How to use specialists wisely?

[couerdelion]

Originally posted by Diadem
Yes but that is just flawed mathematics

You can ignore happiness and health limits if you want, but they will not ignore you. You have to take them into account.

Furthermore if you are using a lot of specialists, then you are running representation. It's the best civic hands down if you are using many specialists. If, on the other hand, you are not using many specialists, then the GPP of the few specialists you do have should definitely not be ignored.

You can't just ignore all the positive sides of specialists and all the negative sides of coast tiles and then conclude that coast tiles are better than specialists.

You mention specific circumstances which break the relationship between food, commerce and beakers. By deciding that you are running Rep. You’ve given the scientist 6b so the equation changes. But Representation is a special case and hardly the norm. It is not even always true that it is the best civic if you are using lots of specialists.

Likewise with happiness, health, these might be problems and, if they are, the value of food in the equation changes. You still have to compare 2f with 1b but now it may be true that 2f < b

Let me introduce some more exceptions.

1) I want to build an academy in my capital so assign two scientists in a library nearby – equation changes because GPP factor more heavily in the equation
2) We have a high science multiplier in a city while others have high gold multipliers. The assignment of scientists in science cities and merchants in gold cities may allow us to make gains from comparative advantage so that the gold and beakers from the specialists become more valuable than basic commerce.

But in this (my) scenario the conditions are

1) No Representation
2) No Happiness/Health problems
3) GPs in the city in question are “valueless”.
4) No significant “comparative advantage”

It’s not that I ignore the downside of coast tiles. Rather that there is not much or a downside and, far fewer than for a specialist.

[Blake]

Well there are always health/happy problems.

The REAL tradeoff, is between specialists and using food surplus to poprush.

I'll make this clear.
Take a fishing village - it has 2 clam tiles and a lighthouse, so in total it has +8 food surplus. Lets say the happy cap is oh... 10 and the health cap is higher.
Now, you could work all coast with population 6 through 10, this gives +8 food and +8 commerce.
Or you could run 4 Specialists, giving +12 commerce.

Now here's the thing. If you are running slavery you can turn that 8fpt into like 12hpt for popping in infrastructure. But if you aren't running slavery, then the food is completely useless because unhappy population is unproductive.

The thing is that it's pretty much always best to run slavery, so we never have to choose between 8c vs 12c, instead it's always 12hpt + 8c vs 12c... or maybe it's precisely that reason we always run slavery. Gives something better to do with food surplus than run specialists.

IMO precisely the best place to run specialists is where you can't do anything better. Like to me, the perfect scientist pump is a tundra city with a deer, fish and half a dozen coast tiles, with everything else being wasteland.

[Diadem]

If you have no happiness / health problems then you have already won the game anyway. If you're already in a situation where you can have unchecked population growth for only 2 food per pop in each of your cities, well, you will be so far ahead of everybody else you always win.

Like I said before: Of course you can think of scenario's where specialists are not useful. But what is the point of discussion extremely artificial situations that never occur in real games?

In real games, you always have food and health problems.

[Diadem]

And regarding representation. The problem is not that you are assuming no representation. The problem is that you are assuming no representation *and* irrelevance of GPP production.

These two assumptions contradict. If you are running a lot of specialists then representation is always the best civic. If you are running only a few specialists in a few cities, then the GPPs these specialists produce matter a lot.

[couerdelion]

Originally posted by Diadem
And regarding representation. The problem is not that you are assuming no representation. The problem is that you are assuming no representation *and* irrelevance of GPP production.

These two assumptions contradict. If you are running a lot of specialists then representation is always the best civic. If you are running only a few specialists in a few cities, then the GPPs these specialists produce matter a lot.

You’re arguing that Representation should be run IF you have lots of specialists. It’s not like that, you either do both or neither. If I can run Representation in the early game then using specialists will be a powerful strategy and scientists will often be a strong tile option for most cities if they can get by with a little production – see Blake post. But this is a Pyramid based strategy and very specifically demands a specialist economy. The “normal” case will be that you do not have Pyramids so assuming that you cannot run Representation should be considered the norm.

As to the ability to unlock Representation later in the game, this has limited impact on the whole game since, at this stage you should already be in a position to have won or lost. So arguing that running Representation and specialists – while it might be the right option at the time – will not give you any game-winning strategy since all the main work should have been done already.

As for the GPP argument, this is a specific benefit from specialists and does not apply to all of them. The fact is that, if you are appointing specialists in many cities, a lot of the GPP will be of no value whatsoever – only that which, at some time or other, will generate a GP. There’s nothing wrong with setting up one city (or maybe two or three) to generate GPs but these are unique specialised cities and do not describe the average city. And here, the specialists are not being used really for their beaker/gold benefit but for their ability to produce GPs. The discussion is entered into involved the case where a city was not being used to generate GPs so the assumption that the specialist generated GPPs were valueless seems reasonable.

Finally, the happiness/health question. In a city where you have coastal tiles to work – and not much else, you’re likely to have several other cities and will have resources to provide some basic health/happiness. After that, buildings are more likely to be whipped leaving you well short of any limits. This will be the case most of the time so, once again, the use of specialists in preference to tiles would be an exception.

