Farming & Growth Meter Effects (for v0.3 ?)

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: 12675
Joined: Wed Oct 08, 2003 1:33 am
Location: Munich

#31 Post by Geoff the Medio » Fri Jul 16, 2004 2:01 am

emrys wrote:If I produce 1 unit per turn, I gain a tech that doubles my output (i.e. has a+1 bonus), and I now produce 2 per turn. To double my output I now need a tech that stacks a +2 bonus on top, at which point I produce 4 per turn, and need a +4 bonus tech to double my output, etc.... Seems simple enough.
I don't think it's necessary to double output to have equal strategic effect. If we start with +1 and +2, then go to +2 and +3, the difference is still 1, so the strategic situation is essentially unchanged.

I understand what you're saying already, and saying it 5 more times won't convince me. (You might be able to outvote me or convince the leads though...).

guiguibaah
Creative Contributor
Posts: 441
Joined: Tue Oct 07, 2003 1:00 am

Hmmmm....

#32 Post by guiguibaah » Fri Jul 16, 2004 2:13 am

I think whether the scale should be linear or logarithmic could also depend on how easy it is to expand to meet that extra bonus...

For example: I could spend a lot money on research to get an extra +1 in farm production. It cost me 50 buckazoids to do that. Now my 10 farms (100 food) now produce 110 food.

Or, I could spend 10 buckazoids and simply go out and get a new farm. In that case, the +1 was not much of a good investment.



Now, come late game, when farming worlds are being plagued by Sariens and space garbage, and I can't expand, that bonus is going to start to look pretty tasty now.

The rate at which I can expand will also depend on the galaxy size. For a small galaxy, an exponential increment may bear bigger fruit if it costs more to research a bonus than to build more farms... Vice versa for a huge galaxy.
There are three kinds of people in this world - those who can count, and those who can't.

grinningman
Space Krill
Posts: 9
Joined: Wed Jun 30, 2004 5:00 am

#33 Post by grinningman » Fri Jul 16, 2004 2:40 am

Geoff the Medio wrote: If we start with +1 and +2, then go to +2 and +3, the difference is still 1, so the strategic situation is essentially unchanged.
It is not the difference between two resource outputs that is important to the strategic situation. It is the ratio of the resource outputs.

e.g. If my empire produces 1000 PP/turn I will crush another empire that produces 100 PP/turn. If my empire produces 100,000 PP/turn I will have a very close contest with another empire that produces 99,100 PP/turn.

The difference is not a strategically important number, it is the ratio that is important. In the first case my production is 10 times bigger than the opponent, in the second case it's only 1.009 times bigger.

Just thought I'd back up emrys on this :)

BTW I've been lurking on and off since the project began; it's looking very exciting! Can't wait to try out the next version.

User avatar
Geoff the Medio
Programming, Design, Admin
Posts: 12675
Joined: Wed Oct 08, 2003 1:33 am
Location: Munich

#34 Post by Geoff the Medio » Fri Jul 16, 2004 4:11 am

Ok...

I'll admit that difference doesn't matter if it's less than the random chance factor (say 10% of the production amount / turn for example). But for similar ranges of values of production, equal additions to both empires do cancel out equally, and what matters is the difference.

If you compare +1000 vs +900 and +1 vs +101, the difference ( +100 ) is the same as the random chance factor in the larger case ( +100 ), so the bonuses don't cancel.

However, for more reasonable scales, equal bonuses cancel out.

+1 vs +2 is the same as +2 vs +3

In either case, the latter empire ends up with a +1 advantage, which is the same advantage since random chance (0.3) in both cases is significantly less than the difference (1).

+200 vs +250 is the same as +250 vs +300

Again, random chance (30) is less than the difference (50)

The 10% is completely arbitrarily chosen, but illustrates the point. In examples like +1 vs +100 compared to +100001 vs +100100, the random chance (1001) completely overwhelms the difference (99), so it appears that only the ratio matters.

