DESIGN: Learning Category

Creation, discussion, and balancing of game content such as techs, buildings, ship parts.

Moderators: Oberlus, Committer

Post Reply
Message
Author
Aquitaine
Lead Designer Emeritus
Posts: 761
Joined: Thu Jun 26, 2003 1:54 pm
Location: Austin, TX

DESIGN: Learning Category

#1 Post by Aquitaine »

This thread contains guided discussion for the contents of the Learning category. It is moderated by Bastian-Bux.

Specific rules for this thread as set by the moderator are as follows:

Stay civilized.
Think.
Try to learn. ;)

Away from that we came to the following requirements for suggestions:

- send them as XML (read design and effects document first)
- requirements rules (no refinement except for refinement...)
- no changes to current_science, or Geoff bites your head off ^^
- only small changes to max_science. If its more then +3 for a late tech you'll have to answer to me personally ^^

PS: again edited by Bastian-Bux
Surprise and Terror! I am greeted by the smooth and hostile face of our old enemy, the Hootmans! No... the Huge-glands, no, I remember, the Hunams!

Bastian-Bux
Creative Contributor
Posts: 215
Joined: Fri Jun 27, 2003 6:32 am
Location: Kassel / Germany

Maximum RP multiplier

#2 Post by Bastian-Bux »

OK, lets get a start.

The first think I would suggest we do, is get a rough feeling for overall RP.

The learning category will be the one category that increases the RP amount producable per population unit.

So while growth and construction will influence how much researchers you have, learning will influence how many RP each researcher produces.

So what I want from you guys (and girls) out there is data.

You all supposedly played many 4X games.

How was it working there?
How much RP could a researcher produce at end game compared to the start?
Did that "multiplier" work out?


Once we have a little base of data on that questions, we can decide on the maximum multiplier for FO. Thus we will have a guideline for the modifiers to be applied to our researchs.
Wenn du die Macht hättest die Geschichte zu ändern, wo würdest du anfangen. Und viel wichtiger, wo aufhören?

If you had the power to change history, where would you start? And more importantly, where would you stop?

LonghornXtreme
Space Floater
Posts: 31
Joined: Fri Jun 11, 2004 1:54 am

#3 Post by LonghornXtreme »

One thing to keep into consideration is that efficient and profitable research is not always based off of population numbers. To a certain extent population does help... but the overall prescence of said research community in each empire is more important than the overall numbers of researchers. Mid to late twentieth century USA springs into mind. Imperial Britian of a few centuries ago... population was never the limiting factor with research... its money and aptitude of the research community.

I think that somehow this should be taken into account when setting the progression of RP's as an empire develops.

If I've overstepped a boundary of a design document I apologize... just trying to contribute my 2 cents.

I always like the Stars! method of research, i.e. converting a percentage of total production/economy to research. MoO II's model of combining population along with facilities is a good idea, but i think that the bonuses awarded by improving the facilities topped out way too early... i.e. I would've liked to have seen a very expensive facility that gave each researcher more than the +15 of the Research Lab + Planetary Supercomputer.

Space Empires IV's model is not population driven... RP's are generated solely by the number of 'research facilities' built... and these facilities are upgradeable.

With that said... I think MoO II was 'too' population amount driven in generation of RP's and I think that SEIV was not enough population driven for generation of RP's.

Does anyone think it would be better to design RP production amounts in FO if a rudimentary tech tree was lined out with approximate turn lengths to completion? Generating RP formula's with that template would be much easier imho i believe....

James

PS Not trying to start any flames :D Just trying to help... and if I've overstepped any design doc's or already settled ideas just yell at me and I'll hush :D

PowerCrazy
Creative Contributor
Posts: 383
Joined: Fri Jun 27, 2003 2:35 am
Location: Texas

#4 Post by PowerCrazy »

I assume that learning will also be the category that will allow tech to procede faster?

i.e. If I have a tech that takes 500 RP for 10 turns, then I should be able to research a tech that allows tech to procede at 1000 RP for 5 turns.

In moo1 the amount of research done was 8 times per pop unit. Also the population of each planet doubled by the end of the game. Thus for a fully maxed out planet you would be doing 16 times the research as beginning planet set to a research focus.

That seems like a good "multiplier" to me. In moo2 I think 1 scientist produced 3 RPs at the beginning, and some thing like 15 at the end, but they had all the buildings that would add a base amount of RPs such that at the end a fully maxed planet would be producing something like 150rp while in the beginning it would produce around 15. Again it may be different, but that is what I remember.

Really though the multiplier can be as high as we want it to be as long as we adjust the tech costs accordingly.
Aquitaine is my Hero.... ;)

LonghornXtreme
Space Floater
Posts: 31
Joined: Fri Jun 11, 2004 1:54 am

#5 Post by LonghornXtreme »

I really think that RP generation numbers really need the tech tree... or at least an alpha of the tech tree to make an attempt at balancing how RP generation increases as the game progresses due to population growth, and prior applied technology...

