DESIGN: Tech Tree (Broad Structure)

Message

Aquitaine · #1 Post by **Aquitaine** » Thu Dec 11, 2003 5:56 pm

Ladies, gentlemen, Nightfishes:

...here we go. :twisted:

I am purposefully dividing the tech tree design into small, chewable pieces. It will be tremendously easy for us to get carried away with this, and we need to be able to separate the game design from the sci-fi stuff. Consequently, we will probably have at least three phases of design for the tech tree.

1) This one - structure of the categories. While the number of categories and which technology lies in which category is probably mutable (or should be), we need to figure out how to define them; that is what this thread is for.

2) What should the categories be, and what parts can we begin to populate now and what parts have to wait?

3) A system for submitting actual techs. This will be last and will be ongoing all the way through the project. I'm going to postpone talking about this since it'd be premature to do it now.

Anyway, on to #1: CATEGORY STRUCTURE.

There are two possibilities that I see here (although feel free to recommend a third). The first is that we have 'generic' science categories ala the MOO series: Physics, Mathematics, Biology, etc; this will result in a small(er) number of categories with lots of techs in each one. This in turn has the effect of reducing the amount of cross-category requirements we can have, depending on how far down a particular branch we let people go without researching everything above it.

The second is that we make the categories more specific when we need to: for example, off the top of my head, we might have:

Fighters
Corvettes
Light Capital Ships
Heavy Capital Ships
Super Capital Ships
Agriculture
Industry
Culture
Espionage
Economy
Political Science / Xenoscience

This allows for many more branches on a smaller tree.

What we are deciding here is which format -- /not/ the specific categories (that will come next). So please, post your comments and arguments in favor of one or the other (or a third way) but remember what we've already passed: it will be based on a tech tree similar to Hearts of Iron. If you don't know what that is, I will see about posting a full description, but I think it should be in the passed features thread or the game design archive if we talked about it after the last forum move.

Let the chaos begin.

(and welcome back to design mode, everyone!)

Sandlapper · #2 Post by **Sandlapper** » Thu Dec 11, 2003 8:14 pm

@ Aquitaine

I have not seen or played Hearts of Iron, please post a description. I have been following FO since the discussion of trying to save the old forum. Hearts of Iron has been mentioned several times, but I don't recall a specific description. Thanks in advance, if you can provide this.

#3 Post by **Tyreth** » Fri Dec 12, 2003 5:32 am

From a programming perspective we should make it so that either system can be implemented via text files if that is what others making their own variants of freeorion wish to do. That's a little off topic.

Can't make up my mind yet, but I will throw out some thoughts. One possible problem is if we want to give specific research based advantages. If we have 6 or less categories (such as biology, chemistry, etc), it is easy to give races or per tech advantages. Eg, +20% to biology research. However, when there are more categories, such advantages are lessened, can become numerous, and not as clean to say +15% Super Capital Ship research.
This point may be moot if we don't offer field specific advantages, but it would be nice if it were possible.

On the other hand, the second category method makes it easier to have many cross dependencies without relying too much in researching out the entire base of the tree. It also allows for more ultra specialisation. That in itself makes it easier for the game to be designed in such a way that some races may never get high in a particular field at all. This would produce a greater variance of technology. One race may specialise in corvettes and heavy capital ships, to the detriment of other spaceship fields.

Then there's aesthetic considerations, I find the single word names to be more appealing (physics, mathematics, etc) than mixed single and multiple word names (corvettes, light capital ships, etc).

Just some thoughts for now.

PowerCrazy · #4 Post by **PowerCrazy** » Fri Dec 12, 2003 6:51 am

I favor a more general tech tree. We can still have numerous cross dependencies, and it is a "cleaner" look.

As for the cross-dependencies
We can say that advanced plasma missiles require lvl 4 propulsion, lvl 2 construction and lvl 8 weapons. for example. Or we can have several branches of a specific field that must be researched before plasma missile is availible.

Basically restricting the tech tree two only 8 primary branches or 6 or whatever will not decrease our options, it seems to just make research more direct, and intuitive. Also like Tyreth said, it would increase racial advantages. +20% to biology is a tangible bonus and its advantage can immediately be seen. Whereas +20% to "capital ships" may take awhile to realize. especially if there are 20+ intial branches.

