SGOTM 13 - One Short Straw

[Dhoomstriker]

One thing that seems to be missing is the turn that you switch from building the Lighthouse to a different build item while we wait for Writing to be 1 turn away. Presumably, you are building a Work Boat, so I'll just make that assumption unless you contradict me

 

[Dhoomstriker]

Also, am I correct in assuming that when you 1-pop-whip the Lighthouse that you will work the Fish + the Clam?


A similar question applies for the Library if whipping it brings us down to City Size 2 (I haven't checked yet). Obviously, I'll know for certain which squares to work if we whip the Library from Size 5 to Size 3.

 

[LowtherCastle]

No, monument. And it should be 2t from completion, or else I think there's some hammers lost to coins.

xpost

 

[LowtherCastle]

Also, am I correct in assuming that when you 1-pop-whip the Lighthouse that you will work the Fish + the Clam?

A similar question applies for the Library if whipping it brings us down to City Size 2 (I haven't checked yet). Obviously, I'll know for certain which squares to work if we whip the Library from Size 5 to Size 3.

Can't remember on the first question. Haven't optimzed the second. I tried waiting for pop5, which seemed okay, but got the GS on T140.

 

[Dhoomstriker]

Okay, so please clarify for me: are we whipping the Lighthouse into the Library or are we manually-completing the Lighthouse? I don't see any indication in the PPP that we're supposed to whip the Lighthouse.

If we are whipping the Lighthouse, on which turn do we whip it?

 

[LowtherCastle]

Yes, we pre-build the lh to something like 78/90h, then build the monument till 1t to the completion of writing and 1pop the lh.

 

[Dhoomstriker]

So, I see that you have a Warrior being built after we whip the Library.

Does that mean that:
a) I should not be sending Warrior 2 over from Paris and you'll have to suffer with any Unhappiness that results from not having Warrior 2 from Paris?
OR
b) You're building an extra Warrior for the fun of it?
OR
c) I should just remove that build item from the test, since we won't need it due to shipping-over a Warrior from Paris?

 

[LowtherCastle]

That warrior is left over from the previous PPP. I only updated what's inside the box. Gold builds the monument, then its own wb, then the forge, but that's beyond my turnset anyway.

 

[Dhoomstriker]

I tried waiting for pop5, which seemed okay, but got the GS on T140.

Pop-rushing the Library at Size 5 compared to 1 turn sooner at Size 4 just before growing to Size 5:

They both end up with equal Food and equal Hammers on Turn 111, 1225 BC. Waiting until Size 5 nets us 4 additional Commerce. Whipping at Size 4 nets us 1 additional turn of working a Library in a City that will likely be making 15 Commerce (that's 15 with a Foreign Intercontinental Trade Route; 14 without it).

Unfortunately, floor [15 Commerce * 0.25] = 3.

So, waiting until Size 5 to whip the Library gives us +1 Commerce over whipping the Library when we are at Size 4 and are about to grow to Size 5.

Since both approaches will give us an identically-sized Foodbox on Turn 111 while we're still at City Size 3, neither approach lets us hire a Scientist sooner than the other, so your approach of whipping after growing to Size 5 is slightly better. I'll change my version of the PPP to use your approach.

 

[Dhoomstriker]

Alright, here are some numbers as of Turn 112, 1200 BC:
Your approach has 24/42 Food (growth in 4 turns), 38/45 H in the Monument, and has earned 208 raw Commerce from citizens

My approach has 21/42 Food (growth in 5 turns), 41/45 H in the Monument, and has earned 225 raw Commerce from citizens

Note that the Commerce values ignore Trade Routes and the City Centre square, since those values will be constant. I also did not factor in the Libraries up until this point, since we got the Libraries at identical times. Also note that I will value a Scientist's Flasks as being equivalent to raw Commerce, but we haven't yet included any Scientist Flasks in the results since the 1st Scientists were only just hired on this turn.

Now, I could keep playing it out, but let's just do some estimates for now instead.

Let's say that your extra 3 Food allows you to work a Scientist 1 turn sooner, giving you +3 raw Commerce.

Let's also say that my later 1 turn of Unhappiness gives you +3 Food and +2 Commerce.

Then, let's say that we don't solve our Happiness problem from whipping unhappiness and that you get 3 additional turns worth of no-whipping-unhappiness (since you whipped 3 turns earlier than I did). That will give you +6 Food and + 6 Commerce.

Now, I'm going to be really generous here and will give you the full (unrealistic) + 25% on this additional Commerce that you earned (likely you won't quite get +25% due to rounding, but whatever).

