Final (Maxed) Designs

Circuit-free solutions of basic factory-design to achieve optimal item-throughput.
Involving: Belts (balancers, crossings), Inserters, Chests, Furnaces, Assembling Devices ...
Optimized production chains. Compact design.
Please provide blueprints!
Forum rules
Circuit-free solutions of basic factory-design to achieve optimal item-throughput
SpeedDaemon
Fast Inserter
Fast Inserter
Posts: 124
Joined: Fri May 22, 2015 3:31 pm

Re: Final (Maxed) Designs

Post by SpeedDaemon » Thu Jun 11, 2015 8:09 pm

DerivePi wrote:
DerivePi wrote:Actually, there are 3 mistakes with my circuit layout schematic:
1. There need to be more inserters from the copper plate belt into the wire assemblers
2. You can't fully load an express belt with inserters. I need to add branch belts and splitters to combine.
3. The colors are too garish. Not Feng Shui at all!
1. added 2nd fast inserter to feed copper cable assemblers
2. added 2nd belt for exporting circuits with balancer in the middle.
3. changed the background to "salmon" for this one time only
final ckt.gif
I've been using a very similar layout, except that instead of trying to bring the iron in through the middle where it gets in the way, I bring it in outboard of the copper belts, through the gaps in the wire assemblers. Cleans up the output area considerably, allowing even loading of both sides of a single belt, or double belts with one long and one fast inserter per circuit assembler (not-quite-even loading of both sides of two belts, but the sum of the respective sides should be equal, so they should merge to a single balanced belt).

User avatar
DerivePi
Filter Inserter
Filter Inserter
Posts: 504
Joined: Thu May 29, 2014 4:51 pm

Re: Final (Maxed) Designs

Post by DerivePi » Mon Jun 22, 2015 3:25 pm

Batsma2 wrote:if you'd like I can set you a challenge as well, try and build an optimized design for building all belt types efficiently. although you might just want them to unload in logistics chests for your bot, the supply lines can be quite tricky.
Belt production is limited by gear input. One full express belt of gears is almost completely used up (27.5 gears per second) by 1 belt, 1 fast belt and 1 express belt assembler 3s. To make it a full 30 gears per second we could just add another assembler 3 for the basic belt production.

In short, any setup that has more than 1 assembler 3 for: fast belts, fast UG belts, fast splitters, express belts, Express UG belts and Express splitters is probably not balanced since it takes 190 gears per second to run all of those assemblers at full speed (Over 6 full express belts of gears which is over 12 full express belts of iron plates).

The trick for the basic logistics department (transport belts and inserters) is not to fully implement the assemblers but to balance the output so that used items are replenished quickly. I notice this problem most when I fill up on Express belt products (belts splitters and UG) and Fast belt products. Replenishing the Express Belts takes a considerable time since, in my system, the fast belt assembler has priority.

Batsma2
Burner Inserter
Burner Inserter
Posts: 7
Joined: Thu Jun 11, 2015 11:14 am

Re: Final (Maxed) Designs

Post by Batsma2 » Mon Jun 22, 2015 5:36 pm

Aren't those difficulties exactly what makes it interesting?

(to be honest I always end up using bots to create belts :( )

DoctorWho?
Long Handed Inserter
Long Handed Inserter
Posts: 80
Joined: Sat Jan 10, 2015 3:42 am

Re: Final (Maxed) Designs

Post by DoctorWho? » Thu Jul 02, 2015 2:07 am

DerivePi wrote:
Batsma2 wrote:if you'd like I can set you a challenge as well, try and build an optimized design for building all belt types efficiently. although you might just want them to unload in logistics chests for your bot, the supply lines can be quite tricky.
Belt production is limited by gear input. One full express belt of gears is almost completely used up (27.5 gears per second) by 1 belt, 1 fast belt and 1 express belt assembler 3s. To make it a full 30 gears per second we could just add another assembler 3 for the basic belt production.

In short, any setup that has more than 1 assembler 3 for: fast belts, fast UG belts, fast splitters, express belts, Express UG belts and Express splitters is probably not balanced since it takes 190 gears per second to run all of those assemblers at full speed (Over 6 full express belts of gears which is over 12 full express belts of iron plates).

