Great Scientist - when do you stop planting and start popping?

[beetle]

One game I think it was rocketry, my latest nanotech because I had poorer culture that game.

Rocketry, and really anything sooner, can work because that still leaves GS available for Satellites or whatever else. Ratty finisher for Nanotech though, probably means little use for faith purchased GS, and two of those are easy to get.

The nuclear plant option first gives me the chance to buy one in Apollo city to take a turn or two off when I can start parts.

That is quite sensible and your way may be faster even if it means that Hubble is irrelevant. I tend to prioritize not letting perks go waste, even at the expense of games taking longer. OTOH, if you are starting Apollo later so the Nuclear Plant can be rush purchased first, it would be hard mathematically to make up for those turns. That is in stark contrast to the effect of SS Factories (and Hubble counts as one) which, by halving the SS production time, are efficient enough to delay part building.

The other factor that I will admit to not accounting for is rush buying. I mean, you rush buy what you can. Some things cannot be rush purchased (most critically, Apollo). With money and Freedom, you can skip the SS Factories completely. But, if you assume that you will be using two or more cities to build SS parts, why and how does the ability to rush buy change how you work the tech tree? You still want Rocketry ASAP, and you still want Robotics before building SS parts in your non-Hubble city. That seems to me to lead plenty of room for GS bulbs and for Hubble to be quite significant. What am I missing?

 

[danaphanous]

Well as I said, if I start bulbing around rocketry then my faith+normal scientists are enough to clear all requisite techs and then it's just hard-building while buying as many spaceship factories and nuclear plants as I need to minimize the simultaneous building of parts.

BUT, after hearing your version, I get the feeling if I could have a better policy acquisition rate then I could actually use hubble by getting an engineer through Pioneers, using it on hubble, and starting bulbing 2 techs earlier back in the modern era, or perhaps for plastics. It wouldn't speed up the time between apollo and launching over what I do now, but it would be faster if it works since I'd have used 2 more scientists. I've found the engineer for the last part to be pretty useless personally the way I do it, because I just save my most powerful production city to start the last part and the others are started around the same time with the current way I do things and if I watch the order I buy Sfactories and Nplants all the parts get built in the same 1-2 turns anyway regardless.

I may mix your way and mine and see how it goes. last game I had trouble getting any wonders because of the huge world AI and lost Kremlin so it threw me off a policy, thus nanotech ratty finisher. This still worked though because those last 2 scientists pushed me through to the last part tech on the top of the tree and were useful. I would like to get better at cultural acquisition so as to have all this stuff finished earlier though. I have found saving writers for world's fair to be the most effective since they will get at least 1000 culture and 1000/2 = 500 which is way more turns then I play the game. But in general have found myself bad at keeping friends with all cultural AI since I have such bad midgame gold with the big empires. Also I'm not really building the middle culture buildings much since I don't have many works to put in. I also have found myself not building musicians guild since the spawning of musicians loses me artists and writers. Is this good practice? I just figured the +6 culture and musician works weren't worth the slowdown in writers and artists for a big empire. Also last game no AI spread religious building religions to me so I had no sources of culture besides monuments till archaeology, so that sucked. I've always wanted to get full liberty, commerce, rationalism, and order through spaceflight pioneers but never managed. I think I'd need good religious buildings and cultural CS help and these things have never aligned on Deity, might be a pipe dream without Someone like Polynesia or Poland lol.

Anyway, some of these problems may have been easy to get around if I'd kept my empire smaller but I was trying to push it last game. I was developing around 12 cities I think and was running the line with gold income most of the midgame. It worked out ok with a solid victory but the synergy felt off. you're supposed to have the gold to buy buildings with order haha and I didn't have as much as usual. Was fun to run a bigger empire then usual though!

Thanks for the discussion!

 

[danaphanous]

If I go with combining my ordering and yours here is my thought for a big empire/order optimal launch, I anticipate having 11-12 scientists AFTER the liberty finisher scientist due to 2 faith, 2 hubble, 1 porcelain tower, 1 pioneers, and 5-6 natural: Sx denotes a scientist and its number, I'm gonna assume late scientist power is around (2000 bpt)x8, which should be reachable.

