SGOTM 13 - One Short Straw

[shyuhe]

I am amazed at the level of detail in this set (sorry this is all way over my head)

 

[Dhoomstriker]

The lh completed T78 and then what I did was build the settler for a turn, so I'm still at pop5, but the settler has 17h. To match your plan, I would have to build the wb for a turn.

Ah, okay, that works. There's no functional difference, since we don't actually pay +1 Maintenance when Paris is at City Size 6 when we only have these Cities. Later on in the turnset, we WILL pay such +1 Maintenance, which is why we pre-built Settler 4 at City Size 5 in my approach.

BOLDED TEXT EDITED IN: ANYWAY, yes, if you have some reason not to whip Settler 3 ASAP, then you can't play forwards past Turn 80 until you make your decision.

Here's the message where I deemed it being worth whipping Settler 3 after only 1 turn of investing Hammers into it, message #916.

I'm roughly following Scenario D), except that I will start on Settler 4 a turn earlier (while still being at City Size 5, just 2 turns before growing to Size 6 instead of 1 turn as per Scenario D)). I do so in order to capture the Hammers from the Work Boat into Settler 4, which, at the time, I thought would help us. MAYBE it will be better just to get those Hammers started into a Monument, which MIGHT allow us to get Warrior 2 out without needing to use a Mine or if we still use the Mine, might allow us to get Warrior 2 sooner... I'll have to look into that possibility while making sure that neither Settler 4 nor Worker 2 would get delayed as a result of not catching those overflow Hammers (approximately 4 Hammers, if I recall correctly). It could be that those roughly 4 Hammers have no difference what-so-ever on being captured by Settler 4 versus being captured by our Monument.

 

[Dhoomstriker]

I guess it doesn't matter if I build the wb for 1t now or in a turn or two, does it?

In which case I can build it now and play forward 1t, right?

Since there are no additional Maintenance Costs incurred at City Size 6 in Paris, then yes, you are correct, you can play forwards 1 turn by working the Work Boat and growing to City Size 6, with a Science Slider of 90% for Mysticism.

You would still have the flexibility of delaying the whipping of Settler 3, should you choose to do so. The calcs that I just linked, though, show that a Turn 80 whip for Settler 3 wins out, but you don't have to decide to agree with the numbers just yet if you only play forwards to Turn 80 (and preferably, leave yourself 1 unit still "active," such as the Warrior in Paris, so that you don't accidentally advance the turn).

 

[LowtherCastle]

Okay. None of them know:
masonry
meditation
polytheism
writing
iron working
metal casting

We can see the edges of VIcky's culture now and we may be able to skirt it down the east side of the snakey island.

--------------

Edit: I suppose it's more likely they're researching masonry than meditation, so we might as well go for meditation first.

 

[Dhoomstriker]

As far as I understand things, here's an updated Test Game as of Turn 79, 2025 BC.

Note that I only had us "meet" 2 AIs, since we only care about HOW MANY AIs know a tech, not WHICH AIs know a tech.

These 2 AIs know Mysticism but do not know Masonry and do not know Meditation. I also took away Mining from both of them, to discourage them from going for Masonry.

The more AIs that we "meet" in the test game, the harder it will be to keep track of who has accidentally researched more techs than the AIs in the real game have researched, which is why I limited contact to 2 AIs.

 

[LowtherCastle]

Okay, Dhoom, I'm not sure there's a huge difference between your new plan and what I've been doing, as far as GOld's concerned. However, you're getting the Marble wb out significantly later, which is a problem. I guess what I should do is update my PPP so you can compare them and understand what I'm doing. I don't seem to be making any progress in comparing the two.

 

[Dhoomstriker]

So, while I'm playing things through, updating recommended Science Slider values and adding-in my latest Gold City approach, you can try to explain wh you think that it is a good idea to delay whipping Settler 3. I showed numerically in message #916 (the message that I linked a couple of messages back) why whipping on Turn 80 is best, but perhaps you have a really strong reason for trying to get extra overflow Hammers or something that you can justify.

 

[Dhoomstriker]

I guess what I should do is update my PPP so you can compare them and understand what I'm doing.

Yes, please!

However, you're getting the Marble wb out significantly later, which is a problem.