[Dominae]

[QUOTE] Originally posted by Blake
Well there are always health/happy problems.

The REAL tradeoff, is between specialists and using food surplus to poprush.

I'll make this clear.
Take a fishing village - it has 2 clam tiles and a lighthouse, so in total it has +8 food surplus. Lets say the happy cap is oh... 10 and the health cap is higher.
Now, you could work all coast with population 6 through 10, this gives +8 food and +8 commerce.
Or you could run 4 Specialists, giving +12 commerce.

Now here's the thing. If you are running slavery you can turn that 8fpt into like 12hpt for popping in infrastructure.

Your implicit assumption here is that building up infrastructure is better than doing research (or generating commerce, or culture). While you are poprushing away, your reserach suffers, both because you are not working as many Commerce tiles, nor are you taking advantage of GPs.

Sometimes you have the luxury of slaving away and having your economy come into its own later on, sometimes not.

But if you aren't running slavery, then the food is completely useless because unhappy population is unproductive.

Not if you are using that Food for Specialists.

[Dominae]

Originally posted by couerdelion
Finally, the happiness/health question. In a city where you have coastal tiles to work – and not much else, you’re likely to have several other cities and will have resources to provide some basic health/happiness. After that, buildings are more likely to be whipped leaving you well short of any limits. This will be the case most of the time so, once again, the use of specialists in preference to tiles would be an exception.

There comes a point when are you are doing by growing your coastal cities bigger is letting them work one more costal tile, and keep growing a bit. Is better to get that extra 2 Commerce, or is it better to stay smaller (save on infrastructure) and work a Specialist for 3bpt and some GPPs? GPPs get worse as the game goes, on but on average they are worth at least a couple cpt, especially if you use GPs for GAs.

For this reason, I will not seek to raise my Happiness/Health caps in my coastal cities as aggressively as I do inland. I'll grow them as big as they can go naturally, then make Specialists. At the stage of the game when this occurs on a nation-wide basis, I'm close to Representation so it makes even more sense.

If I'm Financial or I have Colossus, I'll work the coastal tiles instead (pre-Representation).

[couerdelion]

There is a lot to be said for rushing infrastructure in coastal cities using the coastal tiles 2/0/2. But there seems to be some confusion over what this city actually looks like so I’ll introduce an “average” coastal city which perhaps has one food resource and can work one mine for a little production. With +4 food it is not going to be a decent GP pump so we should assume that we already have other means of acquiring GPs from some food rich site. If this is the case the GPP become almost irrelevant.

Now let us assume another standard assumption – we have not built Pyramids. If we have, then the whole equation for specialists becomes completely different. But most games played will not yield you the Pyramids so it is a safe bet to comment on the use of specialists without making allowance for the wonder benefits. I think it is also safe to assume that The Colossus will not be available either.

Now the question of happiness/health caps. In most cases you will not be running close to these for the simple reason that this city is a pretty average one with nothing alarmingly strong about it. You’ll have other cities and one would assume you’ve bagged yourself a few resources to keep most caps at bay. If not, you’re probably going to be whipping which will be a key part of improving those caps.

True, there will be some circumstances where there are problems that cannot be resolved simply by whipping and the specialist strategy might be desirable. But it is the realm of all the exceptions that I want to avoid because I am looking at general strategic and tactical game choices here.

We can assume also that the city is more of less average for the civilisation so that it has no comparative advantage in producing gold or beakers. It’s likely that there will be some small comparative advantage at all times but if it is close enough to the average then the benefits of specialising in the area of comparative advantage are small (almost insignificant)

So back to our city and let’s assume it is size 5 with +4 food surplus (or +2 surplus with scientist). City 1 grows in 4,5,4,5,5 and 5 turns respectively (to size 11). City 2 grows in 8 turns and then appoints another specialist so growth it halted. Let’s also run a third city that keeps growing but maintains one specialist. This one grows in 8, 9 and 9 turns to size 8.

Over the 28 turn period in which City 1 is growing, the coastal tiles (worked by city worker 5-10 will have generated 202 commerce. City 2 will have generated 140 beakers while City 3 will have generated 84 beakers and 66 commerce. What is more, City 1 can whip three of this population for 90 hammers and STILL produce 8cpt from those coastal tiles – more then the beakers from City 2 specialists and slightly below the current rate of City 3.

So higher production, higher commerce and at least as high in population terms. I know which option I would take.

The one remaining question is what to do as your get close to Constitution. But here the question is clouded a little since you are also just one tech further from Universal Suffrage. I would tend to favour USuff so Representation may never even get a look in here.

[Blake]

I'm going to second Dominae that specialists are EXTREMELY useful for Great People. If you want to see this in action, do a Philosophy slingshot - that is:
Build Oracle and pop Code of Laws (or research it normally)
Run Caste System - in your best food cities run as many scientists as you can - I've managed 4 or even 5 (with 2 food resources at one city). With the first Great Scientist create an Academy in your Capital (to help research mathematics), with the second lightbulb Philosophy.

Run Pacifism.

