Research Upkeep

For what's not in 'Top Priority Game Design'. Post your ideas, visions, suggestions for the game, rules, modifications, etc.

Moderator: Oberlus

Message
Author
Ophiuchus
Programmer
Posts: 3433
Joined: Tue Sep 30, 2014 10:01 am
Location: Wall IV

Re: Research Upkeep

#16 Post by Ophiuchus »

Suggestions we were talking about
  • research-upkeep: slowing research down if you have researched more
  • research-upkeep-as-production-bonus: if you have researched less, an effect increases a bonus to planetary research production
  • research-upkeep-as-production-malus: if you have researched more, an effect decreases planetary research production
  • research-upkeep-as-tech-cost: if you have researched more, all techs cost more RP to research (there is currently no way to distinguish malus from bonus)
  • research-upkeep-as-individual-tech-cost: if you have researched more, some techs cost more RP to research
  • research-upkeep-as-cheaper-enemy-tech: other empires have to pay less for tech, the more one researches
Oberlus wrote: Thu Oct 15, 2020 5:21 pm
Geoff the Medio wrote: Wed Oct 14, 2020 5:50 pm Bonuses are generally more fun than penalties.
The upkeep thingy (more expensive techs the more techs you get, or RPs invested), fluffed as "complex body knowledge", could be formulated as more cheap techs the less techs (or RPs) you have, fluffed as "focused body knowledge", to be seen as a bonus.
The difference between malus and bonus is mostly a effect accounting UI issue, with some secondary complications. Where there is no effect accounting presentation (e.g. tech cost) one cant distinguish between malus and bonus.

For research-upkeep-as-production-bonus:
So you would start with maximum bonus to research production? That ratio could be a game rule. Maybe maximum 500% going down to minimal 100%.

For "its stupid to get tech for (almost) free" tech independent upkeep would need to be based on the number of techs, not on the research points spend.

If we could find a formula for the ratio I think that would be rather a stop gap solution. I am pretty sure we should throw it away when we switch to tiers and themes.

Another completely different idea: research-upkeep-as-special-tech-sale or research-upkeep-as-research-progress:
after finishing researching a tech, the user is offered some options for a special deal for researching certain techs cheaper. the level of cheap being determined by what the other empires already researched (for catch-up) and what your empire researched (the more your researched, the lower the discount). On offer would be directly researchable techs. Minimal "discount" would be 10%. In the KISS implementation if you choose one deal, you would get the progress of the discount (so 10% if tech costs 100RP, research progress would be at 10% and you would only have to spend 90RP). Probably the tech should also be queued automatically.
Fluff would be rather something like "focused research gain". Of course one could remove the optional part and have an effect distributing research progress. That could be almost scripted with the current implementation.

The main problem with all those bonus oriented approaches: fives times almost-for-free may be still almost-for-free. And having huge discounts for scaling - like ten times research production or 90%-off look quite weird i think. And for research-upkeep-as-special-tech-sale, everything outside off sale would be ridicously expensive.

I would prefer something similar to research-upkeep-as-individual-tech-cost, maybe research-up:
For me a tech dependent bonus/malus like a current tech level and researching tech at your tech level is significantly cheaper would make most sense to me (although having the bigger UI challenge).
Any code or patches in anything posted here is released under the CC and GPL licences in use for the FO project.

Look, ma... four combat bouts!

User avatar
LienRag
Cosmic Dragon
Posts: 2148
Joined: Fri May 17, 2019 5:03 pm

Re: Research Upkeep

#17 Post by LienRag »

Ophiuchus wrote: Fri Oct 16, 2020 11:58 am

The main problem with all those bonus oriented approaches: fives times almost-for-free may be still almost-for-free. And having huge discounts for scaling - like ten times research production or 90%-off look quite weird i think.
I think that the idea was not to make the small techs that more expensive by themselves, but to discourage researching them "just in case" because they would make the bigger techs incredibly expensive ?

As said earlier, I don't like the idea, but it's the best we have so far...