I suppose that getting this Work Boat out sooner could be one reason to justify trying to additonal whip-overflow Hammers from Settler 3. The cost is less Hammers in Marble City, which has the potential to impact The Oracle's date.

It could be that if you whip Settler 3 on Turn 81 that The Oracle's date will not be affected, but it might be.

However, The Oracle's date is almost surely going to be affected if you whip on Turn 82 for maximum Hammer overflow.

 

[Dhoomstriker]

Team Member Feedback Requested
Okay, on Turn 104, 1400 BC, we are 1 turn away from completing The Oracle, with a Forest Chop going into it.

Counting this turn's production and the Forest Chop, we will have 230/225 Hammers.

We only have 5 extra Hammers.

Delaying Marble City by 1 turn costs us 3 base Hammers, which translates into 6 Hammers in The Oracle.

So, if we whip Settler 3 on Turn 81 instead of Turn 80, we will settle Marble City 1 turn later and will get 1 less turn of 6 Hammers going into The Oracle. That means that at the end of Turn 104, 1400 BC (i.e. the start of Turn 105 from our perspective), we will have 224/225 Hammers invested into The Oracle. Since our Forest will have been Chopped, we will make 2 base Hammers, which gets us 4 Hammers going into The Oracle or 228/225 Hammers by the end of Turn 105, 1375. So, we would delay The Oracle by 1 turn.

If we instead whip Settler 3 on Turn 82 instead of on Turn 80, then we will have 6 less base Hammers going into The Oracle, meaning 12 less Marble-enhanced Hammers in The Oracle, so at the end of Turn 104, 1400 BC (the start of Turn 105 from our perspective), we would have 218/225 Hammers invested in The Oracle. Unless we "goof around" and lose a Worker turn (build an unneeded Road over top of the Forest?), we will have lost the Forest and will only make 2 base Hammers = 4 Hammers per turn going into The Oracle, meaning that on Turn 105 we will have 218 Hammers with 4 going in by the end of the turn. On Turn 106, we will have 222 with 4 going in by the end of the turn, giving us 226/225 Hammers. So, we will complete The Oracle at the end of Turn 106, allowing us to pick our tech on Turn 107, 1325 BC, which is a delay of 2 turns on completing The Oracle.


Therefore, each turn that we delay whipping Settler 3 in Paris translates to a 1 turn delay on The Oracle.

So, instead of completing The Oracle such that we can pick our tech on Turn 105, 1375 BC (which appears as 1400 BC in the Event Log), we would end up picking our tech one or two turns later.

None of the AIs that we know apparently know either Meditation or Polytheism, so this date could still be a really safe date, but I wouldn't mind hearing if people are willing to accept this delay if it means allowing us more flexibility in terms of Hammers going into Paris. Those excess Hammers will translate into an earlier Work Boat #1 for Marble City, allowing us to grow Marble City faster and to make more Commerce off of Marble City later when it grows to Size 2 faster and can work the 2nd Clam sooner.

However, the tradeoff is a later Oracle date.


At this point, my vote would be to be willing to delay The Oracle for up to 2 turns by delaying the 3-pop-whip of Settler 3 by up to 2 turns (obviously only using 1 turn if 1 turn gets us the Work Boat fast enough but I'm thinking right now, without having tested it, that we'll ideally want both of those 2 turns in order to speed up Marble City's Work Boat as much as possible), so that we can get Marble City growing as soon as possible.

 

[Mitchum]

Based on the fact that the AI we know don't know Mediation or Polytheism, I "think" our Oracle date is quite safe. As long as delaying MC doesn't mess up any of our forge build plans, I'm okay with delaying the Oracle a turn or two if required.

 

[LowtherCastle]

Here is an abbreviated update of my PPP (outlined in black). I 3pop the settler on T81 but this does NOT slow down the ORacle, because on the last turn, I work a citizen instead of the coastal tile.

We can get the Marble wb out 1 or 2t late, but better is probably 2t late overall.

Also note that I left my GOld MM as it currently is. TO me, that's the main debate here.

 

[LowtherCastle]

Btw, after set4-wkr2-warrior, Paris just builds monument-wb normally. To get the wb out a turn sooner, we would have to hire a citizen for 2t instead of worknig a coast at pop6. Probably not worth it.

Also, I didn't do set4 at pop5 as you suggested, but I see how that would be better. Gold doesn't need the wb that soon anyway, at least not in my plan, so that's fine.

