Friday Facts #225 - Bots versus belts (part 2)

[Durabys]

How about a tier 4 belt that has a huge throughput but is too fast for inserters ? It can be specifically used for transporting bulk materials over main bus, then you split it into slower belts for production?

Loaders would still be able to saturate even such a belt. But Loaders would need a nerf. E.g. They can only load/unload from chest types (chests, wagons) and furnace-types. Not assemblers and power structures like boilers.

[hitzu]

I'm not sure how would you make belts more space efficient other than totally getting rid of inserters in belt setups and embedding a direct loading from the belts into containers and assemblers.

The devs proposed upgrading all belts with one research into stack belts as the ir final option that they like the most.

This would affect throughput. Space efficiency is how much space you need to build a production setup. Now the more complex the recipe the more complex and large the belt setup is (unlike the bot based setup which is always 2 chests, two inserters, and a power pole per an assembler). Beacons make this even worse.

I posted this on Space Efficiency underneath that:

Space Efficiency? You should count the neccessary space and resources to build needed Solar Fields and Accumulators or Nuclear Mining&Refining&PowerGeneration into the such an equation too. That is easy to change too. One or multiple of these options: Increase power demand. Nerf the Kovarex Process. Make Solar panels much bigger and requiring more space per panel. Or you could also decrease the roboport logistic radius like you proposed (which hamper Beacon Set Ups). Or increase the size of the robotport.

Solar fields don't directly affect local setups. They are far away where is plenty of space. What really restricts the space is beacon coverage area, inserters reach range and the assemblers perimeter length. All of this has nothing to do with solar setups 'cause you are totally free where to put them relatively to the production setup. Solar field size has a very little impact on these setups and even if you play with biters the more the field the more efficiently you can defend it due to the fact that the amount of laser turrets grows linearly but the amount of sol/accues grows quadratically.

[Caine]

posted this on Space Efficiency underneath that:

Space Efficiency? You should count the neccessary space and resources to build needed Solar Fields and Accumulators or Nuclear Mining&Refining&PowerGeneration into the such an equation too. That is easy to change too. One or multiple of these options: Increase power demand. Nerf the Kovarex Process. Make Solar panels much bigger and requiring more space per panel. Or you could also decrease the roboport logistic radius like you proposed (which hamper Beacon Set Ups). Or increase the size of the robotport.

So basically we are talking about a power usage/upkeep cost metric. Belts are infinitely better here and trains or quite good too. Bots are rather costly in comparison.
The difficulty is that these costs change (decrease) over time. E.g. the cost of a bot in the early game is super high in comparison to its cost when you have a megabase. The table oversimplifies (all models do to an extend), which makes these matters a tad tricky. Solving a twenty-dimensional problem is also too hard to tackle.

The reason that I did not include the power creation in space efficiency is because they can be spatially separated. I.e. the presence of a large solar field does not prevent an optimal beacon design with bots while the presence of belts or rails next to assemblers impacts it much more.

Could stack belt be a solution for late game belts?

Very interesting idea.

[Xecutor]

Pneumatic tubes for the win:

[Escadin]

"5+ times stronger than anything else"

I even believe that robot-only Factory should be possible and not useless, I just don't believe, that they should be 5+ times stronger than anything else.

Considering in your test setup the bots not only have higher tech / progression as you say but are also easily more than 5 times more expensive and have high maintenance costs vs NONE of belts, I do not see why this is a problem. And as you said yourself: It's not "anything else" - Trains are fine.
Could you explain how this adds up in your point of view?



"Why use anything but bots?"

With bots, there is no reason to think about other types of transport (Cable cars or automated vehicles for example), because why would anyone use it when you have bots?

I frequently use bots during late game, yet I have plenty of reasons to want more types of transports.
You say bots are inheritly balanced but that's only true if you mean lane balancing / compressed belts. The only method to control balance of distribution across final destinations is through backpressure - which in a bot net means you have to massively overproduce your demand. Anyone who has tried loading trains with bots know the issue. Not to mention all the problems with large networks.

I would love to have

• a pin-point transport method which doesn't obstruct building space. Something to let me carry 1 item across the factory to one exact destination without having to rearrange space, belts or the speration of my logistic / construction networks. Can be low through-put in return.
• something other than mass-transport trains for irregular outpost resupplying far out in the fields.
• a one-time delivery system for uncolonized areas (like supply drop pods)

None of these tasks can be handled by bots, let alone better than belts.



