"You successfully researched Razor Edge Swords XXIII!!! As a consequence, all your swords do an extra +0.001% Keen damage."

The risk with that kind of tech tree is that it can easily be boring.

If you leave out random techs, then each tech needs to be kind of balanced. Therefore there might be no techs that give you special advantages but instead only slight increases to some standard values each time you research something.

Of course that is not a must. I loved about the MoO2 tech tree, that among the standard increases to certain values there were also several special techs that gave you entirely new options and make a significant difference once you have researched them.

I don't mind that at some point you only get into infinite standard researches but before that there should be some spiced up techs that stand out (maybe with some rule that you at least get one of a certain bunch of techs to keep it fair).

Rolemaster had a nice system for progressive skill development that never capped.  First 10 ranks got you +5, next 10 got you +2, next 10 got you +1, everything after that got you +0.5.  Ranks 31+ were always tempting, because you always wanted to improve certain skills, but the cost compared to improving other, possibly useful skills was equally tempting.

In general, I like logarithmic curves (i.e. diminishing returns) for things with "infinite" in them.

I always assumed this would be the way it would be... I mean, before you can upgrade "Generic Swords Upgrade I" you'd have to research swords to begin with. It's just that once you've actually researched the first stage, you can continue researching that tech for improved results.

This is my preferred route, too - at least over logarithms. The diminishing returns of logarithms (and asymptotic approaches) is that you reach a point where there is just no reason to keep going. Even if you have resources flowing out of your ears and more than enough to throw away, you'll never notice that +0.01% boost that you paid 100,000 gold for. It just becomes a waste of space and they may as well have implemented a hard limit.

Personally, above all these methods I prefer a method without diminishing returns. I would prefer if progressive generic bonuses provide constant bonuses but with increased cost. Whether the cost should increase as a root, linearly, or exponentially I would leave up to testing. My reason for this is that I think this method allows for research to always remain a factor in the game - with diminishing returns there comes a point on larger maps that research simply stops being relevant.

Another downside to diminishing returns or costs rising too quickly is that it encourages you, for the sake of efficiency, not to focus too much on any one tech. And the result of that is, after enough time everyone ends with more or less the same research other than faction-specific or randomized techs. Allowing people to continue researching fields that are particularly important to them without too much of a hit to efficiency allows for much more variability in technology research.

I don't see why it's unrealistic. Look where we were 300 years ago, and look where we are today. Technology has absolutely exploded - we have had increasing returns. Computing is a perfect example, even though that is slowing down. Still, once the next major step into a whole new method of computing is made, it'll likely happen all over again. Not to mention what's realistic isn't so important to me in terms of game design.

And even in the case that such a method often results in 'unbalanced' situations, I think that it could be made to be balanced enough to satisfy anyone but the diehard multiplayer people who are hell-bent on ruining every game they get their dirty little hands on with their obsession with absolute balance

Um, more or less exactly my point. When you start with close to nothing and expand at an exponential rate, at first you're expanding preeeetty slowly. Even slower than linearly! But then you reach a point where the function just explodes and shoots off into the sky. Technology had slowly progressed over tens of thousands of years, and then it reached that critical point and just exploded...

I guess what it really comes down to is the era, the time scale, and the context. In a game like Civ, it makes sense for a farm to produce 100 food, then 300, then 1000, etc... In a medieval game spanning 100 years, not so much. In a fantasy game it depends heavily on the imaginary context. It's been stated that Elemental's technology will all be medieval; but Elemental is a recovering world, not a growing world. In the beginning there are no cities or civilizations - but then they start to form. People come out of hiding in the wild to rebuild civilizations - in fairly short time spans. Presumably some knowledge from the past would have been preserved, either verbally or through writing; and thus vast technological strides would be made - going from nothing to medieval technology within a generation.

Also, not to mention even in the past 10,000 years they had their technological explosions, usually brought about by something specific. Like cultivation, irrigation, domestication, the wheel, metalsmithing, written language... astronomical navigation methods... gunpowder... the printing press. Let's just take irrigation as an example, to continue the farming theme. The difference between going from no irrigation to even the most rudimentary form of it is huge. It allows you to farm where you could never have farmed before, over vast swaths of land. And upgrading from the most rudimentary form of irrigation to something just a little more sophisticated will drastically improve the quality of all that land - making it decent, rather than passable farmland.

