Research observations, experience => ideas

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

Moderator: Oberlus

Message
Author
FettJu
Space Krill
Posts: 4
Joined: Wed Mar 31, 2021 2:13 am

Research observations, experience => ideas

#1 Post by FettJu »

Hello!
I have been following this forum for a couple if years but last week was the first time in a couple of years I looked.

I downloaded the latest version and played it, and I like it alot. But I feel like the research is a little empty. I know you cut it down to not make it too confusing or complicated until you get it sorted out better for later. The names and the descriptions are very good and well thought through.
One thing I don't like, and all games in the 4x genre has the same problem, is that the research is fixed. It makes no sense and is not realistic. It doesn't make it more fun either, but make it more boring and makes you think that it is just a game and that you have to go to the research screen to pick things you know all about and then pretend it's research when a thing gets completed.

I have some ideas I will post here, this will be very long. I searched on the forum for the things I am going to write but I couldn't find anything about it. " Random research" just led to research that leads to getting one of Plasma Cannons or Planetary Ashclouds, and that's not what I'm after. I also found some interesting stuff about the refinement techs, but not exactly what I think would be cool.
These are just ideas, not "I want this in the game! Together with this!" and the numbers are all just examples.

Sorry for the long post!

****Refinement techs****
*All applied techs that result in a number in the game (increase PP by 0.2 per population, increase stealth of ship by 20 etc) should have refinement techs. E.g. Gas giant generator 2, 3, 4.

*There should be no limit to the numbers of refinements of each applied tech. Then it will be possible to keep using a tech until mid / end game, instead of always having to switch to the latest weapon or having to research a couple of new theoretical techs to get to a superior applied tech in that field. Gives variation and choice. (I think still a superior applied tech should be better than an inferior tech tho, like Black Shield 6 should be better than Defence Grid 28)
An easy U.I. to make it work could be that the refinement is sticking out to the right beneath the applied tech (or latest refinement). Something like: [Laser 2] 3] You click on the "3" just like now when you click on a tech to research it. After you choose it and it is being researched it will just be [Laser 3] with no "4" sticking out, until it's finished. If you haven't researched the applied tech yet, nothing will be sticking out beneath it.

*Then such refinement techs can be automatically repeated if you click some button on the tech in the queue window. If the next refinement of the same tech would be automatically placed last in the queue, then new techs can be added and the refinement techs will continue being researched when the other ones are finished without having to think about it. Since components and buildings automatically upgrade there is no micromanagement.

*The upgrade from a refinement tech should be randomized. Laser 2 increases damage of Laser with 1 to 3 points, Laser 3 increases the damage with 1 to 4 points, Laser 4 with 1 to 5 points etc. Diminishing returns will then be baked in: if the latest Laser refinement tech costs 500rp and with a bad dice roll you might only get an increase in damage of 1 point, it could be better to start researching the theoretical tech and applied tech needed to get the Plasma Cannon.
But if you get lucky with the dice rolls in the refinement techs then this empire will be different than the last one you played because you will choose other techs. It changes the refinement techs from something that is needed because Laser 1 is not very good - into something you might want to try because if you are lucky Laser 6 will be really powerful.

*If you later on can steal or conquer or trade research from aliens, maybe you come across an alien Plasma Shield with 12 damage reduction, while your own refinement tech Plasma Shield 4 only gives 6 damage reduction. Then your latest Plasma Shield refinement tech will be upgraded from 4 to 5 and your ships' shields will get 12 damage reduction

*There could be a new theoretical tech that improves the dice rolls for refinement techs. If Laser 4 improves damage of Laser with between 1 to 5 points, then this tech would make it improve between 2 to 5 points.





***New theoretical tech****
When all techs have fixed prices in RP and a fixed amount of turns to finish, and they are already known before hand, then it does not create any sense of discovery which is not fun, neither is it realistic (realism vs not fun)

First change all the techs so that when you start a new game every tech has only one step (the forced amount of turns it will always require to research a tech even if you have 1000rp), so you can only research one tech at a time until it is finished and RP can't be distributed to other techs before then.

(The names are not very good, just desciptions. Give some "lore")

*Tech A ("Delegated attention" or something): a new tech that makes all techs having several steps so you can research many projects at the same time. It makes the research work the same as it does now, but the number of steps/turns required to finish a tech is decided randomly, so a tech that now has 4 steps will get 3, 4 or 5 steps. A tech that has 8 steps will get between 7 and 10 steps. Or maybe totally random, but if a tech that has 4 steps now gets 10 steps it will not be so fun.
Simulates structuring of research and delegating research to different universities in the empire.
It also creates variation when you play.

