My current approach to a 1-car-every-2s factory in 1.07

Post your factory layouts and screenshots here for the world to gawp at!
User avatar
MeduSalem
Supreme Robot
Supreme Robot
Posts: 199
Joined: Sat Aug 22, 2015 10:17 am
Location: Vienna, Austria

My current approach to a 1-car-every-2s factory in 1.07

Postby MeduSalem » Mon Feb 13, 2017 8:25 pm

So here is my current map from 1.07.


First on the throughput stuff... I actually adjust my entire production line around the time it takes to Fit the Front Axles (First Step in the Production line)... which takes 4.08s when fully upgraded with Extra Robots.

I do that by making everything else run slightly faster than 4.08s, either by using more slots where necessary or just by upgrading slots with Robots until they are below the threshold.

That way I never have any nasty backlogs in my factory and everything produces as fast as it can.


I'm aware of the fact that it's not entirely possible to get certain slots to work around ~4s, some of them are probably even working at 2.5 or 3s and therefore are a lot idle, but well, I can live with it if it only happens at a few stages.

At least I have somewhere about 66% Running time... constantly. No sudden drops, nothing:

Efficiency 1.png
Efficiency 1.png (11.56 KiB) Viewed 2870 times


That's the best I can get currently using 2 production lines each of which is able to produce a car every 4 seconds, so effectively exporting a car every 2 seconds. Also it is able to produce 512 design variants. Each of the 2 production lines has a unique set of 8 options that are or aren't installed... effectively creating 2*2^8 = 512 variants. Yeah an awful lot of clicking at first export but once you are through with that it's epic. I can even sell them at 75-100% premium because I seem not to overflow the market too much due to having all these variants.

Factory 1.jpg
Factory 1.jpg (395.56 KiB) Viewed 2870 times


Yeah, it's a custom made map I fiddled together in the text editor because the map editor ingame is too buggy. It's only to stress test the engine and find bugs etc... so no serious playthrough. I know the amount of importers is quite unfair and unrealistic compared to the stock maps the game comes with. xD But apart from that I started with the same budget as a regular medium sized map and no research unlocked.

I might be able to improve the uptime with an even bigger factory which allows me to adjust the amount of slots/robots better to match the desired throughput... but I guess that will have to wait until 1.08 because currently most of my maps get busted quite fast due to the notorious Resource Pending Bug.
User avatar
ProductionPie
Senior Line Supervisor
Senior Line Supervisor
Posts: 22
Joined: Fri Feb 10, 2017 6:32 pm

Re: My current approach to a 1-car-every-2s factory in 1.07

Postby ProductionPie » Mon Feb 13, 2017 9:38 pm

Nicely done. At first I was very confused as to how you even managed to pull that off since importers can only handle so much but then you mentioned it's a modded map. Still well done. I've been playing with the multi-model setup and it's amazing how high you can get the count using various paths and setups.

Also nice management of efficiency. Best I have been able to get on a regular basis is 50% runtime with the rest divided up between waiting on line and waiting on car.
User avatar
MeduSalem
Supreme Robot
Supreme Robot
Posts: 199
Joined: Sat Aug 22, 2015 10:17 am
Location: Vienna, Austria

Re: My current approach to a 1-car-every-2s factory in 1.07

Postby MeduSalem » Mon Feb 13, 2017 10:12 pm

Yeah... the amount of importers is ridiculous in my map. I placed one in the map every 4 tiles or so... so quite a lot. And I also keep all the importers strictly seperated from one another. No resource conveyor connects to 2 importer slots... because that would result in an absolute nightmare due to queue/load balancing issues between the conflicting resource importers. Otherwise one wouldn't be able to keep up with importing all the resources to manufacture the components.

If the resource requirement would be 1 resource item for every component then it wouldn't be that problematic... but some of them require 2-3 resources to craft one component and that's overkill for a single resource importer...


And yeah... the multi-model thing could hypothetically be done for all upgrades... with 34 upgrades being available currently that means 2^34 = 17 179 869 184 possible combinations. NOT ever going to do that because I would never get to do anything else besides clicking the "New design" windows away... also the save game would bloat through the roof, probably using multiple GB or something. It already uses like 3-4 MB just for 512 combinations.