Diminishing returns in gaming typically means each level of a skill gives you less benefit than the last.  What you are describing KellenDunk, while technically accurate means every leveling system from every game I ever played had diminishing returns. takes 2x as much xp to get from level 3-4 as it did from level 1-2!  So yes, this is a diminishing return for xp gained, but it is not a diminishing return of the level, or in this case skill.

It's true, KellenDunk, but there is a significant practical difference between them. It's not only psychological. Spending 1000 gold for a 0.1% increase vs. 10000 gold or a 1% increase is hugely different assuming the time required are in the same ballpark.

Like you said, it's still diminishing return on investment, but it's not diminishing return per research tier. Diminishing return per research tier is what bothers me.

My take on this whole thing is that the magical/unscientific nature of the game will allow us to "justify" pretty much any system we can think of. Take the steel example: in a scientific world, the durability of a metal is fundamentally limited by the internal "stay-together-ness" of its atoms, because all (regular) matter is made of atoms. However, in a magical world, steel can be made of a precise mixture of Earth and Fire, and refining the precision of the mixture is how you make stronger steel. Since there are only practical/economic limits on how refined you can make it, you could have steel that gets 10% stronger each time.

Infinite research can be done amazingly, it just needs to be balanced. diminishing returns is one way to deal with it, and should be part of it, but you should never feel like the effort is useless from then on.

For example, research could give randomized bonuses. Researching infastructure may give you +1 to speed while on roads this time OR it could give you a 25% construction speed bonus to bridges. Weapon tech might give you access to a new sword type, or it could give your current swords +1 damage, or it could let archers carry +5 arrows. You get the idea, this way the game is dynamic AND balanced. Worked on a little, the idea could extend to "if you have more swords than any other weapon, weapon research will far more often effect swortds".

No matter how they do it, im just glad they are doing it. its time for a fresh look at research in the TBS genre.

Really? RNG are some of my favorite things about games.  FIre Emblem for example.  I know a lot of people hate getting screwed over by it, but if a person doesn't happen to be growing very well then it puts me in a situation where I have to rethink who I want to bring.

I think you might have this wrong, RNG in the case of infinite research here would not unbalance the game over much.  Each player should have a roughly equal chance at getting something nice, something they need, or something totally worthless.  Besides, this model would more accurately represent the way raw scientific research works, researching one thing often has spillover effects into some unrelated practical application.  In EWOM, it would be like spending some gold for a group of scholars to take samples of some rocks in a certain mountain range in your kingdom, they find the rocks to be exceptionally resistant to weathering or something, and so make better stones to pave your roads with.  The majority of the process described here is what normally is abstracted into the scientific spending bin that transforms your cash into things that make you win.

1. something nice in one situation is generally quite useless in another, and the coding needed to make the game figure out which is which is expensive and generally does not work all that well, and
2. that still doesn't balance the game: as any statistican will tell you, just because there is a 33% chance of something happening doesn't mean that it will happen 33% of the time. And all it takes is one run of bad luck to essentially ruin your chances of ever getting back on your feet. Finally,
3. It removes a lot of the skill-based parts of research (knowing what to research and when to research it), since you have little to no control over what you are doing.

While you make some interesting points, you look at it the wrong way.

1: If Elemental is what it looks like, there should be very few situation specific techs to be useless. If you don't need income, dont research into infastructure. If your looking for more powerful infantry, simply invest in weapons tech or something similiar. A good random system wont be "you research tech 22: you gain longswords" it will be more like "You research tech 22: Your swords gain +2 vs opponents with shields". What if they aren't using shields? Then use archers. Having a tech never meant that it becomes the fulcrum upon which your empires existance lies.

2: Again, if there are a signifigant number of useless techs dotted around, then we have an issue bigger than thier chance of popping up. One would imagine that the game design would not be so unimaginative as to have too many "wasteful" techs.

3: That would be the point. Alot of people (myself included) hate the tech rush in TBS's. Your in effect desperately trying to sqeeze out a few more research point for that tech you need solely because of a very linear strat. In Civ4 for example, there are a handful of strategies to rush through the game at warp speed, hitting certain techs along the way. In elemental one would hope the game would be more about working with what you have and not about "the rush". The game should feel epic, and part of that should be the mystery.

Random techs are no worse than random lairs and random heroes. They provide an asset that you are givin and must make do with. A skilled player isn't a mathmatician, calculating the turn needed for important tech A, but rather a player who gets "useless" tech B and makes it work in a clever way.

My 2 cents.