do Philo right away, pick CoL as freebee and then do Republic by hand or
• take the risk, do CoL first and hope to still make the Philo race and get Republic as freebee.

(As you see in the earlier posts, we already talked about that in the beginning.)
That's a difficult decision and needs a good assessment of our own research speed (lot's of roaded river-tiles and big towns to use them?) and of how far the AI is already advanced at that point. (Of course, even of the AIs we know don't look too threatening, there's always the risk that there is a yet unmet AI with a very good start location -- food, rivers & lux resources -- who will spoil our plan...)

 

[choxorn]

I usually play without reloads, what's done is done. (Old GOTM habit.)

I generally also favor that, though if we did it would sure make the game a lot harder, and it already looks like it won't be that easy.

 

[MrRandomGuy]

Yeah the goal is to get back in shape and learn more about the game. If everyone is ok with it I'll replay the turnset.

 

[Lanzelot]

No objections.
Perhaps it's sufficient to replay just from the point where you traded Writing? (That would be something like the last 7 turns or so?!)
Except for the trade, everything was quite ok.

 

[MrRandomGuy]

Turn Log

Spoiler :

Pre Turn 2150 BC: Worker set to mine tile

Checking our cities, KK is about to revolt. I chose a tile being worked on without a road to make a scientist. This makes Writing to be discovered in 12 turns (down from 14)

Switched a tile in Alma to have the citizen work on the future site of a road. Changed production from scout to worker.

Switched Kazan citizen tile to where the worker is building a road. Note to self: Remember to switch it back when KK is no longer in danger of revolting.

T1 2110 BC

Ta-Tu produces scout. Production set for barracks. Scouts explore.

T2 2070 BC

Worker moved north to finish road to Alma.

Both scouts explore

IBT

Kazan produces worker, production set to warrior.

T3 2030 BC

Worker from Kazan ordered to mine shared bonus grasland with KK.

Scouts explore

IBT

KK produces settler. Production set for temple.

T4 1990 BC

Switch a citizen in Ta-Tu to work the tile directly south of KK since it is no longer used due to the settler.

Settler ordered to move to spot 1.

Scouts explore

T5 1950 BC

Moved a few citizen tiles around in cities.

IBT

Treasury low notification. The first of many.

T6 1910 BC

Road complete from KK to Alma. Worker ordered to mine tile it completed road on.

Mine complete south of Ta-Tu. Worker ordered across the river in order to improve bonus grassland.

Scouts explore

IBT

Barracks completed in Ta-Tu. Production set for archer.

T7 1870 BC

Tabriz founded on spot 1. Production set for curragh.

Scout north explores. Scout near our cities core pops hut and we now know Bronze Working.

IBT

Kazan produces warrior. Production set for worker.

Scout near our cities dies to barb horseman

T8 1830 BC

Created scientist in KK to avoid civil disorder

Warrior moved to Tabriz for MP.

Scout explores

T9 1790 BC

Mine between KK and Kazan finished. Misclicked the worker to the tile west of capitol. It will build a road here next turn.

Scout explores

IBT

Alma builds worker. Production set for warrior.

T10 1750 BC

Worker near KK ordered to build road

Worker Alma built ordered one tile SE to start building a road next turn.

Road southeast of Ta-Tu completed. Worker ordered to mine.

Scout explores

IBT

An Egyptian warrior comes into sight range of our scout.

T11 1725 BC

Lowered science slider to 80%. Writing will be researched next turn.

Scout explores

IBT

Ta-Tu builds archer. Set to produce another archer.

T12 1700 BC

Writing researched.

Attached is the save. I'm on turn 12 of the playset and researched Writing one turn earlier than I did last time. We'll now need to discuss what we should do next.

Something to make note of that might influence the decision, we've now met Egypt but we don't know where their cities are yet. Egypt has every technology America and Korea have, so they've obviously been trading around among the three of them. None of them have writing. Yet.

I'm thinking we play it safe and go for Philosophy right away so we can get CoL for free.

 

[MrRandomGuy]

No objections.
Perhaps it's sufficient to replay just from the point where you traded Writing? (That would be something like the last 7 turns or so?!)
Except for the trade, everything was quite ok.

I would have, except the autosaves didn't go back that far for me. It was best to just restart and mimic some of the things I did.

 

[choxorn]

Why build a Temple in Karakorum and not like, more Archers and Workers, or really, anything else that's more useful?

 

[tjs282]

I would have, except the autosaves didn't go back that far for me. It was best to just restart and mimic some of the things I did.