Because without that tree we might as well just spout out random numbers that increase...

User avatar
skdiw
Creative Contributor
Posts: 643
Joined: Mon Sep 01, 2003 2:17 am

#6 Post by skdiw »

Crap, forgot to log in before I posted above ><

I forgot to add one more thing, if you wanted to consider the rate of colonization, using the raw data from the other thread, you just have to multiply the formula by e^(.07 x n), because the RP generation approx doubles by end game independent of techs and ln 2 is 0.7, spliting 0.7 into ten percent of the tree gives you 0.7 x n.


Some numbers from GalCiv: ratio of max tech and lowest is 5000, but that number is high since the first set of tech are give-me and last tech is Final Frontier, which you win via research. A more resonable from lvl 2 tech and lvl last tech that acutally does stuff is about 500, which took about 20 turns to research and I generated about 6X more RP compared to beginning of the game. Here, the mulitplier is about .2 and the expo I calculated is about 1.75.
:mrgreen:

drek
Designer Emeritus
Posts: 935
Joined: Thu Jun 26, 2003 8:07 am

#7 Post by drek »

Something else to consider, re: learning category.

In addition to research, the orginal intention was for it to be a catch-all category of the big techs needed to advance from one "age" to another. Also, if we have game-ending techs (like SMAC's tech end game) they'd probably belong here.

User avatar
utilae
Cosmic Dragon
Posts: 2175
Joined: Fri Jun 27, 2003 12:37 am
Location: Auckland, New Zealand

#8 Post by utilae »

drek wrote:Something else to consider, re: learning category.

In addition to research, the orginal intention was for it to be a catch-all category of the big techs needed to advance from one "age" to another. Also, if we have game-ending techs (like SMAC's tech end game) they'd probably belong here.
Even though it is named learning, I called it 'information gathering' when we were finalising the categories. This allowed for some other techs to come under the 'information gathering' category, eg spying, scouting.

So I was wondering, is learning the same as information gathering, so that spying, scouting and any kind of information gathering, processing and storing based techs could come under it.

User avatar
Geoff the Medio
Programming, Design, Admin
Posts: 13587
Joined: Wed Oct 08, 2003 1:33 am
Location: Munich

#9 Post by Geoff the Medio »

I too have been wondering what is the intended scope of the various categories... this one in particular. I assume Aq had a particular idea in mind when deciding the names, however it's not immediately clear to me what the plan was from the one-word names. IMO it'd be helpful if Bastian-Bux ammended the explanitory text in his first post to explain explicitly the scope of the category... including both what is in, and what might be mistakenly thought to be in when it really isn't...

Bastian-Bux
Creative Contributor
Posts: 215
Joined: Fri Jun 27, 2003 6:32 am
Location: Kassel / Germany

#10 Post by Bastian-Bux »

@Geoff I'd like to if I'd know myself :).


AQUITAINE


As I didn't participate in the latest stages of the decision on categories, I have to wait together with you for answers from our glorious team leader. ^^

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

Longhorn poked his fingers in the wound I was to look at together with you first :). Sorry if I was unable to explain that clear enough.

Away from all other implications the learning category will define first and foremost how fast we accumulate RP, and in which way this is done.

As he very correctly pointed out, research is affected by two things: personal genius (population) and usable ressources (money).

Now my intention is to find a consensus together with you, how much each of those two vetors will influence the overall RP development.

I'll try to explain a bit what I mean by those two vectors:

Everyone knows a gauss bell, and also knows that high intelligence isn't the most common in our population. Thus the higher a population of a given nation, the higher the chance to have a shining genius amongst them.

But which parts of research are dependant on personal genius? I'd postulate that theories benefit most from personal genius. Why? Cause to develope a theory you need a very high amount of creativity, which in it self is one sector of Intelligence (read Renzullis three factor modell for more on this).

Looking around me I'd say this vectors is fairly time stable, influenced by only a few factors. The most important for FO in my opinion would be genetics, thus a very uncreative race would have a lower number of personal genius then a creative race. Thus I like the naming in the MoOs ^^.
Another factor would be the way of socialising. You know the PISA study? it shows fairly well which contemporary states have a school system that encourages creativity, and which states are rather on the repetitive side.
A third factor would be genetical again: genetic manipulation for traits like creativity or dullness.

IMHO this vector should be fairly hard to influence in-game, but should also be the only way to lower the turn requirements of research (if we even allow any lowering of them). Only a spark of genius can find a "short-cut" to a given research.


So whats most influenced by usable ressources? Again I make a postulate (open to discussion of course): an high amount of ressources invested into research is most usefull for refinements. Reason behind this is its short term profitability. AFAIK most research money at the moment is invested into refinement rather then into theretical research.
Of course, a new microwave oven that has a 4% faster internet browser then the "old" one is profitable much faster then a breakthrough in temporal physics.

