Galaxy Size and Tech Costs

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

Moderators: Oberlus, Oberlus

Message
Author
User avatar
Geoff the Medio
Programming, Design, Admin
Posts: 12672
Joined: Wed Oct 08, 2003 1:33 am
Location: Munich

Galaxy Size and Tech Costs

#1 Post by Geoff the Medio » Fri Jan 14, 2005 5:32 pm

An issue we haven't really discussed (as far as I know), but IMO need to, is how to deal with greatly varying RP production due to differing universe sizes, differeing # of empires in a game and thus differing area and number of planets or systems (and thus RP production) for empires at otherwise equivalent stages of the game (or same turn).

IMO, we should try to have the approx number of techs being reserached at a given turn be independent of galaxy size, so that in large galaxies, every empire can't research every tech simultaneously due to having huge empires and thus huge RP production (exaggeration, but illustrative). Similarly, in small galaxies, we don't want every empire barely able to research a single tech due to tiny empires and tiny RP production levels.

RP production growth will presumably start to level off after empires expand to fill all available space in a galaxy, whereas RP growth would presumably be roughly exponential for uninhibited growth. After empire growth is roughly finished, the average size of empires would determine their average RP production, and thus number of techs researchable simultaneously for each empire, assuming fixed tech costs.

Correcting this would presumably require adjusting tech costs by galaxy size and/or # of empires in a game.

Games like Civ or (I think) SMAC do this, but I'm guessing EU doesn't, due to having a fixed-sized game map, and thus not having to worry about the problem. (Is this accurate? If EU has variable size maps, how does it keep tech progression in line with time / turns for different sized maps?)

(There are similar problems for unrestricted PP use at a given planet with shared production, but limits on local spending are apparently going to be discussed later, which should resolve that issue.)

User avatar
Bastian-Bux
Creative Contributor
Posts: 215
Joined: Fri Jun 27, 2003 6:32 am
Location: Kassel / Germany

#2 Post by Bastian-Bux » Sat Jan 15, 2005 8:00 pm

EU does this very heavily.

Your research cost is dependent on the number, religion and nationality of your provinces.
Wenn du die Macht hättest die Geschichte zu ändern, wo würdest du anfangen. Und viel wichtiger, wo aufhören?

If you had the power to change history, where would you start? And more importantly, where would you stop?

User avatar
Geoff the Medio
Programming, Design, Admin
Posts: 12672
Joined: Wed Oct 08, 2003 1:33 am
Location: Munich

#3 Post by Geoff the Medio » Sat Jan 15, 2005 8:59 pm

Bastian-Bux wrote:EU does this very heavily.

Your research cost is dependent on the number, religion and nationality of your provinces.
Based on a bit of google research, it seems that EU increases tech cost roughly linearly with number of provinces:

http://www.europa-universalis.com/forum ... adid=88245

There's some debate on that thread about whether adding more provinces generally gives a net benefit or penalty to research (though IMO it looks like more is generally better according to the formulas in the thread). Assuming it does, such adjustment for number of provinces in your empire would not adequately correct for variation in the number of provinces available to be conquered. ie. If every country had twice as many provinces, everyone's research would be significantly faster. The adjustment is helpful, but insufficient to deal with this problem (though the adjustment has other unrelated balancing benefits I imagine...).

User avatar
Bastian-Bux
Creative Contributor
Posts: 215
Joined: Fri Jun 27, 2003 6:32 am
Location: Kassel / Germany

#4 Post by Bastian-Bux » Sun Jan 16, 2005 9:10 am

Well, its not directly linear. And I also don't know how old this thread is. EU is a rather old prohgramm, but Johann (the lead programmer) is still making changes on it. Yes, its true, he id dedicating his free time to support a game thats only sold for a few bucks now.

As far as I know it, each colony increases the tech cost.

On the other hand the income of a colony is dependent on: religion, nationality, land connection and so on.


How could we use that?

I would keep the cost of theories the same, no matter how large the empire is. To research a theory needs a genius, and the number of genius doesn't decrease with an increasing empire.

The tech cost for applications I'd increase with the square root of planets in that galaxy. Why? Because many applications will have an instant effect, and that would be more costly if it needs to be applied at many places. Notify: if we do so, we shouldn't at the same time let the building costs of items scale with galaxy size, cause else applications that result in something to build (weapons and such) would be scaled twice.

The tech cost for refinements I'd scale 1:1 with galaxy size, as they have to fit all existing modells.


