Marathon Scaling seems a little out of whack

Seyren

Chieftain
Joined
Oct 23, 2015
Messages
16
From what I understand, Vox Populi is balanced around standard game speed. Yet, I do think Marathon needs a little love and adjustment. Perhaps it is indeed the easier (or easiest) game speed to play with, but some numbers are quite off.

Lately, I tried Russia which has a bonus of 25% border growth and a bonus yield to Science on border expansion. On Marathon Ancient Era, you are given 70 science for border expansion. That's many turns worth of research which is quite substantial on Marathon. However, when you pair it with the faith for 20% border growth and have a number of cities settled, the science gain on Marathon becomes absurd, doubly so with Angkor Wat and the Tradition policy that increases border expansion growth by 25% respectively.

In the games I tried, I am often as much as 5-8 techs (at one point even 10 techs) ahead of the best AI leaders on Immortal difficulty without having to focus on Science. It might be a Marathon only problem, or it might be that Russia's bonuses are pairing up way too well with the early Tradition policy, wonder and faith, all of which are easily attainable.

But there are also other problems. Shoshone's bonus land grab on settling is somewhat problematic for Marathon speed. Usually it takes a long time on Marathon for a city to expand its borders, partly cause constructing a monument takes a good while. Shoshone's unique trait feels a little broken only because it often takes ages for cities to expand their borders to be on parity with the Shoshone. And even then, that's not really possible as the bonus land grab by Shoshone doesn't count towards as consecutive border growth.

Also, some city state quests are a little too rewarding. During the very early stages of the game, if you are yet to discover a civ, you can get a quest for it. The reward is a huge sum of Science, easily netting you a full tech and more, and encourages a sort of weird play-style where you are rewarded for not deliberately uncovering the location of other civs on the map.

And in Marathon, 1-2 tech difference is a big deal, the equivalent of many, many turns of research, especially if you are embroiled in a conflict or two.

Chandler did quite a wonderful post with https://forums.civfanatics.com/threads/balancing-game-speeds.626400/ on the problems with game speed, but I thought I did highlight more outliers.

I wonder if the numbers can be tweaked a little bit more because some of it is a little out of hand.

Thank you.
 
I was able to be ahead 5 - 8 techs in Standard Speed in inmortal with Russia too. But your advantage is going to decay at some point.

About the other things, you're right, specially with shoshone. But we can't balance that, early game and mid game is really long in Marathon, so, early game units and mid game units are really strong. We can balance things like gg point scaling, but we can't really balance entirely UA, UB, and UUs.
 
Marathon research costs are triple that of standard, border growth costs are triple that of standard, etc. It makes sense that any instant yields should be triple that of standard, because you receive them at 1/3 the speed of standard. The whole point of Marathon is that every turn is less meaningful compared to standard. The synergies you mention with Russia is just as absurd in Standard.

Shoshone's instant border growth is fine. The whole point of Marathon is that it takes ages to do anything. Shoshone in Marathon doesn't have its bonus land for any longer than in Standard when you account for the game speed making everything cost three times as much.

Conflict is the only exception in this, and is why Marathon is balanced around a Huge map. Conflicts generally take the same number of turns in Marathon as it does in Standard, so having more turns at a given tech level benefits the tech leader. However, a Huge map should have more cities and more civilizations to deal with, so there should be an appropriate increase in the number of conflicts.

Chandler's thread about City State balancing is worth looking into, but everything else here is exactly the reason why you play Marathon in the first place.
 
Marathon research costs are triple that of standard, border growth costs are triple that of standard, etc. It makes sense that any instant yields should be triple that of standard, because you receive them at 1/3 the speed of standard. The whole point of Marathon is that every turn is less meaningful compared to standard. The synergies you mention with Russia is just as absurd in Standard.

Shoshone's instant border growth is fine. The whole point of Marathon is that it takes ages to do anything. Shoshone in Marathon doesn't have its bonus land for any longer than in Standard when you account for the game speed making everything cost three times as much.

