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

[bobucles]

Yes, but with iron plates, an object that looks mostly flat. I bet it wouldn't look so neat and clear with items that aren't square and flat, like ores, batteries or science flasks.

Even if it does look silly, when was the last time you needed a full belt's throughput of batteries or potions? Those are very valuable items and stacking them will get very expensive very quickly. It's only natural that players would find no value or need in stacking them. It's one of the reasons I think normal inserters should not have stacking output, partially because there are so many places you will never need stacking power and partially to make sure the stack inserter stays worth the expense.

(I have no strong opinion on allowing inserters to grab stack INPUTS. I think it's safer if they can grab items since the real difficulty is stacking items on the belt in the first place. Also if the player drops items as a stack you don't want that to immediately break your base.)

There's probably a UPS benefit to separating the options as well. A normal inserter only has to look for a viable horizontal space on the belt to place an item. That's one CPU test. An item stacking inserter has to find a horizontal space, or it can find a matching item that can go taller. That's 3 CPU tests. I don't want to speculate on how much CPU time it costs to play with a stack, but for a hyperbase every little bit counts I guess? I don't know much more than that.

With an item stacking belt system the item stacking tech would be unlocked somewhere with stack inserter research. For speculative numbers the stack inserter would peak at 12 item capacity and then 4 or 6 belt stacking capacity (3.0 or 2.0 belt unloads; upgraded inserters drop 3 single items). Super stack Loader research would be FAR beyond that, somewhere near end game tech or post game tech since you don't really need it any earlier than that.

[GenBOOM]

layered belts continued...
this gif

+ layered belts =




this is all you would really need to do because visually the things on belts hang off a little bit, so visually indicating if the side of the belt is full or not can be done simply like this for the hidden belt underneath
so there is plenty of space to do layered belts as long as the structure has a built in indicator for each space on the belt
you just need to scale the existing assets and create a better looking structure to hold up the belt, and might have to tweak the inserter animation slightly so that it does not extend as far and places things sooner

obviously inserters cannot place things on the belt underneath.

this is where a new sorting machine could be introduced that sorts belts vertically.
also a new way to merge belts together is needed

you can take this even further and make the bottom belt be completly indicated via lights instead, giving the artist more room to work with.
again these would have to be powered, but it would be optional to have the structure have a light indicator or not as its not really needed, but visually is nice to have

[deef0000dragon1]

layered belts continued...
this gif

+ layered belts =




this is all you would really need to do because visually the things on belts hang off a little bit, so there is plenty of space to do layered belts
you just need to scale the existing assets and create a better looking structure to hold up the belt

Or as has been mentioned in previous posts, have the belts go underground. Now you only have to have one belt showing with a new texture instead of having to have n layers of belts showing above ground.

[GenBOOM]

Or as has been mentioned in previous posts, have the belts go underground. Now you only have to have one belt showing with a new texture instead of having to have n layers of belts showing above ground.

you miss the entire point of layers is so that you can visually see that more is going on and be able to add and remove things from it and split and merge off of this new belt type
unlike undergrounds that may as well be in another universe because you can't interact with the belt while its underground.

[vampiricdust]

Increasing cost is equivalent to decreasing output per resource.

Sounds like, but indeed this isn't the case with the robots: You can see the costs as an investment into the future to build more bots and increase production with more robots. Which decreases the costs of the bots.

[Moderated by Koub]. Logistic bots do not decrease the cost of anything. You're outright lying or being manipulative to imply that by spending less time building the factory, you can spend more time making mining outposts. In no way shape or form do Logistic Bots make anything cheaper. They take more production steps to make, their parts take longer to make, and they require infrastructure to make them work.

You guys dismiss resources as having no value, but players HAVE TO SPEND TIME TO GET RESOURCES. Without a mod, you have to at least go to the resource patch, use blueprints, build a train station, build the tracks to that station, create a new train or add the station to an existing train, and ensure everything works. My time has value. You guys can argue resources don't matter, but I very much care that I have spend 7 times as much of my time unlocking bots to a decent level versus belts and spend 3.4 times as much time getting resources for those bots as I would blue belts. ALL TO GET BOTS THAT ARE NOT BETTER THAN BELTS. Yeah, you could use 20 times as many bots as belts moving times, but those 20x bots are worth 3x more in belts.

