Friday Facts #224 - Bots versus belts

[TiMatic]

To be honest, if you would give me train bridges, I would almost be willing to forego bots altogether! Bots or belts, that's just flavour. What I really love in this game is the whole OpenTTD feeling you get with a mega base! However, the train traffic in the game is severely hampered by todays standards. Once trains escalate to a certain point, the intersections can't really be optimised any more, and you have to start making individual train lines to handle it all. Trains could do so much more if you weren't limited by the silly intersection lockups we have today, and the most important thing to fix an issue like that would be train bridges! Just imagine how powerful trains could get if you didn't have to cross the main line to get back to that mining outpost with the 16 entity long train..

I think a train tunnel would be more suitable to the game style and is easier to implement. (Its like underground belts)

[Astrateau]

I like bots. I think belts become tedious after you learn how to use them. also bots seem to be overrated. from my use when bases become really big bots become to inefficient.

I can see that they do simplify a lot of the game in a very drastic way. maybe add some complexity to them. instead of charging, make them need battery replacement. used batteries can be recyclable, which can adds a logistical cost to running bots. what makes bot so strong is how they make planning easier - the most important resource in Factorio is your attention, and that's where bots are really overpowered. you got to keep bots strong because their overpowered-ness is fun. you need to make it a challenge to set up properly. make setting-up bot maintenance a huge headache, so this won't be so much simpler than belts. then when you do set them up, their overpowered-ness is well earned.

also, maybe make logistic chests really big, so you couldn't place one for each machine, and you'd have to use belts to transfer items from the requester chest to the machine. that way the logistic bots will replace the main bus, but not the whole factory. the biggest downside of trains is how much space they require after all.

[goliathfan]

I understand that part of making a good game also requires pruning and culling things, in order to arrive at a good game. It is also true that players will choose the way of least resistance, and possibly ruin their fun by going the path of least resistance. I however take issue with the overall feeling of Twinsens FFF, in that he seems to think that a fun game needs to be a complex game, i paraphrase: "Belts are the skilled way too play, bots are too "easy" ".

Let me preface my critique and feedback by saying that overall I like Factorio, and feel the developers are working hard to make a fun game. I for one think the "fun" in factorio is in managing the logistical parts, getting the ingredients for the recipes to the right assembler, or to the right part of your factory. I have played the game for nearly 400 hours now. I haven't yet launched a rocket (because the game complexity spirals out of control for me, and I lose the overview of things which I need to do to get to the rocket). I think one issue with the game as is right now is that the logistical parts are hard to understand and use. Also there is a bit of "busywork" that doesn't add to the fun of playing (like manually upgrading belts, or setting new train schedules after outposts run dry).

On with my points:

On belts:

Belts are a real drag to upgrade from yellow to red to blue, there is no way in un-modded Factorio to easily and automatically update the belt system to a higher speed color belt.
Splitters are also very convoluted to actually use (see: books of blueprints for all kinds of belt balancers).
Train loading and unloading with belts gets convoluted fast also, requiring massive loading and unloading complexes with belt balancers.

Suggested improvements to belts:

I'm going to lump belts, splitters and underground belts into the category "Belts" for this part:
Make it easy to update belts from yellow to red, and from red to blue.
Let me mark a area of belts to be upgraded. Then allow the belts to be automatically replaced as I walk past them, provided i have the better belts in my inventory.
Also allow bots to do "in-place replacement", the belt should keep working while it is marked for upgrade. Bots would pick up better belts from the logistical chests, and replace them.

Suggested improvements to splitters:

Make splitters do the "dumb" work of balancing belts. Preferably make it so belts are smarter, and load balance automatically, without the need of massive belt balancers, which are hard to understand, complex to build by hand, and take up massive amounts of space in order to get any real throughput via belts. I would much prefer to upgrade throughput by just putting better belts in, instead of going wider with belts.

