FreeOrion

An excellent summary. Here are some thoughts.



As previously argued, I think 'no' on this one, but could be convinced; I simply don't see a positive reward for this.



I view this question less as 'many or few' and more as 'specific or broad' - that is, HoI-style specific categories would give us, for example:

Starfighters
Corvettes & Destroyers (Light Capital Ships)
Cruisers (Medium Capital Ships)
Battleships and Carriers (Large Capital Ships)
Xenobiology (Diplomacy/Assimilation)
Land Warfare (Infantry/Tanks)
Espionage
Military Infrastructure (Shipyards, Missile bases)
Civil Infrastructure (Banks, Govt.)
Industry


..and so on. In this sense, the 'categories' are to the individual techs as the theories are to their applications; they are fairly specific 'parents' to which everything belongs. Whereas a MOO2 style, broad category list such as:

Physics
Biology
Electronics
Propulsion

with one or two others -- that is really the choice. We could choose specific categories and still only have 6 or 7, or we could have 14 or 15. This number won't be finalized for quite some time, I suspect, but that's the decision I'm trying to get out of these threads.



I could be convinced about having an OR pre-requisite, but I think we sould establish a system and then play by it; offering 'ways around' the rules is tricky both in that it's an added layer of rules for the human player to remember, and its difficult for the AI to adapt to those rules.



I should say yes on both; which categories exactly and by how much we would have to balance later.



No. With a large tech tree, this would get ungainly as you would have to program in every tech to which the first tech is 'related' to; people would argue relationships where none should be and vice versa; all in return for which we are breaking the 'could not be operated by a human in a boardgame' rule for marginal returns.



I'm not sure I understand this; if it's 'should the tree be visible,' then I'd say yes.



This is fairly HoI-specific and generally relates to the HoI concept of 'organization' in battle. I don't think it applies to us.



HoI grants bonuses to categories based on the ministers you have in your government. Specific techs could be tricky as an easily mod-able tech tree could change techs around and render whole races inoperable. I would say, at least, allow bonuses to categories, and keep our options open for bonues to specific techs. But right now I can't think of a compelling reason why we would need to.



I'm not sure I understand this. I think each race and/or each starting strategic position should have preferred paths, but several of them.



See previous posts.



Two rules of thumb: KISS and Steal, Steal, Steal!

_________________
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!

This is very true, but the official team still has to manage the official release. The fact that we will support ten zillion mods is a wonderful bonus, but there's a lot of work to do.

_________________
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!

1. We need to think what to do about the spare RP's left over each turn (either another use for them (is this the place for refinement?) or bear in mind that they will need to be easily redirected back into the economy somehow.)

We can make rp funnel back to pp. That will be fine and it would make our econ model easier.

My other thought is to allow reinvestment of excessive rp into theory or cat. you have a stockpile of rp that you can use to make incremental investments. By reinvesting you get discounts for all the application if you invested in theory or for all theories in you invest in cat. You can think of it as refinement in theory just as you can refine the application in the game. On top of refining the theory or cat, this idea will also make prereq system easy while maintening mutliple path through the tech tree; you can say that to research a cross-discplinary application, you need certain levels of refinement in total in this and that theory or cat. The total is the key as it allows specialization in a specific field or strategy.

2. Randomly block out different techs each game or not?

There are two groups of techs: basic tech that affects food, min, pp, rp, and other basic elemental that are vital to every race and the game as a whole. And there are techs that are uniques or weaker techs that doesn't mean to do much by itself. For these techs, I think the game is more interesting if it's random.

3. Many categories (10+) or fewer (6-8 )?

If we are going to make cat to represent the multiple paths that one can take then i would prefer a few.

I don't think there's anything wrong with new cats spawning at a later point in the game however.

4. Will it be possible to develop applications without knowing the theory, if all the prerequisites have been met, maybe at some kind of penalty?
Should we have techs that reduce the cost or time of research generally (or for particular categories). ?

No, I would keep it simple as a rule that theory spawns the application. If we agree to this, then there should never been the case where "all the prereq have been met."