So how to influence this vector? Fairly easily: its mostly about making "research" more cost efficent. Developing a set of rules how research is to be conducted is one way. Internet, to allow scientists easier access to their fellow scientists work another. Plagiarism is so nice, especially as it allows many scientists to spill out lotsa new refinements fast ^^.

IMHO this vector should be much more easy to build up then the personal genius one.

So what we are looking at is a 2 vector matrix, that influences severall effects. The stereotypical research race would max both vectors, and achieve the max of RP per turn.

But whats the max? By what factor can intelligence be increased? And by what factor can money increase the research output?

Which effects shall be affected by our researches?

- amount of RP per population
- "free" RP as result of a lab?
- turn requirements of researches?
- generall RP/turn requirements of researches?
- or differentized into theory, application and refinement?
- something I didn't come up with?

Well, I need input, and I suppose I gave you some as well. ^^

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

@Powercrazy

IIRC we had an agreement that the turn requirement of a research shouldn't (should hardly) be changeable.

I myself ain't sure yet if it should be totally impossible, or be possible, but only marginally (10-20% decrease max?) and only due to "creatification/genialisation" of your population.

Lets discuss it, to get a creative result. :P

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

@Skdiw, I have to crunch through your numbers after breakfest ^^. Even a Psilon can't think with an empty stomach.

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

OK, we had a nice start. Powercrazy suggested a multiplier of 16. I myself think that this is a good "size", meaning that I myself would like anything between 9 and 25 (but not 100 for example).

Skdiw brought the multiplier down to a curve (which is also very nice, but which requires me to get the multiplier again ^^).

But more examples are needed.

Oh, and if you are at it, maybe look a bit closer if you can see the two vectors I described, and tell us, how those games did it. I don't need "hard" numbers, descriptions and feelings are enough. Something like:

"It looks that the overall multiplier was some 20ish. And I got the feeling that xy% of that was due to personal genius, and the rest through usable ressources."
Wenn du die Macht hättest die Geschichte zu ändern, wo würdest du anfangen. Und viel wichtiger, wo aufhören?

If you had the power to change history, where would you start? And more importantly, where would you stop?

Tyreth
FreeOrion Lead Emeritus
Posts: 885
Joined: Thu Jun 26, 2003 6:23 am
Location: Australia

#11 Post by Tyreth »

utilae wrote: So I was wondering, is learning the same as information gathering, so that spying, scouting and any kind of information gathering, processing and storing based techs could come under it.
This much I can answer (hopefully accurately) on Aquitaine's behalf, but he is most welcome to correct me :)

Espionage/spying is not in the "learn" category and is scheduled for 0.7. When we reach there it will be assigned an existing category or given its own. For now, assume it doesn't exist.

tzlaine
Programming Lead Emeritus
Posts: 1092
Joined: Thu Jun 26, 2003 1:33 pm

#12 Post by tzlaine »

Bastian-Bux wrote: - "free" RP as result of a lab?
Just FYI, the Effects system does not currently support this; if there is a great need for it we can add it, but it is sort of antithetical to the overall design. Tech buildings (and all other resource-oriented buildings) should usually only modify resource meter maxes, rather than supplying resources themselves.

User avatar
utilae
Cosmic Dragon
Posts: 2175
Joined: Fri Jun 27, 2003 12:37 am
Location: Auckland, New Zealand

#13 Post by utilae »

Some useful 'tools' (variables) for making learning techs:
----------------------------------------------------------------------
CurrentResearch
MaxResearch
CurrentPopulation
MaxPopulation
ScienceProduction
PlanetSize
PlanetType
PlanetEnvironment

Other Useful variables:
---------------------------
ObjectType
StarType
PrimaryFocus
SecondaryFocus
Owner
ID

Effects Document:
http://www.freeorion.org/wiki/index.php ... troduction

LonghornXtreme
Space Floater
Posts: 31
Joined: Fri Jun 11, 2004 1:54 am

#14 Post by LonghornXtreme »

OT and I apologize,

but I'm having trouble finding the .3 design docs... they're not on sourceforge like they use to be... and the wiki is only showing me the undetailed roadmap....

I think I'd be MUCH more useful in this thread if there was a consolidated place to read all that has passed concerning the tech tree and the RP structure we have in place so far....



/end OT

Tyreth
FreeOrion Lead Emeritus
Posts: 885
Joined: Thu Jun 26, 2003 6:23 am
Location: Australia

#15 Post by Tyreth »

LonghornXtreme wrote: but I'm having trouble finding the .3 design docs... they're not on sourceforge like they use to be... and the wiki is only showing me the undetailed roadmap....
It is there. Previously it has been hidden deep underground behind barred walls and laser enhanced shielding, with no less than three hundred Spathi patrolling at any one time. I've now moved it to a place where anyone can find it, on the front page.

But here is a link for your convenience:
http://www.freeorion.org/wiki/index.php ... quirements

Post Reply