What's your research strategy?

[RefSteel]

Unintended consequence of removing the interest-based-on-existing-investment mechanic: Research spending will be delayed and may become weirdly sporadic, because getting an early start on research no longer has value: Until you spend enough to get a tech into the percentages, it doesn't matter if you spent the BC to get there slowly and steadily or all at once. This means opening research before factories are complete will be a rarity, and may create some gamey research strategies later in the game as well (though this would take some experimentation, or at least more in-depth thought than I've given it so far...).

See for instance Civ4's min/max research situation, where (with rare exceptions) it's always optimal to set the research slider at either 0% except when ready to crank it up to 100% and burn through the next research project ASAP.

[Ray F]

Unintended consequence of removing the interest-based-on-existing-investment mechanic: Research spending will be delayed and may become weirdly sporadic, because getting an early start on research no longer has value: Until you spend enough to get a tech into the percentages, it doesn't matter if you spent the BC to get there slowly and steadily or all at once. This means opening research before factories are complete will be a rarity, and may create some gamey research strategies later in the game as well (though this would take some experimentation, or at least more in-depth thought than I've given it so far...).

See for instance Civ4's min/max research situation, where (with rare exceptions) it's always optimal to set the research slider at either 0% except when ready to crank it up to 100% and burn through the next research project ASAP.

Civ 4 is different in that you are only researching one tech at time.

The solution here doesn't address WHEN you should start researching (the Civ4 example), but how you spread your RP across the 6 categories when you do research.

However, I am sure that there will be ways to optimize any system we come up with. My primary concern is that the optimization doesn't require a lot of micro-management nor that it is so overwhelming that everyone feels compelled to do it.

[Psillycyber]

What if you had a small interest mechanic on top of the 1/6th bonus detailed above?

For example, what if each tech earned only 5% interest on already-invested RPs (regardless of contributions this round), IN ADDITION to the 1/6th bonus.

So, for example:
Turn 1: Invest 1/6th of your RPs in planetology (let's say that amounts to 80 RP in planetology). Total: 120 RP
Turn 2: Invest 1/6th of your RPs in planetology (still 80 RP). Total: 120*1.05 + 80*1.5 = 246 RP
Turn 3: Invest 1/6th of your RPs in planeteology (only 8 RP, becuase you switched most of your empire production over to factories this turn). Total: 246*1.05 + 8*1.5 = 270 RP.

And so on. This would slightly incentivize starting research before factory construction was finished. You could tweak the interest percentage as needed from feedback in alpha.

[RefSteel]

Expanding on the unintended consequence I mentioned: In most cases, at the beginning of the game, whenever you get a new terraforming or robotic controls tech (unless you can do a little research and just barely get one or more techs into the percentages at the time) and any time your highest-immediate-priority tech(s) is/are in the percentages, it will make sense to set the sliders for all your worlds to little or no research, in favor of infrastructure or other priorities. This won't happen all the time, and it doesn't seem horrible to me; it's just that when introducing a significant change to the rules of the game (and this would be a very significant change, for better, worse, or both) it's good to know how it will affect things.

What if you had a small interest mechanic on top of the 1/6th bonus detailed above?

Any interest mechanic that doesn't depend on the same-turn investment won't really help; you'll still want to invest 0 rp through (for instance) most of the early turns of the game; it's just that if you'll get enough free interest, you might want to create a one-time research investment very early to get the interest going.

I feel like there's something that can be done based on the previous turn's spending or an average of the previous N turns' spending, but that's assuming we even want to reward consistent research as the original game intended to do, and it's not clear that Ray F likes that idea (and as the one who's actually doing the work, spending the money, and everything else on this free recreation, his opinion is the one that counts!)

[Ray F]

I feel like there's something that can be done based on the previous turn's spending or an average of the previous N turns' spending, but that's assuming we even want to reward consistent research as the original game intended to do, and it's not clear that Ray F likes that idea (and as the one who's actually doing the work, spending the money, and everything else on this free recreation, his opinion is the one that counts!)

Here's my general priorities.

The #1 goal is always to eliminate micro-management without eliminating choice
The #2 goal is to avoid the "this is the best way" puzzles
The #3 goal is to eliminate features that scream "the dev thinks I should play this way"

I think the MOO1 mechanic violates both #1 and #3. We see #3 a LOT in games because it's very difficult for devs, as gamers, to not imprint their personal playstyles into any game they design.

I also do not think there's anything sacred about the research interest mechanic in MOO1. It seems to me that it was added to give players an incentive to spread out their spending across multiple categories instead of just going 100% until they get one tech, then switching 100% to the next. After all, allowing that kind of strategy to work would defeat the entire point of having 6 tech categories.

However, the mechanic that they used to create this interest has the unintended downside of requiring micro in order to maximize the benefit. With all of the other things they were balancing in the original game, I'm guessing that they didn't really have time to playtest the consequences of the mechanic (there are definitely worse problems in MOO1 that also remained unfixed).

[Ray F]

I really like the 1/6th as an idea. It simplifies micromanagement greatly and this is amazing. There are many games I've played using '=' research for the whole game just to get rid of the need to micromanage. But it does have consequences that are not so simple, for example it will promote homogenious research in most games which may not end up being as interesting as the ignoring of certain fields during certain portions of the game that happens now.

Well, the bonus for spreading research as much as possible should never be so great as to discourage players from EVER focusing research. That bonus is easily tweaked and hopefully we can find a good balance in the alpha. It seems like +50% is a good start.

