Friday Facts #224 - Bots versus belts

[GenBOOM]

hahaha



http://lapsedhistorian.com/get-blower-l ... ing-tubes/

this is actually a good tech alternative that requires steam power and solves the problem of ammo, repair packs, and other small orders.

[Industriosity]

The real problem with bots is that they can carry anything.

A great solution would be to consider objects weight, so one single bot could not carry a nuclear reactor, but could carry modules with easy.

Maybe requiring many bots to carry a heavy object would do, or, they would fly slowly, or even could not carry some items.

In my factories i use belts for primitive tech items and bots for expensive items at the end of chain of production.

[ske]

Make it easy to update belts from yellow to red, and from red to blue.

This could be done by not upgrading the belts themselves but by upgrading the belt motor units. An entity that doesn't exist currently but would make total sense. In order to make belts move you place a motor beside them and power it with coal/electricity. Every connected belt starts to move. (Limits may apply.) If you want faster belts, you just put in faster motor units.

[Stlyau]

After reading/skimming most of the comments over the last 2 days, I decided to take a look at all the items that have been brought up and some that haven't. Here's a list:

Technologies:
⦁ Robotics - Effect: Allows Flying Robot Frame. Science Packs: 200 Red, Green
⦁ Construction Bots - Effect: Allows Roboport, Passive Provider, Storage, ConBot. Required Tech: Robotics, Flight. Science Packs: 100 Red, Green
⦁ Logistic Bots - Effect: Allows Roboport, Passive Provider, Storage, LogBot. Required Tech: Robotics, Flight. Science Packs: 150 Red, Green
⦁ Worker Speed - Increases flying speed of LogBots and ConBots. Required Tech: Robotics.
Effect: WS1 +35% Science Packs: 50 Red, Green, Blue
WS2 +40% Science Packs: 100 Red, Green, Blue
WS3 +45% Science Packs: 150 Red, Green, Blue, Purple
WS4 +55% Science Packs: 250 Red, Green, Blue, Purple
WS5 +65% Science Packs: 500 Red, Green, Blue, Purple, Yellow
WS6+ +65% Science Packs: Infinite Research scaling up
⦁ Logistic System - Effect: Allows Active Provider, Requester and Buffer. Required Tech: Logistic Bots, Science Packs: 150 Red, Green, Blue, Purple, Yellow
⦁ Character Logistic Slots - Allows request items from logistic network, items delivered by LogBot. Effect: Logistic Slots +6. Required Tech: Logistic Bots. Science Packs: 100 Red, Green
⦁ Character Logistic Trash Slots - Allows set limits on inventory, excess sent to trash. Effect: Unlocks logistic trash slots. Required Tech: Logistic Bots. Science Packs: 100 Red, Green
⦁ Inserter Capacity Bonus - Allows more items to be moved at once. Science Packs: Starts at 200 Red, Green for Bonus 1, increasing to 600 Red, Green, Blue, Purple, Yellow for Bonus 7

Noticeable issues:
⦁ Roboport, Passive Provider and Storage are under both ConBot and LogBot Techs.
⦁ Roboport should be moved to Robotics.
⦁ Storage should be moved to Logistic System or Character Trash Slots.
⦁ Logistics System should be required for Character Slots or Trash Slots (how can you request pickup/removal without known access to Requester Chest?).
⦁ Worker Speed should require Flight, ConBot, LogBot Techs. (As it states in the description all 3 techs). It is also a stackable bonus.
⦁ Logistic System requires LogBots Tech, but Buffer is designed to be used by ConBots for auto-construction as well. This means Buffer needs to be move or LogSystem needs to be have requirement of both LogBot and ConBot Techs. LogSystem is currently locked behind Purple and Yellow Science.
⦁ Stack bonus demands the need for more items supplied via belt, bot or production. This contributes to the rift between belt speed/compression vs. bot usage (after upgrades) due to no researchable upgrades in belt speed bonus like Worker Speed.