Milk Pacifism for another 3-4 great people who should be settled in your capital (other than the shrine prophet), then switch (back) into Slavery.

This can be an incredibly powerful opening, often better than CS Slingshot, it gets you more religions anyway and you can expand more aggressively than with a CS slingshot. It's much less dependent on having a good capital and it works absolutely great on Highlands maps which are bad commerce but good food & hammer (no pressing need for slavery).

Those ~4 great people generated with Caste/Pacifism will generate a massive amount of income over the game, and can give carte'blanche to expand without regard for upkeep (because with a good ~20 bpt from great scientists in your capital, you can afford to go down to 0% science).

Specialists for Great People can be better than nearly anything else - but specialist who don't generate GP's are not very good.

[Diadem]

Specialists are extremely useful for great people. That is because great people are so extremely useful. But even in situations where the GPP do not matter much specialists can still be powerful. I'll come to that later, first let me demonstrate the power of specialists with a little screenie.

This game I finished 2 days ago or so. My first OCC win on immortal with warlords. I ran a lot of specialists for most of the game, and added great people as super specialists. Of course I used representation. It's simply the best civic. (By the way: In warlords the AI prioritizes the great wall over the pyramids. Pyramids are thus very possible to get even at the highest difficulties. You need stone or chop though, but you don't even need both).

Very interesting to note is that I'm only generating 5 commerce in my entire city screen. Yet I can outtech an immortal AI while having more cash than I can spend. That's the power of specialists.

[couerdelion]

I think OCC is a rather specific case where you do not have other cities to specialise in one thing or another. Hardly a case for justifying working a civic such as Representation as being "the best" or to justify specialists over coastal tiles.

[smbakeresq]

I agree that the main purpose of specialists are GP. The other benfits are nice, but even without them I would still use specialists to generate GP.

I always generate at least 2 GP, the scientist for the academy and the prophet for the shrine. If you can manage to get the GW and Pyramids, then you have to use engineers to get the GP Engineer just so you can rush wonders.

[zeace]

A lot of the debate here seems to come down to

Why would you use Representation and farms when specialists are so bad, therefore assume no Rep and few farms, therefore specialists are bad.

The reverse argument is equally circular.

I usually decided very early whether I'm going to have a specialist economy or not. The two are very different games and require very different terraforming, wonder building, and city site choices.

If my capital has lots of food, and I can find a good 2nd city site with lots of food, and I think I can get the pyramids (Ind, stone, or just decent production in one of my cities) I'll decide at that point, before I even build city #2, that I will have a specialist economy and all cities from then on will be highly specialized. I'll build all specialist improving wonders, and lots of farms. Being Philo increases the chance I'll go this route, being financial decreases.

That said, if I have a good capital site for production or commerce, and my 2nd city is going to be in the woods to get some copper or horses, a specialist economy is not going to work as well, so I'll probably aim for a normal land-working game.

Both stategies work in their respective situations, and I think it's foolish to say one is always better than the other.

[Rancidlunchmeat]

Re: How to use specialists wisely?

Originally posted by Olavi
The most problematic aspect of the game for me are specialist citizens. I can't really handle them. In many games, I end up playing without paying any attention to these special guys.

(This is because assigning specialists is annoying micro-management I truly hate, but since specialists are an essential part of the game, I should probably change my ways).

I can't believe you're managing a single victory on Monarch without utilizing specialists...

There's been lots of good discussion about specialists, specialist economies, and the like.. But the answers are actually complicated, and I'm not sure if you wanted/are ready for that yet. So I figured I'd just answer your questions.

So would you please share your thoughts about specialists?

- At which time do you start using them?

As soon as possible. If you are going the religious route, you're going to want to build the Oracle/Stonehenge and a temple in the same city. Assign a Priest specialist to boost GPP production because you want the GPr as quickly as possible to build the shrine to get the income.

The other important early need is for the GS to build the academy, you want to get this as early as possible as well. That means prioritizing writing and building of a library. Then assigning a scientist specialist.

- Do you ever cancel specialists back to normal field workers (if happiness/health limits go up and allow futher city growth, for example)?

Usually after either of the above scenarios (or both) have paid off, both the priest specialist and science specialist will go back to working the land after they've achieved their goal of creating a GPr or GS.

- Do you plan your research and buildings so that they allow more specialists, or are specialists just a nice by-product?

Planned from the very beginning. Depending on starting location and choice of leader. I always play random leaders so I don't know before the game starts, but as soon as the first city goes down, my tech path is dictated by what specialist I want to be able to assign so I can get the GPP I want.

- Do you mix different kind of specialists in one city?

I never do. I'll build science wonders and assign specialists in one town, religious ones in another and engineering ones in another. I ignore merchants and artists for 90% of my games. But thoughts differ on this.

I prefer specific GPP over more GPP, because I've already know what I'm going to do with the GP before he arrives.

- How important are they, really?

Immensely. Whether you base your economy on them or not, it's very important to plan to get the GPr for the shrine or the GS for the Academy or the GE so you can rush the GL, or the GS or GP so you can slingshot your way deep down the tech path via lightbulbs.