Jump to content

Arbor tree automatic harvest farm 2


Recommended Posts

Hello.
I posted this topic in the past.

I've come up with a small idea to improve the space efficiency of this farm and I'll post it here.

 

Traditional designs required one free space between the arbor trees. The reason is that if the branches are adjacent, it is necessary to make a waterfall of two widths, which is very difficult to control.
However, I realized that if the harvest timing is divided into odd-numbered columns and even-numbered columns, it is not necessary to create two-width waterfalls even if the branches are adjacent to each other.
This eliminates the need for empty space between the arbor trees and makes the farm smaller.

Spoiler

20200919232219_1.thumb.jpg.80082c49f6ed45edebc74009aaa9a3e1.jpg

 

You can use the same water timer as before to control the waterfall, but I used the Timer Sensor because it is easier.
The mechanism is as follows.

Timer Sensor: Green = 10 cycles, Red = 8.2 cycles (total 18.2 cycles timer) [*1]
Signal Counter: count = 1, Advanced Mode = ON, (as an edge circuit, the moment when the timer turns green is detected)
Buffer Gate: 100s (100 seconds waterfall to destroy branches) [*1]
*1: Depending on the game load, more margin may be required for the set value.

Prepare two sets of this circuit and harvest the even-numbered column and odd-numbered column arbor trees at different timings.

Spoiler

20200919232242_1.thumb.jpg.099a7195a7866414f391a65b9549e132.jpg20200919232324_1.thumb.jpg.f28a6d65f18cb5744f76b26f3a86ad11.jpg

 

Dreamland.sav

This is save data.
Maybe it's easier to understand by looking at this save data than reading my English text made by google translate.

Link to comment
Share on other sites

I did your first version of you farm on Badlands, and yes it's a masterpiece and it's doable on a standard game without dev mod. It's a really good setup if you need a lot of water.

I've just change the liquid clock by a standard clock with the new sensor.

Link to comment
Share on other sites

This and and your first farm are technical masterpieces. I love them.

However, I'm having trouble convincing myself they're actually worthwhile in survival. I considered using your first farm on my current ethanol powered run, but ultimately decided against it. The issue is that harvesting only increases productivity of an arbor tree by about 20%. But you pay for that increased productivity with a decrease in density. Compared to the optimal  1:1:2 spacing v1 was about half as dense. v2 here is better, but my unharvested farm is still producing more wood per tile with the minor downside of needing more arbor acorns to accomplish it.

Link to comment
Share on other sites

13 hours ago, ghkbrew said:

The issue is that harvesting only increases productivity of an arbor tree by about 20%

I think probably more like 110%.

13 hours ago, ghkbrew said:

But you pay for that increased productivity with a decrease in density.

Correct, the OP's v2 design has a ratio of 5:9 in density, but it could be improved to 3:5 with the automatic harvesting still working. It's still inferior to the most optimal one, 5:7.

Frankly I hardly see density as a big problem.

13 hours ago, ghkbrew said:

my unharvested farm is still producing more wood per tile

No, not really, since you need more than twice as many trees for the same lumber.

Link to comment
Share on other sites

22 hours ago, TheMule said:
On 9/20/2020 at 11:35 AM, ghkbrew said:

The issue is that harvesting only increases productivity of an arbor tree by about 20%

I think probably more like 110%.

I may not understand the intent of their comments correctly, but what does a + 20% or 110% (+ 10%?) Increase in productivity mean?
We recognize that the amount of wood obtained is the same for both automatic harvesting using waterfalls and manual harvesting using duplicant, and there is no difference in productivity.
Or do you describe the reduction of labor costs by automatic harvesting of waterfalls as a relative improvement in productivity? But if so, what kind of calculation did the + 10% or + 20% numbers come from?
 

22 hours ago, TheMule said:

but it could be improved to 3:5

Is it the following arrangement?

