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

Regular reports on Factorio development.
Locked
Elecen
Burner Inserter
Burner Inserter
Posts: 12
Joined: Thu Apr 13, 2017 5:02 pm
Contact:

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

Post by Elecen »

Can make the bots dependant to wind. If there's a strong wind opposite the direction they're flying, they should go much slower. This effectively would make them untrustworthy replacement of belts.


Another thing would be, to limit the logistic bots per roboport control (red) area. If the roboport would be something like the control center for it's area and it supports say, 80 logistic bots at a time, then only 80 logistic bots can navigate through it's path (the red box). Any logistic bot over that number that enters the roboport aerial control space would actually have to wait before it can navigate (also this logistic bots per roboport number could be increased via research).


Make bots unsafe. For every hour of lifetime the bot has it has an increased chance of dropping dead during flight (boom and you loose whatever it was carrying). This means, logistic bots don't live forever and need constantly be renewed plus if overused and "explode" you also loose whatever it was carrying, like your new Nuclear reactor, or your new MKII armor). I would set a possibility of fault rate - like 2% every 60mis. So, in 120 mins you have 4% chance of the bot *boom* on flight. Plus would be cool effect.

mcvey
Burner Inserter
Burner Inserter
Posts: 13
Joined: Fri May 05, 2017 6:41 pm
Contact:

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

Post by mcvey »

Really liking the Splitter features, can't wait to try it out.
One way to make the compression on belts work would be to use loaders. They could stack items on belts, or even make the containers work inline with belts. With boxing it would be like belt in box out, and reverse it on the other end. It could wait until the box was full before outputting it.
Another way to make belts better is to reduce the number of beacons around a machine to make more room for belts. Speed Modules have no cap unlike the other modules. This would also reduce some throughput issues bots are used for. Removing the cargo upgrades would reduce logistic bot use without hampering construction bots.

GotLag
Filter Inserter
Filter Inserter
Posts: 532
Joined: Sat May 03, 2014 3:32 pm
Contact:

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

Post by GotLag »

If you want to make belts powerful and flexible, then please, please, PLEASE give us per-lane control on the splitter GUI. Just a simple enable/disable checkbox for each lane would be enough.

The other addition you could make would be to include the lane-selecting functionality from Side Inserters in vanilla.

Belt lanes are a huge feature for controlling item flow but the tools we have to manipulate them now are so limited.

This is an old mod I made before belt optimisations killed the method it was using to block output (invisible colliding entity on the unwanted lanes):
Image

chimeric10
Manual Inserter
Manual Inserter
Posts: 3
Joined: Sun Jan 07, 2018 12:54 am
Contact:

Re: An alternative way of balancing bots vs belts

Post by chimeric10 »

Elecen wrote:An alternative to making belts equally powerful.

Imagine the trains have wagons carrying containers (much like real-life trains).
A nice new building (crane) can unload and load containers to wagons.
Containers must be manufactured (similar to barrels) except 1 wagon with 1 container can carry the normal amount a normal wagon can carry.
Unloading a container is super fast (compared to unloading the wagon with inserters/chests, etc) and very compact (doesn't take much space).

The magic catch:
Containers can be transferred deep into the factory using belts (some special belt perhaps?)
This means you can ship items big amounts of items deep into the factory through single belt passages.
The belts moving the containers can be a new type of belt

If I can move 4k iron plates in the length of a wagon (say, 6 belts?) deep into the factory - the bots can never match that.
Not to mention it can be really cool using cranes to take containers off of trains to belts. More cranes to take containers off of belts and onto the ground so inserters can take the items before the crane can move them back on the same (or another) belt. In the end, the container can get loaded onto a belt so it can find it's way onto another train.
I find the idea/concept of containers really cool too:D . Also like the idea of a bot loading station, so bot charges at a roboport (whilst it has no task) then flies to the loading station (container mentioned by Elecen), then deliver. If they use their charge they still have to deliver, no recharge as roboports won't let a bot park (recharge) if carrying items.