Suggested improvements to trains:

I like trains and watching them on my network. What I don't like is that there is no good way to actually manage a network of trains in a smarter way. I would like very much to be able to group trains and issue commands to that group. Say: make a group of Iron Ore outpost trains in a group, then when the outpost would run out, I could en-mass order those trains into some other outpost, or even a train Yard, for reuse. I think that having functionality like the Logistics Train Network in vanilla Factorio would be very very nice.

Please fix the busywork of manually managing schedules, let us automate trains too! I would love a interface where i could select my stations in a map and list view, and then set supply and demand at that station, there would then be a rail-yard with idle trains, they would get a schedule automatically and get to work moving things when there is supply and demand at a pair of stations.

I understand that the game is due for a GUI upgrade, please, please, please also make trains easier to use and manage.

Also I like the idea of containerizing bulk goods. This was a massive, massive improvement in real world shipping, there was no need to mix and match different size goods anymore, just stick those different sized goods in a container, and move that container around. I would very much like big gantry cranes in a rail-yard, which would load and unload trains.

There could even be hoppers to load ores straight into a new wagon. In real life trains don't have inserters moving individual bits of ore around, they would get loaded into a hopper via a conveyor belt, the hopper would load a wagon on the train, and that wagon would also dump the ores into a special unloading station at it's destination.

On logistic bots

As the game is now, with recipes as there are now, they only feasible way for me to manage the growing complexity is to use bots. With bots I can have a assembler, set the recipes, add provider and requester chests, and the bots figure out what is needed, how much is needed, and get the parts there. With belts I'm forced to figure this out myself.
Bots are in this sense a bit of a mitigator of the recipe complexity. I don't like that the proper logistical bots were pushed into the 5 science pack territory. This makes the game unnecessarily hard for me to play. I just want an easy way to hook up things to my assemblers. Now I'm forced to belt things (which was intended behaviour post-patch), but it didn't result in more fun for me personally.

On game flow and progression

I think the early game of Factorio is quite boring. You are constantly busy laying down repetitive things (smelting array, main bus). I would like to have a construction bot "light", and light personal roboport way earlier in the game, so that the early game is faster, and you get to the fun parts earlier (trains, refinery, logistical bots). I think for me the fun is in managing the logistical side of things, and seeing your factory grow, not in manually building repetitive stuff (that's what construction bots are for!).

TLDR:

Make belt upgrading way easier, and allow bots to do this.
Make a better train interface, with trains grouped into similar timetables.
Make "logistical" trains, which figure out themselves where there is supply and demand, and let me set supply and demand in a central user interface.
Maybe make hopper wagons, and have a bulk loader on stations for ores.
Maybe containerize trains, so they get loaded/unloaded via a gantry train which picks up the container.
Make a construction bot "light" with limited personal roboport, so the repetitive-ness of early game is over quicker.
And maybe think about simplifying recipes.

[Jan11]

I have some different solutions to resolve the problem (I guess it will solve the problem without take transport bots out of the game) but first of all I have to say I like both the drones and the belts and I use both of them the whole game. Bots should thay in the game but different beacuse they are necesseary. Example: I could never repair all my laser towers without bots. But yes, I think the game gets to easy with to many bots.

My solutions are:


1 . Drone towers get a limitation how many bots each one can control and to prevent player to build them side by side the commands area from each tower must build outside the other one. If the control areas overlap bots are going out of control and deliver the wrong goods or flying away or crashes to ground. Thats the easiest way to reduce the bot use.
I think a lot of players will now use a belt and bot combination, or they will spread their packed bot bases in seperate bases with own botnetworks and connect the different part with belts or trains. I think this is a good possibility to give player bots and belts.

2. The other function, bots cant use chests and only can use drone landing stations ,which will be let them transport the goods much slower because they have to land at the port and must be loaded with a gripper arm one by one and not all at the same time (like now at chest). And from this stations the goods must be transported with belts to the production buildings. This means player have to spread their bases to because if they want to tranbsport the same amount like belts or trains they have to build big landing ports arround each product type and also have to deliver goods with belts from them to the production buildings.