Spoiler

20200921235618_1.thumb.jpg.01ef7a3a4f8cbf76e0ad7140693630b8.jpg

Yes, this is certainly better because it has a higher density of branches. The number of branches per tree will decrease a little, but I think there is no problem because mass production of acorns is easy.
Actually, when I was making v2, I was thinking a little about its placement. However, I couldn't come up with a good way to get the waterfall down through the gap between the two trees, so I compromised on the current v2 placement.

If anyone has an idea to do this well and increase the density of the farm, I would appreciate it if you could let me know.

Link to comment
Share on other sites

4 hours ago, kbn said:

We recognize that the amount of wood obtained is the same for both automatic harvesting using waterfalls and manual harvesting using duplicant, and there is no difference in productivity.

He wrote unharvested. There's not manual harvesting.

4 hours ago, kbn said:

Yes, this is certainly better because it has a higher density of branches.

Correct. Assuming density as a metric matters, of course.

Link to comment
Share on other sites

On 9/20/2020 at 12:33 PM, TheMule said:
On 9/19/2020 at 10:35 PM, ghkbrew said:

my unharvested farm is still producing more wood per tile

No, not really, since you need more than twice as many trees for the same lumber.

I meant per total space the farm occupies (i.e. taking density into account). Still I'm not sure how you came up with "more than twice" as many trees.  All my trees have the maximum 5 branches.  So, I need at most 27% more trees.

 

9 hours ago, kbn said:

But if so, what kind of calculation did the + 10% or + 20% numbers come from?

I was talking about the decrease in wood production if you leave a tree unharvested until the branches drop on their own.  Branches (and all plants) will drop their product 5 cycles after they mature. For arbor trees this means that you get 300kg of wood per branch every 23 cycles instead of 18 cycles.  Or equivalently you get 138.89 g/s of wood with a harvested tree and 108.69g/s if the tree is left unharvested.  23 / 18 = 1.278. So about a 27% decrease in productivity if you don't harvest the tree versus instantaneous harvesting.

 

9 hours ago, kbn said:

The number of branches per tree will decrease a little

Each tree grows a maximum of 5 branches in 7 possible branch locations.  This means that you can put some trees 1 tile away from each other without decreasing the total number of branches.  The densest planting pattern is the one described as "the tree one" in this post:

https://forums.kleientertainment.com/forums/topic/110299-pip-planting-everything-you-need-to-know

 

5 hours ago, TheMule said:

Correct. Assuming density as a metric matters, of course.

I think we may just disagree on whether it matters or not.  But let my try to convince you with a few points:

1) arbor acorns are not actually a scarce resource once you invest heavily in pips.  You can quickly produce many of them by repeated breaking branches and having them regrow.  Or more passively, if you keep pips in your tree farm your acorns will increase exponentially (the rate of increase is proportionate to the current number of trees = exponential).  This doesn't really mean density is important, only that another metric production per tree less so.

2) When building large structures, finding space for them is non-trivial.  My current wild arbor tree farm is 36 x 44 tiles in size.  There only so many places where you can place a structure that large with overlapping a geyser or POI. The larger the build the harder it is to find a good spot for it.

3) Many of the costs associated with a build scale with the absolute size of the farm, not the number of trees.  Obviously things like the number of sweepers, loaders, conveyor rails, power wires.  But also indirect costs like travel time or obstruction of gas flow.

 

Link to comment
Share on other sites

6 hours ago, ghkbrew said:

will drop their product 5 cycles after they mature.

Not sure about that.

3 hours ago, Yalp said:

Arbor Trees being the exception with 20 days till drop.

Correct.

 

6 hours ago, ghkbrew said:

Still I'm not sure how you came up with "more than twice" as many trees.  All my trees have the maximum 5 branches. 

You can harvest wild trees ever 18 cycles for 1500 kg, 83.33 kg/cycle.

If you wait for lumber to fall, it's 38 cycles, 39.47 kg/cycle.