That's 3 + 2 + 6 = 11 *.25 (I'm going to floor it now, though) = 2

So, in addition to the "to date" totals, I'll give you 3 + 6 = 9 Food and 11 + 2 = 13 Commerce

So, that makes the expected final results to be:
You have 12 more Food than me
I have 3 more Hammers than you
I have 4 more Commerce than you

If we do manage to solve our Happiness problem, though, by getting Confucianism in Gold City, then you wouldn't get a credit of 3 turns of less whipping unhappiness, so the final numbers in that case would be:
You have 6 more Food than me
I have 3 more Hammers than you
I have 11 more Commerce than you


So, based on a totally random factor of where a Religion happens to be founded (although you yourself claim that Gold has a higher random chance to get Confucianism), the vote sways in from being in favour of one approach to the other.


Okay, let's try something else: let's see how many Flasks we will get pre-Writing. As of Turn 102, 1450 BC, counting only the raw Commerce from citizens (i.e. the Commerce from working any of the Fish, Clam, and Gold):
You will have 99 raw Commerce and I will have 114 raw Commerce. Will 15 additional raw Commerce net us Writing by this turn? I'm not sure.

Note that we're both whipping the Lighthouse on Turn 102, 1200 BC.

As of Turn 101, 1475 BC, you have 28 Food in your Foodbox to my 25 Food. However, we will both regrow to Size 3 at the same time after whipping the Lighthouse, we will both grow to Size 4 at the same time, we will both grow to Size 5 at the same time, and we will both regrow to Size 4 at the same time. So, it's only when we grow to Size 5 next that this Food helps you, which is why you got credit for the first comparison for having the Scientist Specialist for one extra turn.

As of Turn 101, 1475 BC, I have 3 more base Hammers than you do (I halved the Lighthouse Hammers for both of us to be fair). However, we will both build the Lighthouse at the same time as each other and will built the Library at the same time as each other. I will complete the Monument 1 turn before you will, but it's not like doing so really helps any (whoopdie doo, 1 extra Culture).

It could be that later, I will complete a Work Boat 1 turn sooner than you or something, so if we wanted, we could assign +2 Food to my approach for whatever City uses that Work Boat... it's a reasonable assumption since Gold City only makes 4 Hammers per turn, whereas 3 Hammers is almost a full turn worth of production. Note that this potential +2 Food has not been factored into the comparisons above, since it's still not a guarantee (just a very likely possibility).


Okay, so there are your numbers. Happy?

 

[Dhoomstriker]

Okay, I put three worksheets in the spreadsheet... the first one having both LC's and my latest approaches. The second one has LC's latest plus the comparison to the old one that he'd uploaded, along with some of the numbers that I'd been collecting for the Gold City comparison test. The third one has a slightly-outdated version of my latest spreadsheet with a ton of colours that contrast it with LC's original PPP, as well as numbers that I'd been collecting for the Gold Citiy comparison test.

Probably, the only useful worksheet is the first one.

For some reason, the converted .ods file was about half of the size, so I wouldn't be surprised if an entire worksheet or two got dropped or something like that. Maybe Open Office just uses an efficient file storage algorithm, maybe MS Excel stored a bunch of temporary changes within the file that got stripped, or maybe some data was lost.

On the first worksheet, I gave a few more turns' worth of Science Slider values. When we're about 1 turn away from learning Masonry, we'll probably want to switch to Meditation, but I'll leave those calculations until around that time, in case we've got an AI that knows Meditation by that time.

 

[LowtherCastle]

Dhoom, I thnk it makes more sense to research Meditation before Masonry, because AIs are much more likely to reserach Masonry and possibly give us a n extra beaker or two. DO you have a optimaztion for that? Or shoud I just go binary?

 

[mdy]

I'm a little concerned that spending all this time worrying about such small details will be counter productive as it will force us to rush the later stages-and the sub-optimal move this will result in will probably far exceed the loss of a couple of beakers at this part of the game.

I think we should stick to the 1375BC plan. The risk of losing the Oracle hasn't chnaged, and I don't think getting a workboat a couple of turns earlier will make up for the other delays.

 

[LowtherCastle]

mdy, I'm ready to play now. Just waiting to see if Dhoom has an optimization for the research slider. I agree though, that we need to get this going. I'm more concerned that team members will lose all interest in the game. On the bright side, though we're the last to play, others seem to be dragging their heels too and I think we're fairly set all the way through to ~500BC, so hopefully things will pick up.

THen our warmongers can take over...

 

[Dhoomstriker]

It is one tight, tight, tight set of Science Sliders. Without any bonus Flasks from the AIs, we should be able to make it to all 3 of Med, Priest, and Masonry learned on Turn 96, 1600 BC, but it's really close.