I guess both solutions don`t forbid transport drones but it will make them more specialized for some tasks( maybe repair or some really rare important goods etc etc over the whole base (and don`t let players transport all goods with them). Players can use them but also have to use belts because the drones are very good but have disadvantages to.

[vaderciya]

While I do agree that bots are powerful, I believe that bots are necessary. Nerfing or removing logistic bots would not only make me and others like me very angry, I believe it would also hurt the game. Then there isn't the big technology rush at the end of the game anymore. Why would I keep coming back to the game if every factory I build could only use belts? Or if using 5k robots equals 5GW of constant power drain? I'd get sick of it.

At the same time I understand the flip argument. They're powerful.
But.
Instead of nerfing the powerful item, which by the way you guys have already done many times..
Why not incentivize belt building? Ultimately, robots are slow until you get the first few levels of the infinite speed research. Which for the average player takes a while.
So while in this end game state of expansion, give the player 2 things in a single package. Very very fast belts, that have incredible throughput.
I don't remember the actual numbers here, but let's say the new belt is green(like the stack inserters) now let's take blue belt stats and increase them. Whatever the blue belts speed is, let's triple it, and provide 4x more items on a single green belt than the blue belt. While increasing cost to require 70% more resources. Again, I don't look at belt stats. But something like 30 or 50 gears and 10 lubricant for standard belts, 4 blue circuits for splitters(in addition to the extra iron and lube), and maybe concrete or stone brick for the underground, as the underground will go much farther (twice as far perhaps) it needs structural supports underground, which is where the concrete or stone brick comes in.

This incentive gives the player a very very good, fast, useful belt for a good amount of resources. It will go much faster than the bots and carry a large amount, and surpass robot flying speeds until the infinite research has got a few levels.
I would honestly love to have this kind of mega belt in the game. As it stands, belts are limiting. Even blue belt is slow, doesn't carry that much, and could definitely be improved.
I hope you consider this request for a feature, instead of more nerfs to our beloved robots.
I really think a mega belt is the way to solve this dispute.

[GaboFDC]

Hey, this is one of my few post in the game,s but reading that FFF, and some replies really waked my interest, and i just want to drop my 2 cents.

I have not read all the thread replies (more than 15 pages) but i believe the vast majority agree in that we should not generalize bots, as construction robots and logistic robots accomplish totally different purposes) and i see Twinsen tried to leave that clear, but sometimes when discussing this we just generalize in "bots". Construction robots are not breaking anything in the game, they just add up to the generalized idea of the game (automate all the things), and also as has been said, allow mega bases and self-constructing things, that are hell of fun, to see, build, and play.

On the other hand the logistic robots are the matter of discussion, and allow "lazy setup" just copy assembler recipe and paste on requester chest, which for some may "break" the game, and for others is just a QoL upgrade. Not to mention that even when doing this "lazy setup" you have to take charge of, resource providing and movement. So if they where as break as some say, you could just move everything to bots, but i believe (may be proven wrong) that even if you maximize logistic bots usage, you will still need to use belts and trains.

So my real 2 cents here, the first and more important thing you guys have to analyse is differentiate a bit more both types of bots, maybe making research tree of them more apart, and giving each different base stats (make construction robots faster, and logistic robots slower). And one big thing more, consider giving robots a life cycle, right now once you made a robot it will serve you almost ilimitedly (until destroyed by a bitter, or some military thing) so the win in the resources spent over the long time are a lot. If you make something similar to the combat bots, that have some life time, or implement some sort of life/use time for logistic robots, will may be a great sollutions, as players will need to setup a bot renovation thing if they want to use logistic robots a lot, and for log time.

[Bi0nicM4n]

P.S.