Now let's look at the Beacon/Roboport issues:
⦁ Roboport - 7 robot storage slots (robots stack is 50 currently).
⦁ Beacon - 9x9 coverage with middle 3x3 taken up by Beacon.
This allows 350 LogBots/ConBots to be stored in ONE Roboport. Roboports only charge 4 bots at a time requiring more Roboports to be placed down exponentially based on LogBots used to keep up with charging. Solution: Reduce Robot Stack size or storage slots, increase charging ports.
Beacon's small coverage (for it's 50% per module effect) doesn't allow belts to be used as intended. Solution: Increase coverage radius, reduce effect percentage.

With Speed Modules being the prodominate Beacon choice, this contributes to the "belt vs. bot" rift too. As it's making production machines faster, putting out more items that the belts just can't provide throughput compared to the research upgraded bots. Solution: Limit number of effect sources an entity can receive based on tier. If over the limit, then no further bonus received.

I personally don't like using logbots for self supply. At times, you want more of a few items while outside the Network. If you have both Player Logistics and Trash setup to manage both the wanted quantity and disposing of excess, when you wanna carry more you now have to go change for the short term, in BOTH tabs. The other downside to PlayLog is when you have an unfortunate train moment and on respawn, while trying to run out of spawn Logi-network (with no bonus movement) you're bombarded with bots supplying you with all the items that you already had enough of, but are now on that roadkill lying on the tracks.

[Zeblote]

Make it easy to update belts from yellow to red, and from red to blue.

This could be done by not upgrading the belts themselves but by upgrading the belt motor units. An entity that doesn't exist currently but would make total sense. In order to make belts move you place a motor beside them and power it with coal/electricity. Every connected belt starts to move. (Limits may apply.) If you want faster belts, you just put in faster motor units.

> in topic about bots being far too OP compared to belts
> suggests making belts more annoying to use

[GenBOOM]

I think he meant get rid of different belt colors and only have 1 belt type that requires a motor to increase the speed of the belt.

this actually would simplify belts quite a lot and make upgrading them much easier

instead of unlocking new belts in the tech tree, you unlock more powerful steam/electric motors with higher speeds
so the trade off is that you have to power belts now, but you never have to replace them and motors can be an infinite tech to increase belt speeds more and more throughout the game.

this would mean bots are there for automated expansion only, which is what most people use it for, especially in single player bots are quite useful

[Caine]

I am not sure you are getting the point of this thread, which is to discuss what you do or do not like about this thought.

Sure, I get that. I merely question the effectiveness of the approach. A general rule of design is to ask your users for problems not for solutions. The majority of the discussion is on how to nerf bots or how to buff belts.

A much more interesting starting point would be "Why are you using bots instead of belts?". That way you can collect scenarios which opens up a design space for solutions. On the other hand, throwing a bomb into the crowd and watch what happens can be more amusing, I suppose. Don't misunderstand me, the information is more or less here, but it will be a pain to distill. This thread is growing quickly with no end in sight. That said, this meta-discussion is not helping either as people do not read anything in the thread anyway.

[DarkyPupu]

I totally agree bots are way too overpowered in many circumstances, however i wouldn't nerf them as it would indeed just bring frustration. The main limitation of belts currently (without mods) are due to their lack of speed and / or vanilla loaders. When you play with mods fixing that, the bots are not so much as a must have but they stay useful. In general i think that belts, inserters, and even trains at some degree totally lack speed in late game.

Add more science levels to improve that without nerfing anything IMO

[Keba]

+1 for “constructions bots should be available waaaaay sooner than logistic bots”.

Is belt braiding considered a feature or a bug? In any case: Please do not remove that or else it won‘t be hard but impossible to construct good beacons-belts setups.

[madpinger]

I think its fine as it is. Simply put, The player can decide which way to go.

I'm doing a mostly only belt run atm vs my usual bot centric factories and I'm finding the thru-put of belts to be so limiting and a problem. Solving it has been entertainment to a point. However, My base has grown to the point in that process that It'll be difficult to continue the map due to FPS/UPS on my hardware. The 0.16.5 update did improve my FPS/UPS a bit letting me go on a little more.

[ske]

I think he meant get rid of different belt colors and only have 1 belt type that requires a motor to increase the speed of the belt.