83.33 / 39.47 = 2.11  That +111%. I wrote +110%. Close enough.

Link to comment
Share on other sites

 

11 hours ago, ghkbrew said:

But let my try to convince you with a few points

No need. I have no set opinion on that. By "Assuming density as a metric matters" I meant as "in your specific case". If it matter or not depends entirely on your map and gameplay style.

I still stand by my point tho: unharvested trees are way worse even density-wise if you consider lumber production.

 

When comparing manual vs. automated harvesting, it's 3:5 vs 5:7 density wise. So if you consider as 35 tile wild space, you can fit 21 trees for automated harvesting vs. 25 for manual harvesting. I'm just pointing out the fact. It's entirely up to the player to make the call, if the extra complexity and reduced trees for the same space are worth the saving in dup labor.

There can't be an general answer here. Even by looking only at my past colonies, some have extra dup labor (they spend time on wheels, with all batteries full), some don't or dups actually contribute to power production. If I have idle dups most of the time and I even have to build facilities to keep them busy (gyms) I might choose to send them to harvest trees. I did use wild trees with manual harvesting in the past. Harvesting a tree doesn't take long and a dup can harvest many trees in one cycle and it happens only once every 18 cycles so even a huge forest is hardly a lot of labor. I've never measured really. Maybe a dup alone can tend to 100 wild trees (I'm assuming it's just the harvesting, and lumber is moved via rails).

But then if you choose not to harvest them it's ok. Just know the (real) efficiency loss you're taking, then you make the call. 

Link to comment
Share on other sites

18 hours ago, TheMule said:

He wrote unharvested. There's not manual harvesting.

OK, I finally understood all of the intent.
Apparently I saw the word "unharvested" and misunderstood it to mean "not harvested using a waterfall".
I was more confused, thinking of strange inferences from my misunderstood premise.
The google translation did its job correctly from the beginning. The reason for the misunderstanding was simply my preconceptions.
I had subconsciously excluded the option of waiting for the lumber to drop from my thoughts.
 

By the way, I checked again about the time it takes for a plant to drop a crop.
The arbor tree is +20 cycles, as some people have already said.
And for the rest of the plants, it seems to be correctly +4 cycles.
Those have been confirmed by wild plants.
I haven't checked to see if the same is true for plants grown on agricultural tiles. However, I don't think there are many people who wait for their crops to fall off while growing on agricultural tiles. Because that is a pure waste of resources.

Link to comment
Share on other sites

6 hours ago, TheMule said:
9 hours ago, Yalp said:

Arbor Trees being the exception with 20 days till drop.

Correct.

Ok, so this was my confusion.  I was working from wrong information.  The wiki states: "A wild Arbor Tree drops a maximum of 65.21 kg of lumber per cycle or 108.69 g/s (unharvested). " That works out to 1500kg every 23 cycles.  Which is apparently just wrong (I tested too). Is that a recent change?

Given the above, I'm gonna change all my wild trees to automatic harvesting (by dupes).  I think the extra labor is worth the +111% increase in production.

 

Link to comment
Share on other sites

Well, I found this today.

 

This is a great job. I was impressed. Thank you to the person who made this.
Looking at the date, it looks like a design made about 4 months ago, but I haven't noticed this until now because I don't have a habit of looking at reddit. I wanted to know sooner...

Anyway, I'm confident that this idea can be applied to a 3:5 automated harvest farm.
I already have a rough idea in my head, so I'd like to make it and post it here.

Link to comment
Share on other sites

An automatic harvesting farm with a density of 3: 5 has been completed.

Spoiler

20200924215513_1.thumb.jpg.df3db31bf07babba77fc476eebe2099e.jpg

The main mechanism is based on this design, with some modifications.