This would be very close to a real life scenaro if amazon ever deliver with bots ;)

crysanja
Long Handed Inserter
Long Handed Inserter
Posts: 73
Joined: Sun Aug 03, 2014 4:47 pm
Contact:

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

Post by crysanja »

I would suggest to make some items not transportable with bots.
Stone, coal and ores come to mind.
I guess all of thouse are usualy transported in a special way, large containers(boats/trains) or belts.

Maybe they shouldnt be (un-)loaded with inserters, but just emptied into a special station?
(https://www.youtube.com/watch?v=xNIlT4YP3vA)
Flying around with explosives seems not a good idea either.

---

Maybe introduce a big item flag(buildings/rocket parts) which cant be transported in stacks.

Marconos
Filter Inserter
Filter Inserter
Posts: 301
Joined: Mon Jun 02, 2014 10:46 pm
Contact:

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

Post by Marconos »

I'm a lover of belt based bases, but I think there is something missing here in the way the devs are viewing the game.

Early Game: Yellow belts, small production runs
Early Mid Game: Red belts, faster production lines, multiple smelting setups: Base is starting to spaghetti around for most players
Mid Game: Main lines are being upgrade with blue belts for increasting thorughput. Bots are being used to move product from one area to another as the spaghetti in early mid keeps you squeezing in yet another set of assemblers. Things are starting to get nuts but the end is in site (1st rocket end). Basic / simple train setups are starting to be used to move raw materials around.
Late mid game; 1st rocket is launched players are now looking at doing the advanced research and seeing the need for big increases in resource comsumption. More trains are beign laid out and now the "belt layout" issues are changing into train layout issues. Bots are preferred as you don't want to spend time optimizing belt lines as you are having to optimize train lines.
End / VERY late game: Mega bases are growing, belts are relegated to very small areas. Bots have take over local production and trains are the new "belts" in the system as most of your time is figuring out how do I increase my train line throughput, do I do extra production in this area with bots or seperate areas connected by train lines. Spreadsheets are made to figure out optimum build configuration etc.

So yes, belts are decreased, limited (dropped in some cases) in late game but the train lines becomes the belt layout issue. Bots handle the local production but your massive product movements are done by Trains(belts) as they are end game evolution of them. To me this is the real factorio progression and you see this all the time in how players play the game. Most players never get to the "end / late" game stage and are done after their first rocket. By nerfing bots, no matter how much I initially thought I liked the idea, is actually something that will negatively affect the progress of the game IMO.

Edit: text cleanup

quickshot0
Manual Inserter
Manual Inserter
Posts: 3
Joined: Fri Jan 12, 2018 11:49 pm
Contact:

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

Post by quickshot0 »

I think that there have been some interesting suggestions here, but I think everyone has ended up over focussing on the original bot-belt part of the discussion and have thus ended up only finding solutions to the original raised matter with in those constraints alone.

Now the real question posed originally was that belts for many seemed a bit more fun and challenging then bots and that bots eventually take over everything and make things a bit samey in look. Some responded that in practise these designs took more work to come up with then is perhaps appreciated and really took quite some puzzle work. And also that bots simply have the lowest UPS at current and that thus any nerf to them basically is effectively reducing how productive the most developed factories can be.


So perhaps part of the issue can be resolved if an extra low-UPS option could be found for base operation? An option which would then preferably leads to new puzzles in factory layout and structure to be considered, and as such might help reinvigorate gameplay in the end game by changing it up yet again. (Rather then repeating old gameplay elements)

Now in this respect the Belt is some what difficult to make work much better it seems like, atleast no one has seemed to suggest something which would really change it up, rather then just extending its early game use. And using it so exclusively in the late game feels kind of like a lost opportunity as well. And obviously same use would mean the UPS hit will be the same. Still it might be possible to specialize it in some specific roles which improve diversity while at the same time not hitting UPS all to much. Like say high speed mid range transportation options. (Assuming one uses some tricks with some of those proposed super express belts) And with that be able to keep belts being kept in use for specific roles in a factory.


