Add "quality priority" to inserters
Moderator: ickputzdirwech
-
Splitframe
- Long Handed Inserter
- Posts: 73
- Joined: Thu Aug 28, 2014 4:44 pm
- Contact:
Add "quality priority" to inserters
Similar to the spoilage priority, please let us select "high quality first" and "low quality first".
Re: Add "quality priority" to inserters
+1
How come this post has no other answers. I tried many combinations with the new selector combinator but could never get it to work...
How come this post has no other answers. I tried many combinations with the new selector combinator but could never get it to work...
Re: Add "quality priority" to inserters
I didn't find an upvote button, but this is definitely useful
Re: Add "quality priority" to inserters
What's the use case here? I very often filter inserters on quality, but I'm not sure where I'd want one that could move any quality depending on what's available.
Re: Add "quality priority" to inserters
I support this idea
Add a setting for inserters to prioritize items based on their quality tier, mirroring the existing "Freshness Priority" logic used for perishable items.
* None (Default behavior)
* Highest quality first
* Lowest quality first
*
When enabled, the inserter would scan the available items (on a belt or in a container) and prefer grabbing the tier that matches the priority setting before taking others.
1. Consistency: Since we already have "Freshness Priority" for Gleba's mechanics, adding a similar logic for Quality feels like a natural and intuitive extension of the UI.
2. Simplification: It would reduce the need for "circuit spaghetti" in quality-filtering setups, especially when unloading mixed-quality belts into machines or chests.
3. Efficiency: It allows for better management of valuable high-quality fuel or ammo without needing to strictly separate belts for every single quality tier.
Add a setting for inserters to prioritize items based on their quality tier, mirroring the existing "Freshness Priority" logic used for perishable items.
What?
Currently, inserters can pick up any quality if not filtered, or a specific quality if a filter is set. I suggest adding a "Quality Priority" dropdown menu in the inserter GUI with the following options:* None (Default behavior)
* Highest quality first
* Lowest quality first
*
When enabled, the inserter would scan the available items (on a belt or in a container) and prefer grabbing the tier that matches the priority setting before taking others.
Why?
In Space Age, managing quality often requires complex circuit logic or multiple filtered inserters to ensure that high-quality intermediates are used where they matter most, or that low-quality items are sent to recyclers first.1. Consistency: Since we already have "Freshness Priority" for Gleba's mechanics, adding a similar logic for Quality feels like a natural and intuitive extension of the UI.
2. Simplification: It would reduce the need for "circuit spaghetti" in quality-filtering setups, especially when unloading mixed-quality belts into machines or chests.
3. Efficiency: It allows for better management of valuable high-quality fuel or ammo without needing to strictly separate belts for every single quality tier.
9k hours in factorio.
Re: Add "quality priority" to inserters
Tried to make it with combinators, it look like this :
1 filter combinator per quality level to isolate signals, and a decider paired to give them ordered values, 5000 for legendaries , 4000 for epic, 3000 for rare, 2000 for uncomon and 1000 for no-quality signals, the last filter combinator is used to only pick the first signal in descending order to set filter on the inserter, if there is anything legendary in the chest, it will be the filter since it will have the highest 5000 value, if there is no more legendary, the highest signals are the epic , at 4000, and so on until everything is gone.
There may be a better way, because it's a lot of combinators , and i'm still unsure about the use case though. "freshness first" doesn't apply on inserter picking from belts, only from chests.
- high quality first.jpg (34.39 KiB) Viewed 512 times
1 filter combinator per quality level to isolate signals, and a decider paired to give them ordered values, 5000 for legendaries , 4000 for epic, 3000 for rare, 2000 for uncomon and 1000 for no-quality signals, the last filter combinator is used to only pick the first signal in descending order to set filter on the inserter, if there is anything legendary in the chest, it will be the filter since it will have the highest 5000 value, if there is no more legendary, the highest signals are the epic , at 4000, and so on until everything is gone.
There may be a better way, because it's a lot of combinators , and i'm still unsure about the use case though. "freshness first" doesn't apply on inserter picking from belts, only from chests.
Check out my latest mod ! It's noisy !
Re: Add "quality priority" to inserters
First of all, this circuit is incomplete! At any given moment, a chest might contain several different types of items—for example, modules and plates of different qualities. To account for all of them, the circuit would have to be three times as large.