Yes

this actually would simplify belts quite a lot and make upgrading them much easier

Yes

instead of unlocking new belts in the tech tree, you unlock more powerful steam/electric motors with higher speeds
so the trade off is that you have to power belts now, but you never have to replace them and motors can be an infinite tech to increase belt speeds more and more throughout the game.

Maybe. At least it would make a lot more sense physically and it'd be easier to adjust.

this would mean bots are there for automated expansion only, which is what most people use it for, especially in single player bots are quite useful

Maybe.

Logistic bots would still be very useful to supply very low-frequency items such as lubricant and tools for assembly machines. Things that don't exist right now. Logistic bots could be limited by item weight instead of purely item count. This would make sense. You wouldn't use a drone to carry ore or four trains in the real world. But you would use a drone to transport some sharp carbide tooling to the respective assembly machine. Those tools could partially replace modules and go into the module slots (or tool slots) but slowly use up over time and need replacement. It's logical also that a carbide end mill cuts quicker than a HSS end mill.

[GenBOOM]

well just like anything else that requires power, you can turn it off when its not being used.
once you unlock wires you can connect it to a circuit network and manage the power usage. a lot of belts are not utilized 100% of the time.
this makes a lot of sense to me.

a motor at the start of the belt that supports a length of X amount of tiles before requiring another motor to increase the length of belts affected.
alternatively you could think about it like a signal repeater requiring a motor every 100 tiles of belt or something

[Stlyau]

+1 for “constructions bots should be available waaaaay sooner than logistic bots”.

So by this statement, you're gonna run back and forth to supply train or factory every time you want to build new things in early game with conbots? The logbots are what is supplying YOU (the requester/buffer chest) with the items those conbots are using to construct said items you've ghosted/blueprinted.

[Yinan]

This could be done by not upgrading the belts themselves but by upgrading the belt motor units. An entity that doesn't exist currently but would make total sense. In order to make belts move you place a motor beside them and power it with coal/electricity. Every connected belt starts to move. (Limits may apply.) If you want faster belts, you just put in faster motor units.

Powered belts (which are a completely different thing) are fortunately never going to be a thing. They have been discussed before and have been completely rejected by the devs.

[_Peter_]

Make it an intended feature to make designs where belts contain two item on one lane in a certain ratio. This allows to have one belt delivering four types of items to a factory. Why is this fair compared to bots? Because requester chests can deliver more than four (eight, or is it sixteen?) independent items to a factory, in a free ratio, including optional buffering ability (very powerful! maybe this could be nerfed in some way, by reducing the number of item types that can be requested at the same time to 2?).

It is already possible, but you need to have some circuit network skills for it. I found this video a while back demonstrating it: https://www.youtube.com/watch?v=8Lp-UJYwclA

Similarly I have been trying to implement belt provider/requester mechanics myself, again all circuit networks. It would certainly be interesting to have this aspect of the game highlighted more. However, it does not provide a solution for throughput limitations when megabasing.

It is possible, I know, but I suggest to make it a more accessible feature, because I think it makes more elegant designs possible. I think a throughput solution for megabases would be to have a research that increases belt speed. And more elegant belt designs would make megabases more elegant...

[Proxy]

never noticed how OP bots are.

but then again, they are a choice, you don't need to use them.

also thing i noticed,
Trains are long-range versions of belts because they transport items on a set path, but trians also need stops where items get loaded/unloaded from.
so where is the long-ranged version of bots? some quadcopter that transfer items from/to a set station in a straight line?

lastly, how about some maintenance feature?
I mean it would be interesting if everything could break down with a chance depending on it's complexicty and how new it is and requires a random amount of it's crafting resources to be repaired.
but in this case i just mean the bots, what if randomly after some time a bot would break down and would need to be repaired/deconsctructed with some of it's parts in either an assembler or some other machine


a lot of people would hate having to maintain their factories tho....

[bobucles]

Beacons are partially at fault for why bots are so much more important than belts in a late game base. The reason players NEED such insane item throughput is because a maxed out assembler can craft out items at upwards of 8 crafting speed or more. This is accomplished with a high density 8x8 beacon build that only has viable space for 2 long belts. Suffice to say that is nowhere NEAR enough belts to get much done. On the other hand that same beacon build can fit a near unlimited capacity of logistic bots and the chests to use them. The endgame kiloscience base is designed around these incredibly high production beacon bases, and bots are the only viable way to feed them.