Just FTR, you can add a line to the config.ini that increases the default number of autosaves.

IIRC, it's simply "Autosaves=XX" — and if you're playing an SG, it's worth setting "XX" to 10, or even 20.

(I also have "WindowsFileBox=1" so that I get a Windows-style... erm... window when I want to save/load, instead of the much simpler default CivIII version).

 

[MrRandomGuy]

Why build a Temple in Karakorum and not like, more Archers and Workers, or really, anything else that's more useful?

I thought the idea was to let the city grow while also producing settlers? Temple will slightly help with citizen moods, especially since we don't have any luxuries anywhere nearby. Was planning on building a settler after temple.

 

[Lanzelot]

Ok, Writing in 1700BC is quite good I think.

Regarding the temple: usually I also don't like temples (not very useful in most cases: you pay one gold in order to save one gold... )
But in this particular situation, it might even be a good idea: Karakorum is already at size 5-6 all the time, while most our other towns are still size 1-2. So making KK happy via the lux slider would indeed waste a few coins in most of our other towns. If the temple helps keeping the lux slider 10% lower than it would otherwise be, it may prove worth the investment.

However, tip for micromanagement:

The temple overruns 4 shields. By setting up KK as follows for the previous 4 turns or so, the temple would still finish at the same time, but we would have earned an extra 4 coins.
(Which may have even made the difference for getting Writing another turn faster, or would at least provide a little more buffer for deficit research and finish our next tech (whatever it may be) a turn faster. 4 gold may not sound much, but at this early stage, it means a 200% increase of our current treasury... )

Philo or CoL next? That's really a tough call. No one has Writing yet, so we have a bit of a head-start. But three demigod AIs who are trading amongst themselves, can research surprisingly fast... So the risk is definitely still high. If I would play this as a solo game, I would probably still take the risk and try for the full slingshot. But in a team-game I can perfectly understand, if none of you are keen on a nerve-wrecking nail-biter...
So if the majority feels more comfortable that way, let's go Philo next.

I haven't played demigod for quite some time, so cannot really remember how often I made the slingshot and how often I missed. @tjs282: you are playing demigod a lot, what does your experience say? And how good is 1700 BC for Writing compared to your games?

 

[Spoonwood]

You won't get The Republic slingshot on Deity or Sid, unless you have a very "cooked" start. I advise against trying to get it, for that reason.

 

[MrRandomGuy]

Yup, I'm bad at micromanagement. Mostly because it's hard to remember to check cities after they grow since the AI is dumb. I'll have to remember to check more often.

Seems like we all agree Philo to CoL is the best choice so I'll finish my turns today or tomorrow.

 

[MrRandomGuy]

Turn Log (beginning from T12)

Spoiler :

T12 1700 BC

Adjust citizen tile for KK as suggested.

Mine south of Alma is complete. Moved worker to KK tile in order to build road next turn.

Scout explores

IBT

KK produces temple and is set to make settler.

Kazan produces worker. Production set for warrior.

T13 1675 BC

Science slider placed at 100%

Road W of KK is complete. Worker sent to finish road to Tabriz

Worker Kazan made sent to one tile north of KK to build a road there.

Scout explores

IBT

A lone spearman comes out of the void.

T14 1650 BC

We've now met the Byzantines. Bad news. They know writing.

Worker near Alma to mine road just completed.

Two citizens in KK made into scientists. Science lowered to 80%. I should have worked more on MM last turn to find the best income/science, but it's an improvement!

Scout explores.

IBT

Curragh built in Tabriz. Production set for another one.

T15 1625 BC

Curragh explores NW.

Scout explores

IBT

Ta-Tu builds archer. Production set for another

Alma produces warrior. Production set for worker

Mine south of Ta-Tu complete. Worker sent to road the tile north of it.

Worker finished road near KK. Will mine next.

T16: 1600 BC

Science back to 100 percent with citizens back to working their tiles in KK.

Sending Alma warrior to Tabriz for MP.

curragh and scout explore

IBT

Kazan produces warrior. Production set for another warrior.

Summerians have built The Colossus.

T17 1575 BC

Worker completes road north of KK and is now mining it.

Worker completed road to Tabris. Is now mining it.

Units explore.

Citizens made back to scientists with slider moved down back to 80% to cover the deficit.

IBT

KK produces settler. Production set for warrior.

Notification about the Koreans building The Oracle.

T18 1550 BC

Science moved to 90%

Settler sent to spot 2 on map.