 

[Dhoomstriker]

Based on the fact that the AI we know don't know Mediation or Polytheism, I "think" our Oracle date is quite safe. As long as delaying MC doesn't mess up any of our forge build plans, I'm okay with delaying the Oracle a turn or two if required.

As you have implied, it's not a straight-forward decision. I played things out (only paying attention to Paris' build items).

Whipping Settler 3 on Turn 81 instead of on turn 80 gets us 17 extra Hammers floating about. Since Paris makes 8 Hammers per turn around the time that we are completing Marble City's Work Boat for Clam #1, it's not hard to believe that we'd save 2 turns on getting that Work Boat out. And that's exactly what happens.

However, there are a number of side-effects.

Settler 4 aka Stone City comes out 1 turn later.

Worker 2 comes out 1 turn later.

Warrior 2 does come 1 turn earlier, which means that Worker 1 will just partially-build a Workshop for 2 turns near Stone City instead of partially-build a Workshop for 1 turn near Marble City and partially-build a Workshop for 1 turn near Stone City (no real functional gain here).

While we don't lose out on starting on the Forge, since we were previously spending 2 turns building the Forge while growing to Size 8 and only really need to spend 1 turn building the Forge before it can be 4-pop-whipped, what happens is that we delay growth to Size 8 by 1 turn. Therefore, the Forge DOES get delayed by 1 turn, since we can't whip it until 1 turn later.


So, all of those delays are in addition to a 1 turn delay of The Oracle. EDIT: Unless we work a citizen as LC has suggested, in exchange for 2 Food and 2 Commerce.

In exchange, we get the Work Boat for Marble City's Clam #1 2 turns sooner, which equates to a gain of 6 Food in Marble City, plus whatever future extra Food or Commerce we can gain by growing faster (growing a bit faster lets us work the 2nd unnetted Clam, giving us +1 Food and +2 Commerce per turn).


Still, it's hard to justify all of those things being delayed.

I'm not sure what to do about it, though, other than my previously-made comment of making Marble City "suck it up."

 

[babybluepants]

@Dhoom,

TBH, I haven't really followed the MM debate this past week. It's very detailed, and I don't have time to even try and understand it, let alone test. My last test, ages ago, was 1325 BC (1350 BC in log) and that seemed good enough at the time. Still, I prefer a better date, if we can. Generally speaking, if going for Oracle, earlier = better. Losing it by a turn sucks a lot worse than a loss of a few WB turns. These 3 not having Medi/Poly is a bit irrelevant - we're worried about the religion founders, no?

 

[shyuhe]

I'm on the fence about delaying the Oracle for a faster WB. We are playing with a relatively late date, although the AIs are admittedly dumb as bricks on our map right now. So I'd like to err on the side of an earlier Oracle so that we don't end up in a situation where we just miss out on the Oracle. Marble fail gold isn't bad, but we obviously want the tech more.

 

[LowtherCastle]

Okay, Dhoom, I can live with letting Marble suck it up. COUld you run one of your patented tests comparing your latest version of Gold with my version, in which we 1pop the granary in GOld at 32/36f->32/33f?

 

[Dhoomstriker]

Btw, after set4-wkr2-warrior, Paris just builds monument-wb normally. To get the wb out a turn sooner, we would have to hire a citizen for 2t instead of worknig a coast at pop6. Probably not worth it.

On what turn are you completing the Work Boat that gets built after the Monument?

In my "Turn 80 whip" approach, it arrives on Turn 109, 1275 BC and we get more Hammers flowing into The Pyramids since part of the Work Boat was Forge-enhanced. If we really wanted, we could complete the Work Boat on Turn 108, 1300 BC, without getting the Forge's bonus on its Hammers, but we'd also have to spend 1 turn at Size 8 with 1 Unhappiness and 1 Unhealthiness so that we could invest more than 0 Hammers into the Forge before we whipped the Forge. I kind of think that waiting until Turn 109 works out better with this approach, since Paris then doesn't have to suffer due just to get +3 Food in Marble City.

When I played out the "Turn 81 whip" approach (which may have been played out differently than your "Turn 81 whip" approach), the Work Boat arrives before the Forge on Turn 106, 1350 BC, which is either 2 or 3 turns in advance, depending upon whether or not you want the "more efficient Paris" method or the "get that Work Boat out as quick as you can" method for the "Turn 80 whip" approach.