Tiny chests are another factor here. If logistic storage chests were 2x2 or 3x3, the super 8x8 beacon build breaks and the NEED for high throughput bots becomes less important. However the only way to make a super RPM base is to achieve these insane 5+ crafting speed assemblers so that entities and CPU cycles can be saved. It's no easy problem to figure out.

I do worry that a new type of "Super belt" has the potential to compete with train networks. Trains move a lot of items over a long distance but the land footprint they require is simply huge. A super belt also moves a lot of items, but the footprint has to be small to funnel items through a base. That gives them a dangerous edge over trains straight up. It'd take some real blundery to make any kind of belt or super belt better than a full speed train.

[Engimage]

Reading another dozen of pages here are some more comments from me.

So the main problem with bots is their iteraction with chests. They have just unlimited throughput. So something should be done to nerf this particular side of them. Limiting chest iteraction to make queues like roboport charging pads have is a nice thing to think about.

Belts are not scalable as they have exactly fixed parameters and throughput. Bots are just like magic wand. You need to limit them somehow to strict throughput to force players scale up builds respectively.

Belts do iteract with stuff via inserters which are hard limit for throughput. Bots can go and just swarm a chest and instansly transfer its contents somewhere.

In current implementation I also think that infinite research for robot speed should not apply to logistic robots (as construction robots have no in-game alternative and are just QoL feature that does not effect gameplay). However if you introduce rate limits this might become less relevant.

I do not think that collision boxes for robots will benefit game as they will hit CPU like there is no tomorrow.

So IMO a good way to think is how to rate limit robots and how to possibly increase throughput for belts. I do find boxing (palletizing, stacking whatever, similar to barreling) being the best solution to increase belt throughput dramatically. Not magically but through a special device (even if assembler).

If bots could be rate limited happiness would be instant. Bots are great to have as a feature but they are waaaaay to strong in every possible aspect. If limited they will still remain an awesome feature and will help majority of players with their laziness but magic swarm must go.

[Noiser]

It's like a car with a manual and an automatic mode (manual = belt, automatic = bots)
EDIT: Not the transmission! I mean a car that can drive on it's own.
Maybe it is/can be more fun to drive manually, but maybe you just want to reach your goal in an easy way and focus on other stuff.
It's always nice to have a choice.

I use logistic robots for moving rarely used items around (Train fuel, barrels, nuclear cells) or for moving Items that are needed occasionally (for the production of satellites, assembling machines, pump jacks, ...) because i don't want to lay an extra belt through my whole factory for such rarely used items. That's neither more fun nor neat!

My Opinion: Please leave the bots as they are. They are useful and admittedly ,if you use a lot of them, overpowered. But everyone can decide for himself what's fun. People like to have choices. (and don't place their science condition behind the first rocket launch, that would break the mechanic to use them as Rare-Item-Transporter in mid game, while still allowing mega-bases). A belt buff would be a better idea.


Greeting from germany

[TheYeast]

In the real world, belts would beat flying bots because of the insane power required for lifting and carrying heavy loads through the air. Greatly increasing power consumption and factoring in weight would make the belt vs. bot trade-off more fair. Power efficient yet complex design vs. simple design that requires an additional power plant and much more fuel. Fuel scarcity would become a major factor. For large scale application, the balance should always tip towards belts.

Use of the logistic network should then be avoided where possible to save power. This would make blueprint building a bit problematic if materials need to be brought in over a larger distance. To alleviate this, construction bots should be able to (and prioritise) picking up items from the trunk of a car, and return them there when deconstructing. That way a car placed near a construction site would minimise bot travel distance. Furthermore, a car placed next to an inserter should be restocked/emptied of junk. You can also consider introducing a larger cargo truck with some robot slots into the game for this purpose.

I think this would actually improve the game, because there would be renewed value in being present at a construction site, additional challenge of building loading bays for your cars/trucks, supplying these, etc.

To move low throughput, large distance items through your base, a train with filtered item slots would be a power efficient alternative to logistic bots, and more fun to build.

For me, logistic bots as they are now make too many fun design challenges obsolete...