For more reasonable comparisons, the difference is what matters.

To avoid random chance overwhelming differences, we can ocassionaly (rarely) increase the the bonuses for techs (subject to balancing, obviously). The meter itself should be linear.

You can't just ignore the importance of the linear difference and use an exponentially increasing effect, because not all bonuses are equally matched. Consider empires with +1 vs +2 and empires with +1 vs +3. In the former, the +1 for the second empire gives a 100% advantage over the first empire. In the latter, the +2 bonus for the second gives a +200% advantage over the first. The ratio of advantage to size of meter bonus is constant! Exponentially growing advantage per bonus would not have this.

grinningman
Space Krill
Posts: 9
Joined: Wed Jun 30, 2004 5:00 am

#35 Post by grinningman » Fri Jul 16, 2004 5:17 am

I agree with all your reasoning, just not the conclusions :). Anyway, this is academic since the resource output in drek's 'Meters and Infrastructure' is linear with the meter level. And as you (and emrys) have said, with a 0-100 scale it's easy to make the later bonuses add more to the scale if it's needed.

User avatar
Geoff the Medio
Programming, Design, Admin
Posts: 12675
Joined: Wed Oct 08, 2003 1:33 am
Location: Munich

#36 Post by Geoff the Medio » Fri Jul 16, 2004 6:56 am

If you convince the leads that drek is wrong and that what he is proposing is unworkable, his proposition might not be accepted. You might also convince drek, I suppose.

Admittedly, I don't expect either of those will occur, as there is sufficient range in 0-100 to have various-sized bonuses, as has been mentioned.

User avatar
noelte
Juggernaut
Posts: 872
Joined: Fri Dec 26, 2003 12:42 pm
Location: Germany, Berlin

#37 Post by noelte » Fri Jul 16, 2004 8:04 am

I can't follow the arguments which state that any newer tech has to double the effects, e.g. +1,+2,+4,... I think +1,+2,+3 is just fine.
grinningman wrote:e.g. If my empire produces 1000 PP/turn I will crush another empire that produces 100 PP/turn. If my empire produces 100,000 PP/turn I will have a very close contest with another empire that produces 99,100 PP/turn.
If your race has an advantage of say +1 which leads after 500 turn to 100,000 PP/turn vs. 99,100 PP/turn, what have you done at early stages? I think your race have to take advantage of it's picks from the beginning. If you face after 500 turn the mentioned situation, your strategy might be buggy.
Press any key to continue or any other key to cancel.
Can COWs fly?

emrys
Creative Contributor
Posts: 226
Joined: Fri Oct 24, 2003 3:44 pm

#38 Post by emrys » Fri Jul 16, 2004 9:12 am

Although dragging this out seems fairly pointless (since those who can see the difficulty, can see that it isn't relevant anymore, now that we've got a bigger range of values to work with, and those who can't see the difficulty will be able to change the values they pick for bonuses when balancing things so that there doesn't end up a problem, so they'll never actually need to see the difficulty explicitly, so all told this is a non-issue) , for cussedness' sake I'll have one last go.

[note about terms of discussion]
There are many factors to be considered here. Empire A produces A(t) at turn t, empire B B(t). There are the relative (a/b) and absolute (A-B)advantages that an empire might have over another at a particular instant, there is the sum of that advantage over time, SUM[t](A(t)/B(t)). and then there is the cumulative advantage an empire might be able to gain after time from starting out in front I.E A(t2) c.f. B(t2).

It's important not to mix these up, overwise you end up thinking you've countered an argument, when all you've done is made an argument about something totally different.
[/note]

@Geoff
My "there are at most six fully stackable techs with equally desireable bonuses" statement was meant to convey just that. i.e not that we should outlaw techs that give bonuses of +7, or that they would be unusable. That's silly. It just meant that if you had +16, and your last bonus was +8, then you'd be looking for a bonus of about +16 for a similar relative cost, to have the same magnitude of effect, anything with a lower bonus would, other things being equal - which they rarely are, be less significant in your considerations.

