Re: Logistics storage fixes
Posted: Tue Jan 28, 2014 8:41 am
10x10 was said in spur of the moment - i was trying to say that even this small addition would make a difference for me.
Page 2 of 2
Re: Logistics storage fixes
Posted: Tue Feb 04, 2014 4:38 am
I was thinking - what if it would be possible to use beacons to boost roboport charging speed?
Re: Logistics storage fixes
Posted: Tue Feb 04, 2014 6:38 am
Cool idea. Yes. Speed and efficiency is clear. Productivity could be the chance, that a robot takes one item more.
To the size: after playing around with them, I'm now sure, that 50x50 is very ok for the logistics. We need to think in bigger dimensions. When I think what must come after finishing the rocket defense, I realized, that the whole production must change to bring the colonists to the surface. And with blueprints another type of game will become more important: "sub factories", which produce one item or so. They will work perfect on their own, but if you connect them to other roboports, it won't work anymore. The space as resource will become much more important with that changes and with the space the need for smaller control units sinks.
The behavior of the logistic bots a chaotic. I'm not a mathematican, but they would nod. Depending on the distance, the bots can fly a while before they need to recharge. The behavior reminds me a lot to the Feigenbaum-diagramm:
http://de.wikipedia.org/w/index.php?tit ... 0518090025&
http://en.wikipedia.org/wiki/Mitchell_Feigenbaum
http://en.wikipedia.org/wiki/Logistic_map
http://de.wikipedia.org/wiki/Logistische_Gleichung
As said, it reminds me, but surely there are more parameters. For example the distance between the ports and the distance, the bots fly. But assumed, that the ports are at one point, I'm very sure, the diagram would look like so. Now my interpretation of that diagram (which I cannot prove yet):
From left you see one line. That is, how much a bot transports in the network without chargers, about 0.6 items, because he must fly sometimes empty. The more bots the network has, the better is the chance, that a bot is near to the next item. That is the behavior with no charger: non-chaotic. We come then to a point, where charging is needed: sometimes the bots are then faster, sometimes slower. With the number of charge-ports (one roboport has 4 chargers) and bots the chaos rises, even if there are "windows" with less chaos. And there is a maximum throughput, such a network can handle, no matter how much bots and ports you have. As said, a very free interpretation. But I'm quite sure, there is some truth in it.
It's this maximum throughput, which astonished me most. I made several tests which proved, that a robotic network with 1000 bots and 20 ports are under constant load not so efficient as two robotic nets with each 500 bots and ten ports.
So my thought for the logistics: no real need to change. There are physicalic rules (not in the game!) which limits their size, and if you ignore them, they hit you.
No game element will change that! We shift the borders a bit, we stretch it, but in the end it will end in chaos. What's needed are some sensors, which are able to recognize that.
For the construction I see a need for larger areas. I think a roboport for construction bots only would be cool. Same size as radar, to make things easier. Two chargers should be enough.
And if those ports can also work as requester chest for repair packs, or the construction bots could distribute all repair packs to the ports equally (and also themselves): fine.
To the size: after playing around with them, I'm now sure, that 50x50 is very ok for the logistics. We need to think in bigger dimensions. When I think what must come after finishing the rocket defense, I realized, that the whole production must change to bring the colonists to the surface. And with blueprints another type of game will become more important: "sub factories", which produce one item or so. They will work perfect on their own, but if you connect them to other roboports, it won't work anymore. The space as resource will become much more important with that changes and with the space the need for smaller control units sinks.
The behavior of the logistic bots a chaotic. I'm not a mathematican, but they would nod. Depending on the distance, the bots can fly a while before they need to recharge. The behavior reminds me a lot to the Feigenbaum-diagramm:
http://de.wikipedia.org/w/index.php?tit ... 0518090025&
http://en.wikipedia.org/wiki/Mitchell_Feigenbaum
http://en.wikipedia.org/wiki/Logistic_map
http://de.wikipedia.org/wiki/Logistische_Gleichung
As said, it reminds me, but surely there are more parameters. For example the distance between the ports and the distance, the bots fly. But assumed, that the ports are at one point, I'm very sure, the diagram would look like so. Now my interpretation of that diagram (which I cannot prove yet):
From left you see one line. That is, how much a bot transports in the network without chargers, about 0.6 items, because he must fly sometimes empty. The more bots the network has, the better is the chance, that a bot is near to the next item. That is the behavior with no charger: non-chaotic. We come then to a point, where charging is needed: sometimes the bots are then faster, sometimes slower. With the number of charge-ports (one roboport has 4 chargers) and bots the chaos rises, even if there are "windows" with less chaos. And there is a maximum throughput, such a network can handle, no matter how much bots and ports you have. As said, a very free interpretation. But I'm quite sure, there is some truth in it.
It's this maximum throughput, which astonished me most. I made several tests which proved, that a robotic network with 1000 bots and 20 ports are under constant load not so efficient as two robotic nets with each 500 bots and ten ports.
So my thought for the logistics: no real need to change. There are physicalic rules (not in the game!) which limits their size, and if you ignore them, they hit you.
No game element will change that! We shift the borders a bit, we stretch it, but in the end it will end in chaos. What's needed are some sensors, which are able to recognize that.For the construction I see a need for larger areas. I think a roboport for construction bots only would be cool. Same size as radar, to make things easier. Two chargers should be enough.
And if those ports can also work as requester chest for repair packs, or the construction bots could distribute all repair packs to the ports equally (and also themselves): fine.
Re: Logistics storage fixes
Posted: Tue Feb 04, 2014 8:04 pm
If we're going to talk about roboports, why not take advantage of tools already present: reduce the area of effect (AoE) of the robotports themselves, and then add a module slot so players can improve the AoE (productivity), the charging speed (speed), or the cost to charge and maintain the roboport itself (effectivity). Then you can add effect beacons around the roboport to boost things even more, giving you a much more useful system and adding incentive to use effect beacons, which I almost never touch.