Page 3 of 3

Re: 0.13 Rainbow Belts and other madness

Posted: Fri Jul 29, 2016 10:44 pm
by BlakeMW
andywankenobi wrote:Could this sort of system be used for sparse ammo belt loops? I've been messing with various ways to keep a set amount of ammo on the belt but I haven't yet found a solution that works.
Yes it could. I normally want my ammo loops to be saturated, but this would definitely permit having a sparse ammo belt.

Note that there are also various ways to throttle a belt. Try wiring two belts in a row together, set the first belt to "enabled: ammo = 0" and the second to "read contents: hold"
throughput limiter.jpg
throughput limiter.jpg (32.2 KiB) Viewed 7317 times
This setup only lets one item through at a time (or two if there are items on both lanes). You can even wire up additional reader belts to further increase the space between items. Good if you want to make sure that only one item enters at once

Re: 0.13 Rainbow Belts and other madness

Posted: Fri Aug 05, 2016 2:29 am
by Prymal
siggboy wrote:I'm also violently opposed to underground belt interleaving, just because it's ugly and cheesy. It only "works" if you use different belt colors, which makes it a poster child case of meta-gaming (bending the rules). If a particularly clever belt layout uses that, it does not count for the record.
The way I like to think about it is this: With the faster speeds, yet still covering the same underground difference; I consider the different colors to go to different depths. They do not collide because the reds are deeper than yellows, and blues deeper than reds. In any event, play how you like, just have fun eh.

Re: 0.13 Rainbow Belts and other madness

Posted: Fri Aug 05, 2016 7:38 am
by dragontamer5788
I've been trying to accomplish something similar except as a general base design. On the one hand, I could just use logistic bots... on the other hand... where's the fun in that? My design iterations are towards a regular, expandable system, a real-life hardware analogue would be something like an CPLD or FPGA.

Image

Sorry for diverging into a totally different topic. I've got several factorio designs but all of them are flawed so I don't feel like sharing. Instead, I'm sharing my real-world inspiration.

In any case: a CPLD / FPGA in real life is a very regular square structure, with programmable interconnects. If "rainbow belts" connected "assembly-blocks" together, then a very logical and regular design could be made (similar to the "Modular Bus", except using fewer belts "rainbow style" like in your example).

Unfortunately, the more and more I understand the patterns of items in this game (ie: the relationship between Inserter -> Fast Inserter -> Stack Inserter -> Stack Filter Inserter), the less and less a "modular factory" seems useful to me. Still, a regular square structure would be an interesting base architecture...

Re: 0.13 Rainbow Belts and other madness

Posted: Sat Aug 06, 2016 9:05 pm
by Gertibrumm
Although you just missed the topic by 99%, what I thought FPGAs are good at is switching the logic and making relatively simple logic parallel without outsourcing tasks to various dedicated chips.

A rainbow belt is a more elegant and configurable single bus system compared to
an actualy programmable and flexible factory: (advertisement) viewtopic.php?f=8&t=29560 which might satisfy your needs of square structured architecture ;)

Re: 0.13 Rainbow Belts and other madness

Posted: Fri Jan 13, 2017 10:01 am
by NoQ
Guys, did you ever see circular belts with splitter input, such as the rainbow belts, locking up when full (stop rotating)? Like, there's a continuous input to both sides of the splitter, and continuous unused output of the splitter, and the splitter thinks it needs to do nothing, versus passing the circle around.

Because rotating the rainbow belt is essential (otherwise it may occur that no item is anywhere near the machine it needs, even though items are present on the belt), i can imagine this leading to an unlikely but possible deadlock (higher chances with more item kinds on the belt).

Upd: yep, i'm seeing my labs deadlock. Need some help, because the behavior of the splitter sounds natural, what am i doing wrong?><
screenshot

Re: 0.13 Rainbow Belts and other madness

Posted: Fri Jan 13, 2017 12:24 pm
by Yoyobuae
NoQ wrote:Upd: yep, i'm seeing my labs deadlock. Need some help, because the behavior of the splitter sounds natural, what am i doing wrong?><
screenshot
You just need to keep belt under certain capacity. Use some logic to stop all inputs once the belt is full enough.