Though this means we have to look elsewhere for a solution. Fortunately there's one device in game already, that I've noticed rather little discussion on, but which can be made to be excellent for close in work and thus replace much of the belt use there and even probably lessen the UPS burden. And this would be the humble inserter. Now in the current game inserters are rather limited in ability wise. But some time ago I played a game of Factorio with Bob's mod, which as I'm sure many know has an upgrade which gets you a 7x7 configurable grid for inserters, where you can select where it picks up and drops items. (With the center one being the inserter itself obviously, and it giving you an effective max range of 3) For quite awhile I just used them as more versatile inserters in a standard belt-assembler layout, and that worked as one expects pretty well. But eventually just for fun I tried using belts as raw material input only and then tried using inserters exclusively in moving intermediate goods between all the assemblers and creating a high end products out of a single fabrication system with minimal to no internal belts.

This turned out to be rather fun and created new challenging puzzles to work on for once. And in the end turned out to be rather more doable then I first expected, with me actually often achieving all intermediate products being handled by inserters alone. (Though it can take quite some work to puzzle out how at times and if some assembler deep inside the structure needed a raw product it was even more of a puzzle figuring out how to supply it) And this rather naturally also made it easier to build rather compact factory layouts with fairly substantial output levels, and if well designed even little idle time on the assemblers. And because all operations were often done by one, or in high volume cases two inserters. This obviously all probably had a rather low hit on UPS. After all a bot system obviously needs not only to handle the UPS for each bot movement, but also the in and an out inserters for each assembler. Being able able to reduce this to one or two inserter should be a step up.


As such I think it might be worthwhile to consider the option of evolving short range work between assemblers in late game Factorio via use of a far more flexible and powerful inserter system. Thus potentially resolving the samey look factor some have complained about, allowing for new potential solutions in factory layout, possibly allowing for the use of all the current game transport systems in their own special roles, and possibly even allowing for a reduced UPS hit from very large factories. (Besides, in real world factories, the arms used for automation have variable length and rotation options. So it would capture some of that real world flexibility as well)

devaking
Burner Inserter
Burner Inserter
Posts: 7
Joined: Mon Nov 02, 2015 8:56 pm
Contact:

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

Post by devaking »

YES!!!!!! filtering inserter!!!!!!!


now if we could get a block that is late game upgrade to the inseter in the way that its a 1x1 block that can receive by 1, 2 or 3 side and outpout in the same way the new ui for inserter in any of the left side.

now that would help late game mega base belt design unwrap those spagetti

again tks for this upgrade

User avatar
Wakaba-chan
Long Handed Inserter
Long Handed Inserter
Posts: 96
Joined: Mon May 30, 2016 6:39 am
Contact:

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

Post by Wakaba-chan »

I completely agree that bots are too strong now and
The idea was to level this a bit so players who want to play optimally can choose to play using belts without feeling like they made the wrong choice.
- yes, this is exactly what I feel every time! I could use bots, but it's really just not fun to
extending the factory is just about plopping more assembler/requester/provider cells without any thought
Thank you to pay attention to this problem!

Actually, I think that just removing the "logistic system research" features will bring balance in game. Logistic bots will still be able to bring items to and take from the character, but not between factory entities. Hey, this is just the job they are best at! No belts could do this. Belts are for in-factory transportation, since factory entities can't move; and bots are for dynamic entities interaction - like players. That's it.

Conventia
Burner Inserter
Burner Inserter
Posts: 10
Joined: Sun Jun 04, 2017 9:45 pm
Contact:

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

Post by Conventia »

Last FFF, my issue was that game devs shouldn't try to tell their users how to have fun playing their games. Game devs should have a vision and should apologetically follow that vision.

Instead of rolling that perspective back, it seems this FFF doubles down and adds something else. Last week, Rseding91 (who I think is still a developer of Factorio) posted: https://www.reddit.com/r/factorio/comme ... s/ds8vxea/

He seems to have a dissenting opinion, compared to Twinsen and Kovarex. I sort of expected his perspective when I saw this FFF claim that other opinions would be given and with the last FFF claiming that it wanted a discussion. Unfortunately, his perspective was not added either in his own words or paraphrased by someone else, in the unlikely event, he refused to participate in this FFF (unlikely, given that he's written lots of them).