4. Add research that requires space science to allow logistics bots to travel within (connected) logistics networks -- or make a more-expensive Logistics Robot 2 (that also requires space science) that has this functionality

Are you even serious?

I was looking at it from a compromise perspective, and building on peppe's idea of 'limit logistic bots to the range of one roboport rather than the entire network'. If we hypothetically went down that road, space science unlocking the full (current) functionality would appease megabase builders who need the full extent of the current logistic bot capabilites (for UPS reasons, for one).

I tend to encapsulate my base in early-mid game in turret wall completely. How am I supposed to provide repair packs to the wall 100 hours prior to reaching rocket science?

If you read the rest of my suggestions, you'll note that most of it actually is intended to make it easier to get logistics bots and take advantage of some of their capabilities, but gating the one feature that makes people replace belts outright behind a much higher requirement. An in-between setup might involve a main bus where each branch pulls contents into a provider chest to serve assemblers within a 1-roboport area. Combine this with research to expand roboport ranges, the coverage area of a single roboport may be sufficient for a lot of uses and there's several creative ways available to expand that.

And again, are there *really* that many players that ditch belts altogether as soon as they get bots? I never use bots to unload trains or operate smelters. If you haven't noticed, bots need to recharge (which is not the case for belts, those are always functional). The sheer amount of roboports I'd need to replace all my belts is what keeps me using belts.

And even *if* there were that many players who did -- it's not like we have any rights to get into their pants or teach them how to live their lives, -- OR tell them how to play the game, in our case.

P.S. Again, to all players who propose nerfing the bots - you either have too much free time on your hands, or you're just being trolls. Either way, I have a job to attend, i don't have that much free time as you do (and many people here do have jobs, I'm certain). For the sake of everything that is good in humanity, don't propose butchering the game mechanic simply because you don't like it. You can always install a mod to disable logistic bots if you want. And I want to play the game the way that I want to play it.

[Mash]

I feel a lot of social pressure from the rest of the community any time I read the forums or reddit where everyone is telling people with spaghetti bases that they're dumb, bad people who are playing Factorio wrong. I even stopped watching some of the YouTubers because they insist on calling anything they don't like "cheaty" and generally being extremely condescending.

These are just some of my personal thought processes and not an attempt to force anyone else to play Factorio in my preferred way.

I usually quit such build servers quite soon, they get boring as fuck, barely a challenge, just copy pasta without the spaghetti, therefore I really like limited/ribbon words, they make BP's unusable, it needs to be messy.

[GenBOOM]

I understand that part of making a good game also requires pruning and culling things, in order to arrive at a good game. It is also true that players will choose the way of least resistance, and possibly ruin their fun by going the path of least resistance. I however take issue with the overall feeling of Twinsens FFF, in that he seems to think that a fun game needs to be a complex game, i paraphrase: "Belts are the skilled way too play, bots are too "easy" ".

Let me preface my critique and feedback by saying that overall I like Factorio, and feel the developers are working hard to make a fun game. I for one think the "fun" in factorio is in managing the logistical parts, getting the ingredients for the recipes to the right assembler, or to the right part of your factory. I have played the game for nearly 400 hours now. I haven't yet launched a rocket (because the game complexity spirals out of control for me, and I lose the overview of things which I need to do to get to the rocket). I think one issue with the game as is right now is that the logistical parts are hard to understand and use. Also there is a bit of "busywork" that doesn't add to the fun of playing (like manually upgrading belts, or setting new train schedules after outposts run dry).

On with my points:

On belts:

Belts are a real drag to upgrade from yellow to red to blue, there is no way in un-modded Factorio to easily and automatically update the belt system to a higher speed color belt.
Splitters are also very convoluted to actually use (see: books of blueprints for all kinds of belt balancers).
Train loading and unloading with belts gets convoluted fast also, requiring massive loading and unloading complexes with belt balancers.