*Tech B("Attention Deficit Hyperfocused Doctorate"): a new tech that requires Tech A. Makes it possible to change a tech so it can be researched in one step or several steps. You can change the tech while it is being researched. Good for when you have lots of RP but haven't uncovered Singularity of transcendence yet, or when you just want to finish a tech faster. Or if you have been focusing on one tech and now when you have researched 50% of it you want to begin research on a couple of other techs at the same time. Maybe there should be a higher cost in RP when you change a tech from having several steps to just one, as it could be unbalancing.

*Tech C ("Research optimization"): a new tech that requires Tech B. Makes it possible to choose how many steps a tech shall have. It simulates great understanding of setting up goals in research and anticipation of what the empires research will uncover.

*Maybe skip Tech A and just have it as it is now, but make the number or steps for each tech random for each time you start a new empire, for variation.








*Make each tech have a random RP cost. Instead of the price of one tech 200rp and another 500rp, have them cost 150-250rp and the other 400-600rp. Or maybe bigger or smaller spread, so the 500rp tech will be 300-700rp for example.
As you reach the lower limit in spending rp on a tech, every new turn there is a chance that the tech will have been finished, until you reach the upper limit when it will be finished automatically. The chance can be a fixed % or follow some formula that increases the chance as it gets closer to the upper limit.
A tech with several steps works the same: a tech's cost is 500-1000rp and has 5 steps; after 100rp has been spent in one step there is a chance every new turn that that step will have been finished, until 200rp has been spent in that step and it will finish automatically.
It would make research more thrilling and simulate sudden breakthroughs. And it simulates that you can't know beforehand what is required to discover something valuable.
To not make research too cheap it might be better to move the spread above the standard price, so instead of 500rp, 450-650rp, or 500-700rp, for instance, and use a formula to sharply increase the chance to have finished the tech every new turn.
It would make it even more realistic and a really new thing (I think) in 4x gaming if the tech cost had no upper limit, so a tech that now costs 500rp would cost 400-->>, so after you have spent 400rp there is a chance each new turn it will have been finished, and after 500rp the chance will increase each turn. "Tech Z will cost at least 400 RP" if it now costs 500rp. Then it will feel like your empire has discovered something when it is finished. Or at least the simulation of that is better.








*Make every tech's result have a spread instead if being fixed like it is now. Instead of getting empire detection level 50 e.g. you will get between 40 and 60.
It would simulate that research is going on until something of value and of use has been discovered or developed, and it is not possible to know exactly what will be achieved.
Gives a larger variation between empires with similar preconditions. Gives less total control and a reason to research refining techs, which makes the research each time you start a new empire different. Also gives lower level components a chance to be usable later in the game. If Laser 1 has a spread of 4-7 and you are lucky and get 7, it might be a reason to research refining techs for Laser instead of jumping to Plasma Cannon.

User avatar
Geoff the Medio
Programming, Design, Admin
Posts: 13587
Joined: Wed Oct 08, 2003 1:33 am
Location: Munich

Re: Research observations, experience => ideas

#2 Post by Geoff the Medio »

You might find this podcast episode about progression systems interesting:

https://player.fm/series/the-game-desig ... en-johnson

FettJu
Space Krill
Posts: 4
Joined: Wed Mar 31, 2021 2:13 am

Re: Research observations, experience => ideas

#3 Post by FettJu »

Yes it was. But I wouldn't agree with the part that the whole tech tree should always be researched tho. For me it's more fun if one empire has these big ships and extra strong armor that repairs itself from the Advanced Damage Control and another empire has those other small ships that use lots of shields and special missiles.

And randomization in tech creates diversity between empires in itself, as it also makes you have to do decisive choices. Will I keep Deflector Shield that only has 4 damage reduction and go on with other research or do I spend more on the refinement of the Deflector Shield to make it better? Or if I get a Deflector Shield that has 6 damage reduction I might wait longer with researching Plasma Shield and spend more on growth tech instead.