"How to improve belts?"

The question of belt throughput improvements

How about you support the main bus pattern with higher throughput? It's the main reason I switch to bots later. I don't fancy building and wasting space on belt lanes 30+ tiles wide. Not to mention how crowded the connection between furnace area and start of bus gets with more than just several houndred fournaces.

Solution: Add crontrolled access highway belts to support main bus throughput
Add some special extra wide, extra high capacity belts which are not meant for feeding assemblers, but rather feed the regular belts leading to assemblers all across your factory. Think of it like controlled-access highways connected to regular streets via interchanges:

[Jürgen Erhard]

The solution to bots is NOT nerfing bots. It is nerfing chest access speed/volume and at the same time also allowing inserters faster ouput out of chests. Either through side inserters or through research upgrades increasing their maximum overall (not just for stack inserters) items per arm swing. Both would have to be implmented: Nerfing chests and boosting inserters. Not nerfing robots.

"nerfing chest access speed/volume". For bots only, because you want to boost speed for inserters (and I can't recall any other means to take things out of or put into chests than bots or inserters).

Which means: nerf bots.

[Caine]

Twinsen certainly trolled the community with success. Well played

This person don't see any problem in this. Tbh Caine please use your brain, you don't see what Twinsen was doing? He played like lil children with the emotions of other.

I use my brain too much and struggle with experiencing emotions as intensively as you seem to do. That is a problem I have to deal with. To imply that I do not use my brain is an insult to me and a worse one than what Twinsen did, I am enjoying his thought experiment. We are different, that is fine. You are rude, that is not. At least not in my opinion.

You are free to do what you want but, based on your words towards Koub, I strongly advise you to reconsider your tone. Unless you are trolling yourself, you will not achieve your intended goal with it. If you are then you have no right to accuse Twinsen.

[wlfbck]

Just another post to finally put loaders into vanilla. Loaders make belts immediately better. All the inserter setups always suffer with continuuity problems.

And something you forgot in the FFF: Bots, especially if massively used, are terrible for seeing at what state your factory is. Just look at your own picture here. I have no idea if those bots are bringing enough copper, if anything else under there is starved for ressources or if anything meaningful is happening there at all.

[vampiricdust]

Logistic bots already cost more than they add compared to belts. Any further nerf without a cost reduction would further make them cost a ridiculous amount for what they add. You could make 25k blue belts just from the research cost difference alone. Each bot costs 3.4 times as much as each additional belt and roboports are another 13 some blue belts ea h. Using bots now is already 35% less cost efficent by Kovarex's throughput numbers just based on research costs.

Ressource cost just isn't the right measure, because of automation: As a player it doesn't matter how much it costs, because the factory will pay for me. What matters is time cost, because I as a player have to pay that myself, and time cost of bots is currently way less than for belts.

And yeah, I've mentioned it before and lots of others wrote the same thing, but just to make sure the devs see it:
Limiting bots per second that can access chests is the ultimate solution.

EDIT: I like Cains metrics, and limiting bots per second that cannaccess cehsts would effectively lead to the second table he posted.

So make everything cost nothing then and remove bots, cause who cares?

[Avezo]

Although I like belts more than bots, after reading lot of suggestions about nerfs, now I think I would rather have devs just leave bots alone... I'm scared.

[PurpleGreen]

when i read stacked belts ... i actually thought about something like this

this would be fantastic

[PurpleGreen]

The filter-splitter buff is amazing and I love it. Can't wait to see it in the next update!

This opens up many new ways to use belts not just within factories, but on mixed mining fields too.

Love the changes to splitters especially the priority functions. TBH I'd rather have a filter on electric mining drills than splitters though.

A tree of such filter-splitters could be used to instantly sort that which has otherwise required large convoluted filter-arm setups prone to backing-up if a particular resource is not consumed quickly enough, or "Black Magic" exploits that take unintended advantage of the inherent "splitting logic" of regular splitters.
If this future splitter buff could send and read from the circuit network I see a future explosion of designs as we attempt to stretch the limits of what this new toy can do.
I even see mixed-item belts becoming a viable option once again as the ability to quickly sort objects into and out of a belt becomes a simple affair. Even filter inserters do not lose usability to this as they fulfull the purpose of being both an inserter that can filter.

...