Secondly, this doesn't cover wagons, trains, cars, tanks, or spidertrons. You might say, 'You can just unload from the wagon into a buffer chest first, and then apply the logic,' but I’d argue that’s a bait-and-switch. You're just unloading one way to then unload another way later. Where's the UPS savings you keep talking about? All these extra steps and logic kill UPS more than a built-in function in the inserter ever would. Plus, in that case, I'd need to implement the following functions in the circuit: 1) detecting when unloading is finished, and 2) a massive circuit to handle receiving multiple signals of the required types.
So, I absolutely disagree with the idea that a 'Quality Priority' parameter doesn't belong in the inserter! In terms of the task at hand, we would actually see UPS savings if the developers added a 'Quality Priority' setting directly to the inserter. Right now, the massive logic chains required to achieve this result in much higher UPS costs.
Please evaluate UPS costs correctly! They should be measured relative to the task being solved by comparing different implementation methods. And let’s not forget about Quality of Life (QoL). Without this parameter, some interactions simply don't feel like proper gameplay, as I've already explained with the tank example above.
Secondly, this doesn't cover wagons, trains, cars, tanks, or spidertrons. You might say, 'You can just unload from the wagon into a buffer chest first, and then apply the logic,' but I’d argue that’s a bait-and-switch. You're just unloading one way to then unload another way later. Where's the UPS savings you keep talking about? All these extra steps and logic kill UPS more than a built-in function in the inserter ever would. Plus, in that case, I'd need to implement the following functions in the circuit: 1) detecting when unloading is finished, and 2) a massive circuit to handle receiving multiple signals of the required types.
So, I absolutely disagree with the idea that a 'Quality Priority' parameter doesn't belong in the inserter! In terms of the task at hand, we would actually see UPS savings if the developers added a 'Quality Priority' setting directly to the inserter. Right now, the massive logic chains required to achieve this result in much higher UPS costs.
Please evaluate UPS costs correctly! They should be measured relative to the task being solved by comparing different implementation methods. And let’s not forget about Quality of Life (QoL). Without this parameter, some interactions simply don't feel like proper gameplay, as I've already explained with the tank example above.
9k hours in factorio.
Re: Add "quality priority" to inserters
No this circuit handles all types of items and quality being mixed in the chest and you can replace the chest by a train wagon just fine.terradus wrote: Sat Apr 18, 2026 10:08 pm First of all, this circuit is incomplete! At any given moment, a chest might contain several different types of items—for example, modules and plates of different qualities. To account for all of them, the circuit would have to be three times as large
There is still no use case presented for the feature.
Check out my latest mod ! It's noisy !
Re: Add "quality priority" to inserters
No one talked about this are you generating your answer with a robot ?
Check out my latest mod ! It's noisy !
Re: Add "quality priority" to inserters
I’ll say it again: no, you’ve built a completely different functionality using combinators. Let me give you another example. Imagine a chest with 20 different items, but you only need 5 specific types, prioritized from highest to lowest quality. If we had a 'Quality Priority' toggle in the inserter, you’d just check the box and set 5 item filters (each accepting any quality!). Your current setup is incomplete and only handles a very narrow task in isolation, because it only focuses on quality and nothing else. In other words, you haven't implemented filters.mmmPI wrote: Sun Apr 19, 2026 3:44 amNo this circuit handles all types of items and quality being mixed in the chest and you can replace the chest by a train wagon just fine.terradus wrote: Sat Apr 18, 2026 10:08 pm First of all, this circuit is incomplete! At any given moment, a chest might contain several different types of items—for example, modules and plates of different qualities. To account for all of them, the circuit would have to be three times as large
There is still no use case presented for the feature.
Basically, you’ve only built one tiny piece of the puzzle. To match the actual functionality 100%, you’d need to add more combinators to fix the missing ability to filter for those 5 items. And if you account for a dynamically changing signal, the setup would become unnecessarily massive (consuming more UPS than if the devs had just added it as a native inserter parameter).
Last edited by terradus on Sun Apr 19, 2026 4:26 am, edited 1 time in total.
9k hours in factorio.
Re: Add "quality priority" to inserters
I think you are misunderstanding the purpose of the circuit i posted which is to replicate the behavior or a "quality priority".terradus wrote: Sun Apr 19, 2026 4:18 am I’ll say it again: no, you’ve created a completely different functionality. Let me give you another example. Suppose you have 20 different types of items in a chest. With a 'quality priority' setting, you would check a box and select 5 item filters (where each item could be of any quality!).