Conflict is the only exception in this, and is why Marathon is balanced around a Huge map. Conflicts generally take the same number of turns in Marathon as it does in Standard, so having more turns at a given tech level benefits the tech leader. However, a Huge map should have more cities and more civilizations to deal with, so there should be an appropriate increase in the number of conflicts.

Chandler's thread about City State balancing is worth looking into, but everything else here is exactly the reason why you play Marathon in the first place.

That's not correct. Many requirements (or yields) are not simply tripled for Marathon games.

Golden age points, great people points requirement as well as gold costs etc to invest are not proportionally tripled because everything should scale up by virtue of a magical multiplication. If it was as simple as that, Marathon would be much easier to balance for.

Also, yields are not correspondingly subjected to a fixed multiplier. I believe Russia gets 30 science on border expansion for standard, but only 70 on Marathon. Instant yields can break the game if all you did was just increased them by a factor of 3.

If anything, instant yields should be reduced so that they aren't too effective in Marathon games.

And no, I don't get your point for Shoshone. It does indeed take ages to do anything in Marathon, and that includes expanding your city by 8 tiles, all of which Shoshone receives for free while not incurring a border growth penalty. Shoshone's trait as well as Russia's absurd border expansion and science gain on Marathon make them far, far easier to play by way too large a margin.
 
Since this Topic is about Scaling for Marathon...

Is the big amount of flat Science from City State Quests really normal? Some, for finding other Civilizations Lands, give up to 430 Science. A Tier 1 Classical Era Tech was 460 Science on my Settings. And if you get 4 of these Quests at Turn 31 (because you found some CIty States but not the Land of overs Civs), you can end up as Maya done with Mathematics before Turn 100. This seems to be complety wrong - and happend two times for me now (Emporer Difficulty). It is also a big RNG factor if you get said quest in the first place. I feel like some AIs get the same kind of Quests, would explain why sometimes random Civs out of the blue 4 Techs ahead to everyone without being any kind of Science Civ.
 
That's not correct. Many requirements (or yields) are not simply tripled for Marathon games.

Golden age points, great people points requirement as well as gold costs etc to invest are not proportionally tripled because everything should scale up by virtue of a magical multiplication. If it was as simple as that, Marathon would be much easier to balance for.

Also, yields are not correspondingly subjected to a fixed multiplier. I believe Russia gets 30 science on border expansion for standard, but only 70 on Marathon. Instant yields can break the game if all you did was just increased them by a factor of 3.
This is observably false. From own my initial working data set on determining differences in game speeds:
Requirements (starting from first)
GAP for Golden Age
Standard = 750; Marathon = 2250
Faith for Pantheon and Great Prophet
Standard = 50, 800; Marathon = 150, 2400
GPP for Great Person
Standard = 150; Marathon = 450
GGP for Great General
Standard = 200; Marathon = 600
Food for Pop Growth
Standard = 15, 28, 43, 62; Marathon = 45, 84, 129, 186
Culture for Border Expansion
Standard = 20, 60; Marathon = 60, 180
Culture for Policy
Standard = 50, 70, 145; Marathon = 150, 210, 435
Production for Shrine
Standard = 65, Marathon = 195

Instant Yields:
Culture from Authority Opener
Standard = 6; Marathon = 18
Food/Gold from Authority Tribute
Standard = 15; Marathon = 45
Science from Russia Border Expand
Standard = 20; Marathon = 60

The only correct differences you've pointed out are that Buy and Invest costs are only ~2x the cost:
Invest Shrine
Standard = 80, Marathon = 170
I don't know why this is the case.

As I've said before, city state mechanics are more complicated, and are worth talking about.

And no, I don't get your point for Shoshone. It does indeed take ages to do anything in Marathon, and that includes expanding your city by 8 tiles, all of which Shoshone receives for free while not incurring a border growth penalty. Shoshone's trait as well as Russia's absurd border expansion and science gain on Marathon make them far, far easier to play by way too large a margin.
Please explain why having extra tiles at the start of the game is a much larger advantage in Marathon than in Standard. The Shoshone receive 8 free tiles in both Standard and Marathon. In Marathon, they get those extra tiles for three times as long, to fulfil requirements that are three times as expensive. It's the same advantage over other civs, regardless of game speed.
 