I realize this is a reversal of my previous comment in #244, but reading 45 pages of comments did open my mind a little to better opinions.
To reference Extra Credit's video on Power Creep, sure we can buff belts to make them a more competitive choice but it will not solve another potential looming issue:

Let us take kovarex's little experiment and imagine a world where both results were the same.
Game's balanced now, right? Belts and bots are equally powerful and its up to the player to use what they want? Perfect world and everyone happy? I doubt it, because from this point onwards, every subsequent change or update or addition to the game would require a finely tuned finger to not disturb this delicate balance. You've also created a much deeper rift between Bot, and Belt players as they simply find smaller anomalies in the other's gameplay to bicker about. In the end, both options become more alike when in fact they should become more radically different, and not in the strict numerical "I am better than you in everything" sense.
Side note: Yes, I get the fact that Factorio is at its heart a "resource manipulation & transport" game, so being able to move more items at a faster pace is a core element not to be ignored.

Perhaps this is not about a war of "what provides the best throughput", but rather "what advantages and solutions do belts provide that a bot cannot", and of course vice-versa. For/Against playstyle-arguments aside, it's what always made Factorio fun whether you're a Timmy or Spike. What the devs are trying to achieve is an equilibrium of incomparables where the solution for a given issue is based on what you could build, not "should" or "must".

It is worth nothing that Bots feel cheaty because they fulfill the purpose of belt-, inserter-, and filter-in-one with the scale of this ability being only limited by the size of a given swarm.
But give a new usability dimension to ground-based logistics and suddenly everyone salivates like Pavlov's dogs ^^

Some of us drive from A to B and others just want to drive around. Those same people may want to do it quickly or slowly. In the end we all still use the same road.

and this is exactly why i am for a complete overhaul of the bot system. viewtopic.php?f=6&t=56549
i think whats missing is the rock paper scissor situation when it comes to the design of the base...
right now you could do a belt only , a train only or a bot only base. so but mixing the systems i find myself really thinking about how to fit these belts or trains ... bots dont really give a challenge. i dont have to plan bots around the other 2 system, i just plop it in and it works. if bots would require you to integrate the design into the thought process of the other systems it would be much more challenging.

[Stevepunk]

Just make a faster belt upgrade. Problem solved.

Adding a new tier of super-fast belt or belt speed research
Inserters would either need a speed increase as well..

No, they don't need a speed increase. The advantage of belts over chests is that multiple inserters can add to and remove from belts. With chests there is a maximum of 4 inserters per chest. Belts can have an infinite amount of inserters.

Not sure how this isn't immediately clear but sometimes in gamedev you get so close to your product you can't see the forest for the trees so perhaps this is the case here.

So bots can keep their current speed and act the same.
Speed 4 belts will be optimal for moving items over short distances but require more complex layouts (as is the fun of belts), to optimise the belt (enough additions and removals that the belt is compressed but never stationary).

Anyway, that seems like the simplest solution to me.
But I've always liked to use a combination of belts, bots and trains.

[Blackheart81]

With the new splitter features I think it would be a good idea to have them locked behind research instead of having them available from the start, especially the filtering. Having full belt filtering available as soon as you build your first splitter is very powerful and almost negates the reason to research filter inserters. I play a lot of Bobs/Angels and one of the early challenges is how to sort without filter insertes and I often make that an early priority for research. But if the splitters can do it from the get go it removes that challenge.

As for the belt improvements my vote would be for paletizing/containerizing items. This would add another step in the process that I would have to take into account for instead of just researching a faster belt or something like that. Adding containers just like barrels adds complexity and and additional requirements while increasing the throughput. It could be a little to powerful so you'd need some limits. Here's the limits I would consider
1. Have to create the containers before being loaded, just like barrels.
2. Limit stack size, maybe to 1.
3. Increase number of inventory slots they take. Instead of 1 slot, maybe they take 4 to account for the huge number of items in it. Or maybe make it so the player can't carry them....theyre just too big?
4. Have them require a special robot to carry them so they don't just become a buff for robots. Maybe a robot that take 4 logigistic robots and some other ingredients to produce. The 4 logistics robots are needed because it a big, heavy object. Once built, these new "heavy lift logistics bots" can only carry 1 cargo box and move somewhat slow compared to regular logistics bots. This would make belting these containers equally sd appealing because we could fit more of them in a similar amount of space and they would move faster but still have to travel along the fixed route.

