Civ choice for the "builder"

[vmxa1]

Originally posted by peterfharris


Umh, you are doing rather better at Emperor than I do. Maybe with a more practice and more adherence to sound strategy I could make the English work for me at higher levels too.

So it wasn't you that was responsible for the Americans and Germans going downhill?

Try using some sort of semi tight city layout with a few camps.
Get a settler pump going and crank out troops in the camps. I had to build a worker every so often to keep the unhappy citizens in check in the camps and even a settler once in awhile. This gave me enough workers to do all the road and tile improvements with those lazy non-industrious people.
In fact I am thinking of stopping as I look out and see all the land I have now and will soon have and contemplate all the effort I will have to go to RR it. I knew I should not have pick a large map, ugh.

No I did not have anything to do with the Americans as they are on another contient. I had no impact on the Germans either as I have not even build a city that prevents them from getting land.
They were always behind in tech and probably had poor start location.
I only had two uprisings of barbs and they were very easy to handle. A sword and a few horses was enough in conjunction with the lay of the land (read hills). It was funny, I had a new city with one spear right next to one uprising and they never attacked it. It was on a hill and they wanted no part of it. They jump one lone warrior from Spain and he just kept killing them.

I now have all units upgraded to Med Inf and Pikes, plus horsemen so barbs are not a concern.

[Dominae]

Originally posted by vmxa1
YS during the early part of the game, increasing the slider would have done nothing, but waste gold.

Not sure what you mean by this. If the satement is to be taken literally, it's false.

The only time you concretely waste Beakers is when you overflow the amount required to research a tech (for instance, you're producing 20 per turn and have already complete 190 of 200).

Even if your research does not appear to proceed any faster when you put Science to 100% in the first few turns of the game, you will (in most cases) get that first tech faster if you do. Here's an example:

You produce 1 Beaker per turn at 50% Science.
You produce 3 Beakers per turn at 100% Science.
You research Alphabet, which costs (say) 150 Beakers.

150/1 = 150 turns to research Alphabet at 50% Science. The game rules decree that it will only take 40 turns, however.

150/3 = 50 turns at 100% Science. Again, the game rules step in and reduce this to 40 turns.

So apparently no matter what you do with the slider, you can only research Alphabet in 40 turns.

But now let's say you put the slider at 100% Science, 10 turns go by and your city grows in size. Now you're producing (say) 5 Beakers per turn instead of 3. Then:

3*10 = 30 accumulated Beakers so far
150 - 30 = 120 remaining Beakers for Alphabet
120/5 = 24 turns to research Alphabet

If you add this 24 turns to the 10 turns that have already passed, you see that your total time to research Alphabet has been reduced by 6 turns (from 40 to 34)!

Like I've said before, if you're really interested in these things, think of research in terms of Beakers instead of turns. This will help you make the right decisions relative to what you want to accomplish.


Dominae

[vmxa1]

Dom I did not say anything about wasting beakers, I said gold.
I also said that I was only looking at the rate up to around 50% and that if you were to go to 100% that would be different.

[vmxa1]

In the example it is +8 at 27 turns. Yes we know about beakers, but we really only want to know how soon.
So at 100% I do not get the break through any sooner, but I do lose out on the extra gold.
So all I was saying is until you get to the point where you can have a reasonable impact, I will stay at the most effective rate.
This may change if I am in a big need of a given tech.
As a non-expansion civ, I may be forced to go to max research to get pottery for instance. Since this was an expansion civ, I know I will get many of the low level techs from huts on this map as it is large.
Once I got to where rate was cutting down the time required significantly, then I moved the slider up. How far is depending on many things, but in this game 50-60% was about right. Lef me to do 10% luxs and put some cash awway for the upgrades.

[vmxa1]

Of course the beaker part is valid, but will not have much impact in this setting as the city is going to be a settler pump and will often be very small in pop. Probably bounce from 2-5 range through the ancient times.
All in all, you can get to the break through faster, but not fast enough to matter for this civ until you get most of the early techs. Between huts and contacts, you do not need to spend on researching. Like I said, if this was a std map and a non-expansion civ, I would do things differently. But that was not YS's question.

[Dominae]

Originally posted by vmxa1
So all I was saying is until you get to the point where you can have a reasonable impact, I will stay at the most effective rate.

If you're going to micromanage the Science slider, might I suggest the following heuristic (from alexman):

Always put your Science at 100% (or as high as possible) when you want to research a tech yourself. Then, when within 1-2 turns of completion, decrease as much as possible while retaining the same time to completion.

Obviously a thorough analysis of a particular situation may show that this is not optimal (it's a heuristic, after all). But in general, in the early-game when your pop is going up/down and tile improvements are being completed, this method works rather well (i.e. it will always get you to within 2-3 Gold of waste).


Dominae

[Yahweh Sabaoth]

I should note that when I used this technique, on wussy level (Regent) with stunning success, I got the first entire tier (I believe) of techs, plus Mysticism and, uh, Iron Working I believe (maybe HR or Writing) from scouting... the rest of the techs I got by founding ~10 cities, almost all on rivers, and cranking my tech ALL the way up (short of running a deficit).