So we can still have as many cross dependencies as we want. And it would have more of a "moo" feel to it.

#5 Post by **Tyreth** » Fri Dec 12, 2003 6:57 am

PowerCrazy wrote:We can say that advanced plasma missiles require lvl 4 propulsion, lvl 2 construction and lvl 8 weapons.

Are you a Stars! player?

Cross dependencies can be done in two ways:
1. Requiring a specific level in an area, like you say here
2. Requiring a specific technology, much like SMAC.
I prefer the latter. With more general categories, I don't think it's all that bad that in order to reach the higher rungs you eventually need to explore most of the lowest technology.

PowerCrazy · #6 Post by **PowerCrazy** » Fri Dec 12, 2003 8:57 am

haha. No i've never played stars. I really should though. Also "level" can be done in a few ways.

1. Each lvl is a tech. Thus merculite missiles are lvl 2, nuclear missiles are lvl 1.
OR
2. There can be multiple techs in a level. And you need about half of them to be at that tech level. Or you need all of the techs in that level to be at that level.
OR
3. There is always the specific tech that is needed to get the next tech. i.e. Civ or SMAC.

I kinda like a little less concrete. Some kind of combination of 2. That way there aren't specific techs that you HAVE to research, and it allows for a little more variety. Thus i don't need to get literacy before i get The Republic, I could get writing, and something comparable to literacy, and then get republic. 2 of course would be harder to balance, but i think it would be more rewarding, and create a more varied playstyle.

Also i should prolly play HoI sometime. But, x-mas break is coming up so we'll see.

Daveybaby · #7 Post by **Daveybaby** » Fri Dec 12, 2003 12:34 pm

I've been thinking about tech quite a lot lately for my own design, especially about where moo3 went wrong, and IMO you should be very careful about tech interdependencies.

The problem with having one specific tech require levels x,y and z in schools A,B and C is that you tend to end up with EVERY school being equally important. i.e. there is no point in focussing extra effort into, say, physics - because on its own physics is useless. In order to get to actual usable techs you will need good levels in other schools as well.

Thus in moo3 you ended up with there not being much significant advantage in favouring one tech school over another. That and the general obfuscation of needing multiple schools just meant that the majority of players just gave up in confusion and plonked all techs at the save level of investment.

Compare this to Moo1, where the tech schools were more applied than theoretical... i.e. force fields, weapons etc. Players could see direct results from focussing on one school over another. I'm not saying that the Moo1 approach is the way to go - its just a comparison.

The final straw in the moo3 approach was in not letting players choose which applied techs to research. The final result being that there was effectively no meaningful player input to tech other than the overall spending level. Woo. :roll:

#8 Post by **Tyreth** » Fri Dec 12, 2003 1:49 pm

Having specific techs as a prerequisite was not a problem in SMAC. It did mean that later in the game you had filled the tree. However, each technology was somewhat special. For example, finally researching missile technology changed the way war was waged. Again, researching aircraft did the same. So later on you did not much care whether your tech tree was full our not. Nor did you think that it did not matter where because they ended up the same. What was important was getting that tech a few turns earlier, and if you were taking too long then to research or steal it off someone else.

You always ended up with tech eventually, whether through espionage, trade, conquest or exploration. I found that the way particular technologies changed the way you played the game, giving a distinct feeling of ages, was more important than whether the tech tree forced you to have all lower techs or not. Having missiles for a few turns more than your opponent was what mattered.

Seems like we're straying mildly off topic, but it's all related and helpful. I really loved the unique feel every tech had. It wasn't "woohoo, level 4 shields". It was "woohoo, high energy chemistry". Each technology had a quote and thought gone into it, and was more than just a new weapon, or a new ship. I like that immersive feel the technology gave, and I think we should reflect that.

I too am unfamiliar with the HoI model, so think it would be worthwhile Aquitaine if you could post up a description. I can't locate anything in the forums, but I did not look too deep either.

I am tending towards more general descriptions, since they encourage a more immersive feel to the technology, rather than just the cold mechanics of "heavy capital ship" or "espionage". I think general technology headings promote the attitude that technology is about sweeping breakthroughs in society, which often include multiple advances in the one "tech". The more specific categories turn the tech tree into something far more specific, representing a smaller unit of breakthrough - the specific tech is produced.