Now, given that we now have a reasonable range of numbers to work with, we have several reasonable choices. Either

a) keep the costs of each additional +1 bonus roughly constant in absolute terms, so that in relative terms getting the extra bonuses becomes less costly, so the cost for an equivalent relative bonus is pretty constant in relative terms. In other words make player fight like hell for piddly +1 and +2's early game, but give away +5's and +6's like confetti in late game...

or

b) increase the mapping function from bonuses to effects roughly geometrically, and keep the cost to get each +1 (or ~+10 given the new range) roughly equivalent in relative terms (i.e. later techs give similar scales of bonuses to early techs, but they cost much more, because they generally stack with the early game ones you are expected to have got, and so get you to ever more dizzy heights of total bonuses).

or, the route we will probably take if we don't explicitly consider the issue

c) a bit of both, late game techs cost a bit more, give a bit better bonuses, the two together make most techs fall on a geometric absolute cost/benefit path (i.e. a constant relative cost/benefit path), and we play with both numbers constantly until they get somewhere near balanced.

Like I said, it really doesn't matter, since the first balancing pass of method c) we probably be to get us back to the effect of a) or b) anyway, from which point on they are all pretty much the same in practical terms, if not conceptually.

p.s. For anyone who really thinks that it's the absolute differences that matter, not the relative ones, imagine what the variety of reponses of people who earn different amounts of money would be to a suggestion that all income or expenditure related taxes should be scrapped and replaced by a fixed tax of $2000 per year... (n.b. yes, this is a realism argument, and as such a bad thing, however the underlying economics principle of 'relative marginal value' applies to the game as well, and the real world example just highlights it.)

drek
Designer Emeritus
Posts: 935
Joined: Thu Jun 26, 2003 8:07 am

#39 Post by drek » Fri Jul 16, 2004 10:44 am

Late game techs should give less, not more.

a) These bonuses are mutliplied by population, of which a late game empire has a lot more. An empire-wide bonus is worth a lot more to a population of 10,000 (compared to circa 20 for a new empire.)

b) We can assume, in late game, that expansion carries with it a lot more risk, since decent worlds have already all been colonized. Therefore, adding more tech would be a certain way of improving economy opposed to risking defeat in war.

c) Maintaince fees limit the number of ships and buildings in the empire. We'd want to define a relatively sane "max realistic cap" for production of maintaince resources (food, money) on each world. Once the player reaches this cap, improvement through technology should become much more difficult to help avoid run-away exponential expansion.

emrys
Creative Contributor
Posts: 226
Joined: Fri Oct 24, 2003 3:44 pm

#40 Post by emrys » Fri Jul 16, 2004 4:45 pm

Drek, I egree entirely with points b and c, essentially this whole silly thread is mostly about point a:

I contend that "an empire wide bonus is worth a lot more to a population of 10,000 (compared to circa 20 for a new empire)" (i.e. a bonus whose effect scales with population) although trivially true in absolute terms, in relative terms is false, the bonus is worth exactly the same to both empires in relative terms, since it has the same proportional effect on their outputs, and hence about the same likelihood of affecting their strategy.

p.s. before we go any further, let me make it very, very clear that I am NOT suggesting we change the proposed resource or meter system, I am happy to debate a point of logic ad infinitum... :twisted:

User avatar
Ablaze
Creative Contributor
Posts: 314
Joined: Tue Aug 26, 2003 6:10 pm
Location: Amidst the Inferno.

#41 Post by Ablaze » Fri Jul 16, 2004 10:39 pm

I've been thinking about it, and I really like the idea of exponential growth being applied to the meter rather then applied as an increased bonus to the technology. This solves several issues that have been nagging at me for a while:

1) A racial bonus (which would be applied as a constant amount (i.e. a constant +5) throughout the entire game in any system I can think of) would now have a significant effect at all points of the game rather then mostly just at the beginning.

