I think I also just figured out why the throughput losses were so hard to detect on the input side: Because of how large this balancer is, and the fact that the belts are split into more belts. The compression waves from a bottleneck would be distributed over the large number of belts. These compres...

Whelp, after testing it 4 times, with increasing numbers of plates per input belt (final test was 1400+700=2100 plates per belt) and leaving in the backlogs from the unused output belt (to reduce time to stabilize). I found that the outputs were indeed not saturated. What was even more interesting w...

I've disproven the current running theory that it is impossible to balance a 4-belt balancer without throughput limit. See: http://i.imgur.com/6wsMrV2.jpg Basically, what I did was divide each of the 4 lanes into 4 more lanes. I then sent one of the four "quarter" lanes to each of the fou...

So I thought I'd give a quick mini-tutorial on my belt balancer testing setup and how to make one for yourself. Here's a picture of the whole thing (sans power source which is off-screen to the east): 20160508062801_1.jpg The middle area is where you build whatever balancer you want to test, be sure...

So I did some extensive testing of the standard 4-belt balancer, trying to find a pair of inputs and outputs which failed to support full throughput. I did not test all (4 choose 2)*(4 choose 2)=6*6=36 possibilities, but that's because I don't need to, thanks to symmetry of the balancer. I tested, u...

I just experimentally confirmed my edit2 to my previous post (nabbed some screenshots showing the failure to saturate belts 1 and 2 on output side (there was also so visible hitches in throughput on belts 1 and 2 on the input side, but that doesn't show in a screenshot; how do you make a gif for fa...

After looking carefully at my previous post, I noticed why there was no throughput problems: Both outputs came from the last splitter. Sure enough, change that, and the familiar throughput issues come right back: 20160507174418_1.jpg This highlights why the forbidden triangle causes the trouble that...

So I decided to test the 4-belt balancer some more to test a prediction of the aforementioned 'forbidden triangle' theorem. Which is that their should exist a trio of input belts and a trio output belts, and among those trios a pair for each, such that the throughput fails to be maximal. However, in...

The worst part is that, in any balancer with at least 3 belts, this problem is unavoidable . There will always be 3 belts of input and 3 belts of output such that the configuration of splitters that affect them form such a triangle, or a configuration containing such a triangle (or else it wouldn't...

So I ran the experiment w/ the 4-belt configuration again, and watched closely at the start to see what happened, and indeed it did backlog into the balancer, specifically along the rightmost lane heading into the final pair of splitters (as was theoretically predicted), once that backlog found the ...

Interesting 1,2,3 -> 1,2,4 experiment of that 4 lane splitter. I feel like that splitter (again, only with the last row of splitters) should be theoretically non throughput limited, but maybe not perfectly non throughput limited in practice because of some quirk in moving discrete items? It seems l...

~snip Is anybody up for a challenge of making a non-throughput-limited balancer for all cases? I was unable to make a reasonable sized one. Or a more compact version of mine (non-throughput-limited in chunks of 4)? ~snip~ The more I think about this problem, the more convinced I am that it is not p...

Decided to ask the related theoretical question on math stack exchange: http://math.stackexchange.com/questions ... m-factorio
For those that are interested to see what comes out of it.

I just experimentally confirmed my edit2 to my previous post (nabbed some screenshots showing the failure to saturate belts 1 and 2 on output side (there was also so visible hitches in throughput on belts 1 and 2 on the input side, but that doesn't show in a screenshot; how do you make a gif for fac...

BTW, the "standard" 4 belt balancer (the one in the middle of this image ) is non-throughput-limited, but only if you include the last set of splitters. Sometimes that last row is omitted because the outputs are already even at that point, but that is not perfectly non-throughput-limited ...

I think there is a relatively systematic theoretical way of knowing if the balancer can suffer from this problem or not. Consider the 8-belt balancer in the first gif. Notice that belts 1-4 on the input are the only ones with throughput happening. A plate on belt 1 can only end up on belts 1,2,3,4,5...

I have just started playing, and I am struggling with the early game, particularly the transition to electricity. After several restarts and some tutorials I think I understand the very first steps I need to take, and I can build up until I have several burner miners and furnaces working for me. In...

Just wondering what people's preferred tactics for clearing Biter Bases at various stages of the game (read tech level vrs. biter level/size of base, etc) are. I've got a rather large biter base near by oil production output that I want to take out, but I definitely can't do it on foot, and even tur...

~snip~ (PS why on earth didn't they just set the crafting speed of assembly3, chemplant, and refinery at 1, and just adjust the ass1 and ass2 times and the recipe durations...) For Assem3, it's because the player character has crafting speed of 1, which is 4/5 what an Assem3 has. With the Chem Plan...

What is the most efficient e.g. fastest way to load / unload three cargo wagons on a train horizontally / vertically? - given one full compressed belt for each wagon is the source/destination? - and how to ideally place the power poles in between so that the inserters are working properly? The fast...