So, at first, my tech rate was at a reasonable, or low, level, and I raised it to research Writing, Map Making, Mathematics, etc. - the "second & third wave" of techs.

It was the quickest I've ever gotten ahead in tech, although I also cranked the tech rate all the way playing as several other commercial civs (Spain, India and others) and achieved much the same results, though certainly less dramatic than with those English scouts.

Once again, I'm not stipulating that this is a very effective strategy overall... I just wanted to throw it out there to see what you pros think. I'll probably give it a shot (as the Hittites I guess) on Monarch level myself eventually.

[Arrian]

If I start an expensive tech, like Writing, when I have 1-2 cities, I will research it at 40-turn pace (10%) and rake in the gold.

Part way through, I may check to see if I can make a serious dent in the time to completion by cranking up the slider, if my pop has grown significantly.

-Arrian

[vmxa1]

Originally posted by Dominae


If you're going to micromanage the Science slider, might I suggest the following heuristic (from alexman):

Always put your Science at 100% (or as high as possible) when you want to research a tech yourself. Then, when within 1-2 turns of completion, decrease as much as possible while retaining the same time to completion.

Dominae

This is what I would do with a std map or a non-expansion civ. Maybe it is a good plan for even that setting, the caveat is where you mention "if you're going to micromanage".
I was not doing that, I was staying at 10% until I had many of the techs from huts and trades.
Then I went to the highest level I could without doing into a defict or at least only a small one.
My thinking was that I did not really expect to finish researching a given tech before I make the switch to the higher rate. I probably should have used 0%, until then. I think it made no difference or very little to the gold saved.

[Dominae]

Arrian, switching from a 40-turn pace half way through is inefficient. If you really want that tech fast, you're better off researching it full speed; if you can wait, you waste Gold changing your mind and speeding up research halfway through.

vmxa1, I'm relatively certain what we're discussing applies to all maps sizes. If you have reason to believe techs will be "accessible" (say, playing a Huge map as Expansionist, or on Deity), then there's no point in even doing a 40-turn pace. Or, if you think you have a chance to research a tech before any other AI but it could be close, you're better off going 100% until the last few turns. Finally, if you're reasonably confident that you a tech will not be accessible and there's very little chance an AI will beat you to it, the 40-turn gambit is the right choice. I'm trying to figure out why one of these three alternatives does not fit your test game.

By the way, the heuristic I mentioned actually reduces micromanagement. Trying to find the right slider setting to get a tech in a certain number of turns is very difficult when your pop goes up and down (as it does in the early-game). So, you might as well just put it at a max value until the last couple of turns, and get that little extra Gold by adjusting the sliders then.


Dominae

[vmxa1]

Originally posted by Dominae

vmxa1, I'm relatively certain what we're discussing applies to all maps sizes. If you have reason to believe techs will be "accessible" (say, playing a Huge map as Expansionist, or on Deity), then there's no point in even doing a 40-turn pace. Then again, if you think you have a chance to get a tech before any other AI but it could be close, you're better off going 100% until the last few turns. Finally, if you're reasonably confident that you a tech will not be accessible and there's very little chance an AI will beat you to it, the 40-turn gambit is the right choice. I'm trying to figure out why one of these three alternatives does not fit your test game.

By the way, the heuristic I mentioned actually reduces micromanagement. Trying to find the right slider setting to get a tech in a certain number of turns is very difficult when your pop goes up and down (as it does in the early-game). So, you might as well just put it at a max value until the last couple of turns, and get that little extra Gold by adjusting the sliders then.


Dominae

I think the first one does fit and is the one I was comtemplating. I went with the 10% as a lame compromise. I did not want to keep checking to see if there was any increased value between 0 and 10, so I just left it at 10. You know I have been known to be lazy some times or rather most times. I was not very interested in the research part, I was interested in the function of the camps.

I think I may take that heuristic tactic next time around as that idea of not dealing with it as much does appeal to me.

[Theseus]

Originally posted by Dominae
Arrian, switching from a 40-turn pace half way through is inefficient. If you really want that tech fast, you're better off researching it full speed; if you can wait, you waste Gold changing your mind and speeding up research halfway through.

Dominae, I am not sure I follow. How is gold wasted?

Do you mean time is wasted?

[Dominae]

Originally posted by Theseus
Dominae, I am not sure I follow. How is gold wasted?

Do you mean time is wasted?

Yes, sorry, you waste time.

Of course, time = money, and this is as true in Civ3 as it is in real life.


Dominae

[Theseus]

Weeeeell, I like having some gold around too.

That's an interesting question that I don;t remember grappling with:

When do you WANT 40 turn research?

[Dominae]

Originally posted by Theseus
When do you WANT 40 turn research?

1. When you're planning to play the tech trading game. That is, when you're not able/interested in getting a tech lead, but just planning to be as prosperous as possible.

2. When it would take you over ~25 turns to research the tech at maximum Science.

3. When you're reasonably confident that an AI will not beat you to the tech. The whole idea here is that you're gambling that the one tech you pick for a 40-turn pace will net you a bunch of other techs when traded around. You are thus able to make a lot of money but still keep up in the tech race. If another AI gets that tech first, its trade value is far lower and you may not be able to get back into the race.


Dominae