In case what I'm saying is not clear... in SMAC there is a technology called "Biogenetics", for example. This represents a breakthrough, "Understanding the human genetic code", and that breakthrough opens up to you all the possibilities that that breakthrough offers. It gives you the possibility to build the Human Genome secret project, and the base facility Recycling Tanks. Under the second proposed category system, the more likely culture would be that the individual components of such a breakthrough are researched, rather than the general category "Biogenetics".

utilae · #9 Post by **utilae** » Fri Dec 12, 2003 10:24 pm

PowerCrazy wrote:Also "level" can be done in a few ways.

1. Each lvl is a tech. Thus merculite missiles are lvl 2, nuclear missiles are lvl 1.
OR
2. There can be multiple techs in a level. And you need about half of them to be at that tech level. Or you need all of the techs in that level to be at that level.
OR
3. There is always the specific tech that is needed to get the next tech. i.e. Civ or SMAC.

Some kind of combination of 2. That way there aren't specific techs that you HAVE to research, and it allows for a little more variety. Thus i don't need to get literacy before i get The Republic, I could get writing, and something comparable to literacy, and then get republic. 2 of course would be harder to balance, but i think it would be more rewarding, and create a more varied playstyle.

I think that to advance to the next tech level you need half the techs in the current level (option 2, I think). Though you could have many different ways to get a tech (some ideas later in this post).

Also if you had missile level 1 and missile level 2, instead of researching them as seperate techs in the tree, you would only research missile. The point is that you could use the refinement part of the tech tree to develop level 2 missile tech. However refinement would work I don't know (maybe automatically).

In this way we could have very unique techs and not repeat the same techs over and over in the tree, ie Shield 1, Shield 2, Shield 3. Instead you have Standard Shield, Heat Shield, Other varients of shield (each able to be advanced in level using the undecided refinement system).

Also about tech dependicies - techs could be dependant (ie you can get the tech if you have ...) in the following ways:
- You have researched certain techs, ie laser to get shield.
- Refinement level, ie laser has been refined to level 5 to get fusion cannon.
- You have researched general tech types, ie researched 40% of weapon techs to get plasma cannon.
- You have reached level X in a category, ie level 35 physics to get mega beam cannon.
- You have a leader of a certain race or are a certain race or have captured a certain race to research race specific techs.

- can't think of any more, anyway some ideas.

Ablaze · #10 Post by **Ablaze** » Sat Dec 13, 2003 12:07 am

Doesn’t the issue of how to overdrive and underdrive development have to be solved for production anyway? It seems to me that the same mechanism used by planets to build things faster or slower could be used in the technology system to research faster or slower. If the system must be made, why not use it more then once.. or are we not going to have a mechanism for overdriving production on a planet?

Krikkitone · #11 Post by **Krikkitone** » Sat Dec 13, 2003 9:07 pm

I'd say several (4-8) Theory categories
each with a few (3-6) types of applications coming from it
each with a few (2-4) types of refinement for each

That provides as many as 24-200 projects you can be working on at one time.

However, I would also give applications a 'category label' so that races could have bonuses on certain 'types' of applications.

INCREASINGLY OFF STATED TOPIC DISCUSSION STARTS HERE

I'd also say allow the start project and stop project simplicty but have a general research 'funding level' that automatically overdrives if the required RPs /turn are more than required (quite possible to do withsome diminishing returns function.

Essentially give every project a # of turns and a RP/turn. If the total RP/turn for all funded projects is greater than needed there is some gain in the turn progress (each project gains more than one 'turn' in that turn). If the total RP/turn is less than needed there is a loss in the turn progress (each project gains less than one 'turn' in that turn).

With the right formula to make those rather seriously diminishing returns, you can more and more constrain the spending towards the all or nothing model of HoI or the RP requirement only of MOO.

The other problem is the 'entangling' of Creativity and research output. If we have a system where you choose to start and stop projects then you can make the choice to move rapidly and get a few techs or move slowly and get multiple techs. For HoI it sounds like a 'race' with a high RP output would be forced to be more 'Creative' researching more projects, rather than being more advanced.

