Using Great People : A short look at the Golden Age

[rjmatsleepers]

Originally posted by darrelljs

In a vacuum (i.e. there are no opponents) my guess is running your last N turns in a Golden Age would be the most efficient use for ALL your GPs. I would not have said that before reading this post.

Darrell

I need some convincing that saving your first Great Scientist for a Golden Age at some point in the future is better than an Academy.

Similarly my intuition says that a shrine is a better use for an early Great Prophet than saving it for a Golden Age.

In my game in progress (see the spaceship comparison thread) my Acadamy is currently (AD 1500) worth more than 80 beakers per turn. I think a shrine would be worth at least 15 gold per turn - can't be exact since I haven't managed to turn out a Great Prophet in the last 1500 years.

In a 9 city empire I can't see a Golden Age getting remotely close to matching these figures.

RJM

[darrelljs]

Originally posted by rjmatsleepers
I need some convincing that saving your first Great Scientist for a Golden Age at some point in the future is better than an Academy.

Similarly my intuition says that a shrine is a better use for an early Great Prophet than saving it for a Golden Age.
RJM

It is highly dependent on the number of tiles you'll be working at the end. I was trying to figure out the value of settling the first three GPs, but I'm at work and can't remember the number of turns in an Epic game. I assumed 1000, and further:

1. My first three GPs were Prophet, Prophet, Scientist.
2. They were born on turn 150, 250, 300.
3. The average production multiplier was 50%.
4. The average research/gold multipler was 75%.
5. The game ended on turn 900.

Hammers = 900-150)*2*1.5 + (900-250)*2*1.5 + (900-300)*1*1.5 = 5100
Gold = (900-150)*5*1.75 + (900-250)*5*1.75 = 12250
Beakers = (900-300)*6*1.75 = 6300

Factor in any bonus from Representation and more importantly the fact that the value integrates over the game rather than one lump sum at the end, and I clearly spoke too soon . You're also right that Shrines and Academies in the proper places will outstrip even these numbers.

Everytime I try to find the correct "formula" for using GPs, I'm reminded they are intentionally flexible, and their value can only be optimized in the context of a particular game.

Darrell

[couerdelion]

Originally posted by Blake
Naturally, altough I made a mistake, it's only 10 turns on epic.
On Epic GP cost 50% more points to generate and all relevant effects are scaled by 50% (ie cash bomb produces 50% more cash). Golden age are 25% longer.

On Marathon they take 200% more to generate, and all effects are scalled by 200%, except Golden Age which is 100% longer.

I thought I recalled that they only ran for 10 turns rather than 12 so thanks for explaining this.

It simply means that Golden Ages are devalued in Epic and Marathon.

[Blake]

[QUOTE] Originally posted by darrelljs


Thanks for taking the time to run the numbers in a controlled experiment. It makes me think of two questions:

1. Is 2250 commerce and 800 hammers in 1500AD worth more or less than 10,000 commerce and 3000 hammers in 1940 AD?

How you have not run into the concept of "turn advantage" is beyond me (Vel! Go write up something on turn advantage in a Civ4 context! ).
The earlier you get something, the more useful it is. 10 commerce at turn 10 might be worth as much as 1000 commerce at turn 100. One axemen in the early game is worth more than 10 tanks in the late game. Your first worker is more valuable than ALL your workers combined at the end of the game.

2. Is it worth holding onto those Ancient/Classical/Medieval Great Engineers/Scientists/Merchants for late game golden ages?

The answer is a definite and resounding NO. While Great People have a fantastic shelf life, every turn they sit around is a turn less of having the advantages they would be giving you.

[couerdelion]

Although the intention of this thread was to talk about Golden Ages, I felt it necessary to first illustrate the other options available for GPs for comparative purposes. The basic value by which each GP should be assessed is it’s “lightbulb” value expressed in beakers output. In considering how to use your GPs you then need to consider whether the three alternatives give a better return.

So here are some basic rules of thumb

1) Earlier in the game, settling will pay off, particularly if this settling is done in the right city. Settled GPs will be enhanced by city multipliers, bureaucracy, representation, civic bonuses etc too.
2) If you have an option of establishing a shrine or one academy then this comes first almost always
3) Sometimes a rush build or lightbulb tech will yield more value over the long term but this is because the opportunity cost of rushed tech/wonder - and not because the beaker gain or hammer gain - was worth more than the value of settling.

Moving into the Middle Ages/Renaissance eras, let’s assume that you have your Academy/Shrine and enough time left in the game for the settling to pay off higher returns over the whole game while opportunity costs fall to low values if you have a good tech lead. Rush building with GE’s become simple hammer bonuses while additionally shrines/academies need to be judged on merit against the long term gain from settling in your OU city (which you should now know). As cities grow around the world, your GM trade mission value will grow commensurately and may become preferable to settling. But by this stage it will probably always remain a better alternative to the standard lightbulb.

