Fractal mine for super-high output

[db48x]

full size

This is a type of fractal known as a space-filling tree, and it ensures that every mine in the grid is exactly the same distance away from the destination. With normal belts at the edges (which are actually crinkled around into the inside as well), and higher-speed belts in the center, this has the maximum possible output capacity. I haven't been able to use the ore fast enough that it doesn't back up, but then I haven't really tried very hard. I'm pretty sure that with this many mines it's capable of optimally filling even the triple-speed belts. Now I just need to upgrade my iron mine...

[rk84]

Cool one. I think 8 miners per lane is enough to fill basic belt. Mayby you can report back to how many is needed to fill express belt. I haven't played much with them yet

ps. Row 5. Column 10 and 11 miners needs rotating.

[Mysteria9]

That's a lot of copper!
Interesting design, but how do you plan to use all that?

[ficolas]

my head hurts D:<

[db48x]

Cool one. I think 8 miners per lane is enough to fill basic belt. Mayby you can report back to how many is needed to fill express belt. I haven't played much with them yet

ps. Row 5. Column 10 and 11 miners needs rotating.

Ooh, I knew there was some reason the output was on the low side

Hmm, so if 8 mines will fill one side of a belt, then a cluster of 16 should fill the belt perfectly on both sides. I can then switch to red, which has twice the capacity, where two clusters of 16 join up. That leg turns a corner and joins up with it's opposite number coming the other way, where it should transition to blue, turn the corner and head out to the outside world. That means that this mine may not actually be large enough to completely fill a blue belt, since the pairs of 16 all overlap the edge of the deposit. Now that I have a hypothesis, I just need to make a bunch of blue science packs and try it out. For Science!

[rk84]

I made a couple tests with 64 miners (field size 8x8) in my test scenario.
if express belt has turns in both ways. Then ~38 miners was enough to fill it(in output flow items have caps between them). one entry point to express belt was enough.
If the express belt was straight then 52 miners were active, but I was not able to fill the belt full. Even with 5 entry points there was empty cap size of 2 item after every 10-15 items in both lanes. mayby I need better layout or some super fast inserters. oh and 16 fast inserters were able to clean the belt (8 both sides)

[db48x]

Alas, the T junctions can't actually keep up:




The outgoing belt ends up being only half-full, so I'd need two of these in a row. Of course, they can each have fewer mines, but it just won't be the same.

Also, you can really see how the game prefers one direction over the other at the junctions, and one side of the belt over the other as well.

[LoSboccacc]

a miner should eat 5x5 so you can have two space between them, allowing double junction

[ficolas]

use high speed transport belts.

[MF-]

use high speed transport belts.

I am afraid that the final output line already uses the express belts.

[db48x]

Here's what I ended up with:



The output is now full, but I'm not using it fast enough for it to run at full speed, so it's not at full capacity.

[Dakkanor]

if you have two spaces between each electric drill then you would get a more efficient output, not as fast certainly, but every square of ore would be mined from only one drill and that setup would last longer before you need to relocate, for greater output you could probably bunch up those drills so you could fit more drills per ore deposit.

[db48x]

if you have two spaces between each electric drill then you would get a more efficient output, not as fast certainly, but every square of ore would be mined from only one drill and that setup would last longer before you need to relocate, for greater output you could probably bunch up those drills so you could fit more drills per ore deposit.

I'm well aware of that. However, if you bunch them up you don't have any place to put the power lines, and you end up with drills that aren't all the same distance from the core. When that happens, the rate must necessarily be lower, because those closer to the core will have fewer opportunities to actually be in use; they'll be blocked by ore coming from drills further up the line. On the other hand, it's probably moot because of the way the game prefers one lane over the other at the T junctions.

Also, the amount of _time_ that a particular deposit lasts is irrelevant; there is a fixed amount of ore in one, and mining it more slowly doesn't extend it. Although I didn't measure the precise rate, I've smelted 50k plates just in an hour or two of testing. My next test will be to ramp that up and try to use it at full speed. I hope the first-level feeders will unblock once I get enough forges.

[Ilnor]

You need more than 1 belt as an output
And it needs to be a faster belt

Since you chose to only use 1 belt you're limited by the capicity and speed of that belt

With an iron deposit that big i'm betting you can do 3-5 outputs and keep the belts packed as full as the one you got coming out

[bosebuck375]

You don't need such a sophisticated design. A bunch of miners around a winding conveyor belt should do just fine. No T-Intersections; all you have to deal with is the corners.

[darkminaz]

aside from the fact that i kinda wonder where you got so mutch ore on your map
seems quite unproductive to only have 1 line + having that one line in slow belts.

16 mines + speed modules are usually way more then enough to fill a single line up.

i'm mostly happy if i can place down more then 6 on single ore spot.

[-root]

Check the dates gentlemen.

These designs are a long way out of date.

[darkminaz]

sorry didn't check that someone did dig out a corpse ..