In other words, you’ve only implemented one tiny part of this massive feature. To be 100% functional, you need to add item filters.
But it doesn't stop there. Your next challenge will be creating a circuit scheme that can handle these dynamically changing filters.
If you want to add other things like only selecting 5 items, you can just add a decider combinator to filter only those signals anywhere.
The dynamic part is harder and already achieved.
Check out my latest mod ! It's noisy !
Re: Add "quality priority" to inserters
No, I understand perfectly how your setup actually works. I’m just saying that your circuit isn’t a full replacement for the functionality everyone here is hoping for. You realize we’re discussing a specific feature idea, not its '10% derivative' in the form of combinators, right? I appreciate the circuit and I'm not criticizing it—it’s great at one thing: pulling items starting from the highest quality. But that’s all it can do. A native 'Quality Priority' parameter in the inserter would do so much more because the inserter can also receive signals for filters.mmmPI wrote: Sun Apr 19, 2026 4:26 amI think you are misunderstanding the purpose of the circuit i posted which is to replicate the behavior or a "quality priority".terradus wrote: Sun Apr 19, 2026 4:18 am I’ll say it again: no, you’ve created a completely different functionality. Let me give you another example. Suppose you have 20 different types of items in a chest. With a 'quality priority' setting, you would check a box and select 5 item filters (where each item could be of any quality!).
In other words, you’ve only implemented one tiny part of this massive feature. To be 100% functional, you need to add item filters.
But it doesn't stop there. Your next challenge will be creating a circuit scheme that can handle these dynamically changing filters.
If you want to add other things like only selecting 5 items, you can just add a decider combinator to filter only those signals anywhere.
The dynamic part is harder and already achieved.
'Quality Priority' would also mean that a single filter type would apply to all quality tiers of that item. In other words, you wouldn't need to build a 40-combinator monstrosity just to work around this problem.
9k hours in factorio.
Re: Add "quality priority" to inserters
I disagree, you may be refusing to consider combinator but it's not the case for everyone here as it was mentionned someone tried to use the selector combinator:terradus wrote: Sun Apr 19, 2026 4:33 am You realize we’re discussing a specific feature idea, not its '10% derivative' in the form of combinators, right?
maharvey wrote: Sat Nov 30, 2024 2:21 pm I tried many combinations with the new selector combinator but could never get it to work...
It is exactly what it was designed for, as an example to refine to suit the more precise use case that could occur, since none was mentionned in particular. It can be used to unload train station starting by higest quality, like you would start by the fresh first.terradus wrote: Sun Apr 19, 2026 4:33 am I appreciate the circuit and I'm not criticizing it—it’s great at one thing: pulling items starting from the highest quality. But that’s all it can do
The first part i agree, but it's not in there in the game, so currently if you are stuck you need to use other solutions, you can use the circuit i posted, it's 11 combinators, but it can control all the inserter in a train station, you don't need to duplicated it per inserter, it's not 40 combinators, it's not a single one, it illustrate the current state which can be useful to weight in the gameplay added by the suggestion.terradus wrote: Sun Apr 19, 2026 4:33 am A native 'Quality Priority' parameter in the inserter would do so much more because the inserter can also receive signals for filters.
The fact that inserter can receive signals for filter is what i used in my circuits, you can replicate quality priority logic with combinator and send the result as filter.
As mentionned , this only require adding a single combinator to the build i posted.terradus wrote: Sun Apr 19, 2026 4:33 am 'Quality Priority' would also mean that a single filter type would apply to all quality tiers of that item. In other words, you wouldn't need to build a 40-combinator monstrosity just to work around this problem.
Do you have a situation where you would use this feature in a game ?
Last edited by mmmPI on Sun Apr 19, 2026 5:34 am, edited 1 time in total.
Check out my latest mod ! It's noisy !
Re: Add "quality priority" to inserters
In general, everything in Factorio is about UPS optimization; you don't even need to wait for it to be mentioned in a specific thread. But it actually was mentioned here—the moderator merged two threads because they were duplicates. You can check it yourself by following the link:
viewtopic.php?t=128100
viewtopic.php?p=691383#p691383
- 04-19-2026, 07-56-21.png (95.05 KiB) Viewed 283 times
9k hours in factorio.