User avatar
Oberlus
Cosmic Dragon
Posts: 5715
Joined: Mon Apr 10, 2017 4:25 pm

Re: Research Upkeep

#18 Post by Oberlus »

One thing to keep in mind is that after the production boosts nerf many techs' costs can/should be rebalanced. In particular, since the growth is not anymore as fast as it was, later techs can be less expensive compared to earlier ones. Currently some or most refinements take around 3x the PP of the previous level. So you choose between 3 earlier techs or 1 refinement. When you have many earlier techs available and good for your strategy, you take those first for the quicker and stronger return. When you don't need the available earlier techs you focus on the expensive refinement and invest any spare RPs for the secondary techs. By just making the later techs cheaper I expect to encourage more diverse tech paths, since some refinements become interesting earlier. Also, that would imply that having the RPs to complete a whole turn of a late game tech (100 RPs per turn instead of 300 RPs) doesn't automatically mean you also have enough RPs to research 5-10 mid game techs in parallel. But all this needs playtesting to be sure.
Note: I'm still focused on that particular issue. I care less about controlling exponential growth or reducing the gap between winning and losing empires.

Ophiuchus
Programmer
Posts: 3433
Joined: Tue Sep 30, 2014 10:01 am
Location: Wall IV

Re: Research Upkeep

#19 Post by Ophiuchus »

Oberlus wrote: Sat Oct 17, 2020 11:13 am Note: I'm still focused on that particular issue. I care less about controlling exponential growth or reducing the gap between winning and losing empires.
Probably by adjusting the tech costs the get-techs-for-almost-free can partly be addressed.

The design "problem" here is that this needs needs a balance pass every time we change the growth curve.

Currently we still have exponential growth which takes over the growth at a certain point time. When that happens, exponential cost of techs become necessary.

So we would need to have an expectation, on which level of technology we are on which part of the curve.
I think rescaling would be easier if we could express that in a formula (to be adjusted by game rules) or at least some tier-based macros.
Any code or patches in anything posted here is released under the CC and GPL licences in use for the FO project.

Look, ma... four combat bouts!

User avatar
Krikkitone
Creative Contributor
Posts: 1559
Joined: Sat Sep 13, 2003 6:52 pm

Re: Research Upkeep

#20 Post by Krikkitone »

That’s why I think it would be easiest for the “cost-output adjustment mechanism” to make tech cost effectively exponential. So it doesn’t need massive readjustments to the costs of all the individual techs each time the growth shifts.

Because we will always have exponential growth in a 4x game.. so the costs should assume that.

In terms of game design, I’d say give techs a certain “technology point” value and technology points have an exponentially increasing cost (maybe not those exact mechanics, but that should be the effect)

User avatar
drkosy
Space Dragon
Posts: 367
Joined: Sat Jul 11, 2020 9:41 am

Re: Research Upkeep

#21 Post by drkosy »

The next colony building becomes 3PP more expensive: additional 6% of the base colony cost (50PP)
The next outpost ship becomes moŕe expensive: additional 6% of the base outpost part cost (50PP)
The next colony ship becomes moŕe expensive: additional 6% of the base colony part cost (120PP)
Also the ships become more expensive if you build a ship: on top of the 6% the ship becomes 1% more expensive (for ship without part based upkeep)
Thanks for explanation :) I always thought it was 6% of the last produced outpost/colony ship which would make an expotential growth. But as you describe it is just linear makes it more easy to forecast the costs.
That’s why I think it would be easiest for the “cost-output adjustment mechanism” to make tech cost effectively exponential.
Maybe another solution could be to split the tech tree into "basic research" and "technologies". The basic research doesn't unlock anything but other techs. It could be an linear tech tree with different themes (e.g. biology, robotics, physics). You unlock tier1 physics you could instandly go for physics tier2 without more prerequisite. That would be like MoO2. For techs you need different basic research and other techs (e.g. you can't go for plasma cannon without antimatter engine for energy supply).

The basic research tree could be without dynamic cost growth. That could be addressed by techs. Techs have to be optimized, devouring some research points. That means, the more techs you have the more "maintenance" you have to invest. Even I (with my fable of planning everything) could live with an expotential growth of cost for tech only because I still can plan progress for the basic research.

I won't split science to much. I still like to have one meter for science, but maybe it should be necessary to have two science-sub-trees (basic, techs).

Hopefully it's not to confusing what I suggest... :oops:
Want some fresh experience? Try Kosymod

Ophiuchus
Programmer
Posts: 3433
Joined: Tue Sep 30, 2014 10:01 am
Location: Wall IV

Re: Research Upkeep

#22 Post by Ophiuchus »

Krikkitone wrote: Sat Oct 17, 2020 3:04 pm Because we will always have exponential growth in a 4x game.. so the costs should assume that.
Yes, but if you have something like 100, 101, 101.01, ... that is exponential growth. And if you have 2,4,8,16.. that is also exponential growth. These curves are slighty different.
Any code or patches in anything posted here is released under the CC and GPL licences in use for the FO project.