Units explore.

T19 1525 BC

Units explore

IBT

KK builds warrior. Production set for settler.

Ta-Tu builds Archer. Production set for another archer.

T20 1500 BC

Worker near Alma moved to tile NW of KK where the future city will have a citizen working on it with intent to build road next turn.

Worker near Ta-Tu finishes road. Ordered to mine.

Scientist citizen made in KK.

Notes

As noted in the turn log we've now met another civilization. They know Writing. Egypt now knows writing as well., America and Korea still do not. The next player needs to make a careful decision about whether to hold onto Writing or trade with America and Korea.

It appears we are in a Pangea style map and our little area is a peninsula kind of like Italy.

Don't build any scouts. Build ships instead to explore or use military units due to barbarians. Don't worry if the scout north meets any barbarians. It'll happen sooner or later.


I believe choxorn is up to play next.

 

[Lanzelot]

With two civs now knowing Writing, it's good that we went for Philo straight away!
Let's wait two more turns and then trade Writing to the remaining civs, hopefully we can get Wheel, Iron Working and America's gold.

We have enough units in Ta-Tu. I suggest sending a warrior to explore the south now, and sending a spearman+2*archer stack to bust at least one of those barb camps in the NE. The closer one should be ours, the other one might get busted by the Koreans, before we get there. Another 25 or 50 gold would nicely speed up our research.

Kazan is wasting shields. Should we do another barracks there?

Assuming we get CoL for free, we can do Republic next. It's a whooping 1280 beakers. 1280 / 50 = 25,6 1280 / 40 = 32
At the moment we make 28 beakers, at a deficit. Of course we will grow a bit and the income will increase, but I don't think we will be significantly faster than 35 turns. Should we do a min run and collect a big pile of gold, or go full speed?

• If I didn't know, that we don't have horses, I would do the min run and plan for an eventual Keshik mass-upgrade. (But unfortunately we already know we don't have horses... )
• Also, collecting gold make it likely, that the AIs will try to extort it from us... And we certainly don't want a war while still in Despotism.
• And if we an get Republic, while it's still a monopoly, on a high difficulty level like Demigod it might buy us the rest of the ancient age. Then we only need to get Monotheism and Feudalism from Byzanz and Korea (dreaming a bit here...) research Chivalry ourselves and then ...
• So I would vote for full steam ahead. But that means, the next players need to make barbcamp busting a priority, so that we can cover the deficit all the time. (Though America's gold, if we can get it, should carry us a good while. Perhaps we can also get some gold here or there, by trading Philo and CoL at the appropriate time.)

 

[Lanzelot]

PS: we can increase our income/research from 28bpt to 30bpt, with a little rearrangement like this:

KK still finishes the settler with 5x6=30 shields, but earns 1gpt more. And then Tabriz can work an extra roaded tile:

(Ok, that extra gold is eaten by corruption, too bad.)

In any case, we will probably need that extra coin (plus the income from the new town to be settled next turn) in order to complete Philo in 3, because our treasury will be empty next turn.

BTW: I would not have built that mine E of Tabriz. We will later need to irrigate that tile in order to bring fresh water to our coastal plains! So better would be to improve the other three tiles (tobacco and the other two BGs) first, and by the time these are finished, Republic will not be far, and then we can irrigate the tile E of Tabriz without wasting any worker turns.

 

[MrRandomGuy]

Gah! I did it again! I really just need to put reminders before starting the turn log and make them with every turn so I don't forget until it becomes habit. Check your cities. Check your research.

 

[Spoonwood]

Gah! I did it again! I really just need to put reminders before starting the turn log and make them with every turn so I don't forget until it becomes habit. Check your cities. Check your research.

It's super easy to miss. I can assure that there are many, many HoF games (for sure including mine) where not everything got optimized early on. And I mean even top place HoF games. Probably top XOTM games also. It's almost always not so bad to qualify as make-or-break. Bad use of units or building everything or not having cities to work tiles is more make-or-break.

Also, it's probably not worth it time wise to optimize everything always. I read in the Master of Orion strategy manual to start at one's home planet and cycle through all of one's planets to check on settings every turn a long time ago, and have applied the same principle to games of civ III sometimes. But, I don't always do it, and doing that late game with a huge empire to maximize score or get more production for culture makes for a serious timesink.

It's also one of those things that's much easier to notice when you load a save and try to figure out what's going on. When you know what the previous 5 turns were, and have more of a ready plan on how to proceed it's harder to spot.