Factorio Forums

www.factorio.com
https://forums.factorio.com/

4-way intersections: Throughput and deadlocks [image heavy]

https://forums.factorio.com/viewtopic.php?t=46855

Page 21 of 36

Re: 4-way intersections: Throughput and deadlocks [image heavy]

Posted: Mon Mar 26, 2018 3:56 pm
by steinio
dood wrote:
mrvn wrote:1) The game is buggy and will sometimes re-path trains inside the roundabout into a 270° turn. This crosses their own way so the chain signals inside the roundabout prevent the train from moving.

NO, that's fixed a long time ago.

Stop telling this lie to everyone.

Re: 4-way intersections: Throughput and deadlocks [image heavy]

Posted: Tue Mar 27, 2018 12:18 am
by Youri
I've been experimenting with my train system recently. Currently I'm looking at the potential pros and cons of *mixing LHD and RHD* together in a single train system. The original idea came from the fact that most intersections have very short/easy turns to one side, while turns to the other side block basically the whole intersection. By combining LHD and RHD I'm hoping to balance it out a bit more which should help overall throughput. Also, I haven't really seen anyone purposefully mixing drive systems before!

The basics for the train system have LHD on one axis and RHD on the other (I don't really use diagonal straights or 45deg turns, which would've required extra crossings). I started out with the other place where trains switch from drive system: intersections. Below is one of the experiments, a 3+3 lane 4-way intersection.
Mixed LHD+RHD
I've compared the design to a version for pure LHD, both have the same number of sections (120). Pure LHD obviously has 'cheaper' left turns, but more expensive right turns. The mixed version has equal costs for left and right turns, both a bit cheaper than LHD's right turns, which was what I was hoping for. Another notable difference is that the mixed version doesn't immediately split all entry lanes, so trains going left or straight share part of their path through the intersection.

I'm curious what kind of throughput I can expect with this type of design, and how much better/worse it is compared to a 'normal' intersection. Also, what do you guys think?
Reference design

Re: 4-way intersections: Throughput and deadlocks [image heavy]

Posted: Wed Mar 28, 2018 1:31 am
by mrvn
steinio wrote:
dood wrote:
mrvn wrote:1) The game is buggy and will sometimes re-path trains inside the roundabout into a 270° turn. This crosses their own way so the chain signals inside the roundabout prevent the train from moving.

NO, that's fixed a long time ago.

Stop telling this lie to everyone.
What fix would that be? If you mean the fix in 0.15.27 then look at viewtopic.php?t=34844#p294095.

The fix only solves the problem for building rails but not for enabling/disabling stations. Afaik there is no fix for trains rerouting inside chain signal blocks when stations are enabled or disabled yet. Please do proof me wrong so I can start using roundabouts again.

Edit: I've attached a savegame from the latest 0.16.x where I purposefully trigger the repath inside a roundabout. Just turn the train on automatic and watch it crash into itself. I can't reproduce the deadlock anymore though. They must have changed something post 0.15.27 that isn't obvious in the changelogs or I'm triggering it wrong. I still get the deadlocks in a 0.15.x multiplayer game.

Re: 4-way intersections: Throughput and deadlocks [image heavy]

Posted: Wed Mar 28, 2018 10:48 am
by dood
Oh so it only happens if I mess around with stations at the exact moment when a train is in a loop then?
Well that would explain why it never happened to me. I never control stations via circuits and if I remove one, the train is most likely not driving but waiting at the station because the ore field depleted.

Re: 4-way intersections: Throughput and deadlocks [image heavy]

Posted: Wed Mar 28, 2018 5:52 pm
by mrvn
dood wrote:Oh so it only happens if I mess around with stations at the exact moment when a train is in a loop then?
Well that would explain why it never happened to me. I never control stations via circuits and if I remove one, the train is most likely not driving but waiting at the station because the ore field depleted.
Yes. The problem I had was with enable/disable stations. Adding rails or removing stations is a relatively rare event compared to circuit controlled stations switching on and off (if you use them). And you would have to remove the station at just the wrong time as you implied.

In the large multiplayer game I'm in the trains deadlock every once in while. Say once or twice in a game night. It happens very rarely. But when it does it sucks.

Re: 4-way intersections: Throughput and deadlocks [image heavy]

Posted: Mon Apr 02, 2018 6:57 pm
by hansjoachim
Youri wrote:I've been experimenting with my train system recently. Currently I'm looking at the potential pros and cons of *mixing LHD and RHD* together in a single train system. The original idea came from the fact that most intersections have very short/easy turns to one side, while turns to the other side block basically the whole intersection. By combining LHD and RHD I'm hoping to balance it out a bit more which should help overall throughput. Also, I haven't really seen anyone purposefully mixing drive systems before!