In other worlds, a +10 to farming would be significant at the beginning when you can only scrape together a +8 through other means, and it would be significant later in the game when you can scrape together +40 through other means.

Let me pull some numbers out of my ass here, I'm going to make them large in order to make the point easier to illustrate: The +8 you would be getting without the racial bonus might give you 1 food/pop while the +18 you are getting with it might give you 2.8 food/pop. So your racial bonus of +10 gives you an extra 1.8 food/pop at this early point of the game.

Now let’s look at later in the game when you have +40 to food production through technology. If the scale was linear (so you got only 4 food/pop from this) then an extra +10 would no longer seem like the massive asset it did early in the game. If, however, the scale is roughly exponential and you got, say, 17.8 food/pop at +40 and 41.1 food/pop at +50 then that +10 that you sacrificed everything to get back at the beginning of the game would still seem like a significant advantage.

2) Applying exponential advancement to the meter frees the balancing to be done at the scale level rather then at the individual technology. IOW if the balancing team is finding that people who concentrate on mining are getting too much of an advantage late game then they don’t have to look at the dozen or so mining techs. (all who have different bonuses for different reasons.) Instead they leave the actual technologies alone (all mining techs give you between +1 and +3 depending on their cost relative to that point in the game and if they give you bonuses in any other areas or not) and adjust the scale so that a +31 to mining will give you 2.4 minerals/pop rather then 2.6.

3) Applying exponential advancement to the meter will free up much of the need for prerequisite technologies. I’ve played many games that resemble freeorion, and I know that once people get to know the game and the tech tree they tend to do a bull rush for some higher level technology and skip as many of the ones in between as possible. This is actually not such a bad thing in and of itself. The problem is that now they’ve climbed half way up the tech tree and those measly +5s and +7s at the base of the tree don’t seem significant at all compared to the +45 techs they can now access. If, however, those fundamental techs gave roughly the same bonuses at a much smaller cost then filling out your base techs after achieving your pet tech would seem like a very attractive option.
Time flies like the wind, fruit flies like bananas.

muxec
Space Kraken
Posts: 152
Joined: Tue Jun 15, 2004 7:55 pm

#42 Post by muxec » Sun Jul 18, 2004 8:49 am

It'd be cool to have pop that depends on health. And health that depends on food consumption (like in eciv). For example health may be a deriviative of population and health must have deriative that depends on curent health and food consumption.

guiguibaah
Creative Contributor
Posts: 441
Joined: Tue Oct 07, 2003 1:00 am

Tech Balances

#43 Post by guiguibaah » Tue Jul 20, 2004 9:54 pm

Well, one other thing to discuss, is if you are researching a +1 to food, or a +10% to food, what are you forsaking in the meantime?

Perhaps your empire now earns 100,000 pp as opposed to the other empire that earns 99,000 pp... Except that 99,000 pp empire chose instead to research 'advanced spying' instead of 'farming tactics' and has now stolen 'plasma bolt', 'biospheres' and just that last turn stole 'farming tactics' from you.

In other words, at what stage is the research (food) bonus available, and what are the similar techs you could research that you had to forsake in the meantime?
There are three kinds of people in this world - those who can count, and those who can't.

muxec
Space Kraken
Posts: 152
Joined: Tue Jun 15, 2004 7:55 pm

#44 Post by muxec » Wed Jul 21, 2004 7:21 am

Maybe we must forget MOO2 and implement non-exclusive technologies?

User avatar
Ablaze
Creative Contributor
Posts: 314
Joined: Tue Aug 26, 2003 6:10 pm
Location: Amidst the Inferno.

#45 Post by Ablaze » Wed Jul 21, 2004 5:06 pm

What do you mean by those, mux? Please go into detail.
Time flies like the wind, fruit flies like bananas.

Post Reply