You guys just piss me off more & more with this topic and the bullshit that gets spewed to justify your points of view without a single ounce of honest fact. Bots are only better where you literally cannot fit in more belts than the value of bots.

[Avezo]

If bots get nerfed, I wonder if you gonna re-buff barrels capacity to match up to their current post-nerf status

[Paul17041993]

- Late game belt stacking on the express belts is what I'd go for, essentially like the tier3 assembler (you have to already have most stuff done before you can get them).
- And/or, a way to provide power to blocks of express belts for additional speed, making them more like the coal belts we have in Australia that deliver tonnes of coal per second...

[mushroomxchen]

Sorry for my English.
I am very surprised to see that bots is aimed to be a temporary solution in factorio. From my experience, bots seems to be designed as a late game transportation solution, especially for megabase.
Belts can be acquired much earlier in the game, use no energy, and most important it's capability do not scale.
Logistic bots become practical mostly after first rocket, cost more to make, have upkeep, and scale with numbers and late game technology.

When I design a megabase, I like to make small modules, like smelting module or green chip module, and connecting them with rails. I think many people are doing the same thing.
With bots I have less trouble in trying to achieve balanced unloading, balanced distributing, balanced gathering in one small module, so every smelter / assemble machine is able to work at their paper numbers.
I don't use belt mainly for these reasons
1. It's very difficult to get the balanced numbers I want. It’s very difficult to evenly distribute items to odd numbers. (And I do believe large belt balancer is BAD for most players.)
2. Inserter is slower and unstable while working with belt.
3. Can achieve balance loading/unloading much easier by using logistic network. It’s nature to use bots with logistic chests.
If I can have simple NxM balancer, and inserter can have same (stable) performance like working on chests, then I think belt would get their chances.

Devs hope the best way to play factorio is to combine every method, leave no one behind. I highly doubt it's possible without some major design changes.
Belt and bots are created for one same purpose, transportation, like train.
There are 2 major factor in this category, throughput and distance.
Long range is covered by train, no matter the amount to move.
Belt and bots are both capable in short and mid transportation, so the one has better throughput wins.
Unless some base design is impossible with one method (like space is so critical that belts simply won't work, or some terrain no allow flight), there is no reason to use both.
Limit bots throughput to only short distance could work, but bots-only base will break, and it may not be more fun.
Maybe we can make bot and belt work together (which pulls inserter into this balance mess), but this will be even bigger change in design.

[batorfly]

[quote="bobingabout"]Thumbs up on the filter splitters

Speaking of stack inserters... you've plastered that FFF with pictures of you using loaders. Put loaders in the base game, they'll help buff belts.

I mean, Inserters can be used by Bots, Trains and Belts... but loaders NEED to use belts... if you want to load a train/factory from a belt, 1 loader between the train/factory, and the belt.
You want to do the same thing with bots that feed from a chest, you first need to use a loader to unload the chest onto a belt, then use another to load a train from that belt.

Faster belts, and inserters should be fine, just one more tier with relatable improvements to existing stuff sounds good to me.

Speaking of belts with stacking items.... definitely better than needing to pack or unpack an item, Perhaps consider this... ANY belt can carry a stack of items, but it must be created by a STACK INSERTER!!! As for graphics, well, each item would need to look different to stack well... so don't bother playing with the graphics, I mean, do stack inserters change the graphics when holding a stack?



So summery:
Love the splitter
Stack inserters should be able to add stacks to belts
a higher tier belt and inserters is a good idea
give us base game loaders.[/quote]

ME GUSTA.100% Legit. One thing, already proposed: belt and inserter speed research, not the next tier.

PS fuel bonus should affect burner inserter (is this a thing? I don't think so, don't remember), maybe speed module LOL slot in inserters maybe YOLO, but if modules for inserters than what for belts? I suppose still speed research.