I'm curious what kind of throughput I can expect with this type of design, and how much better/worse it is compared to a 'normal' intersection. Also, what do you guys think?

Interesting idea, but that intersection needs buffers to have high throughput. As it switches from RHD to LHD it can't be tested on the test bench, but I would imagine that the throughput would be lower than a good 2 lane intersection.

Re: 4-way intersections: Throughput and deadlocks [image heavy]

Posted: Mon Apr 02, 2018 10:20 pm
by mrvn
hansjoachim wrote:
Youri wrote:I've been experimenting with my train system recently. Currently I'm looking at the potential pros and cons of *mixing LHD and RHD* together in a single train system. The original idea came from the fact that most intersections have very short/easy turns to one side, while turns to the other side block basically the whole intersection. By combining LHD and RHD I'm hoping to balance it out a bit more which should help overall throughput. Also, I haven't really seen anyone purposefully mixing drive systems before!

I'm curious what kind of throughput I can expect with this type of design, and how much better/worse it is compared to a 'normal' intersection. Also, what do you guys think?

Interesting idea, but that intersection needs buffers to have high throughput. As it switches from RHD to LHD it can't be tested on the test bench, but I would imagine that the throughput would be lower than a good 2 lane intersection.
What you can do is build cells or loops with LHD and RHD that overlap on the outsides. So a RHD cell has a lane going up on the right side. A LDH cell has a lane going up on the left side. Merge these two so they use the same lane going up. Similar on the other 4 sides of each cell. Now you get a checker board pattern of LHD and RHD cells. Note that you don't need any 2 lane ways, everything is one lane going one way.

The problem with that design is that to go far to the left you have to constantly go up and down so it takes 3 times the distance. Also each cell is a loop. A loop can only hold N trains and if it ever saturates you get grid locks.

Re: 4-way intersections: Throughput and deadlocks [image heavy]

Posted: Wed Apr 04, 2018 8:55 pm
by dragontamer5788
Youri wrote:I've been experimenting with my train system recently. Currently I'm looking at the potential pros and cons of *mixing LHD and RHD* together in a single train system. The original idea came from the fact that most intersections have very short/easy turns to one side, while turns to the other side block basically the whole intersection. By combining LHD and RHD I'm hoping to balance it out a bit more which should help overall throughput. Also, I haven't really seen anyone purposefully mixing drive systems before!

The basics for the train system have LHD on one axis and RHD on the other (I don't really use diagonal straights or 45deg turns, which would've required extra crossings). I started out with the other place where trains switch from drive system: intersections. Below is one of the experiments, a 3+3 lane 4-way intersection.
I've been experimenting with a few designs that mix LHD with RHD actually. Although I also restarted my game to reinvestigate the early / midgame... but here's a basic drawing of the last concepts I was working on.

Image

In effect, "sub-networks" run LHD, while the "main-network" runs RHD. A Subnetwork would handle traffic such as Iron Ore -> Iron Plates, or Iron Plates -> Gears. The purpose of a sub-network is to facilitate bulk deliveries without adding to the congestion of the main-line. This overall floor-plan allows most intersections to only have ONE crossing, while some intersections have ZERO crossings, which would theoretically lead to a huge throughput. The key is switching between LHD and RHD between "sub-networks" and "main-networks".

I never got to actually test this design out for real though. Which is why its still just a drawing.

Its not exactly UPS friendly design though. Realistically, this design only makes sense if you're willing to make crossdocks / centralized dropoffs of ore. A more UPS friendly design would be to convert ore->plates->steel in the field before even loading them onto a train. A crossdock would make it theoretically easier to expand mines: all mining trains go to the crossdock, all sub-base trains go to the crossdock. The main issue is "train traffic multiplication": every delivery of Iron Ore is loaded up into another train, meaning you just doubled train traffic for sake of convenience.

The switch between LHD and RHD means that the "base entrances" that are sandwiched between the two networks (ie: the Iron Ore Crossdock) has very, very few blocking intersections. In fact, if you build all of your sub-bases to the west of the Iron Ore crossdock, the Iron Ore Crossdock entrance would have ZERO intersections with the sub-network, facilitating the highest bandwidth.

Re: 4-way intersections: Throughput and deadlocks [image heavy]

Posted: Wed Apr 11, 2018 7:12 pm
by DisRuptive1
Every 2-way Celtic Cross intersection always had the tracks coming from the North, South, East, and West directions but none existed when the tracks came from the diagonal directions. I made the following and would love your critique:

Image

Code: Select all

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

Re: 4-way intersections: Throughput and deadlocks [image heavy]

Posted: Thu Apr 12, 2018 9:08 am
by mrvn
DisRuptive1 wrote:Every 2-way Celtic Cross intersection always had the tracks coming from the North, South, East, and West directions but none existed when the tracks came from the diagonal directions. I made the following and would love your critique:

Image
Not enough signals.