The reason why I would like this tho is the variation it would bring, you would never know exactly what will happen in research, it would be a conquest in itself

User avatar
Oberlus
Cosmic Dragon
Posts: 5714
Joined: Mon Apr 10, 2017 4:25 pm

Re: Research observations, experience => ideas

#4 Post by Oberlus »

Usually, I don't like randomization if it implies significative chances of bad/good luck. At least when playing against other human players.

In MoO2 I liked the dichotomies between Creative (slow research but you get everything) and Non-Creatives (faster research from ther racial picks but you can't get Holo Simulator & Supercomputers), although that could be diluted with spionage and Creative ended up being a poor choice no matter what.
I prefer FreeOrion's way of letting you research everything but having to choose what to get first. Plenty of interesting and hard choices to make (things can be improved, though).

I don't like letting basic/early techs to be upgraded indefinitely, that takes out part of the difficulty of choosing between more or less advanced techs (just go for the early techs, you can do blitzkriegs as well as keeping your old fleet always competitive thorughout the game, so I think that would take out a lot of variability in strategies, and make research strategy and game in general less interesting).

User avatar
LienRag
Cosmic Dragon
Posts: 2146
Joined: Fri May 17, 2019 5:03 pm

Re: Research observations, experience => ideas

#5 Post by LienRag »

Geoff the Medio wrote: Wed Mar 31, 2021 7:08 am You might find this podcast episode about progression systems interesting:

https://player.fm/series/the-game-desig ... en-johnson
Looking in the source of the page I found this URL https://hwcdn.libsyn.com/p/3/9/7/397c6b ... ohnson.mp3 but wget refuses to fetch it ?
(301 moved permanently to https://traffic.libsyn.com/bounce?url=h ... hnson.mp3 which then gives an error 404 message)
(and with nearly one hour in English it's a bit hard to follow online)

User avatar
LienRag
Cosmic Dragon
Posts: 2146
Joined: Fri May 17, 2019 5:03 pm

Re: Research observations, experience => ideas

#6 Post by LienRag »

Technology is quite predictable IRL. What isn't is theory.

I know FreeOrion doesn't value realism very much, but it should (and does somehow) value immersion, which is brought by having game mechanisms follow expectations coming from the real world behaviour.
I can't see a way to have a real unpredictable scientific discovery system in a computer game, though.

And randomizing the research rates and results of techs that will be there anyway will imho get quite boring after a few games, when all the techs will be known.

What could be done (but it's a lot of work, so basically it would need a team dedicated to it) is to have world generation that chooses randomly from a pool of theoretical possibilities (like, is String Theory true ?) and have each of these theoretical possibility come with a bunch of related techs (again, these could come in a tree : even if String Theory is true, there are still different options for it, so different trees of related techs).
The available techs would be seen only when some prerequisites are met : basically, IRL we know that a Dyson sphere is possible even if we don't know how to do them, and we know some of the theoretical possibilities offered by quantum computing even when, as of now, we don't know if really usable quantum computing is actually possible; but mostly we had no idea about the possibilities of carbon nanotubes before fullerenes were discovered, and to get back to String Theory, we don't really have a clue of what the applications would be if String Theory is true.

So the actual universe, from a scientific point of view (is String Theory true ? Are there exceptions to supersymmetry ? What is Gravity ? Is life just applied chemistry ? How many dimensions is the world ? ) would be discovered along the way, and be different each game (at least until all the possible combinations are explored, but that can be a lot).

What's more interesting imho is that the more advanced player in theoretical research would know what the world is much earlier than the other players and could plan his strategy according to it ("Wow ! It IS possible to mix Lithic and Organic life to create asteroid-dwelling artificial Species ?" Or "So you say that we may get that super-weapon if we build a phase-assembly-demogrificator on a planet inside a Void Rift ? I'll call the geography department and we'll see how to build a supply line to the next Void Rift" or "You mean we can talk to Gaïa ?").

Of course this would mean testing all the possible combinations for equilibrium between techs...

User avatar
Geoff the Medio
Programming, Design, Admin
Posts: 13587
Joined: Wed Oct 08, 2003 1:33 am
Location: Munich

Re: Research observations, experience => ideas

#7 Post by Geoff the Medio »

LienRag wrote: Wed Mar 31, 2021 12:25 pmLooking in the source of the page I found this URL https://hwcdn.libsyn.com/p/3/9/7/397c6b ... ohnson.mp3 but wget refuses to fetch it ?
I just right clicked on the "MP3" link (below "Subscribe" and "Play Later"), and (in FireFox) "Save Link As..." to download the file.