Once you get into the late game then the value of settling really starts to tail away so this option becomes far less desirable. We have to fall back on our lightbulb option and compare as follows

1) For our GMs, we already know that the trade mission is going to take precedence.
2) GE rush buys simply get equated to an early build value or hammer bonus which probably translates simply into a few extra cavalry or infantry units. Nothing special there
3) GS can build an academy but we’ve already decided that the settling option is not really going to pay off so there are unlikely to be that many cities where the academy is going to be hugely more valuable. In the OU city, with representation, your GS will generate +32 bpt/+2 hpt so you need a city with at least 64 pre-multiplier beakers for the academy to be worth more than the settling
4) GA – I’ve never liked them much anyway
5) Prophet – Any late shrines might pay off well but if our GS requires a 64 pre-multiplier beaker for an Academy to pay off then your religion will need at least 16 cities following it (with market+grocers+bank) for the shrine to be consider of similar value.

With these few exceptions, the only option to the lightbulb then is the Golden Age and here were have a crude comparison. Big empire - > Golden Age. Small empire ->Lightbulb.

Also Golden Age loses it’s lustre in Epic games and even more so in Marathon. I fail to understand the logic here.

Thanks, Blake for taking the trouble to compare things. One tip I would suggest is that you go to the finance screen to compare the post-multiplier output. The production output was the trickier one and I wonder also whether some specific build bonuses should be included in the multiplier effect (leader trait bonuses, org religion, heroic epic etc). For this reason I used the hammer figure on the city management screen.

[fed1943]

I think one more important factor should enter in the equation: need. (like market changing prices just due to their quantity).
I see cleary the conceipt but no idea how to factor it.
I guess that GP can be used to equilabrate that as our need and plan.
Best regards,

[darrelljs]

Originally posted by Blake

How you have not run into the concept of "turn advantage" is beyond me (Vel! Go write up something on turn advantage in a Civ4 context! ).

I have actually read Vel's guide to SMAC and understand the principle. When I read a statement like "10 commerce at turn 10 might be more valuable than 1000 commerce at turn 100" I have trouble seeing it. I'm not saying it's not true, I'm sure it is. It's just not intuitive to me (the order of magnitude, that is).

Darrell

[Blake]

Well the worker one should be pretty intuitive, the one worker at the start builds improvements for a much longer period than the end game worker, the improvements he builds are worked for a much longer period.
And when it comes right down to it, food, hammers and commerce can be exchanged via various mechanisms.

Turn advantage is most obvious with hammery/food stuff. But it's there with commerce too, since commerce is clearly very important in granting you new hammery/food stuff to build, like granaries and such.

[trev]

In general I see very little advantage in initiating a golden age, with one exception, and that is in the spaceship race. A golden age initiated at the time of building the spaceship parts and when the required techs are learnt ,or nearly learnt is priceless, because these parts cannot be hurried up by any other method. Generally though I will use Great People on hurrying my science, often used this way, they can net religions like Christianity, Taoism and Islam for your civ and other very good techs like Liberalism, Democracy (for Statute of Liberty), Economics etc..

[Ijuin]

My priorities with Great People are normally:

Prophets: Discover any tech that immediately founds a new religion. If that is not possible (e.g. next tech is Priesthood), then build religious shrine.

Engineers: Rush a Wonder.

Scientist: Build an Academy in my Science city (normally my capital since I run Bureaucracy until Free Speech is available).

Any Great People not needed for the above I can save up for a Golden Age. My favorite time for a Golden Age is when I am building up for my Break Out (i.e. going out to conquer an enemy completely).

[Urban Ranger]

Originally posted by Foxhawk
If you were devious, couldn't you save up 1 or 2 gmercs, and send in a rather archaic looking attack force that shouldn't warrant any particular attention, then blam blam, two turns and a mess for the the other fellow.

You can't upgrade units outside of your cultural border.

[couerdelion]

Originally posted by Blake
Well the worker one should be pretty intuitive, the one worker at the start builds improvements for a much longer period than the end game worker, the improvements he builds are worked for a much longer period.
And when it comes right down to it, food, hammers and commerce can be exchanged via various mechanisms.

Turn advantage is most obvious with hammery/food stuff. But it's there with commerce too, since commerce is clearly very important in granting you new hammery/food stuff to build, like granaries and such.

I thought that the idea of “turn-advantage” was simply that of getting ahead of the AI rather than your description which seems to see it as the relative values of commerce, food and hammers at different stages of the game.

That idea would probably be captured by the general principle of “opportunity cost” which attempt to place a value on any asset which represents the simple fact that it is worth more simply because I have it now and can do something with it.

My priorities with Great People are normally:

Prophets: Discover any tech that immediately founds a new religion. If that is not possible (e.g. next tech is Priesthood), then build religious shrine.

Engineers: Rush a Wonder.

Scientist: Build an Academy in my Science city (normally my capital since I run Bureaucracy until Free Speech is available).