I know this suggestion isn't perfect, but it allows us to mirror some of the complexities.
Wenn du die Macht hättest die Geschichte zu ändern, wo würdest du anfangen. Und viel wichtiger, wo aufhören?

If you had the power to change history, where would you start? And more importantly, where would you stop?

User avatar
Impaler
Creative Contributor
Posts: 1060
Joined: Sun Jun 29, 2003 12:40 am
Location: Tucson, Arizona USA

#5 Post by Impaler » Tue Jan 18, 2005 12:41 pm

I think all tec costs Theory/App/Refinment should scale in the same manor to keep the relitive importance of each proportionaly correct.

All tecs are X RP over Y turns and our goal is to keep tec progress at a constant rate, so keep the Y turns constant at all sizes and only scale the X points per turn portion.

In addition I think it would be wise to use a Bit of the SMAC aproatch of scaling costs by the number of tecs owned. SMAC takes the number of tecs know to an Empire squares it and then muliplies by the Square Root of the # of map tiles (esentialy the hypotenus of the square map).

The explination here is that with so much more knowlage it becomes costlier and costlier to actualy discover new stuff because your scientists have so many new "false leads" and "distractions" and talent is split between so many differnt fields with less usefull "cross polination". In any event it helps to massivly reduce the power of huge empires vs what you would have with a strait linear tec cost.
Fear is the Mind Killer - Frank Herbert -Dune

User avatar
Geoff the Medio
Programming, Design, Admin
Posts: 12672
Joined: Wed Oct 08, 2003 1:33 am
Location: Munich

#6 Post by Geoff the Medio » Tue Jan 18, 2005 8:28 pm

Impaler wrote:I think all tec costs Theory/App/Refinment should scale in the same manor to keep the relitive importance of each proportionaly correct.
Good point. Now that you mention the gameplay justification, it seem that Bastian-Bux's argument about keeping theory costs constant is mostly a realism argument anyway.
All tecs are X RP over Y turns and our goal is to keep tec progress at a constant rate, so keep the Y turns constant at all sizes and only scale the X points per turn portion.
I'm inclined to agree.
SMAC takes the number of tecs know to an Empire squares it and then muliplies by the Square Root of the # of map tiles (esentialy the hypotenus of the square map).
Ignoring the way costs increase with additional techs issue, perhaps we should consider whether or not we want to encourage players to fill up large galaxies, give no bias on empire size, or encourage them to reach an optimal empire. We should also consider that the size of the galaxy has more and more effect on RP production as time goes on, particularly due to expansion.

At the start of the game RP production will be identical for all galaxy sizes, since nobody will have expanded enough to run into anyone else to limit their expansion.

For bigger galaxies, the unlimited expansion goes on longer, while in smaller galaxies, expansion will fill all space, at which point RP production area levels off. (RP still goes up due to growth and tech and such though... (Insert comment about convolving the planetary RP impulse response with the empire growth expansion curve)).

If we want to encourage people to fill all space, we shouldn't vary the cost of techs with empire size. If we want to have a particular size of empire regardless of galaxy size and number of empires, we'd probably want to increase research costs as a function of empire area, like in EU.

A problem might be that if we only vary tech cost with size of galaxy, and not at all with empire, then at the start of the game, tech costs will be huge for huge galaxies, even though empires and thus RP production are tiny, due to not yet having had time to expand... So there should be some sense of empire size perhaps... or an early holiday and then slow ramping up of additional cost factor towards a level that's determined by galaxy size... (which could be partly based on empire size, or on time or somesuch...)

Re: SMAC cost increase algorithm, this would have to be used in combination with the costs we have set per tech, as otherwise you end up with later techs costing just a much as earlier techs... Maybe we could come up with some formula that, as you learn more and more techs, makes the earlier ones than the average "level" of all techs you know cost less, but makes higher level techs cost much more.

Ray K
Krill Swarm
Posts: 12
Joined: Thu Feb 26, 2004 1:41 am

Re: Galaxy Size and Tech Costs

#7 Post by Ray K » Tue Jan 18, 2005 9:53 pm

Geoff the Medio wrote:An issue we haven't really discussed (as far as I know), but IMO need to, is how to deal with greatly varying RP production due to differing universe sizes, differeing # of empires in a game and thus differing area and number of planets or systems (and thus RP production) for empires at otherwise equivalent stages of the game (or same turn).
In the MOO clone I am developing, I specifically addressed this issue in my latest test version. Realism aside, this is a gameplay and immersion issue.