The trick for the basic logistics department (transport belts and inserters) is not to fully implement the assemblers but to balance the output so that used items are replenished quickly. I notice this problem most when I fill up on Express belt products (belts splitters and UG) and Fast belt products. Replenishing the Express Belts takes a considerable time since, in my system, the fast belt assembler has priority.
My solution to the priority issue: simply send the gears from the other direction. Make the gear input at the "end" of the line, that way you prioritize blue belts, fast belts and basic celts, in that order. Hope this helps ;)

MantisShrimp
Long Handed Inserter
Long Handed Inserter
Posts: 84
Joined: Wed Jun 04, 2014 1:35 pm

Re: Final (Maxed) Designs

Post by MantisShrimp » Fri Jul 17, 2015 1:03 am

Very cool design. Here's what it looks like in-game and what's needed to build it, if others are interested. No offense, but the pink/salmon picture wrecked my eyeballs. I haven't turned it on since I need to sort out my train stations, so I have no idea how well it preforms. It looks neat though. Also, I'm not sure what color some of the splitters are supposed to be so I just used the color of the belt that came before it.

Image

User avatar
ssilk
Global Moderator
Global Moderator
Posts: 10393
Joined: Tue Apr 16, 2013 10:35 pm

Re: Final (Maxed) Designs

Post by ssilk » Fri Jul 17, 2015 8:18 am

Guys, it would be really helpful, if you where using the Blueprint String mod ( https://forums.factorio.com/forum/vie ... =86&t=6742 ) to have a real copy of your setup.

Code: Select all

Put the blueprint-string into a code-block like so
Cool suggestion: Eatable MOUSE-pointers.
Have you used the Advanced Search today?
Need help, question? FAQ - Wiki - Forum help
I still like small signatures...

User avatar
DerivePi
Filter Inserter
Filter Inserter
Posts: 504
Joined: Thu May 29, 2014 4:51 pm

Re: Final (Maxed) Designs

Post by DerivePi » Fri Jan 27, 2017 7:42 pm

Here is an updated design for circuit production
CIRCUITS FINAL.jpg
CIRCUITS FINAL.jpg (117.57 KiB) Viewed 3475 times
Advanced Circuits (red)
ADVANCED CIRCUITS FINAL.jpg
ADVANCED CIRCUITS FINAL.jpg (369.99 KiB) Viewed 3475 times
And a smelting area for iron or copper
SMELTING FINAL.jpg
SMELTING FINAL.jpg (128.46 KiB) Viewed 3475 times
Just reworking the designs for the "new" transport belt mechanics.

User avatar
hansinator
Fast Inserter
Fast Inserter
Posts: 160
Joined: Sat Sep 10, 2016 10:42 pm

Re: Final (Maxed) Designs

Post by hansinator » Sun Jan 29, 2017 10:25 pm

Why is this topic even sticky? There are still no blueprints, just awkward CAD drawings and frankly the designs from vanattefeldt for example are even more "maxed" out.

User avatar
DerivePi
Filter Inserter
Filter Inserter
Posts: 504
Joined: Thu May 29, 2014 4:51 pm

Re: Final (Maxed) Designs

Post by DerivePi » Mon Jan 30, 2017 3:51 pm

hansinator wrote:Why is this topic even sticky?
I think the idea of assigning space at the beginning of the game that expands to an efficiently laid out system by the later stages of the game feeds our CDO nicely. How many times have we played multiplayer games where we know that another player is bottlenecking a design and then landlocking it? Wouldn't it be nice if most players know that:
- A balanced ratio of pump/boiler/ steam gen is 1/14/10 (most players know this)
- 70 furnaces will smelt a full belt of copper or iron ore (less furnaces are expedient but never more - ie 72 furnaces is "wrong")
- Reserving space for 8 cells of 5 assemblers for circuit production is good. When fully expanded and running, that circuit production will fully absorb 1 1/2 belts of copper plate and 1 belt of iron. This means it is appropriate to replace 1 lane of copper plate with a full lane of circuits on a multi "copper plate" laned bus. Why would you continue to run an empty lane on your most precious real estate?

I think the real question is "Why there aren't more contributions?"
hansinator wrote:the designs from vanattefeldt for example are even more "maxed" out
Here is the link to vanatteveldt's smelting layout - viewtopic.php?f=202&t=33375&p=236930
This is why I don't promote or recommend modules for smelting - viewtopic.php?f=5&t=5705 - further, if you are going to use modules, don't bottleneck your design by only using 1 blue inserter when greater capacity is available.

