Arbor Trees or Sour gas condensors for water? A thought experiment.

[Zarquan]

NOTE:  Water is the goal.  Other byproducts, including power and dirt, are not metrics for this thought experiment.

I was thinking earlier of whether I should build my volcano in to a sour gas boiler/condenser or in to a petroleum burner.  I decided on a petroleum boiler because I was thinking that arbor trees net a lot of water.  However, I wasn't sure if that was the right call, so I've decided to do a though experiment:  Which approach gives you the most water per tile: sour gas condensers, domestic arbor trees, or wild arbor trees?

I will be ignoring the size of the boiler, as those can be made fairly small and can produce a ton of natural gas.  I will also be ignoring any apparatus used to cool these systems, as there are many ways to cool things.  I will also be ignoring any space taken by slicksters, as they can be in 1 tile regardless of the size of the power plant.  I will also be ignoring the sour gas boiler, as these vary in size, but can have an extremely high throughput and can be made at about the same size as a petroleum burner.

Assume we start with 6,153.9 g/s crude oil.  If we convert this to natural gas, we end up with 4,123.1 g/s natural gas.  If we feed the CO2 to slicksters, we get a geometric series that results in 4500 g/s natural gas, which uses 9 pumps and 50 natural gas generators.  Each natural gas generator takes 16 tiles (if you include the floor) and each pump takes 4 tiles.  So, this powerplant uses 836 tiles.

For comparison, we use the 6153.9 g/s crude oil in petroleum generators.  If we use the slickster loop, we end up with a geometric series that nets us 7033.0 g/s crude oil.  This takes approximately 3.517 petroleum generators and 0.7033 pumps.  A petroleum generator take 15 tiles (including flooring) and a liquid pump takes 4 tiles, so it costs around 58 tiles.

Spoiler

In actuality, it would take around 64 tiles, but this more continuous way of thinking about these buildings helps us understand how it scales without having to find a number of natural gas generators that resulted in an exact multiple of 10,000 g/s crude oil.  For example, if we started with ‭12,307.8‬ g/s crude oil, then we would be coming from ~14,000 g/s crude oil after the slickster loop, which would be 2 pumps and 7 generators, it would actually take 113 tiles which is closer to the 58*2 than the 64*2. 

So, the tradeoff is the difference in space, which is effectively ‭778‬ tiles, for the water production difference.

The 4500 g/s natural gas results 3,375 g/s polluted water.  The petroleum generators consume 7033.0 g/s petroleum to generate 2,637 g/s polluted water. 

My sour gas boiler and generators used 778 additional tiles and gave us 738 g/s additional water, or 0.949 g/s/tile.

Now, we consider domestic arbor trees. 

Spoiler

Since this setup would take the place of natural gas generators, we will assume that any CO2 would be converted by slicksters and burned in a petroleum generator.  This is because if arbor trees are better, then the natural gas generators could be replaced by arbor trees for a better result.

1 arbor tree produces 555.55 g/s lumber if harvested promptly (so in actuality, slightly less).  If you have a setup with 36 arbor trees and 20 ethanol distilleries, we can feed 5 petroleum generator ethanol.  This results in 3750 g/s polluted water and 5,833 g/s CO2 at a cost of 4200 g/s polluted water.    That 5,833 g/s CO2, which, after the slickster loop, results in 6,190.5 g/s CO2, or 3095.2 g/s petroleum.  This takes 1.55 additional petroleum generators to burn, which produces an additional 1,160.7 g/s polluted water, resulting in a net gain of 710.7 g/s water.

A group of 2 arbor trees can be planted in 20 tiles ideally (counting the floor).  An ethanol distiller takes 16 tiles (including the floor) and the petroleum generators take 15 tiles.  For this setup, I will also assume there is one autosweeper for every 4 distillers, each with one loader and one receptacle, taking 7 additional tiles per 4 distillers.  Petroleum generators take 15 tiles.  We also require 0.309 liquid pumps at 4 tiles.  This setup could be adjacent to the slickster ranch, so no gas pumps are necessarily needed.