I don't think I can accurately express how disappointed this makes me.

Visione
Burner Inserter
Burner Inserter
Posts: 12
Joined: Sat Jul 15, 2017 2:32 pm
Contact:

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

Post by Visione »

Marconos wrote:I'm a lover of belt based bases, but I think there is something missing here in the way the devs are viewing the game.

Early Game: Yellow belts, small production runs
Early Mid Game: Red belts, faster production lines, multiple smelting setups: Base is starting to spaghetti around for most players
Mid Game: Main lines are being upgrade with blue belts for increasting thorughput. Bots are being used to move product from one area to another as the spaghetti in early mid keeps you squeezing in yet another set of assemblers. Things are starting to get nuts but the end is in site (1st rocket end). Basic / simple train setups are starting to be used to move raw materials around.
Late mid game; 1st rocket is launched players are now looking at doing the advanced research and seeing the need for big increases in resource comsumption. More trains are beign laid out and now the "belt layout" issues are changing into train layout issues. Bots are preferred as you don't want to spend time optimizing belt lines as you are having to optimize train lines.
End / VERY late game: Mega bases are growing, belts are relegated to very small areas. Bots have take over local production and trains are the new "belts" in the system as most of your time is figuring out how do I increase my train line throughput, do I do extra production in this area with bots or seperate areas connected by train lines. Spreadsheets are made to figure out optimum build configuration etc.

So yes, belts are decreased, limited (dropped in some cases) in late game but the train lines becomes the belt layout issue. Bots handle the local production but your massive product movements are done by Trains(belts) as they are end game evolution of them. To me this is the real factorio progression and you see this all the time in how players play the game. Most players never get to the "end / late" game stage and are done after their first rocket. By nerfing bots, no matter how much I initially thought I liked the idea, is actually something that will negatively affect the progress of the game IMO.

Edit: text cleanup
i completely agree with this

User avatar
Wakaba-chan
Long Handed Inserter
Long Handed Inserter
Posts: 96
Joined: Mon May 30, 2016 6:39 am
Contact:

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

Post by Wakaba-chan »

After all, logistic system features can still be implemented in mods as bonus, but I think they should be removed from vanilla game to bring balance and point real purpose for each of transport system.

Jürgen Erhard
Filter Inserter
Filter Inserter
Posts: 298
Joined: Sun Jun 12, 2016 11:29 pm
Contact:

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

Post by Jürgen Erhard »

You do whatever, and I'll stick with the last version I like. Deal? Because as a not-main-bus-maker (and that's not because of "optimal", I hate even a whiff of "git gud", no, I'm not playing to be "optimal", I play to have "fun") I'm completely out of the orthodoxy.

No, I do not want something "to aspire to". I want *FUN*. And you can't define what I find fun, and you can't make me find something fun that I don't consider fun. But you still sound like you want to.

You know, I'm a big fan of Python, *including* its Zen of Python, and the line in that "There should be one, and preferably only one, obvious way to do it". But since Factorio is not a programming language (and no-one except for me ever has to read my factories), I here prefer Perl's Tim Toady (TIMTOWTDI, There's More Than One Way To Do It).

Maybe bold this: Factorio. Is. Not. A. Puzzle. Game. It has scenarios and different game modes, so why not have *those* to chose if you want a game you have to balance just right? And leave Free(!) Play as, well, Free Play?

"But if you're not about the "optimal" setup, then you shouldn't be bothered about a bot nerf".

Yes, I should, if I get the feeling that bots won't be nerfed, but NERFED. And a lot of the ideas posted here by players, and also some of the ideas in the FFF, sound like "well, let's not just nerf them, let's totally castrate them". Which would not just nerf bots, but make them completely useless. Someone even had the nerf nerve to suggest them not being able (and that of course would likely have to be hardcoded to be efficient: no moddable) to carry some items. Namely, ores.