User avatar
hansinator
Fast Inserter
Fast Inserter
Posts: 160
Joined: Sat Sep 10, 2016 10:42 pm

Re: Final (Maxed) Designs

Post by hansinator » Mon Jan 30, 2017 6:36 pm

Mhh that are some valid points you made there. The content of this thread is not bad at all, but the "how it is presented" could be improved, as some posts before already suggested. See, the CAD drawings are nice, but they are alienating to many players, I guess. Maybe you could add in-game screenshots and wrap-up the latest designs in the first post (by editing it) so everybody that stumbles across this sticky thread can get a quick overview on the first glance?

And of course blueprints would be nice. I personally think it is somewhat mean if people post pictures of their cool designs where everybody thinks "nice!" but won't share their blueprints.

Edit:
And for the contributions and module side of things, I think it'd be nice to link other maxed designs and the thread about modules in the first post. There are lots of player contributions, they are just not in this sticky thread. They are instead cluttered all over the forum. As this thread is already sticky it would make a good place to collect some of these other contributions.

User avatar
DerivePi
Filter Inserter
Filter Inserter
Posts: 504
Joined: Thu May 29, 2014 4:51 pm

Re: Final (Maxed) Designs

Post by DerivePi » Mon Jan 30, 2017 7:55 pm

Sounds good hansinator! You're hired :D

User avatar
hansinator
Fast Inserter
Fast Inserter
Posts: 160
Joined: Sat Sep 10, 2016 10:42 pm

Re: Final (Maxed) Designs

Post by hansinator » Mon Jan 30, 2017 9:34 pm

DerivePi wrote:Sounds good hansinator! You're hired :D
Glad you agree, but it's your thread. I can't edit it. If you have already built your setups you could at least post blueprint strings so I or others wouldn't have to build it just to post blueprints..

User avatar
DerivePi
Filter Inserter
Filter Inserter
Posts: 504
Joined: Thu May 29, 2014 4:51 pm

Re: Final (Maxed) Designs

Post by DerivePi » Mon Jan 30, 2017 9:38 pm

Alright. I wasn't planning on editing. But, you're right - if ssilk wants this sticky then it makes sense to update the top thread and keep things current. Updates forthcoming.

User avatar
DerivePi
Filter Inserter
Filter Inserter
Posts: 504
Joined: Thu May 29, 2014 4:51 pm

Re: Final (Maxed) Designs

Post by DerivePi » Thu Feb 02, 2017 1:49 am

Final Smelting Design

There are several layouts that work well for smelting. The advantage provided by this version are:
- putting two red belts together to fully compact a blue belt for the output is simpler than trying to fully compact a single blue belt (usually using UG belts)
- Only 2 inserters are required per furnace (no long handed inserters)
- When starting, the player can utilize each side separately until faster belts are available towards the late game
- I prefer a stouter furnace layout than the long 35 furnace layout.

Schematic
smelting schematic.jpg
smelting schematic.jpg (307.52 KiB) Viewed 3031 times
Picture
smelting-final2.png
smelting-final2.png (2.05 MiB) Viewed 3305 times
blueprint string
Last edited by DerivePi on Wed Feb 15, 2017 7:53 pm, edited 1 time in total.

User avatar
DerivePi
Filter Inserter
Filter Inserter
Posts: 504
Joined: Thu May 29, 2014 4:51 pm

Re: Final (Maxed) Designs

Post by DerivePi » Thu Feb 02, 2017 2:49 am

Advanced Circuit - Final Design

This one sucks up a full blue belt of circuits, copper plate, and plastic just to produce 20 advanced circuits per second. This production would probably do fairly well with lines of beacons in between the advanced circuit assemblers.

Schematic
advanced circuit schematic.jpg
advanced circuit schematic.jpg (422.43 KiB) Viewed 3030 times
Picture
advanced circuit - final.jpg
advanced circuit - final.jpg (409.13 KiB) Viewed 3300 times
blueprint string
Last edited by DerivePi on Wed Feb 15, 2017 7:59 pm, edited 1 time in total.

