Real base gather rate on Mills/Plantations

[duckzilla]

As we already know, the base gather rates on Mills/Plantations is 0.67/0.5. But as european villagers walk around while gathering the corresponding ressources, their real gather rate is lower than that, since walking around does not count as gathering. It was difficult to find any useful information on real gather rates, which is why I just did an experiment.
My setup:

• Single player treaty 40 game against one AI opponent on sandbox difficulty
• Map was Amazonia to have as few AI interaction as possible
• Use of cheats to get age up fast/get ressources/build 80 villagers
• Put 40 vills on mills + 40 vills on coin
• Start at "35 minutes remaining", end at "15 minutes remaining"
• Checked how many ressources I had at starting point and end point
• No upgrade cards/upgrades in the buildings themselves

Result:

• In 20 minutes game time, 22.486 food and 17.028 gold were gathered by 40 villagers each
• This translates to (rounded) real base gather rates of 0.468 food/sec per villager and 0.355 gold/sec per villager

Please not that gather rates on rice paddies are 0.5 food/sec per villager and 0.34 gold/sec per villager. As villagers do not walk around on rice paddies, this number is already correct. Unupgraded rice paddies are ~7% better than unupgraded mills, while they are ~4% weaker than unupgraded plantations.

I did not test having just 7 vills per mill/plantation. If anyone bothers to find out, just post it here and I will put it to the results (with a reference to the fellow player ofcourse).

Now the interesting question is: whose countries villagers have the potential to be the strongest?
Let's take a look at some examples.

Japan
Japan has lots of upgrade home city cards for rice paddy gather rates. They add up to +81% faster gathering of food and +121% faster gathering of coin. Rice paddy upgrades themselves add a further +95% and + 120% respectively. Hence, fully upgraded japanese villagers on rice paddies will gather ~1.38 food/sec and ~1.16 gold/sec.

Russia
As far as I know, Russia has all european upgrade cards for mills/plantations (+ Economic Theory). They add up to +60% faster gathering of food and +100% faster gathering of coin. Mill and plantation upgrades add a further +95% and +110%. Hence, fully upgraded russian villagers on mills/plantations will gather ~1.19 food/sec and 1.10 gold/sec.

India
India is one of the countries with relatively few upgrade cards. They add up to +30% faster gathering of food and +45% faster gathering of gold. Rice paddy upgrades again add a further +95% and +120% respectively. Hence, fully upgraded indian villagers on rice paddies will gather 1.12 food/sec and 0.9 gold/sec. Additionally, indian villagers get a +10% bonus for being close to Karni Mata which yields 1.21 food/sec amd 0.99 gold/sec for each villager.

European/Asian without upgrade cards
A european country without any upgrade cards from HC but all upgrade at the mill/plantation themselves will gather 0.91 food/sec and 0.74 gold/sec. An asian country without any upgrade cards from HC but all upgrades at the rice paddy will gather 0.975 food/sec and 0.75 gold/sec.

That leaves fully upgraded japanese villagers being

• ~16% faster at gathering food and ~5% faster at gathering gold than fully upgraded russian villagers
• ~14%/~17% faster than fully upgraded indian villagers close to Karni Mata
• ~50%/~56% faster than european villagers with all mill/plantation upgrades but without HC cards
• ~42%/~55% faster than asian villagers with all rice paddy upgrades but without HC cards

Note that the japanese player sent a total of 11 home city cards while the russian player sent 7 home city cards and the indian player sent just 4.

[jgals]

due to the widespread tr belief that tasking only 6 or 7 vile to each building yields better gathering I ran that test in a similar fashion, Ifound that gathering was exactly the same per vil no matter how many vils were tasked to the building 1 or 10.

similarly due to the widespread belief that certain animals are better on shrines I tested that and found no difference. Some people even believe that fattened cows are better believing it has to do with animal size; No difference in my testing. none whatsoever. I believe the figure is .84 per shrine no matter what kind of animal or how fat it is

[edeholland]

Hence, fully upgraded indian villagers on rice paddies near Karni Mata will gather 1.12 food/sec and 0.9 gold/sec. Additionally, indian villagers get a +10% bonus for being close to Karni Mata which yields 1.21 food/sec amd 0.99 gold/sec for each villager.