Re: 4-way intersections: Throughput and deadlocks [image heavy]

Posted: Thu Apr 12, 2018 12:22 pm
by Optera
Mixing RHD and LHD sounded like a diverging diamond.

Image
In TTD those kind of intersections can work fairly well, but i'm not sure how well you could project that into only one layer.

Re: 4-way intersections: Throughput and deadlocks [image heavy]

Posted: Thu Apr 12, 2018 6:13 pm
by hansjoachim
DisRuptive1 wrote:Every 2-way Celtic Cross intersection always had the tracks coming from the North, South, East, and West directions but none existed when the tracks came from the diagonal directions. I made the following and would love your critique:
I added the needed signals and threw it on the testbench. About 40 in set 1. Which is rather low for a 4 way 2 lane intersection.

Re: 4-way intersections: Throughput and deadlocks [image heavy]

Posted: Fri Apr 13, 2018 7:42 am
by Zavian
Throughput of 40 is better than many of the junctions listed in the OP. It is similar to Wide and Wide B and Compact Celtic Knot. Junctions with higher throughput tend to be huge because they have lots of buffers, and hence need a lot of space especially if you run long trains.

Re: 4-way intersections: Throughput and deadlocks [image heavy]

Posted: Fri Apr 13, 2018 8:28 am
by vanatteveldt
I use the 'compact' design which is pretty simple & small and also achieves 40 throughput. But I agree that it seems 40 is the par for 2 lane intersections without large internal buffers. Or did I miss a >>40 design which is not more than twice as large as these compact 40 tpm designs?

Re: 4-way intersections: Throughput and deadlocks [image heavy]

Posted: Sat Apr 14, 2018 2:39 pm
by hansjoachim
I made an RHD 4 way 2 lanes 6 car intersections. Can you test it? Its advantage over the other intersections is the higher throughput with left only traffic (right only for LHD)
https://pastebin.com/uqCM5CZK

Re: 4-way intersections: Throughput and deadlocks [image heavy]

Posted: Sat Apr 14, 2018 10:14 pm
by hansjoachim
I made a larger version that finally beats 100 trains per min without combinators in set 1 and 2.
https://pastebin.com/KpSRsecC

Re: 4-way intersections: Throughput and deadlocks [image heavy]

Posted: Sat Apr 21, 2018 4:33 am
by Divaya
Optera wrote:Mixing RHD and LHD sounded like a diverging diamond.

(picture)
In TTD those kind of intersections can work fairly well, but i'm not sure how well you could project that into only one layer.
Did it for funsies
Image

Not at all optimized, didn't have time today to test it.
  • Quirk of design allows U-turns for Nouth/South trains (can be considered good/bad depending on your preference/design)
  • 2 opposite trains can turn left at the same time
  • Reasonable footprint, doesn't need very many signals (36 pre, 12 block).
BP:

Code: Select all

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
Tried to do it in 4 lanes, but most likely is missing signals. Heavy doubts this one will perform well.
Image

Code: Select all

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

Re: 4-way intersections: Throughput and deadlocks [image heavy]

Posted: Sun Apr 22, 2018 8:09 am
by hansjoachim
Divaya
How long are your trains?

Re: 4-way intersections: Throughput and deadlocks [image heavy]

Posted: Sun Apr 22, 2018 11:22 am
by Divaya
I personally use 1-4 and 2-8 trains in my current save.

Both those designs don't have any buffers or space allocated for trains, it was more of an aesthetic whip-up for fun. It wouldn't be too hard to add buffers though, it'd make for a pretty decent intersection with them though.

Edit: Here's a quick buffered diverging diamond for 2-4 trains.
Image

First 4 way I've made with buffers for straight-thru and left turns. Generally not a fan since buffers make intersections much larger. Can still be improved by moving the right turn lanes further out (longer) but I'm lazy.

Code: Select all

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

Re: 4-way intersections: Throughput and deadlocks [image heavy]

Posted: Sun Apr 22, 2018 5:49 pm
by hansjoachim
Divaya
There is just so many problems with the "buffered diverging diamond intersection" that I don't know where to start.
I threw it on the test bench to confirm what I thought. And the test confirmed my concerns. Both with the throughput and how the trains move. set 1 = 68 per min. For reference: 264 is max for 4 lanes and 132 is max for 2 lanes. With this throughput you are better off with a 2 lane system.
I can elaborate its problems if you want to.
All times are UTC
Page 21 of 36

Powered by phpBB® Forum Software © phpBB Limited