[PadreSperanza]

I think the most useful way to "buff" belts would be to give you the possibility to have "intelligent belts and stuff".

The way belts and splitters work right now is fine. Of course you cannot have the same throughput you could have with bots but you can build a second belt lane for this. I mean the real challenge is to build you base (with the throughput you need) the way you want and the way you can. So some players use 2 x yellow belts while some other thake one red belt instead. This is what Factorio is about. And I think belts are fine.

But you can still improve them a little bit without breaking the way they work.

e.g.

Implement a research for "intelligent splitters". You guys have made splitters where you can priorize left or right side, input or output. That's amazing. So combine this with reasearch. Every level gives acces to a new feature of this. And in aaddition add more features like this: I think about give your splitters the possibility to cahnge splitting ratio. Add fields where you can add a ratio 1:1 by default, 1:4 when you need 4 parts an the one side and 1 on the other side. So you can optimaize and calculate much better for your factory.
Add the feature that splitters can fullfill both sides of a belt's lane. so even when the input is just one-side the output can be both-side in exchange. So you can input a red-splitter with one-side copper and one-side iron and output this to two yellow belts one fullfilled with copper and one with just iron.
Add feature to bigger underground-belts.
etc.

all these features are researched by "intelligent belt-tech". Each level granting much more, much better functions of using belts, while underground -belts and splitters are level 1 and already exist.

These is a research which is independent of fast- and express belts. Because you can resaerch the features without fast- oder express belts or you can research fast belts without you needed to research features too. So this is an additional feature to research to get those features but they are not linked to fast- or express-belt technology. this seems logical because when you have knowledge of variabel ratios then you have even with faster technology.

I don't know how many features are possible to add. But in my opinion the game's funtion of belts are amazing even the amount of different types are good. And I think the game doesnt need much more speed in belts to get something transported. I think the game just needs more features for belts to make more attractive to other. Instead of buffing speed or throughput, increase the features belts can work with to let players have the option to take belts because they easily can. And when there are many features which can make some builds more easy or which are unique I guess much more players would use belts in the late too. So I wouldn't look for optimization or compare bots vs. belts. I would look for unique technology which gives your belts more much freedom and the status to be unique to have their places even in late game.
Add features which only belts have, but the way you need to do something for them and not just for free, and belts will be in factories for evermore.

I would also have ideas for "nerf" bots. But this is not what we all want. We want to have the possibility to play the way we want (like you already said). So we just need much more possibilities in theses ways and features.

But even if you let everything the way it is, there will be many players who will use belts before bots even if this would be to place 4 lanes of belts for just one item to get throughput. And this will be because you guys - oh mighty devs - give us a game where we can do what we want, the way we want as often as we want. It's not the limitation what will restrict players but the player's imagination and fantasy.

[Pascali]

layerd belts - bäh. I wan´t to see what´s going on. Wan´t to feel my iron-plates - not only numbers or led. Just nerf the bots enormously.

[batorfly]

layerd belts - bäh. I wan´t to see what´s going on. Wan´t to feel my iron-plates - not only numbers or led. Just nerf the bots enormously.

You must be fearless man

[MrHick]

Hmm I did not want to get in to this but after doing a little testing it seams that with 4x charging time bots would become usless, just a cloud sitting above the factory hogging UPS..

I always considered bots to be end game stuff, when you shoot a rocket then expand and build with bots.
In my opinion the priorities should be fixing broken cr.. first and then work on buff to belts.

Bots are UPS friendly because they allow for better beacon coverage and require less entities (Inserters) and they allow for calculated throughput which at this time belt's dont as they dont compress.

Few suggestions.
First fix Pipes please. Fluid dynamics are horrible so running pipes around big belt factories you never know how much fluid can you deliver. And they seem to be the largest UPS hog right now.

Fix belt compression, inserters and sideloading should compress the belt. It is really hard to deal with ratios if your belts don't have expected throughput.
Optimize inserters to make the belt bases more UPS friendly. (Belt builds need way more inserters and they seam to be second largest UPS hog just because of the numbers you need.)

