Applied research is utterly broken.

You could but I don’t think you could do it directly in that all the techs work the same way

T1 1 costs +X
TL2 costs +2X
TL3 costs +3X

And so on

Which is a curve but the problem is that the gain from the improved facility keeps up with it

What you could do is change when the facilities are available on the tree. If tl1-10 you got a research facility upgrade every level then tl11-30 one every two levels of tech and Tl 31-60 one every three and so forth would make applied research feel less linear progression even if you did not change the additional research points per facility level

(Nb Brokenin this context is one of my frenzy triggers so you owe a new door )

It is widely known (not by all though) that the stock game is very unbalanced.
Applied Research is just one factor of it.

~~~~
http://wiki.spaceempires.net - Visit the SpaceEmpires Wiki and help improve it.

The text costs are based on formulas in formulas.txt - and it's easy enough to stick in something exponential. Of course, 100 levels and an exponential function can make for some exuberant tech costs.

It might be easier to just increase the base cost of the tech area and/or research centres, making the number of facilities required much higher and introducing a much more prohibitive cost to upgrading.

-----

Space Empires Depot | SE:V Balance Mod

It's somewhat exponential in SE:IV, but the number of levels is typically low so it doesn't quite kind out hand. Here, with 100 levels you're exponent will have to be small (ie 1.01), meaning the first 20 or so levels are really cheap, levels 20-50 are quite reasonable, before you get into the 10M+ to 40B+ per level after that.

Going at it from a resource cost perspective might work better, since you'll need to reduce your research space to accommodate more resource extractors to support the costs for the research centers. Combine that with a higher applied research base cost, and it should work ok for most games.

Also note that the formula cost progression is applied to all techs as well.

-----

Space Empires Depot | SE:V Balance Mod

How big "space" you play? And what research cost?

Becose its realy different when you have 3 systems or when you have 15.
In other words, if number of research centers needed for this was bigger than what you can get, then it would not be big problem.

So maybe linearily increase research cost.

Btw, i hated that in SEIV there were different shapes of cost curves (IIRC low was linear, medium and high quadratic).

That imo makes it unbalancable for both variants at once.

This is not argument aginst non-linear cost, but aginst having different shapes between low/med/high.

Argument aginst nonlinear is that some techs have lot of TLs and some after few TLs opens new one (ship sizes/engines are THE example).
And racial techs needs to be competible with both DU/CM and MB/APB/high tech seekers (ehm, is there any in stock better than (non racial) than CM?).
Of course it is balancable, but it requires work.

Thats why i play balance mod.

Regarding the research facilities, the game lacks the law of diminishing marginal utility. Although allocating more resources to research (i.e. building another facility) will allow more technological advancement, doubling the devoted research *should* not necessarily be 100% more effective.

I remember playing a game called "Capitalism 2" that actually took diminishing utility into consideration. As more resources were devoted to a specific project, the less was the effective marginal increase. For example:

# R&D Units........Effect on Linked Tech Gain........Marginal Increase
........1..............................100%
........2..............................170%................................70%
........3..............................240%................................70%
........4..............................295%................................55%
........5..............................350%................................55%
........6..............................390%................................40%
........7..............................430%................................40%
........8..............................455%................................25%
........9..............................480%................................25%

In that game it was only possible to link 9 units to a single project, but you get the idea.

I would wager that few game designers stayed awake in Economics 101. Too bad, really.

Firstly, I would like to say I completely agree with this idea and it would help make the current method of tech distribution worthwhile. If diminishing returns with higher expenditures on a single tech was added in, that would make spending 50/50 on two techs a more reasonable thing to do. It would help keep supreme research empires from going too high in any single tech, too.

I will support this idea in my next e-mail to MM.

Secondly, I think two ways to help counter the research avalanche are two new abilities: Maintenance Modifier (Local) and Complex To Build.

Maintenance Modifier (Local) would act like the current ability, but only affect the facility/component it is on. We could mod research labs to be significantly more costly to maintain (along with some of the other more elaborate structures) than other facilities. Research shouldn't be all win-win, it should have its costs.

Complex To Build could have two implementations, either increasing the build time, or causes a percent of expenditure to be waste (which would have the effect of increasing build time, too).

-----
Space for rent - please contact owner

You can argue it's already diminishing returns because a) It costs more RPs to gain that same amount of improvement in an item and b) The facility itself costs more at higher levels for the same improvement. So simply increasing these associated costs (RPs/Resources) could make a significant difference to how it plays out.

-----

Space Empires Depot | SE:V Balance Mod

You know, this talk of formulae in this thread (don't know why I wasn't inspired earlier) gave me the thought:

Make a new formula just for Applied Research (and perhaps a few other areas) which essentially reads: "[Lvl]*[Base Cost] + 100000". Basically, you start at Lv 1 per normal, but to get to Lv 2 you need some ridiculous amount of research. Lv 3 is pretty much the same ridiculous amount of research with the extra Base Cost.