My domestic arbor tree setup takes 814 tiles and produces 710.7 g/s water, or 0.873 g/s/tile.

Now, we consider wild arbor trees.  144 wild arbor trees can feeds 20 ethanol distilleries, which can feed 5 petroleum generators.  Actually, all the math for this is the same except the space taken by the trees increases and there is no water cost for the trees. 

My wild arbor tree setup takes 1894.5 tiles and produces 4,910.7‬ g/s water, or 2.592 g/s/tile.

This was a mild surprise to me, but it makes sense.  You do get more water back per tree with wild trees, and the infrastructure per wild tree is significantly less.  Given an efficient setup, this shouldn't even take significantly more dupe time, as the time spent tending trees should be about the same as it was before.

Note that there are obviously other variables in the game. 

For one, power.  Sour gas boilers net an absurd amount of power.  You can expect up to 5 times more power from natural gas than petroleum.  Domestic arbor trees produce more power than wild. 

Another is polluted dirt, which you get ridiculous amounts of if you go the domestic arbor tree route.  For those who want to know, the domestic tree approach creates 7.45 g/s/tile polluted dirt (or 4.47 kg/cycle/tile) and the wild trees produce 3.52 g/s/tile polluted dirt (or 2.11 kg/cycle/tile). 

TLDR:  If you want the most water, wild arbor trees are better than sour gas condensers on a per tile basis.  Sour gas condensers are mildly better than domestic arbor trees for water.

[Xuhybrid]

This would be a more practical comparison if you were using water to generate oil instead of starting with some already.

You'd have to compare wild arbor with wild oil or domestic arbor with extracted oil.

[Zarquan]

8 hours ago, Xuhybrid said:

This would be a more practical comparison if you were using water to generate oil instead of starting with some already.

You'd have to compare wild arbor with wild oil or domestic arbor with extracted oil.

The thing is that the water you start with is not important, as it takes the same amount of water to create the oil as the natural gas.  When you have crude oil, you have have two options.  Build petroleum generators in a smaller space for less water OR buld a bunch of natural gas generators taking more space and more water.  The question is whether that space is worth it compared to arbor trees. 

What is wild oil?  Are you talking about existing oil pockets?

[avc15]

I took it to mean this

It costs water to get crude oil, if you want sustainability at industrial volumes. Not so for wild arbor trees or cultivating with pips.

So, for the oil cases it would be useful to look at net water production after the cost of whatever oil wells you'd need to sustain production.

Also, labor is another metric these approaches differ on, useful to look at.

Personally I have so much water since the introduction of saltwater geysers that I don't need to even tame all my steam vents. Must be feeding 40 dupes, bristle blossoms  or on a map without tide pools

 

[Zarquan]

2 hours ago, avc15 said:

I took it to mean this

It costs water to get crude oil, if you want sustainability at industrial volumes. Not so for wild arbor trees or cultivating with pips.

So, for the oil cases it would be useful to look at net water production after the cost of whatever oil wells you'd need to sustain production.

Also, labor is another metric these approaches differ on, useful to look at.

Personally I have so much water since the introduction of saltwater geysers that I don't need to even tame all my steam vents. Must be feeding 40 dupes, bristle blossoms  or on a map without tide pools

I do take in to account water consumption.  The main point is that the water consumption of oil wells doesn't matter, as the natural gas and the crude oil costed the same amount of water, so they cancel out in the math.  This is because I am considering the difference between burning the petroleum in a petroleum generator and natural gas. 

In the first approach, I looked at the difference in space and water production between petroleum and natural gas generators to determine how much more water you get per tile.  Since the petroleum took less space, but the natural gas produced more water using the same resources, the natural gas approach effectively used that larger space to produce more water.

Then, I compared that to domestic arbor trees. 

Then, I compared that to wild arbor trees.

With dupe labor, I agree that there is dupe labor involved in harvesting trees.  But I consider dupe labor to be not as bad as others do.  You can hire a dupe who specializes in that task and lock them in with the trees.  I mean, you have dupes, what are they there for?  To hang around idle, or to do work?  And if a dupe's job produces more resources than they consume, why not have them?