Suggested improvements to belts:

I'm going to lump belts, splitters and underground belts into the category "Belts" for this part:
Make it easy to update belts from yellow to red, and from red to blue.
Let me mark a area of belts to be upgraded. Then allow the belts to be automatically replaced as I walk past them, provided i have the better belts in my inventory.
Also allow bots to do "in-place replacement", the belt should keep working while it is marked for upgrade. Bots would pick up better belts from the logistical chests, and replace them.

Suggested improvements to splitters:

Make splitters do the "dumb" work of balancing belts. Preferably make it so belts are smarter, and load balance automatically, without the need of massive belt balancers, which are hard to understand, complex to build by hand, and take up massive amounts of space in order to get any real throughput via belts. I would much prefer to upgrade throughput by just putting better belts in, instead of going wider with belts.

Suggested improvements to trains:

I like trains and watching them on my network. What I don't like is that there is no good way to actually manage a network of trains in a smarter way. I would like very much to be able to group trains and issue commands to that group. Say: make a group of Iron Ore outpost trains in a group, then when the outpost would run out, I could en-mass order those trains into some other outpost, or even a train Yard, for reuse. I think that having functionality like the Logistics Train Network in vanilla Factorio would be very very nice.

Please fix the busywork of manually managing schedules, let us automate trains too! I would love a interface where i could select my stations in a map and list view, and then set supply and demand at that station, there would then be a rail-yard with idle trains, they would get a schedule automatically and get to work moving things when there is supply and demand at a pair of stations.

I understand that the game is due for a GUI upgrade, please, please, please also make trains easier to use and manage.

Also I like the idea of containerizing bulk goods. This was a massive, massive improvement in real world shipping, there was no need to mix and match different size goods anymore, just stick those different sized goods in a container, and move that container around. I would very much like big gantry cranes in a rail-yard, which would load and unload trains.

There could even be hoppers to load ores straight into a new wagon. In real life trains don't have inserters moving individual bits of ore around, they would get loaded into a hopper via a conveyor belt, the hopper would load a wagon on the train, and that wagon would also dump the ores into a special unloading station at it's destination.

On logistic bots

As the game is now, with recipes as there are now, they only feasible way for me to manage the growing complexity is to use bots. With bots I can have a assembler, set the recipes, add provider and requester chests, and the bots figure out what is needed, how much is needed, and get the parts there. With belts I'm forced to figure this out myself.
Bots are in this sense a bit of a mitigator of the recipe complexity. I don't like that the proper logistical bots were pushed into the 5 science pack territory. This makes the game unnecessarily hard for me to play. I just want an easy way to hook up things to my assemblers. Now I'm forced to belt things (which was intended behaviour post-patch), but it didn't result in more fun for me personally.

On game flow and progression

I think the early game of Factorio is quite boring. You are constantly busy laying down repetitive things (smelting array, main bus). I would like to have a construction bot "light", and light personal roboport way earlier in the game, so that the early game is faster, and you get to the fun parts earlier (trains, refinery, logistical bots). I think for me the fun is in managing the logistical side of things, and seeing your factory grow, not in manually building repetitive stuff (that's what construction bots are for!).

TLDR:

Make belt upgrading way easier, and allow bots to do this.
Make a better train interface, with trains grouped into similar timetables.
Make "logistical" trains, which figure out themselves where there is supply and demand, and let me set supply and demand in a central user interface.
Maybe make hopper wagons, and have a bulk loader on stations for ores.
Maybe containerize trains, so they get loaded/unloaded via a gantry train which picks up the container.
Make a construction bot "light" with limited personal roboport, so the repetitive-ness of early game is over quicker.
And maybe think about simplifying recipes.

this is one of the best posts in this entire thread.
on belts- there is a mod called upgrade planner that does this exact thing and needs to be added to vanilla.
on splitters- I think those splitters should require a higher tech level to do the same thing as a massive belt balancer, there is also mods for this but I think they are a bit too easy to unlock.