You don't want a situation where, in a large galaxy, your race is halfway or 2/3rds up the tech tree before you encounter your first race simply because you can fully colonize 10-15 systems before first contact.

Conversely, you don't want small-map galaxy to be always on low-tech because they have fewer colonies.

My solution was to simply scale the research cost for techs according to ratio of star systems to races, which is determined at start time and remains fixed throughout the game. This means that a 200-system game with 4 races would have the same tech research costs as a 600-system game with 12 races.

In MOO1, the cost for techs scaled according to difficulty level which makes sense only if acquiring the end-game techs makes the game easier to win. Since the ratio of stars from the smallest to largest galaxy in MOO1 was less than 10:1, galaxy size was less of an issue. My largest map size is currently 960 stars, which made it a more pressing issue for me to address.

User avatar
Geoff the Medio
Programming, Design, Admin
Posts: 12672
Joined: Wed Oct 08, 2003 1:33 am
Location: Munich

Re: Galaxy Size and Tech Costs

#8 Post by Geoff the Medio » Wed Jan 19, 2005 4:17 am

Ray K wrote:My solution was to simply scale the research cost for techs according to ratio of star systems to races, which is determined at start time and remains fixed throughout the game. This means that a 200-system game with 4 races would have the same tech research costs as a 600-system game with 12 races.
The problem with that is that at the start of the game, 4 empires in a 200 system galaxy are the same as they would have been in a 600 system or bigger galaxy. This remains the case until empires start filling up the space in the galaxy by running into other expanding empires.

In this case, during the initial expansion phase, tech costs will vary for no apparent reason, since empires are the same size, but galaxy size varies, thus changing the cost multiplier. If you start in a 200 system galaxy, it will take much less RP for a tech than it would in a 600 system galaxy... But there's no reason for us to want to make the tech cost vary at that point... the need to adjust it only applies after the smallest possible galaxy is filled up.

uruguru
Space Krill
Posts: 3
Joined: Fri Jan 09, 2004 4:46 am

Re: Galaxy Size and Tech Costs

#9 Post by uruguru » Wed Jan 19, 2005 4:43 am

Geoff the Medio wrote:In this case, during the initial expansion phase, tech costs will vary for no apparent reason, since empires are the same size, but galaxy size varies, thus changing the cost multiplier. If you start in a 200 system galaxy, it will take much less RP for a tech than it would in a 600 system galaxy... But there's no reason for us to want to make the tech cost vary at that point... the need to adjust it only applies after the smallest possible galaxy is filled up.
But there is no variation within the actual game. Once the game starts, the ratio is set.

Complaining about variable tech research costs in different games makes about as much sense are complaining about variable galaxy sizes or race locations. Most players intentionally change startup configurations for additional challenges.

Unless you are fine-tuning the starting locations of each race every game, then you cannot reasonably predict when the "expansion" phase ends. In fact, having a higher variability in starting positions contributes greatly to replayability. Sometimes you'll start 5 ly from the nearest race, sometimes 50 ly. That initial variability requires the use of different strategies, which makes the game less predictable and more enjoyable.

User avatar
Geoff the Medio
Programming, Design, Admin
Posts: 12672
Joined: Wed Oct 08, 2003 1:33 am
Location: Munich

Re: Galaxy Size and Tech Costs

#10 Post by Geoff the Medio » Wed Jan 19, 2005 6:19 am

uruguru wrote:But there is no variation within the actual game. Once the game starts, the ratio is set.
I wasn't talking about variation during a game. I was talking about variation between games. If tech costs scale with galaxy size, at the start of a game in a big galaxy, techs will cost more than at the start of the game in a small galaxy. But at the start of the game, empires are the same size, no matter how big the galaxy is. Thus RP production at the start of the game doesn't depend on RP production. If tech cost vary, and RP production doesn't, the number of techs that can be researched for the same size empire (before the player can do anything about empire size) will vary significantly. This, to me, is not desirable.
Most players intentionally change startup configurations for additional challenges.
I have no problem with an option to speed / slow research overall as an option at the start of a game. I have a problem with variation in tech progress due to galaxy size as a significant additional factor, separate from the explicit tech speed control. If galaxy size alters tech progress in addition to the "tech speed" control, it's not intuitive for the player.