I would appreciate it if you played it out once in a test game first to confirm my numbers (using the Test game from Turn 79 that I uploaded previously should work, but you'll probably have to remove Masonry from Vicky whenever you want to research that tech, since she keeps trying to re-research it). Note that it only takes her 1 turn to relearn it, since the World Builder only removes the "last turn" worth of research for a tech. So, every time that I remove Masonry from her, I also remove Mining and Mysticism, which buys me 2 turns before I have to do it again (more turn if she chooses to research something else).

I only went as far as the start of Writing, but the Science Slider values SEEMED to work for me. Then again, I reloaded so many times that it would be nice that you double-checked that the numbers worked out.

I only optimized for my build order... if you want me to do it for a different build order, let me know, since the numbers will change if you whip on a different date in Gold City.

 

[Dhoomstriker]

After this point (i.e. starting with Writing), it'll probably be mostly or all Binary Science anyway, so as long as these numbers work out, hopefully this Science Slider optimization stuff will be over and done with.

 

[Dhoomstriker]

I'm a little concerned that spending all this time worrying about such small details

Then the team needs to take a more realistic approach to our goals. Saying: "I don't really want to compromise on our early Oracle date" translates into "we had better make sure that we have researched all of Meditation, Priesthood, and Masonry by Turn 96, 1600 BC."

It's all well and good to get there with a test game with a less-than-solid whipping approach (delayed whipping usually translates to more citizens working more Commerce squares, meaning more Flasks), as well as a test game where the AIs know more of the techs that we want to research and thus give us additional bonus Flasks, but when the team chooses a plan with good whipping timings and when the real game reflects a situation with backwards AIs, then it becomes a real pain to try and meet the desired goals.

I'm pretty sure that we got it now, but like I said, the timing was very tight and several run-throughs had me coming up anywhere from 80 to 8 Flasks short, with just "that much" too much excess Gold. As it is, if I calculated the numbers correctly, we will have exactly 1 Gold in our treasury when we start researching Writing, and we're only going to have a small amount of overflow Flasks (less than 10 Flasks).


I'm not complaining, because I enjoyed some of the optimization work, nor am I digging for compliments or praise, but if we don't want to go to this level of detail (I'd rather do far less micromanagement for future turnsets as the game goes from micro-mode to macro-mode anyway), then we have to allow for more realistic goals. We can't have it both ways; either it's optimized to the hilt and we meet really aggressive goals, or else we reduce the intensity of the micromanagement by allowing for much slacker goals (such as "Oracle by 1200 BC would be fine, and if we miss building it, no big deal" and "Great Scientist by 300 BC would okay, too, even if we risk missing Taoism and Confucianism").

As it stands, people feel a bit uncomfortable with delaying The Oracle for even a turn or two, while also wanting our Cities to grow a lot (i.e. having good whipping timings for our Granaries). I think that we've got a plan that does all of that, but go easy on LC if things don't work out perfectly, because even a small change in the Science Slider values could easily mean delaying The Oracle by a turn or two, while even the best of plans rarely survive the "real world test."

 

[LowtherCastle]

Okay, it works as you predicted, with 1g on T96. You then show 100% for writing, but I assume that's some carryover from the previous PPP. I'll go ahead and start playing. If you want to give me a heads up on the writing slider %, I'll check back in. Otherwise, we'll cross that bridge when we get to it.

 

[LowtherCastle]

I'm playing on. Planning to send our western scout southward, assuming that Cathy to the north if we don't find her in the south. Turnset write-up is here.

 

[Dhoomstriker]

Okay, it works as you predicted, with 1g on T96. You then show 100% for writing, but I assume that's some carryover from the previous PPP.

Yeah, I didn't optimize anything for Writing onwards... I coloured the cells differently, but who knows if the colouring survived the file-conversion operation.

Basically, I imagine that you'll set the Science Slider to 0% when you start work on Writing and then will just use Binary Science after that point.

I don't see us coming into many future opportunities where non-Binary-Science will even work.

Otherwise, we'll cross that bridge when we get to it.

If you want, you can always stop play at Writing and we can then update a test game to match the real game. The only thing that might matter from then on for the Science Slider would be "which turns" should be 100% and "which turns" should be 0%... i.e. if you make 35 Commerce on one turn and 38 Commerce on another turn, then you could choose to take either 35 Gold plus 1 free Flask or 38 Commerce plus 1 free Flask... and then it becomes a more macro-level version of micromanagement... but only if we care about a date of researching a tech.

We probably care enough about Writing's date, but let's wait until you are about to start researching Writing before crossing that bridge, as we may very well have picked up a few bonus Flasks along the way that could change any results that I'd be able to get with a test game now. So, I don't plan to provide any Science Slider values for Writing and beyond, at least not until we get an updated test game.