|
Are you, in fact, a pregnant lady who lives in the apartment next door to Superdeath's parents? - Commodore |
|
Bank Productivity
|
|
As came up in the Adventure 38 discussions, I've written an analysis of the productivity of banks in Civ 4. You can find it here and hopefully achieve enlightenment:
http://www.dos486.com/civ4/index/bank.shtml
Interesting. Thanks.
mh
"You have been struck down!" - Tales of Dwarf Fortress
--- "moby_harmless seeks thee not. It is thou, thou, that madly seekest him!"
Nice!
My favorite part was how you said if it isn't obvious which is better (between University/Observatory and Bank), it doesn't make much difference. That means when I am running a CE I don't have to do any number crunching. Actually, I'd love to know when it is better to whip a Bank or whip a Courthouse, if you are still with thinking cap on 8). This highlights an interesting difference between an FE and a CE. With an FE, you can build Banks in the cities with Grocers/Markets. These cities run Merchants and pay the bills. Most cities will simply get the research buildings and run Scientists (actually not generally true, since Grocers/Markets give health/happiness). The ability to choose between Merchants and Scientists is essentially giving you an individual slider for each city, especially under Caste System. Darrell
I would also suggest taking a look at Mutineer's comments (post #200) after a rush buy variant SG. Its definitely situational and doesn't change your conclusions, but its not bad to have in your toolbox.
Darrell Quote:This highlights an interesting difference between an FE and a CE. With an FE, you can build Banks in the cities with Grocers/Markets. These cities run Merchants and pay the bills. Most cities will simply get the research buildings and run Scientists (actually not generally true, since Grocers/Markets give health/happiness). The ability to choose between Merchants and Scientists is essentially giving you an individual slider for each city, especially under Caste System. Eventually you'll reach Rep though, in which case libraries enter the equation, and it'll resolve to a similar solution to what T-hawk calculated. T-hawk: Corporations? Nice treatise, but I'd have to disagree with the part that says rushbuying is only useful in defense and mop up etc, but that's a production variant. Your comments about research are very good, especially this bit: Quote:This effect is a significant part of why large empires always eventually outresearch smaller empires. Small empires with low expenses get next to nothing out of banks. Large empires essentially get an additional set of expense-reducing buildings that small empires do not. And smaller empires can have a greater proportion of their expenses paid by direct gold such as shrines and settled prophets, leaving even less expense headroom for the bank to work in. A bank can even break down completely; a small empire researching at 100% slider with surplus gold (negative net expenses) leaves the bank completely worthless for beakers. Any way to fix this?
Sorry, I'm still confused. I can't get my head around why the slider is irrelevant to whether commerce or gold buildings are better. I'm sure you're right, but I'm going to present this argument anyway for you to shoot down: The slider for break-even science rate is directly affected by the expenses. If your expenses increase on one turn then your slider must drop. So the decision over what to build, being affected by expenses, must also be affected by the science slider. The simple rule of "Science slider low = build bank. Science slider high = build university." must still work because high expenses causes a low slider.