In that case, even if we have diminishing returns, that means a 'Creative' Race would be one that has a high RP output, and a 'Research' race would be one that decreases the turn requirement for getting a tech. (possibly by their techs advancing more than one turn of project progress per game turn.)

Aquitaine · #12 Post by **Aquitaine** » Sat Dec 13, 2003 10:29 pm

Ablaze wrote:Doesn’t the issue of how to overdrive and underdrive development have to be solved for production anyway? It seems to me that the same mechanism used by planets to build things faster or slower could be used in the technology system to research faster or slower. If the system must be made, why not use it more then once.. or are we not going to have a mechanism for overdriving production on a planet?

There is no mechanism to do this at the moment, and while we might add one, there is no explicit reason to use it twice as they are completely separate interfaces; you build a thing on a planet, but research is done empire-wide.

Aquitaine · #13 Post by **Aquitaine** » Sat Dec 13, 2003 10:30 pm

Guys,

Off topic posts on this board will be deleted. Please do not go off topic in these posts at all, because it makes it very difficult for the design team to figure out what a consensus is when we have to read irrelevant material. I'll leave what's been posted, but in the future, all posts with any 'off topic' section will be entirely deleted.

Impaler · #14 Post by **Impaler** » Sat Dec 13, 2003 11:14 pm

Atempting to stay on Topic here:

Aquitaine refers to MOO style Catagories (physics, Ecology, Shields, Computers) and a large set fo more specific fields (Corvettes, Idustry, espionage ect ect). I dont see why we can not mix these two lists together and have a relativly large number of fields that cosnist of very broad terms like "mathimatics" and "weapons" and more specific things like "Gravitonics" or "Genetics". Specific Tecnologies would then become avalible for research once a combination of two or atmost three of theses "fields" is reached.

For Example:

Level 6 Optronics and Level 2 BioEnhancment give access to the Aplied Research goal Ocular Implants.

The critical differnce from traditonal MOO based fields would be that new fields open up from level combinations of other fields much like Aplied Tecs open up. This brings in some of the elements of Civ and SMAC as these new fields spice up the game, give a sense of acomplishment and wow (I have discoverd the Gravitonics field!). And to keep things from getting cluttered some of the early fields would "finish", the number of levels in a field could varry some having just a hand full, others around 20 that are importnat the whole game (like Mathematics).

So in total we would have roughly.
~30 fields witch can have several levels each, (maybe half open at start of the game)
~100 levels total for the fields
~300 Aplied Tecnologes and goodies resulting from combinations of field levels, so each field has on average 10 or so major goodies, with many of the goodies being refinable and usefull for much of the game.

As Tyreth points out the greatness of the SMAC tree is due to the highly focused tecs which have a great deal of personality and they give "portfolios" of goodies. In this model the whole field is like the SMAC tec and you individual reserach the items of the Portfolio as Aplied Tecnologies.

At anyone time the player has the option of Reseraching an Aplied Tecnology for Which they have the nessary field levels OR advancing any open field to the next level OR refining an existing aplied tecnology to a higher level.

mart7x5 · #15 Post by **mart7x5** » Tue Dec 16, 2003 1:40 am

I thought maybe a discussion pertaining to how broad science is supposed to be in FO could have place also. What i mean is whether stress will be put on technologies or we will have also better developed sociology topics. I am just listing some ideas for editing......:

-Mathematics
-Physics
-Informatics(computers) (maybe together with math?)
-Chemistry
-Biology(maybe including ecology, or not?)
-Geology(also planetology and similar-in general pertaining to rocks)
-Astrology(like cosmology also)
-Sociology
-Psychic abilities(mind reading, thought control, star trek Betazoids...)
-Fine Arts ??? that would make people more peacefull and create talents for research
-Mechanics(like ship mechanisms or Androids)

Weapons technologies would be mostly combination of development of physics, chemistry and computers. biological weapons, of course Biology. planets handling, like Geology...

There might be subgroups or maybe separate groups, e.g in sociology:
-crowd psychology(usefull when planet gets dense with people)
-military psychology
-interracial relations
-trade psychology

in computers:
-artificial inteligence
-Algorithms
..........