Just thinking of more creative formulae.

---Sig---
Playing Touhou games (Go here if interested . No, nobody else is that good/insane as that replay). No rush for SE V bug fixing.

I didn't mean diminishing returns for the individual levels, I meant on the expenditure toward a single technology.

Suppose Research was set up so that from 0-50% you would get 100% of your expenditure as point toward a given tech. Beyond 50%, you would get less for your effort, perhaps 0.5% loss per percent. 70% expenditure would earn you 60% points, 100% would earn you 75%. If you have all the best minds in the world acting toward one goal, it's rarely too much better than a small amount of the minds in the world.

As of now, percent-based spending only adds to the per-turn tedium. If the actual percentage being spent had an impact on how effective the extra spending was, that would make thinking about the actual levels more worth the effort, and it would still be faster to focus all effort on a single tech for increasing that one technology.

-----
Space for rent - please contact owner

Any sort of Diminishing Returns based on what percentage of Research you're devoting to a discovery will only make researching more technologies at once the efficient way.

Right now ... as long as the RP's wont be spent on a tech level that you cannot gain (such as putting 100000 RP's into a tech when you only need 40000 more to finish the last level of it) it is impossible to waste RP. Whether you focus all of your efforts on one tech at a time, gaining levels of that tech at once, or in several areas at once, gaining a tech level in those in several turns rather than in one all at once, the end result will be the same (if there is no increase in RP in the middle).

Having some kind of diminishing returns only forces people to spread out research rather than focussing on one area, in order to avoid the losses imposed by diminishing returns. Instead of letting you put all your RP into Small Hulls until its maxed and then 10 levels of CSM, it'd force you to put 50 percent into CSM and 50 percent into Small Hulls ... getting you 10 Small Hull and 10 CSM on the same turn you would have finished both working on them one at a time.

It only deprives people of options, I think.

The Level Cost field in TechAreas.txt cannot accept formulas.

-----

Space Empires Depot | SE:V Balance Mod

As of now, the most advantageous method of research is focusing all your points into an area so that you get it sooner, rather than later. It is almost always better to get one thing each turn than three things after three turns, for example.

With diminishing turns based on expenditure there is some amount of tactics to spending points. In general, researching multiple tech areas will be the way to go. However, if a tech is really necessary and must be leveled as soon as possible, sacrifices have to be made.

And as you say, splitting the research 50/50 behaves the exact same way as the current system, so I only add tactics to the current system.

-----
Space for rent - please contact owner

"You can argue it's already diminishing returns because a) It costs more RPs to gain that same amount of improvement in an item and b) The facility itself costs more at higher levels for the same improvement. So simply increasing these associated costs (RPs/Resources) could make a significant difference to how it plays out."

One could argue that point, but one would be incorrect. Let me explain:

The first argument (a) is that for each increase in tech level, there is an additional amount of research required. However, the additional research applies to a different tech level (i.e. tech lvl 2 as opposed to tech lvl 1). Those are two different "goods." Diminishing returns applies to the same good. Supposing tech 1 was an apple and tech 2 was an orange, there is a limit to how many sticks would be useful at knocking the apple out of the tree, even if it would take more to knock out the orange (which I like better, but hey, I haven't found that tree yet).

I'm not quite sure what you meant by (b), but I would wager it was apples and oranges again.

I understand that certain things are unlikely to be changed, at least until SE VI. I hope the next version will take more realistic economics into consideration. After reading one good book, the designers could integrate an economic system to the envy of their peers. IMHO they are already ahead of the pack when it comes to turn-based sci-fi strategy.

For me it is a pity the AI is so lame, as I cannot organize my play time to fit other people's schedule. However I still play SE5 much like someone would solitaire. I too have been much bothered by the fact that it is too easy to escalate my research, resulting in maxing out and then loosing interest in my game.

What I have done is to mod my TechAreas.txt to increase the level cost for Applied Research to 50000 . I also increased the level cost for most other technologies by a factor of between 2 and 5. I suppose for balance's sake I should have probably modified everything by a factor of exactly five. I find this completely changes the way I handle research. I still do research Applied Research, but it is later in the game and I only research ten or so levels at a time before spending many turns on other techs.

I also recently increased the max levels for many techs by a factor of 10 (this in addition to multiplying the level cost by 5). I have yet to play to the point of seeing this break the game. I hate hiting ceilings. We'll see how that goes.

KOKOPUFS - I believe it would be diminishing returns, but only under certain conditions. Say the tech cost per level is 50,000, and the additional research per facility per level is 50. If you've got 20 facilities, that's 1,000 extra research, but you've had 50,000 extra points needed. Ultimately, the cost is going to increase by 49,000 tech in real terms under these conditions.

However, as I'm sure we've all discerned by now, if you have MORE than that threshold (In this instance, 1,000 research labs) you will gain with each level you research. Say you have 1,010 labs - that means you get 50,500 tech points per empire-wide upgrade, which means you're gaining 500 points per tech level.

---
"No greater love hath a man than he lay down his life for his friend. Not for millions, not for glory, not for fame... for one person. In the dark. Where no one will ever know or see"

Jack.