Please explain why having extra tiles at the start of the game is a much larger advantage in Marathon than in Standard. The Shoshone receive 8 free tiles in both Standard and Marathon. In Marathon, they get those extra tiles for three times as long, to fulfil requirements that are three times as expensive. It's the same advantage over other civs, regardless of game speed.

It takes longer for other civs to get tiles on Marathon than on Standard.
 
This is observably false. From own my initial working data set on determining differences in game speeds:
Requirements (starting from first)
GAP for Golden Age
Standard = 750; Marathon = 2250
Faith for Pantheon and Great Prophet
Standard = 50, 800; Marathon = 150, 2400
GPP for Great Person
Standard = 150; Marathon = 450
GGP for Great General
Standard = 200; Marathon = 600
Food for Pop Growth
Standard = 15, 28, 43, 62; Marathon = 45, 84, 129, 186
Culture for Border Expansion
Standard = 20, 60; Marathon = 60, 180
Culture for Policy
Standard = 50, 70, 145; Marathon = 150, 210, 435
Production for Shrine
Standard = 65, Marathon = 195

Instant Yields:
Culture from Authority Opener
Standard = 6; Marathon = 18
Food/Gold from Authority Tribute
Standard = 15; Marathon = 45
Science from Russia Border Expand
Standard = 20; Marathon = 60

The only correct differences you've pointed out are that Buy and Invest costs are only ~2x the cost:
Invest Shrine
Standard = 80, Marathon = 170
I don't know why this is the case.

As I've said before, city state mechanics are more complicated, and are worth talking about.

Please explain why having extra tiles at the start of the game is a much larger advantage in Marathon than in Standard. The Shoshone receive 8 free tiles in both Standard and Marathon. In Marathon, they get those extra tiles for three times as long, to fulfil requirements that are three times as expensive. It's the same advantage over other civs, regardless of game speed.

Are those values precise? I have not cleanly looked at the flat requirements in Standard Vs Marathon. Instead, I gleaned that information from Civ5gamespeeds.xml which you can open and edit with Notepad+ under \XML\Gameinfo. In there, the numbers controlling game speed, especially Great People, Golden Age etc are not multiplied by 3. A number of values are at 350% of standard (at 100%) instead of 300%, with a few at values much lower than 300% Vs Standard's 100%. As far as I know, this file is not overwritten by Vox Populi. How that adds up eventually as flat numbers isn't clear to me, but that file seems to control overall values as far as I know.

As for Shoshone, being able to work those tiles of your city immediately in Marathon is a much larger advantage than you would get in Standard. A number of Marathon values don't go up by a factor of 3 (if that file is correct), and since the start is the most important part of your momentum, being able to work most of your cities' tiles immediately leads to a larger snowball effect that few Civs can match, in my opinion of course.

Russia's Science gain on Marathon doesn't seem correct. I am playing a game as Russia in Marathon and I am pretty sure they get 70 science on border expansion, not 60. I will get back when I am at my computer.

Edit: Had time to do a quick test. It's 70 science on Marathon for Russia's science gain on border expansion (of any city) unless than announcement is wrong. In Marathon that's crazy at the start since I am only getting 9 science a turn.
 
Last edited:
It takes longer for other civs to get tiles on Marathon than on Standard.
Yes, and?
The bonus is multiplied by 3 (you have more tiles during 3 times more times), like almost everything else. So that is not an argument.

However, I always had the feeling that settling is quicker in marathon than standard (meaning less than three time longer), and if it is true, then yes it makes the Shoshone stronger, because aggressive settling is stronger in general.
 