User avatar
LienRag
Cosmic Dragon
Posts: 2146
Joined: Fri May 17, 2019 5:03 pm

Re: Research observations, experience => ideas

#8 Post by LienRag »

Geoff the Medio wrote: Wed Mar 31, 2021 5:21 pm
LienRag wrote: Wed Mar 31, 2021 12:25 pmLooking in the source of the page I found this URL https://hwcdn.libsyn.com/p/3/9/7/397c6b ... ohnson.mp3 but wget refuses to fetch it ?
I just right clicked on the "MP3" link (below "Subscribe" and "Play Later"), and (in FireFox) "Save Link As..." to download the file.
D'oh !
Thanks.

He says interesting things (notably on being careful not to take agency out of the player, or that the tech tree is here to serve the game's logic) but not that much on the topics mentioned by the OP ?

User avatar
LienRag
Cosmic Dragon
Posts: 2146
Joined: Fri May 17, 2019 5:03 pm

Re: Research observations, experience => ideas

#9 Post by LienRag »

Oberlus wrote: Wed Mar 31, 2021 11:27 am Usually, I don't like randomization if it implies significative chances of bad/good luck. At least when playing against other human players.
It entirely depends on the stakes at play for each "roll of the dice".
For Universe generation, the thing is one more good planet in the Homeworld's vicinity is a huge difference early game, and due to the snowballing effect it will bring a huge difference to the whole game - especially in multiplayer as you pointed.
With multiple "dice rolls" with limited stakes each (like getting a +2 bonus rather than a +1 for a Mass Driver refinement) the odds even themselves in the long term, but that still would force adaptation of strategies (you get a +3 with good pilots, go for Blitzkrieg; if it's only a +1 you'd better plan for longer peaceful development before going to war).

Oberlus wrote: Wed Mar 31, 2021 11:27 am I don't like letting basic/early techs to be upgraded indefinitely, that takes out part of the difficulty of choosing between more or less advanced techs (just go for the early techs, you can do blitzkriegs as well as keeping your old fleet always competitive thorughout the game, so I think that would take out a lot of variability in strategies, and make research strategy and game in general less interesting).

This is a valid concern, i guess it depends on the mathematical formula for it ?
If Laser 12 costs much more than Plasma 3 (but can be researched more quickly if you have Laser 11) I don't think it would take out the variability in strategies, on the opposite it would give more choices on when to switch from one tech to the other.

FettJu
Space Krill
Posts: 4
Joined: Wed Mar 31, 2021 2:13 am

Re: Research observations, experience => ideas

#10 Post by FettJu »

LienRag wrote: Wed Mar 31, 2021 7:46 pm If Laser 12 costs much more than Plasma 3 (but can be researched more quickly if you have Laser 11) I don't think it would take out the variability in strategies, on the opposite it would give more choices on when to switch from one tech to the other.
This is what I meant, and not meant, with my post. If indefinate refinement techs would work like refinement techs work now, then Laser 12 would cost much more than Plasma 3. If instead refinement techs were more like addons to applied techs then Laser 12 maybe would cost something similar.
If indefinate refinement techs would work like now, that Laser 12 would be pretty damn powerful. But "light-touch" refinement techs wouldn't increase the power that much. Well, I would prefer a powerful Laser 12, that I had some interesting breakthroughs in research to get to. If I just got Laser 1 with 5 damage like it is now I would go straight to Plasma. But if I got a Laser 1 with 6 or 7 damage I would refine it more to make it stronger. It would be like I found a planet with a special. If I got a weak Laser 1 I would either refine it to get it to normal strength, as if a bad mishap in the game forced me to change my research plans, or just go to Plasma and hope I wont meet aliens until then.

I also read something about adding extras to all the weapons, I don't know how the plans are going. But if Lasers are more short range and armor penetrating and Plasma longer range and easier to hit, then there would be another reason to stay with a lower level tech.

As it is now all the aliens use Laser then Plasma then Death Ray. That is not fun.

I never play multiplayer, I go out to meet with people. If multiplayer is played like chess then yes, all techs should be exactly the same every round. But playing against the a.i. gets boring.

FettJu
Space Krill
Posts: 4
Joined: Wed Mar 31, 2021 2:13 am

Re: Research observations, experience => ideas

#11 Post by FettJu »

LienRag wrote: Wed Mar 31, 2021 1:00 pm Technology is quite predictable IRL.
Is it? So someone setting out to make a new microprocessor or a new hairwax knows how long time it will take and how much it will cost to get it?