Either way the game should work well and if any problems crop up, it's an easy readjustment. But are you sure about changing the mechanic in the first place? Research is so central to the game that any change there will have a big overall impact and the game will no longer be a clone with exploit fixes.

Well, I've always thought of it as a "feature clone". What you can do in MOO1, you can do in ROTP. But there are different stages of deviation, in increasing order of significance:

1) Cosmetic changes - generally illustration, sound and color schemes (as much as possible in ROTP)
2) Useability changes - rearrangement of the UIs and controls for ease of use (quite a bit in ROTP.. MOO1 has an old UI)
3) Bug fixes - blatant errors in code (it's a complete rewrite, so presumeably everywhere in ROTP)
4) AI Balance - addressing areas where the AI is insufficient or handicapped (a key focus of ROTP)
5) Exploitable mechanics - typically min/max strategies used to compete at the most difficult levels (recalibrated only if they require micro)
6) New features - things you could not do in MOO1 (very, very minimal. generally only to equalize player & AI abilities)

Everything I've done in ROTP fits into one of those six categories. There may be some debate about whether certain changes are #3 or #5 because the distinction between a bug and an exploit is always not clear. The #6 category, "New Features", is the line I am extremely reluctant to cross for obvious reasons. Fixing exploits (#5) often feels like a gray area since they are usually necessary to compete at Impossible difficulty so experts have become very accustomed to using them.

You will lose some of your hard-thought and cool exploits, but you will also lose micro. I hope it's a fair trade. I have no doubt that you guys will find more exploits once you get your hands on the source code, lol.

To be perfectly honest, I always knew that there was a research interest mechanic, but not that there was a way to methodically exploit it with micro. In that sense, this thread really opened my eyes and reminded me once again of the great value of the experts in this forum.

This exploit is very similiar in concept to the ultra-rich reserve exploit in that it provides a huge bonus in exchange for micro, and the AI doesn't do it. Those are both major red flags, imo.

[Psillycyber]

Any interest mechanic that doesn't depend on the same-turn investment won't really help; you'll still want to invest 0 rp through (for instance) most of the early turns of the game; it's just that if you'll get enough free interest, you might want to create a one-time research investment very early to get the interest going.

I'm fine with the idea of players just putting one or two turns of investment seed into techs and then riding that interest the rest of the way. It would certainly be a lot less micro, which is kind of the point. And there would still be the decision of WHEN to seed your techs. (By the way, at 5% interest, just riding that interest from a small seed would take a long time. 5% interest had a doubling time of about 15 years).

[RefSteel]

You will lose some of your hard-thought and cool exploits, but you will also lose micro. I hope it's a fair trade. I have no doubt that you guys will find more exploits once you get your hands on the source code, lol.

I don't think we have any sacred cows among the game mechanics here, least of all if they're exploits. I don't see anyone saying, "But how will I play if I can't do my special research micro!" The worry is that a significant change in the rules for research will lead to a significant change in the feel of the game and might narrow strategic options., and we want to make sure that doesn't happen.

I love the 1/6th research mechanic you came up with and I like the +50% starting point. What I don't want to lose from the original system is not the power of fully exploiting the mechanic - it's the incentive to perform research continuously.

For me, this is in large part a matter of aesthetics: I don't like the idea that doing zero research empire-wide for several turns, then dumping everything beyond eco Clean into tech for a turn or two, will in many circumstances have the same research effect (or better if some techs are in percentages at the beginning and the rest aren't yet close) as steadily researching, adding the same number of RP over the same total number of turns, but spread out instead of concentrated at the end.

It's not just aesthetics though: Since many other things you might build in that time (factories, terraforming, ships or defenses that might set out for battle or scouting/colony missions in that timeframe) are more valuable if built ASAP, while research (until a tech comes in) is not, well-timed zero-and-binge research will not only be just as good as steady research, but the most optimal play: A new exploit, introducing new micro, since knowing when to use it and how to time each stage depends on monitoring how close you are to the percentages in each field, and of course each stage means changing the sliders around on many or all of your worlds.

This won't happen all the time of course; there will be many times in the game when the advantage will be small or nonexistant. It will apply at the beginning of every game though, while the homeworld is still low on factories, and sporadically through the rest of most games.

[RefSteel]

(An alternate suggestion so that I'm not just whining: The basic research bonus could be half as strong as you described, with an additional bonus equal to the smaller of the current or previous turn's regular bonus. This would incentivize steady spending without overwhelming other priorities.)

[Psillycyber]

I agree, an incentive for continuous research is good.

So, here's an idea: in addition to the 1/6th bonus, have a variable 1 to 10% interest each turn on RP already invested on a tech. This interest could vary with respect to how this turn's base RP spending on a tech compares with the previous turn's base RP spending on a tech.

If this turn's base RP spending on a tech is > or = to 100% of last turn's base RP spending on that tech, then the interest rate will be 10%.
90-99%: 9% interest
80-89%: 8% interest
70-79%: 7% interest
.....
10-19%: 1% interest
<10%: 0% interest

(Edit: or even better yet, make the sliding scale as continuous as possible with the capabilities of the game. So, for example, if this turn's spending is 74.8% of last turn's spending, then 7.48% would be the interest applied to it).

In this case, the player is rewarded for having each turn's base RP spending on a tech be at least as large as last turn's RP spending. If you spend a turn dialing research way down, you lose that interest bonus that turn. Then, if you jack your base RP spending back up, the interest is still capped at 10%.

So, under this scheme, the optimal strategy for pure output of RPs (disregarding tactics, in other words), would be to equalize all fields, and always make sure that every turn your research output is never decreasing from the previous turn. Pretty simple to manage, eh?