Friday Facts #224 - Bots versus belts

[isaacnewman]

Would it be possible to add a customizable limitation to number of bots?

Perhaps you could add a new element to logistic networks: computing. And make the computing requirements of your network increase in a non-linear way according to number of bots (which is realistic). Add a new building type, a "supercomputer", that needs to be part of your network and its computing power would work like electricity (PFLOPS instead of MW). As a very simple example, a 100-bot network could require 10 PFLOPS while a 1000-bot network could require 1,000; just make that formula adjustable with the other game settings. Then you could add research to "optimization algorithms" or some other silly things like that.

[Tricorius]

I am I the only one that think this whole argument is pointless?

... One of the things you buy with a game is the game design, and an important part of game design is balancing. By your logic, you could stuff whatever inside the main game since "it's a sandbox game". ...

I would agree in a competitive game. But in a non-competitive game balancing really *doesn't* matter. At all.

The devs could bring back alien artifacts, alien science, which does nothing except unlock a rocket insta-launch building, and it wouldn’t effect the fun I have with Factorio either way. (If there were an achievement attached to it, I’d perform that chain exactly once, then be happy I had the achievement. If it were tied to a feature that added $5 to my steam balance every time I started a new base and ran through that alien science / insta-rocket chain, it would become tedious within n number of playthroughs—for me “n” would probably equal 2 or 3.)

You can *literally* stuff anything inside of Factorio and it is perfectly fine. It would still contain the exact same potential of enjoyment for the vast majority of people who have fun playing the game.

[meganothing]

You literally said "rules and constraints of the game" about a sandbox game. Which is like saying "We need violence in kid's games".

Even sandbox games have rules and constraints. For example: An assembly machine doesn't work without energy and source materials. An assembler and a pipe can't occupy the same space. A nuclear reactor needs uranium. A train needs tracks. Electricity is transported by wire. One copper plate is created from one copper ore

If you think these constraints are not necessary for a sandbox game, maybe Factorio isn't a sandbox game by your definition.

I would agree in a competitive game. But in a non-competitive game balancing really *doesn't* matter. At all.

Not quite correct. If assembly machines1 were faster and better than assembly machines3, nobody would research or build them. If one coal delivered the same energy than one uranium, nearly nobody would build a nuclear reactor, except once out of curiosity. If the shotgun did 100k damage every shot over the range of the rocket launcher, would you ever build or use a rocket launcher?

Sure, extreme examples, but even a single player game needs a sensible progression and (with the exception of *pure* sandbox games) challenges that grow with the abilities.

[bartekltg]

I didn't read all 41 pages of this thread, so, if I'm repeating - sorry.
One of the topics here was how to nerf bots in a sane way. Many ideas, like ground bots and interacting bost, were unacceptable because they strip boots off their main advantage - computational effectivity. A bot do almost nothing in most of the ticks. If (has power) go in the direction of the target; More advance logic run occasionally (start, reach a target, run out of power...).

Bun on page 4 user ske mentioned the interesting idea. We can limit the number of bots that can land/interact with a chest in a given period. The logic is still simple (boot wanting to land check a permission from a chest), maybe even logic from roboports can be reused.

This does not change anything for bots that carry relatively rare items but reduces the throughput of a single chest to a single chest highway. The speed of chest can be upgradable;> If I correctly understand bots implementation, it would not increase computational cost significantly. If waiting bots are not "floating", but as soon as they request landing pun into a queue, then the waiting bots will be even cheaper;-)

Of course, I'm not claim that is the only path or even we should nerf bots now:) But as an option to consider (alongside belts with stacked items*)).

As a bonus, If we add a condition, that a new bot is sent only if the provider chest is _not_ on a full workload, it should give us nice ant-pathways with a sane number of bots.
I would like to see an option in requester chest that limits, how many bots can be sent concurrently to fulfill requests, it would be probably even simpler to implement and give us nice manual control.

*) faster belts - more computation per tick, probably proportional to the speed.
belts with stack - a constant cost increase regardless the belt's level.

[Tricorius]

Not quite correct. If assembly machines1 were faster and better than assembly machines3, nobody would research or build them. If one coal delivered the same energy than one uranium, nearly nobody would build a nuclear reactor, except once out of curiosity. If the shotgun did 100k damage every shot over the range of the rocket launcher, would you ever build or use a rocket launcher?