Re: Add "quality priority" to inserters
That's not because another player mention it may help UPS that i can answer your question lol, i'm only a showing a circuit that fit what some other player than you tried to doterradus wrote: Sun Apr 19, 2026 4:54 amIn general, everything in Factorio is about UPS optimization; you don't even need to wait for it to be mentioned in a specific thread. But it actually was mentioned here—the moderator merged two threads because they were duplicates. You can check it yourself by following the link:
In the other thread it's also mentionned how should this interract with the inserter if you select both "freshness first" and "quality first" ?
And still there is no real use case in an actual factory that sounds like it would require such feature, do you have an example where you would use such things ?
Check out my latest mod ! It's noisy !
Re: Add "quality priority" to inserters
I’ve run into a specific need for this feature while building factories that switch between recipes (using quality modules). I struggled with the sheer size of the logic required because the system lacks an 'Any Quality' signal for specific items.mmmPI wrote: Sun Apr 19, 2026 4:51 am Do you have a situation where you would use this feature in a game ?
A new 'Quality Priority' parameter for inserters would imply that developers either add an 'Any Quality' signal for specific items in constant combinators, or teach the inserter to ignore an item's quality to make this priority work (otherwise, the feature wouldn't function, so it's a logical expectation).
When I tried to implement quality priority using combinators, I realized it’s just not feasible—the build becomes massive. That’s why I’m here to support the proposal for adding a 'Quality Priority' parameter to inserters. If combinators were four times smaller, I wouldn't be complaining because the logic-based approach would be acceptable. But right now, it takes up too much space—it's easier to just build separate factories that don't switch recipes.
Just try to implement a fully functional setup with all the nuances and test it yourself. You’ll see so many 'hidden rocks' that people overlook. At first, the circuit seems small, but as you account for all the edge cases and fixes, the build bloats. You end up trading one problem for another: the massive footprint of the combinator logic.
9k hours in factorio.
Re: Add "quality priority" to inserters
Space is tight and time is short in the 'Mount Fortress' scenario (comfy factorio - mount fortress) server name in server list: "COMFY Mtn fortress SA". I literally have no more than 4 hours to produce a large amount of Rare modules. On top of that, there are many nuances involving other players and enemy waves. There’s no guarantee there will be enough room for a blueprint, which means it has to be versatile and modular (split into parts).mmmPI wrote: Sun Apr 19, 2026 5:03 am And still there is no real use case in an actual factory that sounds like it would require such feature, do you have an example where you would use such things ?
I have a very rough prototype with many flaws. You can try playing with it in Mount Fortress with the goal of providing the entire team with Rare modules and filling all the production inside the train with them as well.
I’d be glad if you enjoy it, and maybe we’ll run into each other there. Just a heads up—I don’t talk much when I’m playing.
- 04-19-2026, 08-46-34.png (1.73 MiB) Viewed 210 times
9k hours in factorio.
Re: Add "quality priority" to inserters
Edit : I had not seen the large blueprint in previous post before answering
, I don't know what is the acceptable combinator footprint, i feel this is a personnal opinion . I only try to demonstrate the size it takes, maybe it can be made smaller. I think it's closer to 10-15 than 40 combinators, i understand it can be too much sometimes, in this case my build is a proof the suggestion is useful . This is also why i ask for use cases, because i want to try to make the machine if that can help, like if the suggestion is not implemented. Also sometimes it's possible to use different logic to achieve same result for the same reasons.
Here is an example where i tried to add a filter for "wood" only, and use it for a train station unload :
It adds 2 combinators , and it's not perfect, when an unloading inserter is blocked by what the others have already placed on the belt it can accumulate material of high quality, and only release it later when there is room for it, mixing this a little, which would also occur if the suggestion is implemented, "hidden rock" ! i like this idea If you have other idea for edge cases, please let me know, i don't want to overlook anything, and if the resulting build is a 40 combinator monstruosity , so be it , it's ok it will show the merits of the suggestion, or maybe different ways to handle them, i don't know i'm curious
I find the "any quality signal" or "pure quality signa"l you can send to the inserter to make it grab only that quality level would be useful on its own when i tried to use the posted setup a bit more. I'm not sure i understand when in a factory you would want to use that "quality priority" setting on an inserter as personnally i would use splitters to filter the material and drop mixed quality things on belts in games, so i'm always curious to hear what other people do.terradus wrote: Sun Apr 19, 2026 5:29 am I’ve run into a specific need for this feature while building factories that switch between recipes (using quality modules). I struggled with the sheer size of the logic required because the system lacks an 'Any Quality' signal for specific items.