User avatar
hansinator
Fast Inserter
Fast Inserter
Posts: 160
Joined: Sat Sep 10, 2016 10:42 pm

Re: Final (Maxed) Designs

Post by hansinator » Thu Feb 02, 2017 12:21 pm

Thank you <3

vanatteveldt
Filter Inserter
Filter Inserter
Posts: 846
Joined: Wed Nov 25, 2015 11:44 am

Re: Final (Maxed) Designs

Post by vanatteveldt » Wed Feb 08, 2017 5:22 am

DerivePi wrote: Here is the link to vanatteveldt's smelting layout - viewtopic.php?f=202&t=33375&p=236930
This is why I don't promote or recommend modules for smelting - viewtopic.php?f=5&t=5705 - further, if you are going to use modules, don't bottleneck your design by only using 1 blue inserter when greater capacity is available.
I agree that smelters are one of the least efficient places to use prod modules looking at consumption per second. However:
- smelters are at the basis of all other chains, giving 20% resource discount on everything else
- smelters are most likely to be running the full 100% of time

And frankly, any non-beacon design are boring/trivial because there are no space constraints. You can set yourself a constraint or opt to aim for the most compact design possible, but I prefer using beacons everywhere so the beacon cost and range limit creates a real optimization problem. The utility/cost function I try to optimize is throughput/#modules with constraint that all plants have max prod modules (or in other words: first optimize output/input, then optimize throughput/#modules).
hansinator wrote:the designs from vanattefeldt for example are even more "maxed" out
For convencience some other links to "my" designs (which in all cases steal or improve on existing designs by other users)
links

User avatar
MBas
Long Handed Inserter
Long Handed Inserter
Posts: 55
Joined: Fri Jan 06, 2017 12:57 pm

Re: Final (Maxed) Designs

Post by MBas » Wed Feb 08, 2017 8:34 am

vanatteveldt wrote: ...And frankly, any non-beacon design are boring/trivial because there are no space constraints. You can set yourself a constraint or opt to aim for the most compact design possible, but I prefer using beacons everywhere so the beacon cost and range limit creates a real optimization problem. The utility/cost function I try to optimize is throughput/#modules with constraint that all plants have max prod modules (or in other words: first optimize output/input, then optimize throughput/#modules)...
Hello. There is actually huge benefit in energy saving with using blue modules in becons if you use it in combination with red modules. One blue moduled beacon rise speed by 50% and consumption by 70%, however this stats are summed with red modules statistic linearly. What does it mean? if you use assembly machine with four red modules, its base speed and energy consumption is:

energy consumption = 420%
speed = 40%

If you boost it with one simple becon your boost is very significant

new energy consumption = 490% = 1,16 × higher than no beacon version
new speed = 90% = 2,25 × higher than no beacon version

So each becon with blue modules gives you a 125% production boost (against base value) in cost of 16% energy consuption which very sigificatly beat a green module and in most cases strongly beat beacon's own energy consumption.

This is main arument for me. It practically corespond to your idea to have compact factories (different reasons, same results). I realize the energy problem when my solar panel plant start to be 20 times larger than rest of my factory when i start with "big" produtions.

(You start this topic allready in here: viewtopic.php?f=202&t=40886 , you can find my answer there about my "beacon rating")

vanatteveldt
Filter Inserter
Filter Inserter
Posts: 846
Joined: Wed Nov 25, 2015 11:44 am

Re: Final (Maxed) Designs

Post by vanatteveldt » Wed Feb 08, 2017 10:30 am

@MBas: Yes, it is well known that the combination of sp3 (in beacons) and prod3 (in plants) gives very good results. What made you think I was arguing otherwise?

User avatar
MBas
Long Handed Inserter
Long Handed Inserter
Posts: 55
Joined: Fri Jan 06, 2017 12:57 pm

Re: Final (Maxed) Designs

Post by MBas » Wed Feb 08, 2017 11:16 am

vanatteveldt wrote:@MBas: Yes, it is well known that the combination of sp3 (in beacons) and prod3 (in plants) gives very good results. What made you think I was arguing otherwise?
We both try to optimize different factors. My goal is to reach minimum energy consumption per production, your goal is to reach minimum number of modules per production. Am I right? And what i try to say is that this two ways of optimalization has nearly similar results.

Post Reply

Return to “Mechanical Throughput Magic (circuit-free)”