Sure, extreme examples, but even a single player game needs a sensible progression and (with the exception of *pure* sandbox games) challenges that grow with the abilities.

My point is: why does it matter if no one builds them? It matters 0% to me if you or anyone else builds assembly mark three. I already don’t build or use a rocket launcher. It’s existence (or current lack of balance) doesn’t matter to me at all.

What you do in your own sandbox doesn’t have a single effect on what I do in mine.

So I seriously don’t care if your shotgun destroys every biter on the map. To me, that is less fun, so I’d use the rocket launcher. And you’d be having quite a bit of fun obliterating biters with your shotgun.

I personally like playing with biters. Does the existence of all the biter config settings, or the ability to play peaceful effect me? The balance between a game with biters and one without is *significantly* off.

Does that bother you? It doesn’t bother me. I have fun playing with my preferred settings.

[Cellfire]

I read the first few pages and only seen faster belts suggested. Maybe items on belts could stack after you do some research for it. This can increase throughput of each belt type and reduce the number of belts lanes need for large setups.

Current 1 iron plate or product image on a belt is 1 iron plate or product. If the image contained 2 iron plates, right away you can half the number of belts you need or double a production line.

For visual side of it maybe the iron plate or product image could even be a box with the icon of the iron plate or product.

[Winter837]

Here's an idea:

Have roboports able to insert logistic bots (with cargo) onto belts, then have the bots act as vehicles that can add their own speed to the belt's speed. At the end of the belt, the bot is taken off the belt, back into a roboport and returns to flying again. Bot's would just need to be able to pathe using the belt.

This could create bot highways that very rapidly move items around the factory, but require the player to think about how to place them.

Result: A combination belt/bot build which is superior to either individually.

[jschadewald]

I wish we could define multiple distinct, overlapping logistic networks and decide which chests and roboports to add to them. In my mind, a chest or roboport could belong to multiple logistic networks simultaneously (just like machines can belong to both a red and green circuit network). The purpose of doing it this way would be that I could assign each logistic network to a particular supply chain or need, such as a dedicated network to deliver fuel to train stations and a dedicated network to bring ammo to turrets and artillery. Another thing: logistic bots would only be able to interact with roboports in their network, meaning they couldn't network-jump unless they enter a roboport that belongs to multiple networks. So, this would give players the option to manage the number of logi-bots within each network. Construction would work the same, though.

[Engimage]

Regarding sandboxes and balance.
Balance is not irrelevant for sandboxes. People still do compete and share their achieved designs on reddit/forums as they actually feel that achievement.

In any game goals can be achieved in different ways. There is a way for “”hardcore “ players which has more efficient result but requires a lot more effort. There is a way for casuals which will generally achieve the same but is less efficient but more simple.

The flaw comes when both ways are generally the same. In this situation people achieving their goals using other ways are considered “weird” or newbs or whatever.

And in this case devs can be offended if players who actually achieve something in their opinion are considered weird.

[Proculis]

Let's step back and see what would happen if you were to remove the bots:

'Bot haters' would be happy, but they didn't use bots anyway. No difference for them.
'Bot lovers' would be pissed, install mods and have their bots whether you like it or not.

Net change: Zero. You'd just add more nuisance for players who like bots. Instead you've wasted your own precious time you could have spent developing the cool stuff which makes Factorio such a nice game.

PS: Sorry if someone already came up with this argument. 40+ pages are simply TL;DR

[VFaalcatnodriiro]

I don't get you dev-guys.


On the one hand you are seeking for ways of "encouraging" us players to use belts...
On the other hand you break splitters by removing the "per-item-type"-splitting feature. Why? How are we supposed to have balanced multi-item-mixed belts, when you remove the option/feature to sort them per item?

So now (as of .16.16) if we want do distribute a variety of items to assemblers we no longer can use splitters, but are more or less forced to use bots.

*slow clap*

[Bauer]

I really don't understand why the devs find this important enough to set the forum on fire like this.
In vanilla, everyone starts every game with a pick axe. You are litterally forced to start with belts. IMHO, logistics bots are useless until you send your first rocket (or maybe very shortly before that, if that is your only/primary goal). Hence, the average player is forced to >10 hours of bot-free play, which is very good. A beginner will play for >30 hrs before being able to build the first logistics bot. Things like "there is no spoon" guide the player to optimize belt logistics before even thinking about bots. Or have you seen a speed run with bots?

