The system of advances in Civ2

SlowThinker · December 16, 2000, 16:18

Content: This post is related to the way how Civ2 determines selection of advances offered for next research ("What discovery shall our wise men pursue?"), and some more questions about advances. It includes a link to a program that helps to plan science research.

Problem 1

Update:
This problem is discussed at another thread yet: oedo´s unfinished . The formula described lower is known as Oedo's law.
I think that problem is solved.

I was not able to find out how Civ2 specify set of advances allowed for next research on any Civ-oriented web pages. There are some opinions that two advances from the same category (military, economic...) can't follow one another, but I think it is false. I thought about it, and perhaps I found out how it is done:

If difficulty level is Chieftain then following statement is true:

(1)	Advance ADV is in selection even as <NotYet(ADV)> and <TreeOK(ADV)>

If difficulty level is not Chieftain then following statement is true:

(2)	Advance ADV is in selection even as <NotYet(ADV)> and <TreeOK(ADV)> and (not <CaughtByFilter(ADV)> or <First(ADV)>)

where

<NotYet(ADV)>	=	ADV was not discovered yet
<TreeOK(ADV)>	=	both prerequisites (preq1 and preq2) were discovered (or taken any other way) yet
<CaughtByFilter(ADV)>	=	<Ord(ADV)> = (<NumberOfAdvances> mod 3) + 1
mod	=	modulo (1 mod 3 = 1, 2 mod 3 = 2, 3 mod 3 = 0, 4 mod 3 = 1, 5 mod 3 = 2, 6 mod 3 = 0, 11 mod 3 = 2,)
<NumberOfAdvances>	=	Number of advances you acquired yet (by research, diplomacy, conquest (I did note prove this), goody hut). Advances obtained just on the beginning of the game are not included.
<Ord(ADV)>	=	Placings (number of row) of ADV in rules.txt (Ord(AFl)=0, Ord(Alp)=1, Ord(Amp)=2...
<First(ADV)>	=	Ord(ADV) is minimum of a set {Ord(X) \| NotYet(X) and (TreeOK(X))} In short, ADV is the first on a list of advances that have both preq1 and preq2 yet.

Notes:

???

it may be helpfull to plan scientific research ahead

http://home-1.worldonline.cz/~cz0456...n.htm#civ2plan

Problem 2

Problem 3

Problem 4

Problem 5

(If you will alert me to my bad english expressions I will be thankful.)

[This message has been edited by SlowThinker (edited January 31, 2001).]

Ken Hinds · December 16, 2000, 17:54

I'm not up on the latest theories for 1 and 2, therefore I will let someone else answer thoses.

Item 3. When looking at F6 (your advances) white means that you were the first to discover that advance. Blue is the default(normal) color. When looking at an opponent's advances (F3, check Intelligence) the white are those that You do not have and can trade for or steal. Blue is the normal color again.

Item 4. Yes you can get advances from huts without the prerequisits. Just like you can steal them without having the required techs.

Ken

Scouse Gits · December 17, 2000, 03:16

KEN - your comment on Item 4 above is wrong.

You MUST have the prerequisites for any science obtained from a hut.

Think about it! If you didn't need the prerequisites you would have stupid situations. If at 4000BC you pop a hut and get Invention ... no more science from villages. The game designers don't want your first wonder to be Leo's either!

-----------------

SG (2)

Paul · December 17, 2000, 04:33

To answer your second question: the number of beakers is based on how many advances you have already discovered (not including free techs at game start), difficulty level and how far you are ahead of (or behind) the AI. I don't have a formula for it and I don't know if one exists.

Ken Hinds · December 17, 2000, 07:16

SG(2),

Maybe I just haven't paid enough attention, but I'm sure that I've gotten things like Currency and Map Making without having Bronze or Alphabet. I don't normally find things like Philosophy, Monarchy or Literacy in hut anyway.

Ken

Cam · December 17, 2000, 08:59

Sorry - my experience is the same as Ken's (n.b. I use Fantastic Worlds vers.) - I've certainly got advances in huts without having the prerequisites.

Scouse Gits · December 17, 2000, 11:48

I seldom play version 2.78, so I ran a few hut tests. As far as I tell it behaves the same as 2.42 and MPGE.

A game on a Large World was tested playing Romans/Hordes/Deity. Starting techs were Masonry and Bronze. Both settlers opened huts on the same turn during 3800BC. In all 20 Scrolls were recorded - none had different prerequisites from science already discovered. One turn both settlers found wisdom. The first one was Currency, but the second was Construction - this seemed to emphasise my point that once Currency was in F6 Construction was possible.

