Friday Facts #203 - Logistic buffer chest

[factoriouzr]

We were also concerned of people segregating their logistic networks for more control, it seems to us it was a workaround to a problem we should fix.

Still need to do that when building perimeter wall. Since bots don't path too well around large areas of no coverage. They go straight for the destination and when they run out of power they go to closest roboport and that can be starting roboport. But buffer chests are nice addition that will help to keep things organized.

I would love this too. I posted a suggestion for it long time ago:
viewtopic.php?f=6&t=38036

It's unfortunate that the first reply is from a dev:

Postby Rseding91 » Fri Dec 02, 2016 10:16 am
Solution: don't build roboports that cause them to flow over biters.

Also this suggestion would be great to get implemented:
viewtopic.php?f=6&t=35512

[factoriouzr]

Decouple draw logic.

It already is.

Decouple routing decisions. Keep a separate path network (robots, TRAINS) and handle pathfinding on separate threads.

Not feasible. Routes become invalid during game updates which means you can't calculate a path while the game updates. Routes also effect the cost of the next route so they can't be done in parallel.

Decouple AI At least high level (i.e. groups of biters).

Exact same problem as trains.

If routes become invalid during game updates, then how are train paths kept? How are biter paths cached (this was an optimization you guys did and explained in a previous Friday facts)? Even if routes become invalid at game updates, that just suggests that you should use a different algorithm that doesn't suffer from the problem. I don't see any reason why you can cache all these paths and update them only when the appropriate entities are added or removed, and just keep using the cached path info most of the time.

[factoriouzr]

Oh man, this is probably not feasible, but after reading this again something occurred to me:

When the main base is so far away, it can take a long time for robots to arrive with repair packs, so the biters might be able to break through. Using the buffer chest, it will be easy to setup nearby supplies to quickly repair the walls when needed.

What if these chests could request and deploy locally cached construction bots?

This is possible already: you can insert bots into a roboport and this will cause them to be deployed. I often have a requester chest set to insert more bots if the number of available bots drops below a certain level.

I use this too, but the current implementation is flawed as it can block permanently.

Lets say you have a roboport where you insert bots at and only one roboport (adding more still has the same issue it just reduces the likelyhood). Then you want to insert construction bots because lets say some died. If the port is full of bots (logistics or construction), then your new bots would potentially never be inserted. Easy example is the port is full of logistics bots because you don't need all of them and then you can't insert construction bots.

Another issue is that you have no reliable way to automatically remove bots from a network. You can make the conditions, but if bots never land at that roboport where you have the logic, you will never remove them. The solution to this part could be to add a setting to a roboport for always landing x logistics and y construction bots at this particular roboport and keeping them landed there. These bots would still count towards the total active bots available.

I posted issues about these long time ago. It would be great if the solutions to these would be implemented.

[greaman]

Another issue is that you have no reliable way to automatically remove bots from a network. You can make the conditions, but if bots never land at that roboport where you have the logic, you will never remove them. The solution to this part could be to add a setting to a roboport for always landing x logistics and y construction bots at this particular roboport and keeping them landed there. These bots would still count towards the total active bots available.

I posted issues about these long time ago. It would be great if the solutions to these would be implemented.

You can add some logic to remove transport bots if the need to insert construction bots arises and the port is full of transport bots, its messy, but works to the extend where I can get construction bots in.

The only issue I had with that solution is, that it kind of messed up using one roboport to control both populations and it makes things more complicated.

[bripi]

Well, I like the new flashy bot designs...but I can't figure out what they are! From L to R, I see Logistic bot, then Construction bot...but what are the other 3 bots???

[Mylon]

Well, I like the new flashy bot designs...but I can't figure out what they are! From L to R, I see Logistic bot, then Construction bot...but what are the other 3 bots???

Combat robots.

[Selvek]

Couldn't you just use active providers to "push" the items into the filtered storage chests?

No, because if all your items wind up in unfiltered storage chests there is nothing motivating them to move into the filtered slots that you intend to use as buffers. Even if filtered slots have higher priority than unfiltered slots to solve that problem, I prefer the buffer chest because I don't want to have all my factory outputs be active providers - it makes it easier to keep track of how much stuff you have.

I was playing yesterday* ...

*Or maybe is was the day before; time flies when you're having fun.

My guess? You started the day before, ended yesterday, and didn't notice the transition in between

[psychomuffin]

Thank you so much! I had been wanting to request something like the buffer chest, but I wasn't sure if it was just me doing logistic chests wrong. I'm glad to see it wasn't just me. As always, your responses to the community results in changes I (and I'm sure many others) love! Thank you!