But as a crazy programmer, I always have some default that sets it to something cuz crazy bugs always pops up.

5. Should developing related (but not prerequisite) techs reduce the cost /time of developing a tech?

No. You can just make the related field cheaper. I think it just make it more complicated if you do it that way.

6. Do we want to repeat the outline of the tech tree to extend the research game?

I like recurrsion of the basic techs or Eras if that is what you mean. I highly doubt that these basic techs won't fit well with some function.

7. Should we try to hardwire choices with race bonuses to particular techs, or try and make it dependant only on strategic position.

Why not both?

8. Do we want to achieve multiple paths by balancing all techs perfectly, or by deliberately introducing several 'prefered paths' or tech families?
How can we produce genuinely alternative gameplay styles and support this with the tech system?

For the ease of balacing, I think most ppl here wants to balance the tech tree inherently meaning that the tree is balance irrespect to other parts of the game. You balance the paths by balancing the paths, not necessary each tech. I would prefer a few paths with a few other paths that are more risky. For simplicity, however, I'm happy with all the paths balanced. (See question 1)

9. How do we do this whilst keeping things relatively simple and fun?
how big/small do we want the steps of research to be, ie.e how much of adifference will each discovery make to your strategy and tactics?
Do we want to design in 'bursts' of discovery?

Ahh...One of the most basic question that must be answer I think before any other stuff about tech.

We all agree that we don't want rushing games. So I propose that the overall research path, ei. growth in rps terms, be prefered. Later in the game, the research can be more even and eventually be less cost-effective in the end game.

I say a powerful tech should in large discrete investment that yields a 10% to 15% gain in your overall game. In the end-game it can be a little less to promote moping up the game. It's hard to make research interesting if it's continuous and not is bursts, not to mention weird.

For the basic techs, I don't know how are ppl going to make them unique. I don't think making farming tech lvl 1 thats gives you 10% benefit and the next one be 50% at the same costs. No, it going to more or less follow a formula like a computer. So I say make it recurssive for those die-hard research fans that wants a game to go on forever.

None of you have criticize about the eras thing. It does a lot to the game if you entertain it for a moment. Instead of cats or on top of it, you can break the techs up in eras. Eras adds a lot more versitility (its a added on bonus, not additional pain). Eras gives you a reason to change UI skins. Eras allows you to change path if we wanted make it so. Eras resets the values and cost so you don't see +3495millino pp you generate every turn. Eras can save some work ahead of us by cutting the number to techs that we need to come up and balancing each of them. It also has other benefits as well.

10. What happens to research on projects you can no longer afford, or want to cancel?

To keep it simple and more commitment from the players, just say you can't cancel to project. And you must have enough rp in stockpile in total to even begin the research and after that you are locked-in.

We can do it other ways, but think by keeping it simple, we also encourage a better macro game.

11. Do we want more research points to mean more creativity (wider research) or highwer tech (more advanced research)? If the latter, how do we get it with this model?

Both. To start, we can just make the higher techs give more benefits. Say you refine "industry + 1" to "industry + 2."

12. How much do we want to change the relative costs between techs as a result of previous decisions? Will this just complicate things?

I say code the techs that can be controled by a general formula and specific user input. Just make sure that you don't have to reload the game every time you mod something--that was really time-consuming and annoying about modding moo3.

The larger the relative cost, the emphasis on rps. I think we just have to play the game a few times.

13. Should available refinements be strongly based on the prior decisions?

Don't know what exactly you mean. We can make refinement a choice or somewhat have prereq.

14. Would developing fundamentally distinct unit types (which serve separate purposes and are not strongly substitutable) and differentiated terrain encourage varied strategy?

I think so.

15. Should we design around the classic moo start (one planet), or a predeveloped empire start?

I personally prefer a whole system. But the game is really relative so it doesn't matter from a stragetic point of view.

_________________

WOW, this thread took off!

Some clarification is in order for my earlier suggestions(way back now!)