If someone then decides to go for bots because belts are tedious and impractical for a post rocket mega base, I am perfectly ok with that.

When I first read this FFF, I thought that they are making a good point saying that bots are a little too easy. But then I remembered the mess I created with bots the first time I (tried to) put them to use. Efficient, high troughput logistic with bots is NOT EASY. You can easily end up with '000 of bots sitting around robo ports waiting for a charging slot sucking the energy out of your power net like shit.

Besides this, bots are not the best answer to every logistics problem in factorio! They are good for high throughput, short distance. (Eventually also for low throughput, large distance -- but this wouldn't be my personal choice). So even a mega base will use different logistic solutions, i.e. belts and/or trains, depending on map settings.

It seems to me that this belt vs. bot discussion is dominated by ppl who forgot what the game is like in the players' first hours. And an experienced player should be allowed to make his own choices, which seems to be 99% consensus in this discussion.

[bobingabout]

what's a loader? A belt feeding directly into a factory? An extra entity serving as an interface between belt and entity, essentially replacing inserter?

Basically, Yes. you put a belt leading into a loader facing a factory, chest or train to high speed load them with items. literally puts items in as fast as the belt can move them.
works the other way around too, so it unloads a factory, chest or train straight onto a belt.

Personally I don't understand this "buff the belts" argument. What is the problem with current belts? You upgrade yellow to red, red to blue and then you put down multiple blue.

Yeah, belts aren't the thing that needs buffing, its belt -> container insertion.

LOADERS!!!

Except people complained that belts would be overpowered and it would make robots useless.

Pointless anyways ever since Stack Inserters.

Other people are also suggesting faster belts too, which then brings the issue that you already have slower inserters that can't grab items off faster belts, you would need to make even faster inserters capable of grabbing items off the even faster belts.

Besides, Stack inserters were the balanced answer between standard inserter and loader, because people thought loaders were overpowered. and honestly is a horrible hacky solution compared to loaders.

And also keep in mind loaders don't NEED to be a 2x1 entity, they were designed to be that big because it was part of the balancing issues. you can fit this super awesome mega entity in the same space as an inserter? (Well, you can't, because you need to split the belt first, then specifically load it from the end, which takes up more space) better make them bigger.


Just think about the logic for a moment... Loaders weren't added because they were too powerful and would make bots useless... now we're removing logistic bots because belts aren't used often enough... the big issue isn't that they're adding the new logistic method of flying robots, it's that they don't add anything new to the later game to compete with the robots.

I would much rather have loaders than either a nerf to logistic robots, or their removal entirely from the game.

Now if you want even faster insertion the Assemblers/Recipes just can't keep up.

Also keep in mind I play modded. The big issue most people have with my mods is NOT that your machines can't keep up with insertion speed... but that the inserters work so fast you can't insert/remove items fast enough. but again, modded, I could just mod in loaders.

My big issue is more of a worry that whatever they change, I would want to be able to change it back in my mods.


Which brings me back to page 4... nobody has commented on my modders perspective of possible changes to bots.

A 3rd type of Robot (Does some jobs of construction, and some jobs of Logistics) doesn't sound like a bad thing. Perhaps merge them into a single type with flags such as...

Code: Select all

can_repair = true
can_construct = true
can_provide_requester_chests = true
can_provide_player = true
can_attack = true

Basically, if the first 2 are true, it's a construction robot, if the second 2 are true, it's a logistic robot. construct and provide player sounds like a good starter robot, and a decent mid-way point between construction and logistic.
And yes, add an attack zone to the logistics network, and when enemies enter this zone, attack robots go and shoot at them. turning on that tag will obviously require you to specify a weapon for the robot, and should be able to be configured to require ammo(Like most weapons used by a player, or vehicle, and the gun turret), or energy(Like laser turrets, and personal defences). and although I don't think it's needed... just shoot (Like combat robots, or worm turrets)

Basically what I'm saying here is... Give the modder options. Merge the definitions for construction and logistic robots to just "Flying robots" as suggested in the news post, but allow that entity to be configured (by a mod) to perform the function of a construction robot as it is now, a logistic robot as it is now, or even (and this is a suggestion that I've seen a few times) some kind of sentry robot.