Everything that'd be hardcoded is a… Death Spell for Factorio. Like Kovarex' suggestion "completely remove the Worker robot cargo size research"… it could be left as an attribute so mods could modify it, but… why? Granted, the not-necessary-to-be-hardcoded (but could be) "multiply charging several times"… damn, no, since that's hardcoded now anyway, it'd be hardcoded, not moddable, and probably no research (since that would be a variant of the evil "just put down more roboports")

You go that route, sure. And I'm gone, but again, as with the previous, what do you care: I paid, I'm not even on YT so my influence is nil.

Sad times. But I'm really not surprised, most Early Access I followed fucked it up at the end in some way.

mergele
Fast Inserter
Fast Inserter
Posts: 158
Joined: Sat Aug 20, 2016 5:45 am
Contact:

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

Post by mergele »

The additional configuration for splitters could be unlocked via a laegame tech (just remove the GUI elements if you don't have the tech). After all it's just a software update.

And following up on GotLag his idea may then be the second tier software-update for splitters.
quote

vorax
Burner Inserter
Burner Inserter
Posts: 7
Joined: Sat Dec 31, 2016 4:51 am
Contact:

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

Post by vorax »

I believe the real problem with bots, their game breaking ability, is their inventory access speed. Think about it, if I have a chest which needs to be filled with resources, there is a hard limit on how quickly I can supply items to it using belts. Intuitively that limit is the number of inserters I can place around it times the inserters throughput. Bots do not have any such limit. No matter how slowly bots move, or how few items they carry they will always be able to transport items faster than belts given that I have enough bots.

I think chests should only be accessible by one bot at a time, and this access should take a fixed amount of time. This would cause bots to queue up at chests, and put a limit on their throughput. This kind of limitation "makes sense" to me, and wouldn't feel annoying like lengthening bot charging times would.
Bots would still be incredibly handy for the low volume tasks and "housework" which can be anoying to do using belts, but it would create a compelling reason to invest the time into developing a really impressive belt base.

I also think it would be important for bots to still be able to access a players trash slots, or probably their entire inventory without queuing up.

Omarflyjoemacky
Fast Inserter
Fast Inserter
Posts: 104
Joined: Tue Nov 15, 2016 10:56 pm
Contact:

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

Post by Omarflyjoemacky »

Jürgen Erhard wrote:You do whatever, and I'll stick with the last version I like. Deal? Because as a not-main-bus-maker (and that's not because of "optimal", I hate even a whiff of "git gud", no, I'm not playing to be "optimal", I play to have "fun") I'm completely out of the orthodoxy.

No, I do not want something "to aspire to". I want *FUN*. And you can't define what I find fun, and you can't make me find something fun that I don't consider fun. But you still sound like you want to.

You know, I'm a big fan of Python, *including* its Zen of Python, and the line in that "There should be one, and preferably only one, obvious way to do it". But since Factorio is not a programming language (and no-one except for me ever has to read my factories), I here prefer Perl's Tim Toady (TIMTOWTDI, There's More Than One Way To Do It).

Maybe bold this: Factorio. Is. Not. A. Puzzle. Game. It has scenarios and different game modes, so why not have *those* to chose if you want a game you have to balance just right? And leave Free(!) Play as, well, Free Play?

"But if you're not about the "optimal" setup, then you shouldn't be bothered about a bot nerf".

Yes, I should, if I get the feeling that bots won't be nerfed, but NERFED. And a lot of the ideas posted here by players, and also some of the ideas in the FFF, sound like "well, let's not just nerf them, let's totally castrate them". Which would not just nerf bots, but make them completely useless. Someone even had the nerf nerve to suggest them not being able (and that of course would likely have to be hardcoded to be efficient: no moddable) to carry some items. Namely, ores.

Everything that'd be hardcoded is a… Death Spell for Factorio. Like Kovarex' suggestion "completely remove the Worker robot cargo size research"… it could be left as an attribute so mods could modify it, but… why? Granted, the not-necessary-to-be-hardcoded (but could be) "multiply charging several times"… damn, no, since that's hardcoded now anyway, it'd be hardcoded, not moddable, and probably no research (since that would be a variant of the evil "just put down more roboports")

You go that route, sure. And I'm gone, but again, as with the previous, what do you care: I paid, I'm not even on YT so my influence is nil.