Are you counting Karni Mata twice here?

When playing offline, or against bots, you have almost zero lag which really influences the gather rate. Offline you can get a couple of hundred more score at the end of a (long) treaty. I think testing this online against a player would yield different results.

@jgals Do you have your testing documentated? I'm pretty sure llamas yield different results than normal hunts on shrines.

[duckzilla]

Are you counting Karni Mata twice here?

When playing offline, or against bots, you have almost zero lag which really influences the gather rate. Offline you can get a couple of hundred more score at the end of a (long) treaty. I think testing this online against a player would yield different results.

My bad! I did not count Karni Mata in the first calculation and just corrected it in the original post!

This whole analysis is rather about the relative strength of villagers between different civs. Any findings in this relative strength are independent of lag, because lag and other reasons for slower gathering are identical to all players within the same game. Hence, the exact levels of gathering per second may be considered "in the optimal case", while the results for relative villager strength will always hold.

[edeholland]

I don't think lag has the same influence on all civs, because I think the lag makes it worse for villagers that have to walk around. @Cometk or @dicktator_ probably knows more about this though, this debate has come up before.

[Gichtenlord]

There is a huge score offset when you boom offline or vs comps online for nilla and twc civs, while tad civs' score offset is fairly similar.
Obviously this is caused due to walking on mills, farms and plantation, while vills dont walk on rice paddies.

[duckzilla]

Ah ok, that is bad. Especially as these losses due to lag are quite difficult to quantify. At least we can conclude that any relative gathering speeds between euro/asian civs are at best (from euro perspective) as calculated above, while any lag interferences benefit asian civs or the dutch (due to banks and less vills).

[_H2O]

Might be nice to stop the walking and fix the rates on mills and plants. Seems wrong for treaty balance to be impacted so severely by lag.

[jgals]

Hence, fully upgraded indian villagers on rice paddies near Karni Mata will gather 1.12 food/sec and 0.9 gold/sec. Additionally, indian villagers get a +10% bonus for being close to Karni Mata which yields 1.21 food/sec amd 0.99 gold/sec for each villager.

Are you counting Karni Mata twice here?

When playing offline, or against bots, you have almost zero lag which really influences the gather rate. Offline you can get a couple of hundred more score at the end of a (long) treaty. I think testing this online against a player would yield different results.

@jgals Do you have your testing documentated? I'm pretty sure llamas yield different results than normal hunts on shrines.

no i have not documented that, but I could. In fact i have not tested with Llamas, but I think the test I did was with hunts/sheep/cows. The hunts were probably serow or the orinoco tapivara. I think I also checked in Deccan once with eelphants vs deer/sheep. Llamas being better might be somethign I didn't know about. I guess that means only with a spanish team mate can japans economy be maximized? Kind of a rare occurrence TBH

[_H2O]

Shrines don’t need to be tested. The gather rate is on the UI and larger animals do provide more trickle. Larger means amount of food not physical size. Livestock provide even more.

That’s my world view but it could also be wrong.

[princeofcarthage]

Animal gather rates for shrines: Without any upgrades
Pronghorn: 0.07/sec
Capybara: 0.07/sec
Deer: 0.07/sec
Caribou: 0.07/sec
BighornSheep: 0.07/sec
Turkey: 0.07/sec
Rhea: 0.07/sec
Reindeer: 0.07/sec
Guanaco: 0.07/sec
Saiga: 0.07/sec
Nilgai: 0.07/sec
Ibex: 0.07/sec
serow: 0.07/sec
MarcoPoloSheep: 0.07/sec
MuskDeer: 0.07/sec
Bison: 0.08/sec
Elk: 0.08/sec
Tapir: 0.08/sec
Moose: 0.08/sec
MuskOx: 0.08/sec
Seal: 0.08/sec
WildElephant: 0.08/sec
GiantSalamander: 0.08/sec
Sheep: 0.11/sec
Goat: 0.11/sec
GoatFat: 0.11/sec
Llama: 0.13/sec
Yak: 0.13/sec
Cow: 0.14/sec
SacredCow: 0.14/sec
WaterBuffalo: 0.14/sec