Are those values precise? I have not cleanly looked at the flat requirements in Standard Vs Marathon. Instead, I gleaned that information from Civ5gamespeeds.xml which you can open and edit with Notepad+ under \XML\Gameinfo. In there, the numbers controlling game speed, especially Great People, Golden Age etc are not multiplied by 3. A number of values are at 350% of standard (at 100%) instead of 300%, with a few at values much lower than 300% Vs Standard's 100%. As far as I know, this file is not overwritten by Vox Populi. How that adds up eventually as flat numbers isn't clear to me, but that file seems to control overall values as far as I know.

As for Shoshone, being able to work those tiles of your city immediately in Marathon is a much larger advantage than you would get in Standard. A number of Marathon values don't go up by a factor of 3 (if that file is correct), and since the start is the most important part of your momentum, being able to work most of your cities' tiles immediately leads to a larger snowball effect that few Civs can match, in my opinion of course.

Russia's Science gain on Marathon doesn't seem correct. I am playing a game as Russia in Marathon and I am pretty sure they get 70 science on border expansion, not 60. I will get back when I am at my computer.
Your argument is predicated on your observations that Marathon costs are less than three times the Standard costs, while instant yields give more than three times they would in Standard. I agree that this would give disproportionate advantages to some civs over other. My argument is predicated on my observations that Marathon costs and Standard costs are directionally proportional to their instant yields. Given that we're arguing from different observations, it would be a good use of time to figure out why we are seeing such discrepancies.

My Marathon Russian games give 60 science on border expand in Ancient, exactly three times that of Standard. This is with a clean install of VP 3-20b and nothing else:
18-3-20 Russia Marathon Border Expand.png

Turn 32, I have progressed 33 Science into Trapping, with +7 Science per turn. Turn 33, My border expands, gaining me 60 Science; coupled with the 7 per turn, I end up with 100 Science.

Edit: Had time to do a quick test. It's 70 science on Marathon for Russia's science gain on border expansion (of any city) unless than announcement is wrong. In Marathon that's crazy at the start since I am only getting 9 science a turn.
The first technology in Marathon costs 198 science. With 9 science per turn, you will get that tech in 22 turns. Upon border expand, (60) science is gained, equivalent to 6.67 turns of research at +9 science, and will progress 30.3% through the tech. Compare the same situation in Standard: the first tech costs 66 science; with 9 science per turn, the tech is reached in 7.3 turns; upon border expand, (20) science is gained, equivalent to 2.2 turns of research, and will progress 30.3% through the tech.

Just because the amount of instance science gained is "crazy" compared to the per-turn science doesn't mean that the it's not appropriate for the game speed. Instant yields need to fulfil the same percentage of a given bucket no matter the game speed, because by definition the same absolute number of them are supposed to occur between game speeds. For example, a city can only expand a maximum of 84 times, no matter what game speed it's on, and each tile is gained at the same rate because Marathon tiles cost 3x that of Standard tiles. As another example, cities grow to the same size on Marathon and Standard; Marathon cities don't get to peak city size less than three times faster than Standard, given the same strategy and environment.

However, I always had the feeling that settling is quicker in marathon than standard (meaning less than three time longer), and if it is true, then yes it makes the Shoshone stronger, because aggressive settling is stronger in general.
Marathon Settlers cost three times the production, but settlers don't need to travel three times as far to get to a new city site. In a given game, how many turns "gained" does this grant a Marathon player? In the basic case, a settler travels four squares through grassland, costing two turns, settling on the third. Three marathon turns are worth one standard turn, so the Marathon player's new city has gained 6 Marathon turns, or is equivalently two standard turns ahead of the new Standard city in the same location. Is this a meaningful difference?
 
Marathon Settlers cost three times the production, but settlers don't need to travel three times as far to get to a new city site. In a given game, how many turns "gained" does this grant a Marathon player? In the basic case, a settler travels four squares through grassland, costing two turns, settling on the third. Three marathon turns are worth one standard turn, so the Marathon player's new city has gained 6 Marathon turns, or is equivalently two standard turns ahead of the new Standard city in the same location. Is this a meaningful difference?

Not that meaningful, in fact probably less than the effect of having earlier ancient ruins which is huge boost to Shoshones.
 
Top Bottom