I'm not certain what the calculation is, but that site is no longer accurate. It was changed at some point between release and now. I believe it changed with BNW.
Now it factors in the other civ's science.
Evaluating my peaceful SV progress
- Thread starter trixcit
- Start date
Now it factors in the other civ's science.
Gold rushing a library for NC or one university for Oxford makes perfect sense to me. But I understand people to be arguing that from about mid game on, one should be buying science buildings at every opportunity. That would leave very little left over gold for RAs! Strong cities only take a handful of turns for buildings, so a gold rush is maybe 10 turns of a 2 extra beakers from one city. How is that cost effective, when that same gold could pay for 2 RAs? If 500 gold nets you 30 turns of CS ally status, and you have unlocked the half-science tenant, that seems like it would be pretty cost effective too! I would love to hear from folks good at crunching the numbers!
I think its more like 3-4 turns of science per RA. The payback for how long it takes is kind of poor, but you can be running multiple RAs at the same time. That 1000 gold isnt observatory or not, it means observatory sooner. So for a 1000 gold you get +50% science from one city for about 10 turns. Not bad, but if that 1000 gold could be 3 RAs, thats 6-12 turns of science from your whole empire. This is just a back-of-envelope type calculation, but I dont see how the rush buying comes close. Buying science buildings has it place to be sure, but as a general rule? I remain skeptical.
My last game, I was earning about 320-350 BPT on turn 152. On turn 153 I open up Scientific Theory(?), bought 3 public schools and had 530 BPT. That is more than 2 extra beakers. On my primary city, I might have made one in 10 turns, but not on my 2ndary cities. But let's figure it was worth ~70 BPT per public school for 10 turns + bit more for those 2ndary cities which would have spent 15-20 turns.
Note: I do not buy all my libraries. I'll buy 1 at most. I try to buy universities and on. At the very least, I buy it in my primary science city, as it often nets more than half my total science.
I've been trying to do this because all the fastest science victories I see, are done from people who do this. If the best players do this, and are successful, it has to be pretty solid. Though they may also use RA's.
Three public schools makes a huge difference! But let’s net up the beakers:
200 bpt x 15 turns = 3000 beakers.
But what if that gold had been spent on RAs? Is 6 RAs at 3 turns per RA okay? To get that many, some of them were early, so how about figuring they mature when your empire is averaging only 300 bpt.
300 bpt x 3 turns per RA x 6 RAs = 5400 beakers.
300 bpt x 2.5 x 6 = 4500 beakers (using Browd's example)
200 bpt x 15 turns = 3000 beakers.
But what if that gold had been spent on RAs? Is 6 RAs at 3 turns per RA okay? To get that many, some of them were early, so how about figuring they mature when your empire is averaging only 300 bpt.
300 bpt x 3 turns per RA x 6 RAs = 5400 beakers.
300 bpt x 2.5 x 6 = 4500 beakers (using Browd's example)
You get 3 turns of science from an RA? It's more like 2 in my experience and good luck getting 6 friendships. I honestly am fuzzy on how RA's work now. They are not based on your BPT, but the person you sign with, or an average of you and theirs.
It's quite possible that the best is to buy a mix. Perhaps get the 2 RA's you are capable of (fat chance getting 6) and getting 2 universities in your strongest cities.
The one thing that has me wondering is you admit to much slower victories than I get and I get a little slower victories than those who buy all their universities and public schools. They may also get some RA's, but the point is, those who are more successful buy their science buildings.
In BNW, the base output of an RA (before Porcelain Tower and Scientific Revolution) will equal 5 turns of the average beaker generation rate of the RA partner with the lower beaker generation rate over the 30 turns of the RA.
BNW research agreement math:
RA beakers = n% * min(a, b)/3,
where
a = sum of your beakers over RA length
b = sum of your RA partner's beakers over RA length
n% = RA modifier % (base 50% + 25% from Scientific Revolution + 25% from Porcelain Tower)
For example, if a = 6000 beakers (average of 200 beakers/turn) and b = 3000 beakers (average of 100 beakers/turn), the base RA output (without Porcelain Tower or Scientific Revolution) for both RA partners will be 500 beakers, which represents 5 turns of average beaker production for your RA partner but only 2-1/2 turns of average beaker production for you. Accordingly, in BNW you have a strong incentive to focus your RAs on civs that will generate at least as many beakers as you.
I am still on GnK, and I understand RAs were devalued in BNW. But even at 2 turns of science, the RA calculation still beats rush buying (although its much closer). You do not need 6 friendships, that could be 2 rounds of 3 RAs. The big difference I see is turns. Those 4500 beakers could take 60 turns, whereas your 3000 beakers took only 15 turns! But I think the limiting factor is gold, and if that is the case, then 4500 beakers in 60 turns is better than 3000 beakers in 15 turns.