Variation due to galaxy size will potentially also destroy game balance. If we have controlled variation, it can be used to correct for effects on balance and tech progress rate, while unpredictable variation cannot.
Unless you are fine-tuning the starting locations of each race every game, then you cannot reasonably predict when the "expansion" phase ends.
I'm thinking about the general / average cases, so when each individual race's expansion tails off doesn't matter. What matters is the total / average size of empires. After expansion is finished, the average empire size is equal to the size of the galaxy divided by the number of empires... the individual starting conditions of each empire are averaged out.

uruguru
Space Krill
Posts: 3
Joined: Fri Jan 09, 2004 4:46 am

Re: Galaxy Size and Tech Costs

#11 Post by uruguru » Wed Jan 19, 2005 1:17 pm

Geoff the Medio wrote:
uruguru wrote:But there is no variation within the actual game. Once the game starts, the ratio is set.
I wasn't talking about variation during a game. I was talking about variation between games. If tech costs scale with galaxy size, at the start of a game in a big galaxy, techs will cost more than at the start of the game in a small galaxy. But at the start of the game, empires are the same size, no matter how big the galaxy is. Thus RP production at the start of the game doesn't depend on RP production. If tech cost vary, and RP production doesn't, the number of techs that can be researched for the same size empire (before the player can do anything about empire size) will vary significantly. This, to me, is not desirable.
And therein lies the difference. To me, that is a small price to pay for the gain of assuring that the gameplay experience regarding tech is meaningful each game.

If you do not slow down the tech rate on large maps, the players' early isolation inthose games is going to allow them to use the same strategy to win each time. This, to me, is not desirable.

User avatar
Geoff the Medio
Programming, Design, Admin
Posts: 12672
Joined: Wed Oct 08, 2003 1:33 am
Location: Munich

Re: Galaxy Size and Tech Costs

#12 Post by Geoff the Medio » Wed Jan 19, 2005 5:09 pm

uruguru wrote:And therein lies the difference. To me, that is a small price to pay for the gain of assuring that the gameplay experience regarding tech is meaningful each game.
My point is that we can have meaningful tech experiences AND avoid paying the price that a simple solution (eg. cost increase dependent only on galaxy size) would entail.
If you do not slow down the tech rate on large maps, the players' early isolation inthose games is going to allow them to use the same strategy to win each time. This, to me, is not desirable.
I don't follow your logic there... explain?

To clarify: I'm not saying we shouldn't change tech speed for larger map... figuring out the <i>best way</i> to change the tech speed on larger maps is the whole point of this thread. I think it will be possible to find an adjustment that both keeps the early game tech costs the same before the size of the galaxy can affect RP production, as well as keep the game balanced later when different galaxy sizes result in drastically different RP production.

User avatar
skdiw
Creative Contributor
Posts: 643
Joined: Mon Sep 01, 2003 2:17 am

#13 Post by skdiw » Thu Jan 20, 2005 7:50 am

Simple solution: most of you are thinking a scalar factor like (a)(x) where x is the cost of of the tech and a is scalar. Since tech costs grows more like x^b, stick the a into b so the formula becomes x^(a+b).

In other words, you want the a be 'a' function of the growth of the game. 'a' can be approximate with x^a. So when (x^a)(x^b) simplifies to x^(a+b).
:mrgreen:

User avatar
Geoff the Medio
Programming, Design, Admin
Posts: 12672
Joined: Wed Oct 08, 2003 1:33 am
Location: Munich

#14 Post by Geoff the Medio » Thu Jan 20, 2005 8:23 am

skdiw: You'll have to explain that a lot more clearly...

Some specific questions (to get you started):

What do you mean by "function of the growth of the game"?

What do you mean by "'a' can be approximate with x^a" ?

Does your suggestion resolve the issues mentioned above for other "simple" modifications?

What advantages (or disadvantages) does it have, in your mind, compared to other possibilities, besides being "simple" ?

User avatar
Black_Dawn
Space Floater
Posts: 43
Joined: Sun Jan 09, 2005 4:00 am
Location: Canada

#15 Post by Black_Dawn » Thu Jan 20, 2005 6:25 pm

How about we end this argument by doing the easy thing:

Put in a changeable setting at the beginning of the game which will allow you to change scientific progress from "very slow" to "very fast" and everything in between. You could set it so that when you change galaxy size, the research speed changes appropriately (very fast for tiny galaxies), but could then be changed to whatever speed the player prefers.

It might be interesting to have a huge galaxy in which research progressed very fast, for instance, so that by the end of the game everybody has uber-tech. This option, combined with a "starting research level" a la MOO2 (pre-spaceflight, normal and high-tech) will allow players to create any kind of research meta-game they desire.
Professor Hernandez, Human ambassador to Silica:
"Hey, rocks are people too!"
Black Dawn

Post Reply