I suggested the elimination option BECAUSE Aquitaine requested a way to avoid a set path in the tree that would be repeated over and over, game after game. By having all the techs known and available in a static tree, one will always use the same path, the strongest one (deepest part of river as suggested earlier). So we try to balance them, but I find it improbable that we could acheive such perfect balance accross the entire spectrum of the tech tree. I think a prefered path will still emerge. We need a catylist to spur a change in path direction, hence my elimination suggestion. If a tech that is normally in a prefered path is NOT available then a new path is required. In the river analogy, think a dam being built diverting the river left, right or both ways around the normal path. If the elimination option is not used, then we STILL need some type of catylist to spur change in direction, and chose a lesser tech over a prefered tech.

Additional info for elimination proposal, I would allow 60-70% of needed essential techs to be exempt from elimination so as to not disturb basic gameplay. The eliminated techs would be PREDEFINED to insure balanced play. Several DIFFERENT combinations of eliminatable techs would be predefined and balanced. At game start the computer would randomly select one of these predefined combo groups. My suggestion to hide advance elements of the tree was to promote a short term strategy, which I feel would create MORE diverse paths chosen.

Also the discussion on interdependentcies between techs, when I mentioned the computer selecting them at random, I was NOT implying that ANY OBSCURE tech be tied into ANOTHER OBSCURE tech to produce ANOTHER OBSCURE tech. I was considering that ALL interdependentcies be PREDEFINED into the game. AGAIN, I would create multible paths from level to level in the tree and have the computer randomly select from these PREDEFINED paths which way to go from a specific tech to other RELATED tech. In ANY game, tech A is related to techs G and L at a paticular tech level. In one game tech A may only be dependent on tech G, in another game tech A is only dependent on tech L,, in a third game tech A can be dependent on BOTH techs G and L. Repeating, ALL paths through the tree are PREDEFINED, and RELATED, and hopefully reasonably balanced ; and SOME of these are randomly selected at onset of gameplay, from level to level to alter gameplay from one game to another, and to INTERJECT DIVERSE PATHS through the tree.

Hope this clarifies things some, I think ALL areas of the tree (and other game values) should be predefined. And when I mention random computer selection, it's from a predefined value list with minimum and maximum amounts.

Could we calrify/decide something here.

There seem to be two subtly different possible interpretations of the model as described so far by Aquitaine, which might be getting us all to talk at cross purposes a bit. To clarify things I'll describe how the alternatives as I see them would work out

1) Projects costs are e.g. "500RP and 10 turns". Your empire generates say 25 RP each turn which pile up in a notional 'RP stockpile'. When you had 500RP in the stockpile, you could start the project. You then have 0RP left in the stockpile and the research will be ready in 10 turns. your Rp total builds up each turn and when you have enough for another project, you start that off.

Pros: no wasted research in the short term, every RP you generate gets you one step closer to starting another project.

Cons: the 10 turns part of the cost is just a delay here, the model essentially the same as perfectly transferable research under a normal moo model, but you have to wait after you've researched something.

2) Project costs are e.g. 50Rp/turn for 10 turns. The same empire as before (25RP a turn) can never start this project, since it doesn't generate 50Rp in a turn. when it grows a bit, so that it is generating 50 spare Rp a turn, it can start the project, and must maintain this production of 50Rp each turn for 10 turns, and then it will get the tech.

Pros: I think this was what Aquitaine was talking about. The turn delay bit seems less like an arbitrary delay and more like a game element under this system, as the need to keep the rate of investment up means you are making a medium term rather than short term decision when chosing what to research.

Cons: under this model, if an empire grows each turn, so it generates 50rp on the turn it starts the project, then 51 the nect turn, 52 the next etc. until 60 on the last turn of the project, and there are no projects which take less than 10 rp a turn, we have to decide what to do about the (mental arithmetic) about 50 Rp that have been generated but not used. i.e. RPs could be 'wasted' in the short term.

Also, under this model we'd need to decide what happens if an empire doesn't produce the needed research points each turn. Does the lowest priority project get put on hold indefinetely (essentially underspending with 100% efficiency) or does it get cancelled and all the research invested so far lost (bit mean if you're only short by 1 rp for 1 turn on a 10,000 rp for 100 turns project!) or some other choice. Equally can the player cancel or suspend projects by choice, even when they do have the RP's, maybe in order to start another project (does this make the model much different from straight "pile up the rp's" fullly transferable research?