[TiMatic]

Ignoring specific solutions for the moment, what do we want this table to look like? As an example take the following table which is nicely balanced, each solution is strong in one metric, weak in another and average in the remaining.

Now we are in a much better position to evaluate individual changes to belts and/or bots. Do they take us in the right direction?

I want to pose some questions to the community

• Are these the proper metrics? If not, which ones should add/remove and why?
• What do we want the table to look like?

I really would like it, if the transport systems would be balanced like that.
Then logictics bots could become again a mid game feature instead of an end game feature like it is now.

[jodokus31]

I want to share my experience with logistic bots (and modules), after reaching the tech. I was suddenly able to create those really strong assembler lines with requester/provider chests and it took a lot of belt challenge, which was really helpful at the first glance. Now I realized, that all my poor belt designs are just too weak and I want to replace all of it. In the middle of it, I realized how boring it went, despite the fact, that the overall throughput started to grow. (This felt for me like a trap and reduced my enjoyment)
I took me some time to figure out, that this is not my playstyle, I want to go in a big scale. After that, it took me also some time, that for mall like structures to create machines and stuff, logistic bots are the most reasonable way, because belt routing for those structures, with a very low throughput and very diverse inputs are just not feasible. (especially in Bobs and Angels). I also learned, that I really like the challenge of a train network, esp. with LTN-Mod, which can be quite a interesting substitute for bot logistics
This is quite a steep learning curve for me, how to find out, what's my playstyle. I think, we shouldn't take away these kind of experiences in such a complex game like factorio. It is quite indiviual, which kind of challenge is enjoyable.

That said, I now, there a many, many other opinions about playstyle and factorio has the big task to give all the possibilites, which went quite successfull in the past, i think.
I don't think, that the bots should be changed, since many people like them, how they are. For me, it doesn't matter, because I made my decisions, how i want to play. I also might change my mind.

But I think, there should be an "official" game mode, which is allowed to rebalance bots, similar to "marathon". (With an explanation, what can be a "trap" with too powerful bots.)
The invidual changes could be changed in map settings, just like f.e. tech multi
The community can and certainly will try it out and give feedback.

You could also do a voting after a half year trial, which mode should be the default mode and leave the current mode still available.

[ajm.nethunt]

Problem: Whereas bots don't take up space on the ground, belts do - therefore less throughput-per-area.
Proposal: allow building belts on top of other belts.
Descriptive advantages:
1. Takes 1 new sprite, a covered 'belt tunnel' (think like a server rack).
2. Clicking on a 'belt tunnel' square brings up a new UI showing however many layers of belts it has, and what items are moving across them.
3. power capped by research item limiting number of belts stacked.
4. less lines of belts, less spaghetti belts, inserters can simply interact with a 'belt tunnel' exactly like it is a chest.

Summary: Allows bots to be unchanged, but solves the real underlying problem of belts, which is throughput-per-area/spaghetti belts.

[DerGraue]

Dear Devs,

whatever you guys do, please keep the most important thing in mind: fun

The game is supposed to be fun, the more fun, the better.

Removing things is in most cases no fun. Making existing bases stop working, reducing the maximum SPM you can reach with your PC, or lower the UPS of said bases, is no fun at all.

Buffing on the other side is fun in most cases and there are certainly ways to buff belts to make them more UPS or throughput efficient.

Just keep it in mind, please: fun

[POPISowyNumer]

I have run a little tiny experiment of my own, and you know what.

INSERTER CHAINS have around 72% of throughput of BLUE BELTS
GOD DAMNED INSERTER CHAINS
THE OLD CARGO WAGON TRICK HAS 150% THROUGHPUT OF TWO BLUE BELTS FOR ADDED BONUS OF NOT HAVING TO FUG AROUND THE COMPRESSION AND SHIT

ITS NOT THE BOTS THAT ARE OP
ITS BELTS THAT ARE ATROCIOUSLY UNDERPOWERED

BELTS ARE WEAK BECAUSE YOU, DEVS, MADE THEM WEAK

[KriKri]

If you think about a extension with other planets, you could say, that bots need air to fly and on the other planets is no air so you can´t use them, that´s at least what i thought while reading the friday facts.

It´s generally better to encourage the behaivor that you want by buffing belts and not discourage the use of bots by nerfing them, that would make many people very angry.

I guess many players would use the belts more often, if you make them a bit faster and add the loader research to the normal game.