Look, ma... four combat bouts!

wobbly
Cosmic Dragon
Posts: 1880
Joined: Thu Oct 10, 2013 6:48 pm

Re: Research Upkeep

#23 Post by wobbly »

Ophiuchus wrote: Wed Oct 21, 2020 9:00 pm
Krikkitone wrote: Sat Oct 17, 2020 3:04 pm Because we will always have exponential growth in a 4x game.. so the costs should assume that.
Yes, but if you have something like 100, 101, 101.01, ... that is exponential growth. And if you have 2,4,8,16.. that is also exponential growth. These curves are slighty different.
They are both growth but only the 2nd is actually exponential growth. Pretty sure the 1st is logarithmic growth. My maths is a little rusty these days.

Edit: Unless the numbers you are giving are rates of changes I guess. I guess research pts/round sort of is? The amount researched is exponentialing?
Last edited by wobbly on Thu Oct 22, 2020 8:16 am, edited 1 time in total.

User avatar
Oberlus
Cosmic Dragon
Posts: 5715
Joined: Mon Apr 10, 2017 4:25 pm

Re: Research Upkeep

#24 Post by Oberlus »

wobbly wrote: Thu Oct 22, 2020 6:58 am
Ophiuchus wrote: Wed Oct 21, 2020 9:00 pm
Krikkitone wrote: Sat Oct 17, 2020 3:04 pm Because we will always have exponential growth in a 4x game.. so the costs should assume that.
Yes, but if you have something like 100, 101, 101.01, ... that is exponential growth. And if you have 2,4,8,16.. that is also exponential growth. These curves are slighty different.
They are both growth but only the 2nd is actually exponential growth. Pretty sure the 1st is logarithmic growth. My maths is a little rusty these days.

Edit: Unless the numbers you are giving are rates of changes I guess.
I don't think the first one is exponential:
x_0 = 100
x_1 = x_0 * 1.01
x_2 = x_1 * 1.00009901
So the rate of growth diminishes.

This one would be exponential:
100, 101, 102.01, 103.0301...
x_t = 100 * (1+0.01)^t

Maybe it is logarithmic indeed or something more complex?


Anyways, the growth in FreeOrion is always of the second kind, exponential growth, with a growth rate that increases over time, because of the planet expansion, as Krikkitone commented. We have several sources of growth: planets, population per planet, output boosts, with at least the planets being able to grow faster and faster. No need to worry about other forms of growth, they won't happen in FreeOrion.

Ophiuchus
Programmer
Posts: 3433
Joined: Tue Sep 30, 2014 10:01 am
Location: Wall IV

Re: Research Upkeep

#25 Post by Ophiuchus »

Oberlus wrote: Thu Oct 22, 2020 8:14 am exponential:
100, 101, 102.01, 103.0301...
x_t = 100 * (1+0.01)^t
This i meant. This has a constant part and an exponential part. As have we (once-per empire bonus like on the homeworld is responsible for non-exponential part).

Also it grows almost linear for a while. (100 + 1*t) = 100, 101, 102, ..
With that i mean the differences are so low 0, 0, 0.01, 0.0301 that until the explosion happens, it does not matter much in practice if that curve is linear or exponential.

At some point the differences caused by the non-exponential part becomes meaningless.

With your nerf of techs, the exponential explosion comes later.

And saying that tech costs should scale like this 100, 200, 400, 800, because that is exponential and we have exponential growth misses the point.

What I want to say "exponential" means different curves and we have to look at the actual curves happening in freeorion to figure out how much tech should cost.
Any code or patches in anything posted here is released under the CC and GPL licences in use for the FO project.

Look, ma... four combat bouts!

User avatar
Oberlus
Cosmic Dragon
Posts: 5715
Joined: Mon Apr 10, 2017 4:25 pm

Re: Research Upkeep

#26 Post by Oberlus »

Ophiuchus wrote: Thu Oct 22, 2020 11:40 am And saying that tech costs should scale like this 100, 200, 400, 800, because that is exponential and we have exponential growth misses the point.

What I want to say "exponential" means different curves and we have to look at the actual curves happening in freeorion to figure out how much tech should cost.
I think we all agree on that. The (exponential or not) growth that we apply to the tech costs should fit the growth that they can bring to the game. Or, from another perspective, should imply costs that require similar percentages of RP output: early game you can afford getting N early techs in T turns to gain +P% of production, late game you should afford getting N late game techs in T turns to gain +P% (or less) production.
We can't aim at achieving that literally, exactly, because of the multiple factors that affect growth in this game.

Post Reply