Aq, could you say how it was you had envisaged things working.

As far as I see it the choice is between

a) trying to balance the tech tree such that ANY path is as good as ANY other. By this I mean that at every point when you choose between techs, each choice has the same cost/benefit ratio, so a player is indifferent between them. "Choose whichever tech you like, they're all intrinsically as good as each other". The danger of this is that it is very hard to do (I've never played a game yet that has managed it), so you in practice you tend to end up with "normally I'd go for this route up the tree, it's usually the best" instead.

b) accepting this is probably impossible to do in general and deliberately designing a tech tree from the outset that is designed to have 'good' choices, i.e. "you'll probably want to go for this tech, then this one, then that and that", but so that there are several parallel sets of these i.e. "From where you are now, If you take that one, then you'll probably want this set to go with it, but if you take that, then you'll probably want that set".

n.b. in both cases I expect that the exact choices players make will depend on the strategic and tactical position of the player, the difference is between the general

The second approach requires the prerequisites of techs to be chosen such that it is possible to run down one branch for a while without needing stuff from the other(s).

If you design a tech tree like this then you could legitimately prune away branches at random to create variation between games, without annoying the hell out of players in the way that simply hiding techs in a supposedly evenly balanced tree would, which I think was where Sandlapper was going. But I don't actually think you'd need to, so long as the different paths were sort-of linked to different strategies, since players would want to choose between them for that reason, and would be strongly pushed to pick different choices depending on the course of the game. Equally players could choose to take high-risk high return strategies by taking certain non-typical (note, that's not the same as non-optimal) routes up the tree, if we specifically bear in mind that we want to create them (rather than let the players find them in a tree we think is perfectly balanced).

This is exactly correct - the type of categories you want will have a massive influence on the type of tech system you choose. The two are so interrelated it is impossible to separate them.

Since we seem to have a consenus on a point (miracles never cease) shall we have a quick yay or nay on the tenet:

"We want categories that directly relate to gameplay elements, not broad categories unrelated to the mechanics of the game. The number of categories is irrelevant and will be determined by the number of distinct gameplay elements that we identify and design tech to affect."

Any nay sayers?

Maybe the 50RP per turn should be an upper limit on what can be invested. i.e. if an empire can only find 25RP per turn for a tech then they will still get the tech, it will just take twice as long. But of they have 100RP they can still only invest 50RP per turn and it will still take 10 turns. What happens to the spare 50RP is another matter.

I dont like this approach as it imposes a hard cutoff point in tech spending. How about having decreasing returns on investment? Give each tech a base cost and a base time to completion. This gives a base investment level of cost/time. So investment at or under the base level gives a 1:1 return. Above this level you will get gradually decreasing returns on your investment - to the point where it is impossible to complete a project in less than (say) half the base time.

e.g. if a tech has a base cost of 200RP and a base time to completion of 10 turns, then:

* investing 20 RP per turn (the base investment level) will complete the project in 10 years.
* investing 5 RP per turn will complete the project in 20 years (linear relationship below the base investment level)
* investing 30 RP per turn would complete the project in (say) 9 years
* investing 50 RP per turn would complete the project in (say) 7 years
* investing 100 RP per turn would complete the project in (say) 5.5 years
* investing 1000000 RP per turn would complete the project in 5.001 years

This doesnt have to be decided by the player directly - it can be calculated based on the total RP generated and the number of projects running concurrently. All the player needs is the option to start a project, and possibly some way of weighting spending towards some projects more than others.

Not here.
But please stop visiting these boards at the same time as me... its getting confusing

Dont tell me you are also based in the UK (or similar timezone) and are bored at work?

Don't mind me, just sitting in an office in london, being bored.