[duckzilla]

Given that shrines base gather rate is 0.14/0.10/0.10 per second (food/wood/gold) and the fact that this rate can be changed by the age IV upgrade (+350% for food, +300% for wood/gold) + toshogu wonder (+50%) + heavenly kami (+50%) (--> +400-450% at max), the fully upgraded late-game base gather rate is: 0.77 / 0.50 / 0.50.
In the ideal situation (e.g. team game with indian player), the maximum possible gather rate with cows/sacredcows/waterbuffalo on shrines is: 1.33 / 1.06 / 1.06. Hence, the ideal shrine is gathering 73%/112%/112% faster than the base gather rate without animals. Comparing with late-game villagers, this is roughly equivalent to 1 - 1.1 vills which totals 20-22 villagers for 20 shrines.

*updated numbers as mentioned in post below

[guyshir]

Given that shrines base gather rate is 0.07/0.05/0.05 per second (food/wood/gold) and the fact that this rate can be changed by the age IV upgrade (+300%) + toshogu wonder (+133%) + heavenly kami (+50%) (--> +483% at max), the fully upgraded late-game base gather rate is: 0.41 / 0.29 / 0.29.
In the ideal situation (e.g. team game with indian player), the maximum possible gather rate with cows/sacredcows/waterbuffalo on shrines is: 0.97 / 0.85 / 0.85. Hence, the ideal shrine is gathering 136%/193%/193% faster than the base gather rate without animals. Comparing with late-game villagers, this is roughly equivalent to 0.8 - 0.9 vills which totals 16-18 villagers for 20 shrines.

I think it's even more. I have seen 1.09 with sheep, and 1.21 with cows in treaty.

[princeofcarthage]

Given that shrines base gather rate is 0.07/0.05/0.05 per second (food/wood/gold) and the fact that this rate can be changed by the age IV upgrade (+300%) + toshogu wonder (+133%) + heavenly kami (+50%) (--> +483% at max), the fully upgraded late-game base gather rate is: 0.41 / 0.29 / 0.29.
In the ideal situation (e.g. team game with indian player), the maximum possible gather rate with cows/sacredcows/waterbuffalo on shrines is: 0.97 / 0.85 / 0.85. Hence, the ideal shrine is gathering 136%/193%/193% faster than the base gather rate without animals. Comparing with late-game villagers, this is roughly equivalent to 0.8 - 0.9 vills which totals 16-18 villagers for 20 shrines.

Shrines gather food at 0.14/sec and coin and wood at 0.1/sec. Age 4 upgrade for is 350% for food and shrine bonus is 50% afaik so that's basically a max of 450 for food and 400 for wood and coin

which means a max of 1.33 with food and 1.06 with wood or coin afaik

[duckzilla]

I found this here online and interpreted it as the base gather rates:

<ProtoAction>
<Name>AutoGatherCoin</Name>
<Rate type ='Gold'>0.050000</Rate>
</ProtoAction>
<ProtoAction>
<Name>AutoGatherFood</Name>
<Rate type ='Food'>0.070000</Rate>
</ProtoAction>
<ProtoAction>
<Name>AutoGatherWood</Name>
<Rate type ='Wood'>0.050000</Rate>
</ProtoAction>
<ProtoAction>
<Name>AutoGatherXP</Name>
<Rate type ='XP'>0.340000</Rate>
</ProtoAction>

*edit: this is apparently old. I just tested it and will now correct the numbers in the post above!

[princeofcarthage]

Never trust the internet.

[jgals]

So the cows need to be fattened? Try getting that Done as japan

[princeofcarthage]

They don't. Also you don't really need to test everytime, you can just open up the game files and look into it

[pecelot]

So the cows need to be fattened? Try getting that Done as japan

you can make your teammate send them to you