1. Plant Liberty finisher scientist for capital academy, save the rest
2. Open industrialization first, buy/build quick factories for ideology: skyscrapers+happiness
3. Spend S1 to claim early plastics, buy labs in all the best cities using skyscrapers+mercantilism discounts
4. beeline railroads next for Kremlin and production boost, start building Kremlin in the 2nd best city (1st reserve for apollo as it comes before I can finish Kremlin). Put all marginal cities on research for the next 8 turns.
5. beeline apollo next, spend S2 to claim ballistics, wait to pick up radar 1-2 turns later
6. Spend S3 to claim rocketry, overflow into atomic theory, faith-buy 2 scientists. start apollo in best city (should be seven turns pre-plants)
7. time Kremlin to finish and use free policy to claim the rationalism finisher same turn or turn after to get satellites, or if slow advanced ballistics later, start hubble in highest pop city, hurry with engineer from space procurements. This should lower it to 3-4 turns.
8. spend S4 to finish atomic theory and overflow to nuclear fission, immediately send workers to hook up uranium or trade for it.
9. spend S5 to finish nuclear fission and overflow to advanced ballistics
10. spend S6, to finish advanced ballistics and overflow combustion to 1 turn
11. spend S7, to finish combined arms, overflow to computers for 1 turn finish, hubble is finished.
12. spend faith S8, to finish robotics, overflow to nanotech
13. spend S9, faith, to finish nanotech, overflow to penicillin for 1 turn, apollo finishes now, start 3 boosters, cockpit, and stasis chamber, use gold you've saved or borrowed to cash-buy first spaceship factories, then nuclear plants in these cities.
14. spend S10, faith, to finish ecology and half of telecommunications.
15. spend S11, hubble, to finish telecommunications and end mobile tactics on 1 turn.
16. spend S12, hubble, to claim particle physics, start last part 3 turns later then your others, but it should be in your best city that just finished apollo so the timing should be similar to the others. Continue to cash-buy spaceship factories and nuclear plants as needed to ensure all parts finish around the same time.

This is ignoring RA's and you'll probably have a 4-6 turns removed by late-game RA finishes as well which will help with the overflow timing, but hard to reliably factor them in. If you anticipate having 11 rather then 12 Scientists after hubble just don't bulb plastics or start the bulbing at rocketry. Having hubble scientists to work with would enable me to claim early plastics to help with end-game natural scientist spawning and faster science to railroad and kremlin, and also start a tech or two earlier before rocketry. I agree with you now that the only way to get and use hubble if you are going this fast is to hurry it. I've never been able to use it hard-building.

 

[beetle]

That looks really good to me! I am going to try it myself.

P.S. Where I wrote “ballistics” before I meant “Advanced Ballistics,” but I think you picked that up from the context.

 

[psparky]

Earlier in this thread the bulb cap was mentioned. I was vaguely aware of this and always avoided bulbing when few turns where left on the current tech or the next tech was very cheap. However I was inspired to look into this with a view to improving my understanding and hence (hopefully) my play.

I read a number of threads that mention the cap and tried to apply the information to specific cases. What I have found in my tests is that the cap is always dependent on the base cost of the current tech. I didn't find any cases where the cap was related to "five turns of research". On a side note, does that refer to 5 times the current research, the last 5 turns or 5/8 of the bulb value? In any case the overflow I saw was more than any of those.

Here is an example:

Current Science: 674pt
Bulb Value: 4865
Current Tech: 5953/6314
Potential Overflow: 4504
Actual Overflow: 4100

Five turns at current rate would be 3370. 4100 is the base cost of the current tech (Electronics).

Now maybe my tech rate was low for where I am in the tech tree, but it appears that often you can bulb without being capped with much less than 5 turns left in the current tech. In the above case I would have needed only about 1.14 turns left in the current tech to bulb without cap.

 

[lindsay40k]

I'll still plant them shortly after Scientific Theory if they'll be getting boosts from Freedom, Historical Landmarks, or being Korea. If I'm going wide, I'll start saving them a little sooner, as their bulbs will have higher yields and the NC city's total output is not quite so important.

 

[Matthew.]

Anything not an outlier (i.e., Babylon's Writing GS), then math usually comes down to plant 2, save the rest. However, timing is also something to consider. For example I play a lot of warmonger games, and getting to artillery at a competitive turn is more important than saving them for later.

 

[Acken]

@OP : 1 or 2 planted rest is bulbed. You can safely always bulb plastics if you're ready to build/buy labs.

I pop em 8 turns after building research labs in my main science city/cities.
I'm not 100% this is completely optimal

The answer is given to you after the bulbs. If you end up waiting for techs after you bulbed then you bulbed too soon. If on the contrary you end up with too many you could have bulbed earlier or earlier in the game for key techs. It's often the case that a wide game should bulb earlier techs because in the end your scientist are way stronger than a 4 city tradition game needing less GS for the last 20 turns.

Current Science: 674pt
Bulb Value: 4865
Current Tech: 5953/6314
Potential Overflow: 4504
Actual Overflow: 4100

Five turns at current rate would be 3370. 4100 is the base cost of the current tech (Electronics).

Now maybe my tech rate was low for where I am in the tech tree, but it appears that often you can bulb without being capped with much less than 5 turns left in the current tech. In the above case I would have needed only about 1.14 turns left in the current tech to bulb without cap.

It is the max of current tech or 5 turns iirc. Try to redo your example with a very cheap tech.

 

[psparky]

It is the max of current tech or 5 turns iirc. Try to redo your example with a very cheap tech.

Ah, from how it was worded in other posts I assumed it was supposed to be the minimum of those values. I'll have to find another example to demonstrate it to myself, but I'm sure you will be proved right. Thanks!

Still, I have learnt that I can bulb without loss in many more situations than I previously thought.