Though there's still the option of making a turret that shoots distractor robots. (I did make some ammo for that, but it is currently limited to the "Artillery" ammo category... and although it was intended to be fired from the artillery weapon I added to my MK2 tank... it MIGHT be possible to load that ammo in the big artillery wagon used on a train, and likewise load artillery shells into my tank. I'm not sure if it's a good idea to separate the two or not yet, but there's a huge difference in firepower.)

[Caine]

If I understand the arguments of the megabase players correctly, then a major problem of belts in the late game is that in beacon layouts the belts are lacking in throughput. How would a loader solve this? The belt throughput limit is still the same and space requirements are not affected either. It would buff belts in a way that there is still no balance between belts and bots.

[seePyou]

Why is everyone trying to remove a part of the game that they do not use? I don't get that basic premise!
All the people arguing to remove the bots are the same people that are not using them! Keep on not using them please and let the rest of the people that want bots continue to use them.

Arguments for the removal made:
* it takes away development cycles.
To that I say, "Entitled much?" You are not the only player in the world that has paid for Factorio. Developers cannot develop solely for your benefit.
* it is not balanced
But you do not care! You also just go around arguing how much of a bad ass you are that you are not using the OP way of developing factories with bots! This is just a wish to "punish" the people that have it easier because they should all just "get good". How is this fostering a community?

There are less nuclear options out there, made by others which are great ideas, without removing anything. Making bos consumable, and require consumable resources are amongst the best.

I also think that allowing only one bot to access a chest ist ridiculous, as four inserters can access a chest. I would instead propose that bot access to a chest is indeed reduced to 4. If I can go a little further, I'd like to see them come at the box from a different angle, to make the shapes the bots make less of a straight line, but more "organic", but that is looks, not substance.

[jvandillen]

Myself I never use bots. But would it not be more interesting if bots where constraint to one robotport and its own network. You would be able to determine the size of the network manually with the maximum decided by technology.

[daemonworks]

<sarcasm> Clearly, if anything, what needs to go is blueprint storage and sharing. I mean, if /anything/ is making the game too easy, uncreative and whatnot, it's the ability to simply plunk down whatever layout was put together by somebody who bothered to learn the system. Clearly there's no fun to be had in using a smelting, circuit production or whatever layout made by somebody else! It makes things too easy and ruins the game, and results in cookie-cutter bases! </sarcasm>

Seriously, the bots are fine. Nobody's pointing a gun to anyone's head making you use them, any more than somebody's pointing a gun to anyone's head making them use somebody else's blueprints.

[Yinan]

Myself I never use bots. But would it not be more interesting if bots where constraint to one robotport and its own network. You would be able to determine the size of the network manually with the maximum decided by technology.

Why would it? You just admit that you never use bots. So why are you arguing for restraining the very thing that you don't use for players that do use it?
Seriously, I don't get you people that want to nerf bots when you yourself don't use them. Why must you hinder the fun for other people just because they don't play like you?

[jvandillen]

Myself I never use bots. But would it not be more interesting if bots where constraint to one robotport and its own network. You would be able to determine the size of the network manually with the maximum decided by technology.

Why would it? You just admit that you never use bots. So why are you arguing for restraining the very thing that you don't use for players that do use it?
Seriously, I don't get you people that want to nerf bots when you yourself don't use them. Why must you hinder the fun for other people just because they don't play like you?

I am definitely not adversing for removing the bots and I am actually quite happy with the status quo.
I like the fact that in factorio if you do not like something simply do not use it.

Thus let place it differently. I would like a new feature: the capability to isolate robotports from other ones.

[Pascali]

I think that there is only one Way to get both factions satisfied: Make it an Option play robotfree Games.

but thats a lot of more programming-time. You have to make 2 games, if you wan´t both to be funny. Let the fun win vs. the bots. Bots is like bying a keyboard an pressing the autoplay button. Or you can gat to another planet where completely other fabrics are. You can deal some things between the two planets. And on the other planet there are bots availaible. But there you have maybe 10% playing time. And 90% on the fun(=belt)planet.

Noone has startet playing factorio because of the bots do all thing alone...