The techs gained were:
Pottery (8)
Warrior Code (4)
Alphabet (3)
Currency (2)
Horseback Riding (1)
Ceremonial Burial (1)
Construction (1)

During the research on starting techs we did a fair bit of research (still ongoing!) into huts in 2.42. Checking through those records I cannot see any time when a hut yielded science without the prerequisites already in place.

If your game is not heading for Democracy quickly (thus needing Invention) many quite advanced techs are available from huts. (University/Theory of Gravity/Chemistry/Medicine/Sanitation - amongst others.)

---------------

SG (2)

Ribannah · December 17, 2000, 20:45

quote:

Originally posted by SlowThinker on 12-16-2000 03:18 PM
Content: This post is related to the way how Civ2 determines selection of advances offered for next research*...
... [/b]Could you tell me if I am right?

Oedo posted exactly the same formula on the Strategy forum just a few weeks ago ("Oedo's unfinisehd"). There seem to be a few glitches though, especially with Warrior Code, but perhaps we should call that a Myth

.

------------------
If you have no feet, don't walk on fire
[This message has been edited by Ribannah (edited December 17, 2000).]

SlowThinker · December 18, 2000, 18:53

I finished my program (it may help to plan science research). http://home-1.worldonline.cz/~cz0456...n.htm#civ2plan

SlowThinker · December 18, 2000, 22:10

quote:

Originally posted by Ribannah on 12-17-2000 07:45 PM
There seem to be a few glitches though, especially with Warrior Code, but perhaps we should call that a Myth

.

I have read Oedo's thread, but i didnt understand a result of their discussion. (I think they all died one week ago.)
Is "Oedo's law" correct in all situations or not? Do you know any resistant science-thread (for exxample with Warrior Code)?

[This message has been edited by SlowThinker (edited December 18, 2000).]

Daddy-O · December 19, 2000, 02:04

re: your second problem - beakers.
I think I saw this being discussed in a thread on OCC games in the strategy forum. The short story is they are not sure how this works.

re: your english.
Don't worry about it. First of all, it is about 97.5% correct. Second, I am sure that your English is much better than any of our Czech will ever be.

------------------
'I want to be hand counted and checked for dimples.'

East Street Trader · December 19, 2000, 07:30

Problem 2, I'm pretty sure that if you get no tech by means other than research it starts 10 beakers, 20 beakers, 30 beakers and then the arithmetic progression takes on a bit more of a geometric aspect. But that is almost wholly based on feel, not calculation.

I suspect that the practical benefit of knowing what the underlying beaker rate is would be small. OCC experience has taught us how critical the adjustment made for the tech lead over the AI is. That knowledge both demonstrates how complex the formula will be and reduces the value of what detailed calculation would reveal.

Problem 5, I play 2.42 and IIRC what you describe matches my experience.

Early on the AI will often accept a different tech from the one sought and some techs are highly prized by some civs throughout. But, generally, once I am well ahead, they are keen to trade for my most recent advance - especially critical ones like gunpowder or advances which allow wonders to be built. If you offer them something else they turn their noses up. If you are not far into the dialogue though (and they are not being abrupt with you because of your dominance) you can often repeat the offer to trade and the original deal will still be on offer.

Curiously advances which allow WoW don't seem to lose their appeal just because the wonder has been built.

If you are researching a critical tech, want to trade, but don't want to give that advance away it may pay to embark on a round of diplomacy just before the advance comes in.

SlowThinker · December 19, 2000, 09:03

quote:

Originally posted by East Street Trader on 12-19-2000 06:30 AM
Problem 2, I'm pretty sure that if you get no tech by means other than research it starts 10 beakers, 20 beakers, 30 beakers and then the arithmetic progression takes on a bit more of a geometric aspect.

I noticed that in the beginning of the game ](I never continue playing further

, you now, I slowly

start to think

about the system of the game everytime. I prefere thinking over playing

) beakers grow up arithmeticaly with some "jumps". One of these jumps is between 5th and 6th advance, so it is very recommended not to risk any goody huts before the monarchy.

quote:

I suspect that the practical benefit of knowing what the underlying beaker rate is would be small.

If there are "jumps" through all game, it would be very usefull to know were jumps are. But, I am probably too radical micromanager and perfectionist...

Problem 5: Thank you for detailed answer. I was surprised i got an advantage over the computer. Can I pick from the list of computer's technologies at end stages of the game too?