A new 'Quality Priority' parameter for inserters would imply that developers either add an 'Any Quality' signal for specific items in constant combinators, or teach the inserter to ignore an item's quality to make this priority work (otherwise, the feature wouldn't function, so it's a logical expectation).
I tried tooterradus wrote: Sun Apr 19, 2026 5:29 am When I tried to implement quality priority using combinators, I realized it’s just not feasible—the build becomes massive. That’s why I’m here to support the proposal for adding a 'Quality Priority' parameter to inserters. If combinators were four times smaller, I wouldn't be complaining because the logic-based approach would be acceptable. But right now, it takes up too much space—it's easier to just build separate factories that don't switch recipes.
Just try to implement a fully functional setup with all the nuances and test it yourself. You’ll see so many 'hidden rocks' that people overlook. At first, the circuit seems small, but as you account for all the edge cases and fixes, the build bloats. You end up trading one problem for another: the massive footprint of the combinator logic.
, I don't know what is the acceptable combinator footprint, i feel this is a personnal opinion . I only try to demonstrate the size it takes, maybe it can be made smaller. I think it's closer to 10-15 than 40 combinators, i understand it can be too much sometimes, in this case my build is a proof the suggestion is useful . This is also why i ask for use cases, because i want to try to make the machine if that can help, like if the suggestion is not implemented. Also sometimes it's possible to use different logic to achieve same result for the same reasons.Here is an example where i tried to add a filter for "wood" only, and use it for a train station unload :
- train station caption.jpg (155.92 KiB) Viewed 207 times
It adds 2 combinators , and it's not perfect, when an unloading inserter is blocked by what the others have already placed on the belt it can accumulate material of high quality, and only release it later when there is room for it, mixing this a little, which would also occur if the suggestion is implemented, "hidden rock" ! i like this idea
If you have other idea for edge cases, please let me know, i don't want to overlook anything, and if the resulting build is a 40 combinator monstruosity , so be it , it's ok it will show the merits of the suggestion, or maybe different ways to handle them, i don't know i'm curiousCheck out my latest mod ! It's noisy !
Re: Add "quality priority" to inserters
That is very different than what i had in mind, thanks for posting the blueprint, it illustrate well enough for me why a dozen combinator is too much in your use case ( somewhat of an automall ) with your current architecture, it's not a train station where they may serve several inserters , i understand the need for modularity and how it makes it difficult/impossible to attach 5 filter combinator anywhere the logic is required. There are a few things that are rotated away or disconnected which makes it hard to use but i imagine it's made to be tweaked during the game, i heard of the scenario but going fast and not talking is the opposite of how i play multiplayer x)terradus wrote: Sun Apr 19, 2026 5:41 am Space is tight and time is short in the 'Mount Fortress' scenario (comfy factorio - mount fortress) server name in server list: "COMFY Mtn fortress SA". I literally have no more than 4 hours to produce a large amount of Rare modules. On top of that, there are many nuances involving other players and enemy waves. There’s no guarantee there will be enough room for a blueprint, which means it has to be versatile and modular (split into parts).
I have a very rough prototype with many flaws. You can try playing with it in Mount Fortress with the goal of providing the entire team with Rare modules and filling all the production inside the train with them as well.
I’d be glad if you enjoy it, and maybe we’ll run into each other there. Just a heads up—I don’t talk much when I’m playing.
What it illustrate that wasn't mentionned is the absence of robots too, i can understand why from the scenario, it helps understanding the context for the suggestion.
The way that would feel the more natural to me to handle such problem would be to have separate side of belts for separate items and quality level and try to mostly control the assembly's recipe instead of inserter filter, because inserter are smart enough to pick the right materials for an assembly when its recipe switches, i'm still unsure i understand why the particular need for quality priority, how it would help for the setup you are making, but i understand it's like an advanced feature, you're trying to optimize a complex build where there are possibilities to do differently. Sorry if my post added confusion, it was unclear at the time what was the context, so i went with an attempt at illustrating, now i realize it missed the point for you and why.
Check out my latest mod ! It's noisy !