I like all of your train station ideas. forget belts. forget bots. make trains even better and who cares.

[TiMatic]

When I use logistic robots in my games, I also feel that they are overpowerd, because the are no real limits. More bots simply requires more roboports and more solar panels.
So I think the key factors are the limits of robots in comparison to the limits of belts, because belts have a lot of limits, but robots don't.
So I would prefer to balance that limits. For example:

Add limits to robots like:
- Logistic bots may consume rocket fuel, instead of electric energy, or both.
- Logistic bots may required maintenance (e.g. limited battery life time).
- Max. number of logistic robots could be limited like combat robots, and then increased by research.

Remove limits for belts:
- Make items stackable (stack size increase by research)
- Add sorting options to belts
- Add very long underground belts (e.g. 24 tiles).

[AcolyteOfRocket]

Oh dear I am posting without reading all the 14 pages that came before - naughty me, oh well.

I tend to beeline bots for the construction bots - "bot-strapping" as I call it. Because solar powered personal roboports are pants that means beelining portable fusion power plant. And with marathon settings that means using logi bots unless I want to go on a mammoth manual construction spree.

So if you want to make logistic bots redundant, stop making us use them to get what we really want - the ability to use convenient automated blueprint construction.

I don't use mods QOL mods myself, but the nano-bots mod is popular for a reason.

[GenBOOM]

re "stacked belts"

I would love to see this. Give me a whole new category of belt where each individual "item" moving along the belt can actually be a stack of 2+ items. If an inserter tries to drop an item on top of another item of the same kind, they get stacked. If you pull an item from a stack, the stack gets smaller. This wouldn't just change total throughput, but would allow new and more complex things to be done with "smart" circuit+belt designs, etc.

woo, we are like 4 people now.

huh, thats not actually what I meant by stacked belts, but is still intriguing idea. maybe red belt can handle 2 stacks and blue can handle 3?
we would need a machine we can put onto a belt and organized the stacked resources vertically. this way we can still place things on top of other things, making blue belts go from 6 to 18 items per belt, but the UPS should not be really affected because 3 items hold the same position on the belt.

I meant literally a belt on the ground with a box around it supporting a belt above it so that you have 1 tile but 2 belts. inserters would still be able to grab from both.
stacking higher would require you to merge the top belt to a lower one or just have them be dumped into an assembler. but if the assembler doesn't need that many items per second it would be required to regulate it with a circuit network or a hopper.

[Ilesere]

You say that bots isn't fun.... but that's simply an opinion, one view on one way to play the game. By removing or even nerfing bots you remove/nerf one way to play the game. If you want to do belts you go and do belts... no one is forcing you to use bots if you want the challenge of a botless base. By contrast my last base was an attempt (once built up to bots) to go completely belt free.... believe me that is also a challenge. NOt a challenge on how to move stuff but one of how to manage bots locally across a large base, which presents a whole new type of game vs belt management. Also you say belts are a bigger challenge... but I have to say not really. Once you have a main bus and a set of blueprints for different types then expansion is just the same - stamp blueprint and let the con bots build it... so belts can get to the same point of not being a challenge.

Fundamentally by removing or nerfing this option you are saying that one way of playing the game is 'Not Fun'... but it's just different fun.

Now making logistic bots harder to get... that's a different matter. Making con bots and log bots different levels (so con bots can be obtained earlier anf log bots later... thats a different matter. Making it so those log bot upgrades are more expensive to reach or that there are more belt upgrades thats another option (as long as you also improve the belt upgrade process as well.... PLEASE!!!!!). But making it so log bots are always an inferior option (and I know that's not quite what you've said but I hope you take my point)... that's just saying that your way of playing the game is 'the right type of fun'.