Re: 0.13 Rainbow Belts and other madness

Posted: Fri Jan 13, 2017 6:00 pm
by Optera
NoQ wrote:Guys, did you ever see circular belts with splitter input, such as the rainbow belts, locking up when full (stop rotating)? Like, there's a continuous input to both sides of the splitter, and continuous unused output of the splitter, and the splitter thinks it needs to do nothing, versus passing the circle around.

Because rotating the rainbow belt is essential (otherwise it may occur that no item is anywhere near the machine it needs, even though items are present on the belt), i can imagine this leading to an unlikely but possible deadlock (higher chances with more item kinds on the belt).

Upd: yep, i'm seeing my labs deadlock. Need some help, because the behavior of the splitter sounds natural, what am i doing wrong?><
There are some tricks to prevent Rainbow Belts from locking:
  • The belt measuring and the belt blocking flow should have identical travel time of items to the splitter merging into the circle.
  • Include measuring at the belts before/after the ideal measurement spot.
  • Split the Belt at one spot so items can always flow. Ideally where you measure.
2017-01-13-18-49-34-3333624.jpg
2017-01-13-18-49-34-3333624.jpg (75.3 KiB) Viewed 6918 times
I'm using sci pack = 0 as condition for adding more

Re: 0.13 Rainbow Belts and other madness

Posted: Fri Jan 13, 2017 7:53 pm
by NoQ
There are some tricks to prevent Rainbow Belts from locking:
  • The belt measuring and the belt blocking flow should have identical travel time of items to the splitter merging into the circle.
  • Include measuring at the belts before/after the ideal measurement spot.
  • Split the Belt at one spot so items can always flow. Ideally where you measure.
Thanks, this works!
You just need to keep belt under certain capacity. Use some logic to stop all inputs once the belt is full enough.
Couldn't get this one to work though, without significant (and therefore probably unintended) amount of complexity. The question's answered though :)

Re: 0.13 Rainbow Belts and other madness

Posted: Mon Jan 16, 2017 3:42 am
by roy7
NoQ wrote:
You just need to keep belt under certain capacity. Use some logic to stop all inputs once the belt is full enough.
Couldn't get this one to work though, without significant (and therefore probably unintended) amount of complexity. The question's answered though :)
If you want to see some old school solutions to this, search forum or reddit for "passive logistical loop". Maybe obsolete with circuits on belts these days, but was cool long ago. ;)

Re: 0.13 Rainbow Belts and other madness

Posted: Sun Jan 22, 2017 3:29 pm
by huliosh
Very useful, thank you.

Re: 0.13 Rainbow Belts and other madness

Posted: Thu Apr 27, 2017 4:49 am
by Tricorius
I'm trying a few adjustments to try to get seven types on a belt. I'll know soon if it works. I'm pretty sure it will. (Mathematically it should be fine.)

Re: 0.13 Rainbow Belts and other madness

Posted: Thu Apr 27, 2017 5:09 am
by DaveMcW

Re: 0.13 Rainbow Belts and other madness

Posted: Thu Apr 27, 2017 5:20 am
by Tricorius
Haha! Clever! I like how you load from both sides instead of splitting.

Re: 0.13 Rainbow Belts and other madness

Posted: Wed May 10, 2017 7:05 pm
by galen_harvell
Oh god, please post the string for this!

Re: 0.13 Rainbow Belts and other madness

Posted: Mon Jun 05, 2017 7:48 pm
by amz3
galen_harvell wrote:Oh god, please post the string for this!
+1

Re: 0.13 Rainbow Belts and other madness

Posted: Tue Jun 06, 2017 6:05 pm
by pieppiep
amz3 wrote:
galen_harvell wrote:Oh god, please post the string for this!
+1
It's not really hard to build it yourself :P
The belt going round should be red to reach all science labs.
The 'input'belt-piece has enable/disable on science pack = 0, and the belt pieces after that are 2 that are read belt contents hold.
The blueprintstring from how I copy this design is this,
blueprint