SlowThinker · December 19, 2000, 09:16

quote:

Originally posted by skrobism on 12-19-2000 01:04 AM
re: your english.
Don't worry about it. First of all, it is about 97.5% correct. Second, I am sure that your English is much better than any of our Czech will ever be.

Firstly, I wanted to look modest.

Secondly, it would improve my english, if anybody would point to some big mistake.
I dont want to tell you you should start to analyze my posts and mail blunders in english to me

kcbob · December 19, 2000, 12:52

Ken, are you sure about Problem 3? I was hoping the white advances on my science advisor screen were those that some other civs did not have and consquently, it was alerting me to trade possibilities. I guess I just assume that white advances on my screen sometimes turned blue later on.

Slow, you are too modest. Your English is very good. Not perfect, mind you, but very good. Just be aware that not all of those who offer tips to you (me included) speak and write in perfect English themselves.

* Note to Apolyton - Any chance of getting a spell checker on the forums? *

------------------
Frodo lives!

Paul · December 19, 2000, 16:34

kcbob, Ken was right about #3. I think it is even in the manual. I have never seen white advances turn blue on my F6 screen. You may be confused with the F3 screen where white advances of other civs will become blue when you get them.

SlowThinker · December 23, 2000, 23:28

New version of civ2plan.exe was added to http://home-1.worldonline.cz/~cz0456...n.htm#civ2plan

Sequoya · December 28, 2000, 18:45

Problem 2

This was discussed in an old thread. I believe it is archived under CivII Strategy and is called "A List of Data."

That thread had three important bits of information.

First, the number of beakers required for each tech is an integer multiplied by the number of the tech to be researched. First tech is A1*1, 2nd is A2*2, 3rd is A3*3, 4th is A4*4 ... 90th is A90*90. Starting techs are not included in this calculation. (The starting value for the integer is dependant upon map size and difficulty level.)

Second, the integer can be increased each research event and is increased even more if you are ahead in techs. This is the theory behind the crumby tech trading strategy and the gifting of techs in OCC. (The integer is increased every third tech after the 20th.)

Third, the integer ALWAYS takes a massive jump after the 19th tech. The 20th tech costs significantly more that the 19th. This hasn't been discussed in a while but is important (along with oedo's Law) for streamlining research paths. It is nice to get your critical research done by the 19th tech (Gunpowder, Democracy, Theory of Gravity or whatever you prefer.)

Sequoya

[This message has been edited by Sequoya (edited December 29, 2000).]

La Fayette · December 29, 2000, 12:38

ST,
5 problems in one post! How can we have a life outside civ?
Problem 1: It seems that oedo or you or both are close to the truth. I cross my fingers.
Problem 2: Sequoya's answer is OK for me. Perhaps a clever OCCer like Paul would be able to give more precise ideas about the lag between your research and the AI's.
Problem 3: Ken answered
Problem 4: SG answered
Problem 5: I havn't got MGE. In the version I play (copyright 1996, is it 2-42? I must check) I know which tech the AI wants and, if I have an embassy or MPE, which techs it is willing to offer. Otherwise I know one tech offered immediately, but I must say yes before discovering the other techs available, if any. Then the AI gets the tech it asked for and I get the one I choose.

------------------
aux bords mystérieux du monde occidental

SlowThinker · December 29, 2000, 15:23

quote:

Originally posted by La Fayette on 12-29-2000 11:38 AM
Problem 5: I havn't got MGE. In the version I play (copyright 1996, is it 2-42? I must check) I know which tech the AI wants and, if I have an embassy or MPE, which techs it is willing to offer. Otherwise I know one tech offered immediately, but I must say yes before discovering the other techs available, if any. Then the AI gets the tech it asked for and I get the one I choose.

La Fayette,
I works the same way as my MGE. I dont understand it: I have heard that AI often cheats, and then it gives me that advantage. And why computer offers a concrete advance, and then let me choose any? Its not reasoned.
[This message has been edited by SlowThinker (edited December 30, 2000).]

SlowThinker · December 30, 2000, 01:29

quote:

Originally posted by Sequoya on 12-28-2000 05:45 PM
This was discussed in an old thread. I believe it is archived under CivII Strategy and is called "A List of Data."

First tech is A1*1, 2nd is A2*2, 3rd is A3*3, 4th is A4*4 ... 90th is A90*90.

Sequoya,
Unfortunately, "A list of data" is lost. I found only
archive/Civ2-Strategy/The beaker formula
with following reference:
There are some information on this topic in the thread "A list of data" by Caesar the Great with the latest update on May 13 in this forum.

But, hopefully

, it wasn't lost out of your memory.