Sad times. But I'm really not surprised, most Early Access I followed fucked it up at the end in some way.
This post made me sad a little as well. The last line is too true.

Someone posted earlier to give belts insane speed, use a loader to load a chest and then use stack inserters to feed what's needed. Sounds simple and effective to me...
"And then Bender ran."

User avatar
vampiricdust
Filter Inserter
Filter Inserter
Posts: 314
Joined: Wed Jan 14, 2015 1:31 am
Contact:

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

Post by vampiricdust »

I am very disappointed to have 2 devs now completely and utterly ignore not only the straight up resource costs of logistics bots, but the incredible amount of research cost.

Research cost for full blue belt usage: 194,395 raw resources (liquids reduced to petrol, light, or heavy)
Research cost for 240% anf 4 capacity: 1,486,755 raw resources.

This makes the research costs ~7.65 times more expensive. So bots being 2 to 5 times more effective makes sense and actually points to them being underpowered. Couple this with the fact all those roboports alone would have paid for all the blue belts you used in the example, take how many bots you had times 3.4 to get how many blue belts all those bots could have been made into and you start to see that bots are not as overpowered or cheaty as your test makes them out to be. If bots are going to be no better than belts, then the research costs and the costs of bots & roboports needs to be cut down to match. They are overpriced for what they do.

EDIT: By the way, with the difference in resource costs (ignoring duplicate research), you could have made 25k blue belts with those resources instead of getting logistic bots up to capcity 4 and 240% speed.
Last edited by vampiricdust on Sat Jan 13, 2018 8:18 am, edited 1 time in total.

User avatar
Jon8RFC
Filter Inserter
Filter Inserter
Posts: 553
Joined: Tue May 10, 2016 3:39 pm
Contact:

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

Post by Jon8RFC »

If the charge time is increased, that will only make the need for more effective "charge port selection" necessary.

It stinks, even when there are a cluster of roboports, they'll collect around a few and have tons queued up, instead of using close and available charging ports.

The changes to trains now disallowing simple same-named-station dispatching is what already exists for bots, but I'd argue that a similar 2000 penalty would be necessary if charge time increases. It'd be nice if that existed in vanilla already. Poor bots wait to charge when they could go 10-20meters and charge on of the the many available charging ports.
Image

User avatar
Xecutor
Filter Inserter
Filter Inserter
Posts: 260
Joined: Mon Jun 23, 2014 10:15 am
Contact:

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

Post by Xecutor »

vampiricdust wrote:I am very disappointed to have 2 devs now completely and utterly ignore not only the straight up resource costs of logistics bots, but the incredible amount of research cost.

Research cost for full blue belt usage: 194,395 raw resources (liquids reduced to petrol, light, or heavy)
Research cost for 240% anf 4 capacity: 1,486,755 raw resources.

This makes the research costs ~7.65 times more expensive. So bots being 2 to 5 times more effective makes sense and actually points to them being underpowered. Couple this with the fact all those roboports alone would have paid for all the blue belts you used in the example, take how many bots you had times 3.4 to get how many blue belts all those bots could have been made into and you start to see that bots are not as overpowered or cheaty as your test makes them out to be. If bots are going to be no better than belts, then the research costs and the costs of bots & roboports needs to be cut down to match. They are overpriced for what they do.
This! Also, belts do not have any operating cost. Zero. They are working completely for free. While the network of roboports in the moderately complicated base consumes very noticeable amount of power.

No love for pneumatic tubes?

download13
Manual Inserter
Manual Inserter
Posts: 2
Joined: Sat Jan 13, 2018 6:34 am
Contact:

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

Post by download13 »

Having the strongest approach be a combination sounds like a good benchmark to shoot for.

A nice quality of life upgrade might also be to let bots attach to the play while they're charging so they don't disconnect whenever the player moves. Unless this has already been solved by the changes to robot charging. I haven't tried those yet.

Locked

Return to “News”