25% increase in blue belt speed 4y=2r=1b should help greatly with throughput but not add complexity.
Add stack longhanded inserted, the would create more options for belt builds that require a lot of materials or multiple items.

Expand beacon coverage by 1 tile which will make belt beacon builds more viable, there is just not enough room for input in large builds.
Add loaders in, they are already created and should help with belt throughput and UPS.

New splitter options are cool, really usefull in staring bases but they are almost useless in large bases which is why this was brought up. if I need 52 lanes of iron plate from 3600 steel furnaces (10000+ inserters) for space science that wont help me at all, none of the belts in the build would be mixed that will just lover throughput but 25% faster blue belts would make it 38 belts of iron.

Currently I am building 2500SPM Spaghetti BOT and Rail factory with steel furnaces running on rocket fuel so bot changes will not afect me much as bots are used only to feed fuel on to the belt for furnaces.

[Avezo]

Why are you rushing to 1.0 so much anyway? There is so much that could still be added to the game, but proposed bot nerf is like cutting corners in hardcore mode.

[Deadly-Bagel]

Thumbs up on the filter splittersSpeaking of stack inserters... you've plastered that FFF with pictures of you using loaders. Put loaders in the base game, they'll help buff belts.

To specifically buff belts it should be an item (or item pair) that only loads and unloads from/onto belts. Like a 2x1 item where 0.5 tiles is an actual belt (the rest transports items like a belt but is covered). Not only does this prevent "teleporting" items along a string of chests but it makes it more difficult to load from a chest directly into a machine or something as you would need an unloader then a loader.

A more sensible belt buff might be the ability for Inserters to "splat" items onto a belt. Rather than having a single drop point they could simultaneously drop perhaps 2-3 items onto each lane, or perhaps just the entire belt tile. Likewise, some sort of magnetic or vacuum technology should be able to more quickly grab items from a belt. Ideally 1 or 2 Stack Inserters at max research should be able to fully compress and consume a blue belt. I think this is more intuitive than having yet another "inserter" class of item.

[RebelNode]

I think one good way to buff belts would be to add a belt balancer into the game. If I have a 8-lane belt bus delivering iron ore, I need to build a belt balancer to make sure each output is fed from each input. Currently having to build the balancer manually from belts, underground belts and splitters means it will look ugly and be very complicated to build if I don't have construction robots yet. Logistics robots just balance themselves out automatically so maybe end-game belts should have an easier way of doing that too.

[Wackerstamfer]

Making / inventing balancers for your needs is part of the fun! Factorio would not be Factorio without some crazy belt spaghetti !

[csdt]

I would like to throw some ideas into the wild:

I really think that belts could be improved in term of usability, but don't really have to have a higher throughput.
In that respect, this modified splitter looks great with configurable priorities.
However, I don't think that filtering should be handled directly via splitters. While I completely agree that filtering is important, I would prefer having a filtering belt that you could put after the splitter if you want. That way, the splitter doesn't become an overpowered universal tool that will resolve all your belt problems. And this would require more place for the filtering belt while still being reasonable (1 tile more).
With a new filtering belt, we could envision more configurable filters (like different filters for left side and right side of the belt).

Another improvement belt setups could benefit from is configurable inserters that could handle and prioritize the drop point (left or right, closer or further).
Also, it should be possible to compress a belt simply with a bunch of inserters, or by merging belts.

And finally, one way to nerf robots for high throughput would be to implement some kind of traffic jam for robots. For instance, a robot cannot enter the collision box of another robot and needs to get around to continue.
That way, one would need physically larger robot highway to keep a high throughput, while it should not affect much low throughput traffic.
The plus of this solution is it's very intuitive to understand because we all have experience with actual traffic jams.
It would even be possible to have some researches to increase the number of robots in the same spot in order to increase the throughput, if we really want.

I hope my ideas could inspire some of the following discussions.

[Uffi92]

How about an bulk storage container you can only transport by train and belt and pour out only to belts(fully compressed in best case). Increase the container size by research to upgrade.

[thelordodin]

That would be nice if all this splitter features could be controlled from circuit (settings filters, altering priority setting).