  •  Without using visco-gel, make a liquid tower from 4 kinds of liquids and harvest the branches.
  • The liquid tower is reconstructed by dropping water droplets from the upper vent every time necessary, instead of dropping it directly from the mechanized airlock and moving it to the lower stage. The reason is that when the liquid tower is dropped from the mechanized airlock, the timing of the four types of liquid dripping is too close, so the order is easily disturbed when it reaches the ground, and the liquid tower is not stable.
  • The four types of liquid that have fallen to the bottom layer are collected together by a small liquid pump there and transported to the upper layer, where the liquid is re-sorted according to the same principle as the mechanical filter. In order to make this work, four kinds of liquids are pre-filled in the valve and the pipe immediately before it by 30 kg each.
  • Harvesting with a liquid tower is carried out in odd and even columns with vertical lines. The reason is that if all branches are destroyed at the same time, position collisions may occur during branch regeneration and production may decrease.

The drawback is that sequence control by automation is a little complicated, but I think the operation itself is generally good.

 

Also, although this is a by-product, the same mechanism can be applied to farms with an optimum density of 5:7 if it is "semi-automatic".
In other words, let Duplicant harvest only the branches directly above the tree that cannot be harvested by automation, and harvest the side branches by automation.

Spoiler

20200924215650_1.thumb.jpg.bc4958d2ae48008776e6e22ae87e28a2.jpg20200924215712_1.thumb.jpg.256e8ca52c2b218ed0a2079826d6d9ba.jpg

However, there is one caveat to this.
As you can see from the image, Duplicants entering the farm must wear EXO suits.
This is because if the farm air is mixed with another gas, there is a risk that the construction of the liquid tower used for automatic harvesting will fail.

 

Please refer to this save data for details.
arbor_tree.sav

 

Link to comment
Share on other sites

22 hours ago, kbn said:

Duplicant harvest only the branches directly above the tree that cannot be harvested by automation, and harvest the side branches by automation.

I was thinking the same, glad to see that you beat me to it!


Hybrid farms are definitely the best of the two worlds, you reduce the amount of labor to the minimum while still retaining the best density tile-wise and tree-wise.

 

Now, to be 100% honest, I don't believe much in these type of builds, outside of looking at a perfectly optimized colony on your screen or being in cycle 3000 and not knowing what to do next, or building a base to be featured in the next 'Base Loving Video'. Which, BTW, are perfectly legit ways of enjoying the game. Just not my thing, or rather I'm still enjoying other parts of the game more (like how to face various challenges different maps can offer).

Point is, if you're at cycle 300 and want to do large scale wild farming, you probably can go with an vastly simpler design (just trees, with the 5:7 ratio), and have one or two dedicated farmers. The price you're paying is 100g/s of oxygen and 1000lcal/cycle per dup (well there's some infrastructure costs, which may or may not be triggered by one dup - more aptly, they're opportunity costs). Which isn't much, in the grand scheme of things.

The nice thing about this specific build, is that you can build your farm w/o any autoharvesting and add it (much) later.

 

 

Link to comment
Share on other sites

13 hours ago, TheMule said:

Point is, if you're at cycle 300 and want to do large scale wild farming, you probably can go with an vastly simpler design (just trees, with the 5:7 ratio), and have one or two dedicated farmers. The price you're paying is 100g/s of oxygen and 1000lcal/cycle per dup (well there's some infrastructure costs, which may or may not be triggered by one dup - more aptly, they're opportunity costs). Which isn't much, in the grand scheme of things.

Yes, I generally agree with that.
The benefits of automation are labor savings, but the Duplicant workforce isn't that expensive.
Construction costs are high due to the large amount of airlocks used, and it will be difficult to recover the initial investment paid with the savings of cheap labor.

Maybe this is a future build.
A future world where Duplicants have revolutionized and become independent of the players, gaining human rights and forming their own society, and the cost of labor is exalted.
In that world, this build will finally be truly useful. (Is that world coming? Yeah, I don't know.:listening_headphone)

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

Please be aware that the content of this thread may be outdated and no longer applicable.

×
  • Create New...