I think T-Hawk is simply fixing commerce and drawing that conclusion...after all if commerce is fixed and you are maximizing your beakers, then expenses sets the slider, not the other way around. If you fix expenses instead of commerce, it might look more intuitive to you. For example, fix expenses to 10 and assume no multiplier buildings:
Case 1 Commerce is 20. Your slider is at 50% to break even. A Bank allows you to drop the slider to 33%, giving you a net beaker increase of 3.3. A University will give you a net beaker increase of 2.5. Bank is better. Case 2 Commerce is 30. Your slider is at 33% to break even. A Bank allows you to drop the slider to 22%, giving you a net beaker increase of 3.3. A University will give you a net beaker increase of 5. University is better. Case 2b Same as case 2 but you have a Library and an Academy. Now the net beaker increase of a Bank is 5.8 and the Bank becomes better. Case 3 Commerce is 100. Your slider is at 10% to break even. A Bank allows you to drop the slider to 7%, giving you a net beaker increase of 3.3. A University will give you a net beaker increase of 22.5. University is better. Note in all cases the bank boosts beakers the same (independent of commerce => independent of slider). The exception is 2b of course, where we change the existing science multipliers, increasing the relative return of a Bank. Either way you look at it though, you'll end up drawing the same conclusion on which building is better. Darrell P.S. Krill, I don't completely get your comment re: Representation and Libraries, can you elaborate. sooooo Wrote:why the slider is irrelevant to whether commerce or gold buildings are better. That's not quite the right point. The slider is irrelevant to how many beakers a bank produces - Darrell gave a great concise example. The slider is relevant to how many beakers a science building produces, and transitively then is relevant to the science vs gold building question. Darrell's case 2b is especially wonderful - that's a point I mentioned but didn't quantify, that the value of a bank depends on your science multiplier, which might be counterintuitive but is absolutely true. darrelljs Wrote:Actually, I'd love to know when it is better to whip a Bank or whip a Courthouse, if you are still with thinking cap on 8). Whichever building saves more commerce relative to its hammer cost. Both amounts of saved commerce then multiply to beakers at the same rate, so that factors out. The commerce freed up by a courthouse is (City Maintenance * 50% / CashMultiplier) * Inflation. The commerce freed up by a bank is described in my article, (Expenses / Old cash multiplier) - (Expenses / New cash multiplier). It's not trivial to calculate precisely, since you need to work out the change to your national cash multiplier, which is the average of each city's cash multiplier weighted by that city's commerce production. In practice, you'll get close enough by looking at the city's gold numbers, and subtracting just a bit from the bank to account for the self-defeating effect, and adding a bit to the courthouse savings to account for infation. So if you can't decide, the courthouse is probably better.  Krill Wrote:Eventually you'll reach Rep though, in which case libraries enter the equation, and it'll resolve to a similar solution to what T-hawk calculated. Since Darrell and I are answering each other's requests, I'll field this one.  Krill means that once you have Representation, you'll no longer have a merchant city with pure cash. Because Rep produces direct beakers, now this city wants libraries in addition to its presumptive prior cash multipliers.Quote:T-hawk: Corporations? Nice treatise, but I'd have to disagree with the part that says rushbuying is only useful in defense and mop up etc, but that's a production variant. Not sure what you mean about corporations? If you mean rushing executives, I definitely agree that's often worthwhile. Rushing one exec begets more execs quicker, and the payback time on execs is usually quite quick. However, I still tend to rush my execs more by whipping (especially with Kremlin) than with cash. Quote:Any way to fix this? Delving into this a bit deeper: No mature empire, with all courthouses and cash multipliers built (and barring corporate expenses), will ever have a science slider below 75%. That number actually matches my experience pretty closely. In the recession phase, you cannot expand beyond the point where 100% of your commerce goes to expenses (equivalent to 100% cash slider.) Courthouses reduce that to 50% commerce-to-expenses, and bank/market/grocer then let you run only 25% slider-to-cash by doubling that cash. (That's not strictly true, you can expand beyond 100% cash slider by paying expenses with alternative cash like pillaging and tech sales. But it's generally true as a concept.) So, how to fix this? Raise expenses not just with city count and population, but also with the infrastructure maturity level of the city or civilization. In other words, charge maintenance expenses for buildings. Or exactly the way it worked from Civilization 1 all the way through SMAC and Civilization 3.
Ah yes...Representation means Merchants give beakers, so the conclusions from the report do apply.
T-hawk Wrote:So, how to fix this? Raise expenses not just with city count and population, but also with the infrastructure maturity level of the city or civilization. In other words, charge maintenance expenses for buildings. Or exactly the way it worked from Civilization 1 all the way through SMAC and Civilization 3. Sirian definitely not lurking, or he would have commented on this by now . I'm a bit surprised Sulla hasn't...Darrell |