What happens to research on projects you can no longer afford, or want to cancel?
The way to avoid abuse without annoying the player would be imo to have automatic halt of the project if RPs are not enough. Also let any project be canceled at any point, but losing the RPs invested - which is of course unrealistic, but better for gameplay

Randomly block out different techs each game or not?
Will it be possible to develop applications without knowing the theory, if all the prerequisites have been met, maybe at some kind of penalty?
I think both of these would only add to the unneeded complexity, and the former is an excuse to make the tech tree carelessly

Should we have techs that reduce the cost or time of research generally (or for particular categories). ?
A well-developed system of prerequisites is enough.. maybe no more than one such tech per field, and with no decisive effect; someting like one tech that decreses time 7.5% and another that does the same to cost per each field.

how big/small do we want the steps of research to be, ie.e how much of adifference will each discovery make to your strategy and tactics?
I'd like to see both big difference and not that much difference techs.

Would developing fundamentally distinct unit types (which serve separate purposes and are not strongly substitutable) and differentiated terrain encourage varied strategy?
In short: varied strategy, good.

Should we try to hardwire choices with race bonuses to particular techs, or try and make it dependant only on strategic position.
Racial bonuses to specific fields would be nice, but that depends a LOT on what kind of general categories we have.


About the 'wasted' RP I don't minf them much - but I think some players that like to optimise everything would waste a lot of time making sure there are no excess RP, so something needs to be done about that.. the way excess food was handled in MoO2 comes to mind - 2 excess food made 1 AU, like PP when producing trade goods, and both ratios were increased to 1:1 with the fantastic traders pick.

The system is really model #2; if a project costs 50 RP for 20 turns, then having 10 RP, according to the strict HoI model, does nothing for you; this is something upon which we are flexible, though.

The idea is that you have to specifically 'start' a project, which you cannot do without the required number of RP. If your RP # increases by 1 or 2 per turn, then we do have to answer the question of 'what do you do with just a handful of RP.' There are several possible solutions:

- Have a smattering of very small research projects;
- Allow a project to be 'half-started' in which you take a 50-75% effeciency penalty but get to research things you can't afford; optionally this could also be a racial trait (like 'creative') where this efficiency penalty is reduced. However, as a systemic thing, we wouldn't want this to be the norm.
- (my favorite) allow excess RP to back into some other Empire pool; we could have it go into taxes and money (which we haven't designed yet); we could have some sort of literacy rate; we could say that certain refinements come out of just general tech pool stuff; basically, excess RP would be your 'basic science' budget, such that it is sometimes preferable not to be working on specific projects. If we ever wanted to incorporate 'eras' into the game, we could allow people to advance eras either through certain techs, or through enough progress in general science.

_________________
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!

Just to check, by this you mean that after the 20 turns, you will have spent 1000RP total on this project, yes?



And also the question 'what if you fall short whilst a project is running'.

Lastly could I expand/clarify that list of possibilities:

- Have a smattering of very small research projects;
- Allow a project to be 'half-started' in which you take a 50-75% effeciency penalty but get to research things you can't afford; optionally this could also be a racial trait (like 'creative') where this efficiency penalty is reduced. However, as a systemic thing, we wouldn't want this to be the norm.
-we could have it go into taxes and money (which we haven't designed yet)
-we could have some sort of literacy rate
-we could say that certain applications/refinements come out of just general tech pool stuff
-boost a general refinement level which improves properties of items slightly
-increase the chance of lucky breakthroughs
-allow it to be stockpiled and invested in improvements to the research rate.
-allow it to be stockpiled as an insurance against future shortfall whilst a project is running

other ideas?

The 1000RP total is barely relevant. You spent 50 RP for 20 turns. You cannot spend 100 RP for 10 turns or 1000 RP for one turn; if you have 1000 RP, you must still spend 20 turns on it.

The 'realistic' argument in this case is that you cannot simply come up with things overnight; they require both time and money, and after you invest so much money in something, you hit the point of diminishing returns.

The gameplay argument is that this model works better, IMO, than anything else I've seen.

In HoI, if you fall short while a project is running, it gets suspended until you generate enough. You can re-prioritize your projects, as falling short is a fairly common occurance as your shipping lanes get raided, as the world market tumbles after you go to war.

_________________
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!