Do you want to say that A1, A2, A3 are not found out in general situation/game?

SlowThinker · December 30, 2000, 12:28

Sorry, there was a freezing bug in Research planner version 0.2. New repaired version was added
to my page (link) now.

[This message has been edited by SlowThinker (edited December 30, 2000).]

Sequoya · December 30, 2000, 16:16

SlowThinker,

I searched for that old thread too, but I couldn't find it. It must of got lost in the archive disaster. (Just like the burning of the Great Library of Alexandria.)

Fortunately, I had pasted the actual "list of data" into a Lotus 1-2-3 spreadsheet for my own analysis. This data is listed below with my comments.

This data was based upon a cheat mode test. A game was started (I believe 2.42, large map, diety, raging hordes.) Cheat mode was enabled and techs were gifted one at a time and the beaker cost of each tech was recorded. Doing it this way caused a massive lead in techs that the game penalized Caesar for. These would not be the real costs in an actual game. Different game settings would give different starting values for A (as well as increment values.)

I have verified the behavior of beaker cost at emperor, king and chieftain levels. This was to confirm the beaker cost jump at the 20th tech.

To test the beaker cost in an actual game, set your research rate to 0% and check the number of turns required to research the present tech. Don't forget to reset your research rate.

"A List of Data " with notes

1. 12 = 1 * 12, A increment = 0
2. 26 = 2 * 13, A increment = 1
3. 45 = 3 * 15, A increment = 2
4. 60 = 4 * 15, A increment = 0
5. 75 = 5 * 15, A increment = 0
6. 102 = 6 * 17, A increment = 2
7. 126 = 7 * 18, A increment = 1
8. 144 = 8 * 18, A increment = 0
9. 180 = 9 * 20, A increment = 2
10. 210 = 10 * 21, A increment = 1
11. 231 = 11 * 21, A increment = 0
12. 264 = 12 * 22, A increment = 1
13. 299 = 13 * 23, A increment = 1
14. 350 = 14 * 25, A increment = 2
15. 390 = 15 * 26, A increment = 1
16. 432 = 16 * 27, A increment = 1
17. 459 = 17 * 27, A increment = 0
18. 540 = 18 * 30, A increment = 3
19. 589 = 19 * 31, A increment = 1
20. 820 = 20 * 41, A increment = 10

Note big increase in research cost.

21. 903 = 21 * 43, A increment = 2
22. 946 = 22 * 43, A increment = 0
23. 989 = 23 * 43, A increment = 0
24. 1080 = 24 * 45, A increment = 2
25. 1125 = 25 * 45, A increment = 0
26. 1170 = 26 * 45, A increment = 0
27. 1242 = 27 * 46, A increment = 1
28. 1288 = 28 * 46, A increment = 0
29. 1334 = 29 * 46, A increment = 0
30. 1440 = 30 * 48, A increment = 2
31. 1488 = 31 * 48, A increment = 0
32. 1536 = 32 * 48, A increment = 0
33. 1680 = 33 * 50, A increment = 2
34. 1700 = 34 * 50, A increment = 0
35. 1760* = 35 * 50, A increment = 0

I believe Caesar the Great confessed a possible error in the 35th data point. It was decided that the correct beaker value was probably 1750.