Also, I didn't do set4 at pop5 as you suggested, but I see how that would be better.

Yeah, there's no point growing to Size 6 immediately when we don't have the Happiness to work a Coast, as we will pay 1 Gold in Maintenance. However, at least with the "Turn 80 whip" approach, it is worth it to grow to Size 6 once the Unhappiness is about to wear off, which gives us two turns of Working Worker 2 at City Size 6 with the extra Coast.


Contrasting my approach with yours, you spend 1 more turn building Settler 3, while I spend 1 more turn building Worker 2. One difference is that when I'm building Worker 2, I get to work that Coast square, while you have an Unhappy person when you're building Settler 3.

Another difference is that your later Settler 3 costs us 3 Commerce, as well as 3 Hammers in Marble City, as per message#916.

The other difference is the amount of turns that things are different:
- Settler 4 coming 1 turn later
- Worker 2 coming 1 turn later
- Paris' Forge coming 1 turn later
- 3 turns saved on the Marble City Work Boat for Clam #1 (or 2 turns saved if you don't want to be efficient by building the Forge prior to that Work Boat).

Gold doesn't need the wb that soon anyway, at least not in my plan, so that's fine.

No, it doesn't need the Clam netted as soon, but we also lose out on Commerce from not working the Gold square as much, as well as working the Clam less.

I guess that I'll be on the hook for comparing the two approaches to each other pretty soon.

Okay, I guess I have to get back to updating my version of the updated PPP so that we can more directly compare our PPPs to each other.


EDIT:

Okay, Dhoom, I can live with letting Marble suck it up. COUld you run one of your patented tests comparing your latest version of Gold with my version, in which we 1pop the granary in GOld at 32/36f->32/33f?

Hahaha, I knew it! Well, I have very little excuse not to do so now, since you produced your PPP. Okay, give me some time (likely, you'll go to bed before I'm done).

 

[LowtherCastle]

When I say compare, I'm thinking way more long term and I'm not talking about comparing T80 set3 versus T81 set3. I'm interested in knowing how the earlier granary 1pop compares to the grow-to-pop3-then-1pop granary. Assume that the initial stuff all follows your plan (T80 3pop set3, etc.) Only compare when to 1pop the granary in GOld CIty. (Sry, I didn't make that clear before.)

 

[Dhoomstriker]

When I say compare, I'm thinking way more long term and I'm not talking about comparing T80 set3 versus T81 set3. I'm interested in knowing how the earlier granary 1pop compares to the grow-to-pop3-then-1pop granary. Assume that the initial stuff all follows your plan (T80 3pop set3, etc.) Only compare when to 1pop the granary in GOld CIty. (Sry, I didn't make that clear before.)

That's what I understood the request to be. Note, however, that the testing would be in isolation: I'd only look at Gold City. I wouldn't worry about when to whip in Paris because I wouldn't follow the entire PPP. It's the only "sane way" of doing my "patented testing," otherwise you end up repeating a ton of actions that might end up getting thrown away or are already relatively-decided upon.

As I alluded to previously, you'll know the relative raw Commerce difference but you won't know how many turns this extra Commerce can save you on techs. Still, this method of isolated comparison has worked for us up until now.

Since different levels of Unhappiness are involved, I'd have to compile the results of both approaches until after the Unhappiness wears off, which is why I said that it would take a little while to do.

 

[Dhoomstriker]

When I say compare, I'm thinking way more long term and I'm not talking about comparing T80 set3 versus T81 set3. I'm interested in knowing how the earlier granary 1pop compares to the grow-to-pop3-then-1pop granary. Assume that the initial stuff all follows your plan (T80 3pop set3, etc.) Only compare when to 1pop the granary in GOld CIty. (Sry, I didn't make that clear before.)

Nice work on whipping the Granary on Turn 92, 1700 BC, one turn after the soonest turn that we can possibly whip the Granary, in order to manage to get +3 Food overflowing.

It also looks like you have tried to emphasize working the Gold square as much as possible pre-Writing.

That's as far as I have gotten so far (I told you that it would take a while), since I'm recording more details that I probably will have to do record, but so that if I do end up wanting those details later, I won't have to repeat the exercise.