[aober93]

I played a self made marathon mod, where the early game is expanded hugely. There it became evident to me, that bots are a very good implementation of progression. They dont actually take away belts. Its just that in default games you reach end game very fast. And it is there that most gameplay takes place cause its an endless game.

Talking mid/early game: The only thing that felt out of order is the stack size going from 1 to 2 increasing bots power by 100% in 1 shot.

[ralphioli]

What in my opinion would be the best, is if you could turn bots on or off when creating the game map. Just like you can enable peaceful mode, for example. This way, people who like them can have them, and if you don't like them, you can turn them off.

[emk]

I would be really sad to lose my bots. Please don't nerf them.

My major uses of bots:

• Construction bots and blueprints. Personally, I find that Factorio without construction bots and blueprints is a lot less fun, at least after the first playthrough. Factorio is all about scalable power and progression, and bots are a key part of that aesthetic.
• Automatic production of weird things with complex dependencies, for use when placing blueprints. Factorio has all sorts of odd little items with complex inputs, which you only need in small quantities. For example, lights, or different kinds of chests, or radars. It's not fun at all to build belt-based automation for these things. But it's much more fun to use blueprints if your factory can produce anything in modest quantities.
• Bots are fun. It's rewarding to mix belts and trains and bots, and it gives the puzzle a desirable complexity. Sure, it's possible to use bots to eliminate interesting complexity, too, but you have to be pretty deep in the end game to produce enough bots to outperform blue belts.
• Bots provide an intoxicating sensation of power.

[Bi0nicM4n]

When I use logistic robots in my games, I also feel that they are overpowerd, because the are no real limits. More bots simply requires more roboports and more solar panels.
So I think the key factors are the limits of robots in comparison to the limits of belts, because belts have a lot of limits, but robots don't.

Oh, sure. Let us completely forget about robots recharging at roboports, which have only 4 recharge slots, which makes you put tens of roboports in one place to maintain productivity at any serious task (like ore unloading/smelting).

So I would prefer to balance that limits. For example:

Add limits to robots like:
- Logistic bots may consume rocket fuel, instead of electric energy, or both.
- Logistic bots may required maintenance (e.g. limited battery life time).
- Max. number of logistic robots could be limited like combat robots, and then increased by research.

Then again, why not make belts require periodic lubrication? 'Cause that's how it works in real life!

Remove limits for belts:
- Make items stackable (stack size increase by research)

You mean a gear on a belt would be like 10 gears in reality? That will mean 10 times more of resources will stay uselessly on belts.

- Add very long underground belts (e.g. 24 tiles).

You may not have known this, but there are already mods for that. Since, like, 2014.

P.S. I don't like to ask, but have you actually played the game, at least for 100 hours? Or did you ever plan the base able to launch a rocket every 5 minutes?

[Squingy]

My 2 cents is very simple, it's never a good idea to remove a solid polished feature of any game, especially after it's been in the hands of players for so long. Beside that if people don't like how bots play you simply don't have to use them, perhaps to thwart temptation for some maybe I suggest you make it an option during world creation? Which obviously would have the bots enabled by default. The balance of bots VS belts is trivial I feel, both options have their appeal, and removing them completely would be a terrible mistake.

Thanks for reading.
-Squingy

[Meerletalis]

I tend to lean toward belt use, even in the end game. However, there are times when the bots are vital. I suspect, from the plethora of mods making them available sooner, or more powerful, that bots will remain even if they are removed.

[DpEpsilon]

An idea:
- Allow fine control over which robo-ports are grouped together.
- Limit the number of robo-ports that can be grouped together.
This would encourage more interesting base design rather than just a giant logistic network, which I think fairly elegantly solves the problem in a way which preserves logistics bots.

If this were not implemented, I think the only acceptable simple nerfs to logistics bots are:
- Increasing resource cost to build
- Increasing electricity consumption
- Increasing cost to research logistics bots
- Increasing cost to research capacity increases

In general, I think it's preferable to add interesting design challenges over straight nerfs.