Well, that is why I am asking! Those people who buy all their universities and public schools are doing so many other things better than I that I don't see any hard conflict. But those expert players can usually so objectively explain their habits, I am confused as to why my rough math doesn't seem to warrant this particular practice.
There are a few other factors to consider here. RA's are delayed beakers. You pay for them and wait 30 turns for their value to kick in. Mean while, you could get to new techs sooner with purchasing science buildings, which snowballs into more science. And when I said 2 turns, that is just what appears to be the best case, it could just have well been 1.5 turns, you just see the rounded number.
Anyways, I suspect there is some value in RA's, but there is definitely value in science buildings as well.
I plan to test out different methods of play to see if I can see results. Playing the same way every time will result in never improving.
This is a significant point, but it is rare for me to be in a situation where I cant spend gold fast enough!
I think it is the reverse situation though, since compared to buildings, RAs are less expensive and available more frequently. In actual play, you end passing on RA opportunities because you are trying to save up to rush purchase science buildings ASAP. But the net effect is less beakers, including those that unlock science buildings, so you end up getting those buildings later rather than sooner.
Browd calculates it at 2.5 when you are pulling 50% more bpt than your partner, and dont have other multipliers. So 2.5 is a conservative estimate. It can be worse, but RAs are more often better than that.
It comes down to opportunity cost.
Please do report back! Again, I assume that I am missing something here.
How can it be the reverse? You spend your gold, and wait 30 turns for that investment to kick in. The whole time you are X turns behind. Or rather, X beakers behind, because after 10 turns, my initial scenario was 3000 beakers ahead, after 20 turns, you can add another 1000 beakers for the slow cities (just rough numbers here). Now, for that whole time, you got to new techs sooner allowing for more growth and other things. When 30 turns is up and the RA's value kicks in, you are making up what you lost, or at least hoping too. The question is, do you actually make up that ground, because it is more than the beakers you have to make up, it is the several turns in being behind in techs as well.
It is complicated, and like I said before, there likely is a middle ground.
You want all your National Wonders. Fallout was saying dont bother building World Wonders.
The fact that you could rush buy three public schools tells me you were sitting on a pile of cash. What turn did you start saving? How many RA opportunities did you pass up in the meantime, especially had you been willing to loan money? Basically, how many AIs were you friends with on turn 123? But your RAs might have started even earlier than that. With 5 RAs, at just 2 turns per RA, you unlock Scientific Theory on turn 143. At turn 153 you still have 3 (or 4) Public Schools, 530 bpt, and gold left over for two more RAs.
I think gold spent on science buildings is easier to see, since the payout for RAs is so delayed. Or the buildings might actually be better, but it takes a spreadsheet to demonstrate.
Given X amount of gold on turn Y, buildings might be better, but that’s not how the game unfolds. One point I am trying to make is that RAs let you spend money sooner than later.
To be honest, I took one RA, and passed up 1. I only had 2 that would declare friendship, which is typical in my experience.
But don't forget that I also bought 3-4 Universities earlier, at the time when RA's would be first available. So the first set of RA's are being setup in replacement of universities, not public schools. And seriously, when was the last time you got public schools built on turn 143?
The other problem with RA's, is it is far more costly than you realize. When you are the leader in tech, they want lots of gold or resources as well, or they won't sign, which is especially insulting, as they cause you to get very little returns on the RA, due to the BNW rules.
Well you need Universities for Oxford, so I think rush buying there makes more sense. Would you be in a position to experimenting by reloading a save shortly after that? The point of the exercise is to see if you can complete 4 RAs by turn 143.
I do plan to try, but there is no way in hell I'm going to be capable of buying 4 RA's at that point. At least at higher difficulties. Civ's tend to be far more difficult to convince to make a declaration of friendship. I'm also not quite so keen on building Oxford so soon. I usually wait on Oxford until a big tech point. Like Scientific Theory. I'll build it until 1 turn before completion before hand, and when a big tech is up, and I want to rush it, I'll complete it on the turn the new tech can be started.
Radio, Scientific Theory and Plastics are my most common uses of Oxford. I treat it like bulbing a GS.
My premise isn’t that rush buying science buildings isn’t a good deal, it’s that RAs are even better. But I would think to test this assertion, you would want to keep as much as you can the same. If you think you can only manage 2 RAs between Education and Scientific Theory, maybe the problem is more that there is not much opportunity for RAs to make a difference?
I never meant to say RA's were not important, only that science buildings were top priority. I'm not sure how important they are, but they are definitely a useful bonus.
I'm also thinking that DoF's are harder to get on BNW, unless some of the reason I only tend to get a couple is due to my worker farming, but of course worker farming is extremely helpful for growth.