Or if they already have a set time plan and budget, do they always know exactly what the result will be?

Think of SpaceX, they just throw money on engines and designs and try it until they get something that works about as good as they want it. Tech in a game should be like that to simulate science and technology. Otherwise it's just some thing a 4x game is supposed to have. Like they said in the end of that podcast, Civilization has stopped progress in strategy games. They must all have a fixed tech tree now. Because people wish to believe that they enjoy habits that were pushed on them by someone else.

User avatar
LienRag
Cosmic Dragon
Posts: 2146
Joined: Fri May 17, 2019 5:03 pm

Re: Research observations, experience => ideas

#12 Post by LienRag »

FettJu wrote: Thu Apr 01, 2021 2:03 am
LienRag wrote: Wed Mar 31, 2021 1:00 pm Technology is quite predictable IRL.
Is it? So someone setting out to make a new microprocessor or a new hairwax knows how long time it will take and how much it will cost to get it?
No, but they know they will get there.

I remember a boargame simulating WWII where there was "research points" that could be allocated to improve airplane, ship or tank technology. Every turn the player could decide to just add these research points or to throw the dice - if the dice roll was under the research points accumulated, then the research succeeded and the tech was improved.
If you can imagine something like that which would not imply micromanagement and would be fun to play, why not ?

User avatar
Oberlus
Cosmic Dragon
Posts: 5714
Joined: Mon Apr 10, 2017 4:25 pm

Re: Research observations, experience => ideas

#13 Post by Oberlus »

MoO2 does that.

User avatar
human2
Pupating Mass
Posts: 97
Joined: Tue Mar 10, 2020 3:22 am
Location: USA, Utah, Salt Lake City

Re: Research observations, experience => ideas

#14 Post by human2 »

I like the tech tree but the predictability of it is something of a downer to me. I usually research the exact same things in the exact same order. Certain things will modify it a little bit, but I really wish there was more randomization and less work. That is, I wish I could have the predefined research lists just like ships are predefined.

And just like how galaxy age, size, etc are randomized, it would be nice if the tech tree had either random missing techs, or that the costs were randomized. But not every game. More like how you can pick X different galaxy ages, you could pick tech balances, like favors military tech, favors psionic tech, regardless of the race the player picks. Basically you pick the way it's balanced.

The reason I would like this is because every time the balance changes, it's like a brand new game. I experienced this last night when I just played a new build that has totally different ship values. But it has to be consistent so that the game types can be played repeatedly. If it was different every game, it takes away the experimental poking and prodding part that I guess is what I like in the game.

User avatar
Oberlus
Cosmic Dragon
Posts: 5714
Joined: Mon Apr 10, 2017 4:25 pm

Re: Research observations, experience => ideas

#15 Post by Oberlus »

Oberlus wrote: Sat Apr 03, 2021 11:00 pm MoO2 does that.
For those that didn't play MoO2, this is the way it worked:

Each tech has a fixed cost, C, and the player invests research points on it, in a counter X.
Probability of breakthrough (discover the tech) is equal to (X-C)/C.
When X<C chance is 0, when X>=2C chance is 100%.
So when you get the tech shortly after reaching minimum cost C you are lucky, and when you reached 2C you get it for sure (unlucky).
That allowed for some micromanagement: if you reach a chance of say 25% or 50%, you can move all scientist to workers and just wait a couple of turns until you get the tech without any more investment (except from flat bonuses).

FreeOrion could do something similar. This would allow for randomization of how fast you get the tech, but not what the tech gives you.

human2 wrote: Mon Apr 05, 2021 1:32 am you could pick tech balances, like favors military tech, favors psionic tech, regardless of the race the player picks. Basically you pick the way it's balanced.
Maybe each tech category could have a cost multiplier, default to 1.0, adjustable via game rules before game start, allowing also random values between 0.75 and 1.5.
human2 wrote: Mon Apr 05, 2021 1:32 am I wish I could have the predefined research lists just like ships are predefined
I think that's been suggested already, to be able to save and load research queues.
Implementation isn't as straightforward as ship designs, because upon load the queue should be modified by whatever the player has already researched.
But it is a good idea nevertheless. I would use it indeed.

Post Reply