[Beuteschema]

if logistic bots fill a requester chest or a buffer chest, does the storage chest have a bigger priority than the passive provider chest for pulling out, so do bots prioritize the storage chests when getting items? and with the buffer-chest: who gets prioritized now when supllying a normal requester chest? thx for answers

[kovarex]

if logistic bots fill a requester chest or a buffer chest, does the storage chest have a bigger priority than the passive provider chest for pulling out, so do bots prioritize the storage chests when getting items? and with the buffer-chest: who gets prioritized now when supllying a normal requester chest? thx for answers

storage and buffer chest has the same priority for pulling out. The closest is chosen.

[Beuteschema]

if logistic bots fill a requester chest or a buffer chest, does the storage chest have a bigger priority than the passive provider chest for pulling out, so do bots prioritize the storage chests when getting items? and with the buffer-chest: who gets prioritized now when supllying a normal requester chest? thx for answers

storage and buffer chest has the same priority for pulling out. The closest is chosen.

and will buffer chests have a bigger priority than storage and provider chests for delivering to requester chests or is it coming down to distance as well?

[Engimage]

I have been thinking about buffer chests again and will throw an idea.
We could implement a setting for all logistic chests called "logistic subnetwork ". This would be set to Default (0) on newly placed chests. You can compare this to VLAN and would only effect automatic logistic bot transfers. When it comes to construction all chests are equal in their roles.
This would allow solving many tasks such as forming "logistic tunnels" etc, removing logistic loops and mimic buffer chests with a pair of requester/provider chests placed in different VLANs. You could also place a requester chest in isolated VLAN to force deconstruction to a specific chest and then inserting it into provider in global network for logistic bots to handle it.
Overall I should probably make it a suggestion in suggestions forum but I am on a mobile now and too lazy for that.

[nevniv]

Love the update! That buffer chest makes logistics so much better, solves all the problems I always wish I had a solution for (I only play vanilla)

[_alphaBeta_]

So when will buffer chests become available in the research tree? Requester chests are way down the line now, and it would be nice to start filtering storage much earlier in the game.

[DaveMcW]

Buffers can act as requester chests, so probably the same tech.

How to filter storage with only green science:
1. Build a set of empty storage chests
2. Put 1 item in each chest to act as a filter
3. Deconstruct your mixed chests

[RobertTerwilliger]

I think one chest-type logistics still misses is path-chest. As in when I want a simple replacement for a belt when it's impossible or difficult to build one between two locations. Kind of like a fully separate logistics "pipe" that still draws from the same pool of idle drones to do the transportation between my two chests.

For requester chest (and incoming buffer chest) you should build your new chest, copy old one's settings to new one and quickly ORDER deconstruction of old chest. This will cause a bit of a mess, scince part of new chest will get supplied by construction bots emptying old chest for deconstruction simultaneously with logistic bots bringing stuff from providers and storages. Thus all excessive stuff from old chest will go to storage. However system will fulfill items that you take away from your new chest from storage before passive provider, thus after some time you'll get, in fact, moved old chest.
For storage box it's even simplier, as construction bots will empty chest to nearest storage if no requests for an item made.
Fir provider this does not work at all)

Another way is, well, belting away all items. Or using chain of stack inserters+small chests. You should build steel chest, and replace your old ligistic chest with steel one as well to make sure logistic network won't make a mess from process, and after items transfered, simply replace new steel chest with logistic one. You'll have to manually set up all filters though, but there's an option to build temporary chest outside of logistic network just to copy-paste settings.

What I'm saying, this "path" chest you're talking about isn't needed at all, considering it would be very temporary and rare used entity, while there are another simple and harmless ways to do the job.

[Faark]

Why do they want to add a new chest type? Can't we just use the storage chest with "requests" set for this? As long as nothing is set, it keeps the old behavior, once items are specified (maybe even by circuit network) non matching stuff is cleared and... but i guess having the feature somewhat hidden might be less user friendly than having more options to understand initially.

Anyway i really like the idea of finally getting a chest we can fill and clear based on circuit logic.

[Karamel]

Speaking of logistic system, would it also be possible to set per-roboport upper and lower counts for constructor and logistic bots the system will try to house in that port before more will move there or move to other ones.

I've sometimes ran into problems where newly constructed construction bots couldn't be automatically deployed because they sat around, instead of spreading out to other ports. And while I usually play with no biters, it seems like being able to keep a force ready to deploy those repair packs would also be a pretty important part of the equation.

[vanatteveldt]

Maybe it would be good to set a preferred number of bots of each type, and move bots out if there are too many, and in if there are too few. Alternatively, set a prefered minimum and maximum, or choose from [exactly/at least/at most] [N].

That way you can specify that your border roboports prefer say 20 conbots, or that a port you want to be able to insert into has at most 50 logibots, and a port you want to be able to extract from has at least 50 logibots, etc.

[djjaeger82]

Yes!!! So looking forward to this buffer chest feature add, thank you!