36. 1872 = 36 * 52, A increment = 2
37. 1942 = 37 * 52, A increment = 0
38. 1976 = 38 * 52, A increment = 0
39. 2028 = 39 * 52, A increment = 0
40. 2080 = 40 * 52, A increment = 0
41. 2132 = 41 * 52, A increment = 0
42. 2310 = 42 * 55, A increment = 3
43. 2365 = 43 * 55, A increment = 0
44. 2420 = 44 * 55, A increment = 0
45. 2520 = 45 * 56, A increment = 1
46. 2576 = 46 * 56, A increment = 0
47. 2632 = 47 * 56, A increment = 0
48. 2784 = 48 * 58, A increment = 2
49. 2842 = 49 * 58, A increment = 0
50. 2900 = 50 * 58, A increment = 0
51. 3060 = 51 * 60, A increment = 2
52. 3120 = 52 * 60, A increment = 0
53. 3180 = 53 * 60, A increment = 0
54. 3294 = 54 * 61, A increment = 1
55. 3355 = 55 * 61, A increment = 0
56. 3416 = 56 * 61, A increment = 0
57. 3591 = 57 * 63, A increment = 2
58. 3654 = 58 * 63, A increment = 0
59. 3717 = 59 * 63, A increment = 0
60. 3900 = 60 * 65, A increment = 2
61. 3965 = 61 * 65, A increment = 0
62. 4030 = 62 * 65, A increment = 0
63. 4158 = 63 * 66, A increment = 1
64. 4224 = 64 * 66, A increment = 0
65. 4290 = 65 * 66, A increment = 0
66. 4422 = 66 * 67, A increment = 1
67. 4489 = 67 * 67, A increment = 0
68. 4556 = 68 * 67, A increment = 0
69. 4830 = 69 * 70, A increment = 3
70. 4900 = 70 * 70, A increment = 0
71. 4970 = 71 * 70, A increment = 0
72. 5112 = 72 * 71, A increment = 1
73. 5183 = 73 * 71, A increment = 0
74. 5254 = 74 * 71, A increment = 0
75. 5400 = 75 * 72, A increment = 1
76. 5472 = 76 * 72, A increment = 0
77. 5544 = 77 * 72, A increment = 0
78. 5850 = 78 * 75, A increment = 3
79. 5925 = 79 * 75, A increment = 0
80. 6000 = 80 * 75, A increment = 0
81. 6156 = 81 * 76, A increment = 1
82. 6232 = 82 * 76, A increment = 0
83. 6308 = 83 * 76, A increment = 0
84. 6552 = 84 * 78, A increment = 2
85. 6630 = 85 * 78, A increment = 0
86. 6708 = 86 * 78, A increment = 0
87. 6786 = 87 * 78, A increment = 0
88. 6864 = 88 * 78, A increment = 0
89. 6942 = 89 * 78, A increment = 0
90. 7290 = 90 * 81, A increment = 3

Thanks to Caesar the Great, and others, for their original research.

As some may be able to guess, I have been lurking around Apolyton for a long time. I registered for an account only a couple of months ago and don't post often, which is why I am just a settler.

Sequoya
[This message has been edited by Sequoya (edited December 30, 2000).]

SlowThinker · December 30, 2000, 21:37

sequoya,
did they do the same test (starting position saved and then loaded ) twice?
You know, different results may be caused
a) by different conditions (size of map, tribe...)
b) or increment (0,1,2...) is always random.

SlowThinker · January 1, 2001, 16:19

I have found "A list of data" thread: http://apolyton.net/forums/Forum3/HTML/000329.html

Sten Sture · January 2, 2001, 20:43

I can't recall if it is recorded in the first List of Data thread, but someone noticed that the current number of AI civs also impacts the research cost in addition to the map size, diff level and relative advance level.

East Street Trader · February 1, 2001, 09:39

There is another point which emerged from CtG work on beaker count. The progression is slightly different for large maps and for small maps. Thanks to Carolus RexThe numbers and the discussion is available from two of the threads he has added to page 3 of the Great Library.

What remains unquantified is the precise effect of the tech lead modifier. It may be that testing that aspect will prove a challenge even for the extraordinarily dedicated band of scientific investigators whom we now boast (and for whom lazy fellows like me give heartfelt thanks). OCC experience, though, pursuades me that the modifier has a very substantial effect, not just a trivial one. And, suspicion only, I think it is a balance between the human player and all the A1 civs which is relevant, not just the difference between the human player and the next most advanced. I feel reasonably confident that all of my (numerous) tech gifts have some effect, not just the gifts made to one particular A1 civ.

SlowThinker · February 3, 2001, 14:46

Beakers count:

I tested with 2 civs and cheat menu:

1. Results are not correct using Ctrl-shift-F6 (for example first advantage was ignored (like a free advance)).
2. There is a sparsely applied random factor (i.e. there is no explicit formula ).
3. shift-F6 combined with shift-F3 gives other results than pure shift-F6 (i.e. remote shift F6 for the opponent)
or point 2. applies
4. It depends on attack, civilize, expand factors (from rules.txt) of both civs or point 2. applies

SlowThinker · February 4, 2001, 08:34

BTW, I changed a line

10 ; Tech paradigm (higher # slows research)

to 20 and to 50 and nothing happened within first 10 advances.
[This message has been edited by SlowThinker (edited February 09, 2001).]
[This message has been edited by SlowThinker (edited February 09, 2001).]

debeest · February 14, 2001, 03:08

Slow, in my 2.42, nothing happens when I press Ctrl or Shift along with an F-key. Can't tell what you're talking about in your next-to-last post.

Here's a small tip to enhance your fine English: I think you're using "yet" in places where most English speakers would use "already." Take a look in your Czech-English dictionary (which, for political reasons, didn't exist when I visited your lovely country and your lovely city in 1986.)

[This message has been edited by debeest (edited February 14, 2001).]