On the efficiency... it could be better, even in my factory. There are some stages, especially in the Fit Body stages where there are several slots idle a lot of the time, where I could try to do it with Robot upgrades... or where the timing is so crappy that I need 1 additional completely under-utilized slot because otherwise it would create backlog over a long period of time just because of 0.1-0.2s difference in timings accumulating over time. And the backlog then creates other problems elsewhere.

The upgrades for various slots adding up to the timing is a problem too in some cases... where a slot without upgrades performs better, or where a slot with upgrades is better utilized... it really depends on the exact setup.

But that's part of what I love about such logistic games. I love fiddling around and messing with micromanagement in them... even though it is not required to do that to be successful. It would be different if it would be forced on me to do that... then I would probably hate it. But since it's my choice on how much time I want to invest in that I like it. :D
ch_lemr73
Line Supervisor
Line Supervisor
Posts: 18
Joined: Tue Jun 09, 2015 10:02 pm

Re: My current approach to a 1-car-every-2s factory in 1.07

Postby ch_lemr73 » Wed Feb 15, 2017 4:06 am

Yeah, it's a custom made map I fiddled together in the text editor because the map editor ingame is too buggy.


Thank you on this one ! I tried it out and figured quite easily how to alter the map settings this way, which is easier and way safer. I think it's a great way to proceed on testing purposes.
Floppy09
Junior Line Worker
Junior Line Worker
Posts: 4
Joined: Thu Feb 16, 2017 12:05 am

Re: My current approach to a 1-car-every-2s factory in 1.07

Postby Floppy09 » Thu Feb 16, 2017 12:16 am

Wow! That is amazing!
User avatar
GenBOOM
Supreme Robot
Supreme Robot
Posts: 100
Joined: Sun Feb 19, 2017 6:19 pm

Re: My current approach to a 1-car-every-2s factory in 1.07

Postby GenBOOM » Sun Feb 19, 2017 11:33 pm

thats pretty good!

I think it would be interesting to see whats the best running time % you can get. this would likely find the ideal factory size for efficiency rather than production output. I think it would be a good challenge.
User avatar
MeduSalem
Supreme Robot
Supreme Robot
Posts: 199
Joined: Sat Aug 22, 2015 10:17 am
Location: Vienna, Austria

Re: My current approach to a 1-car-every-2s factory in 1.07

Postby MeduSalem » Sun Feb 19, 2017 11:45 pm

GenBOOM wrote:thats pretty good!

I think it would be interesting to see whats the best running time % you can get. this would likely find the ideal factory size for efficiency rather than production output. I think it would be a good challenge.


Thanks... and I thought about the best possible running time % too already...

... from extrapolating the map from 1.07 to 1.10 and the changes/additions I made to it I think that the cap is about 80%-90% efficiency. (But I'm not that far up yet!)

You can drive the efficiency up by 2 ways:

  • Make more components in-house than importing. Those slots push the running time up a lot due to how they are working non-stop.
  • Make the pipeline wider to match the throughput more closely so that less and less slots are idling around due to overcapacity.

I say 80-90% efficiency is the cap because the time it takes for a car to leave a slot and for the next car to enter a slot would be the final efficiency/performance wall. On most slots 10-20% of the time is wasted because of the time it takes for the car to move in/out of slots... even with conveyor speed upgrades.
User avatar
GenBOOM
Supreme Robot
Supreme Robot
Posts: 100
Joined: Sun Feb 19, 2017 6:19 pm

Re: My current approach to a 1-car-every-2s factory in 1.07

Postby GenBOOM » Sun Feb 19, 2017 11:51 pm

MeduSalem wrote:I say 80-90% efficiency is the cap because the time it takes for a car to leave a slot and for the next car to enter a slot would be the final efficiency/performance wall. On most slots 10-20% of the time is wasted because of the time it takes for the car to move in/out of slots... even with conveyor speed upgrades.


thats the idea, at some point the size of the factory could actually decrease efficiency to the point that it becomes less cost effective and kinda of pointless to upgrade any larger or extend a conveyor system past a set distance. this is going to be more important as the game gets balanced more. * for early game decisions anyway.

Return to “Layout Gallery”

Who is online

Users browsing this forum: No registered users and 2 guests