Any Great People not needed for the above I can save up for a Golden Age. My favorite time for a Golden Age is when I am building up for my Break Out (i.e. going out to conquer an enemy completely

GPs are so valuable that you don’t want to waste them. As a general rule, your early GPs are better used settling in the city for the higher long term advantage. The main exceptions to this are the founding an Academy and a Shrine which will give you more than the settling option.

When rushing (a tech or a building) you really need to consider if it is going to be better over the long term. It is rare that the settling option would produce less than the hammers or beakers generated so the only way that the difference will be made up is through “opportunity” cost. You should also note that the opportunity cost of the first religion is greater than that of the second so, as a general rule, a shrine will give you more than a second religion.

For a building, the opportunity cost is both that you have the wonder and that others don’t and if it is a strong wonder (Pyramids) then that opportunity cost might be huge. But beware those wonders where you have a safe chance of capturing it. In this case, rushing a wonder only gives it to you earlier. You get a reasonable hammer gain and a turn-advantage (X-turns saved from building). That benefit is often not so huge so another general rule is not to rush wonders that you can build anyway (particularly if you have the resource and/or are industrious.

Rushing techs at the early stage is simply buying turn advantage and perhaps an opportunity cost (I have Macemen, you have archers!!!). If you do not use the power of your rushed tech, then the act of rushing is simply a free shot of beakers which we already know is going to worse than settling in the early game.

So, in summary, in the absence of any significant opportunity cost or turn-advantage, rushing with your early GPs will undervalue them.

[Mr Justice]

i almost never use GP to lightbuld techs.

possibly only time i could really use em is real early in the game and i always go for acadamy/shrines first.
by the time i have thoes two my reaserch needs generally arent as imediate and the lightbulb just doesnt offer the long term benifits that settleing will.

also i rarely use GMs for the gold bomb, i prefer putting them into a city wherre the +1 food allows the City to grow larger than it otherwise would be, primarly this ends up being either my main production city or my OU/WS city. i know that putting a GM into my production city means i am giving up probably 6 gpt, (the 100% bonus from WS), but as a result in one case i was able to work an extra lumberjacked + railroaded plains forrest which gave base +4 sheilds pur turn, which with all the multipliers turned into 4 * 200% or 8 sheilds per turn. that was better than what the settled GE brough in. note that city kept growin and shrinking before i settled GM so the actual benifit is not as high as it seems becuase for periods of the game the city could have the pop to work that tile.


all this does is illistrate that there is no real "ideal" use, and that you should examine your empire before you decide what you can do with the GP.

[PaganPaulwhisky]

I tend to use my GP almost exclusively to lightbulb techs. I think the compounded turn advantage of obtaining crucial military techs early by far outweighs any other use with a few exceptions (in domination games anyway).

Exceptions:

Obviously the early academy for bureacracy and shrine are important, however I will often delay these if I can get philosophy with my first scientist or prophet. The early benefit of pacifism will speed up the GP generation anyway.

GM - I will usually use the cash bomb to fund future wars unless the tech is really crucial.

extra prophets - if obtained early I will settle these guys.

GE - I really only use his special to construct GL. They can lightbulb some really nice techs early which IMO outwieghs the wonders of this period. This decision does not get made often since I get so few of these early.

[drsparnum]

Great thread.


Question: Why is it so hard to calculate the value of a GA in terms of commerce beakers? Example: Let's say you're running at 70% science and your next tech will take 8* turns and you're making 12g/turn (a reasonable scenario). If you drop down the science slider to 10% you see it will take 42 turns to get your tech and you're making 150g/turn (I'm making up numbers here, so bear with me....). Now induce GA and compare.

Suddenly your g/turn increases to 250g/turn. The tech that was going to take 42 turns now will take 28 turns. This tech costs 4200 beakers (remember I'm making this up). So you know that the GA is giving you 100g/turn (250-150), and 50 beakers/turn (before 42 turns for 4200 beakers = 100b/turn; now 28 turns for 4200 beakers means 150 beakers/turn). As mentioned before hammers can be done**. Multiply by length of golden age and voila.

*Admitedly this only works well when you're just about to start researching a new tech, so you know the number of turns left is working off the full cost of the tech.

** Is there an easier way to get total empire hammers than just summing them in the city spreadsheet? I'd like to know that number because for me, it is a nice single indicator of empire potential.




As a rule myself, I usually love all my GP except the GAs. I don't think ending resistance is a good enough use for them. I don't like the culture bomb. Settling them in the city makes some nice culture, but I like to think that someday in the future, that edge of my civilization where I settle Michaelangelo, will eventually be buried deep within my empire which has expanded. So basically I'll use a GP up immediately except a GA, because I don't like the loss in turn advantage of saving them. Then, late in the game if I'm not so inspired with my options with another great person I'll pair him up with my GA, and make a golden age. You can also have too many Gr Pr, if you've already built all your shrines, but I just don't seem to have this problem often - I don't think I value the early wonders that make Gr Pr as much as the other wonders and the AI usually beats me to 'em.