Creating custom robotic networks

[MaMo]

In many situations in this game, it would come handy to have possibility of roboports close to each other which are parts of different robotic networks. This game sure offers many workarounds but since I love using bots and love having many separated networks with different purposes, trying to limit the "air time" of every single one of these little buggers to minimum, sending short-range "wawes". In current state, I have to separate Rob.Networks by significant distance and sometimes necessity of "geographic" design keeps me from having enought ports to recharge. This leads me to inability to make some very smooth systems. Of course I could solve this by workarounds but that would decrease glory of robots for me For the moment, this limits spreading of my madness, which causes too much real-life awareness!

Can we have the possibility to force separation of networks sometime?

[posila]

I thought about it and the main challenge to solve would be that logistic members (logistic chests, player, anything can connect to logistic network and set condition) can be assigned to only one network in the current system.

This is how I think it would work: each roboport would have channel (default 1), and only roboports with the same channel would be able to connect. Then in areas that overlap you would have to go over all logistic members and check they "listen" to roboport channel you want them to.

[Xeanoa]

I thought about it and the main challenge to solve would be that logistic members (logistic chests, player, anything can connect to logistic network and set condition) can be assigned to only one network in the current system.

This is how I think it would work: each roboport would have channel (default 1), and only roboports with the same channel would be able to connect. Then in areas that overlap you would have to go over all logistic members and check they "listen" to roboport channel you want them to.

That sounds great and would be incredibly useful. Please consider putting this into the game.
And, if possible, look into a way to make a chest listen to more than 1 system. Although I know this could be tricky, but otherwise it would be hard to move resources between system. It could even be a special, 5th chest exactly for that purpose.

It also would allow us to separate the personal roboport area from the system, and allow us to select which robots do the construction/deconstruction when zones overlap.

[MaMo]

Well my idea was to be able to select roboport and redraw its Area of Influence (logistic and construction separatedly) with standard AoI as its maximum. This AoI then could be copied/pasted to other roboports and blueprints would remember it.

[ssilk]

Added to viewtopic.php?f=80&t=18093 Roboport/Logistic Network/Robot enhancements

The argument with the multiplayer here is new. And good! Cause it changes a lot of thoughts in the former threads.

[TI-89]

I have used lognet pretty extensively, as I agree it is a log of fun to optimize and I love 'future tech' tree in all games like this. I've found that it's actually an incredibly flexible and powerful system, and that most of the thoughts I had like these go away once you understand the constraints that are placed upon the network. Unfortunately, it is very unclear to the player how they are expected to use networks, and this expectation of how they should be used (from my experience reading forums before figuring out lognet mechanics) is very different from how they are able to be used in the game.

While variable size networks (or AoI) is one of the things I still think has some use, the zone size and shape are one of the few hard limitations on port placement, and your network topology is the only thing truly limiting potential throughput. In other words, bots are currently fairly overpowered in many ways. I think that this change eliminates one of the few aspects that really makes you consider how you place things. This has led to no shortage of frustrations in building my network; but careful planning and placement (and spaghetti when your needs change) is at the core of Factorio.

Can you provide any more specific info of what you're attempting to accomplish? In my experience most of the inefficiencies this leads to are incredibly minor in comparison to the size of the robot network in question by the time they become an annoyance.

[provet]

Adding channels controlled by logic would be awesome and possibly change gameplay in some mayor ways especially in regards to sorting