Suggestion about inserters

[FatMcK]

Hi,
coming from the robotics industry I really enjoy to have robots (inserters) in the game doing all this picking things for me.

This suggestions starts with a big wall of text. If you don’t care about my introduction skip past the line of ----------------‘s.

The first thing I wonder about is why they are called inserters instead of just calling them robots. But that’s not the reason I’m starting this thread.

Today’s industrial robots are capable of handling things much faster and much more precise then humans could. Robots evolve and get smarter each year using 3D vision systems, force sensors and so on. Compared to this the Factorio inserters are really dumb, even the smart one. And we are talking about a game with a futuristic setting.

Another thing that bothers me ingame which I already disliked in Minecraft for example are the North-West rules. What I mean is that if an inserter has the possibility of placing an item left or right onto a belt he will always prefer left. Or on the top side if the possibilities are top or bottom. I understand that there needs to be such rules. But it bothers me that you create a ‘factory’ that’s working well but doesn’t work anymore when rotating the whole thing.

What I want to suggest here is to redo how robots work ingame, making them more functional and flexible to use. I don’t want to go as far as to implement something incredible complicated like a fully programmable robots doing complicated stuff. But I want to suggest to make the ones we have smarter, letting them do more things, giving us the possibilities of new ways how to handle all those solids.

I’m aware that there are already suggestions about new type of robots and functions that are part of my suggestion. But I worked out a concept of reworking the existing robots completely, while keeping the existing functions as standard. Also I think that with my concept all the new features are really easy to configure, being intuitive.

I also saw that there are already mods including new kinds of robots for all different purposes. But I would prefer to have the count of different robots as small as possible while still having the ability to have most flexibility out of them.

I hope it is ok that I start a new thread for my suggestion. And sorry for that long introduction into my suggestion.

---------------------------------------------------------------------------------------------------------------------

Here my suggestion:

Here is a scheme I made which I am referring to after.

factorio_robots.jpg (285.33 KiB) Viewed 3690 times

Top row (1.x) is for normal and fast robots, bottom row (2.x) is for long handed robots.

1) When you build your first robot in the game without having researched anything and you place it somewhere it will simply behave like the actual robots. The change would be that you could open a robot settings menu with a left-click. What you would see will be its program which you couldn’t change yet. See figure 1.1 and 2.1. All options are greyed out and the checkers are set and couldn’t be changed. Like now it will prefer to take from the inner side of a belt and place it to the outer side of another. In curves etc. it will still use the north-west rule.

There will be two things to research: ‘Robotics flexibility upgrade’ and “Robotics smart lane picking’.

2) With the Robotics flexibility upgrade you get the ability to change the output direction of the robot. If you open the robot settings menu you will see that the outputs are not greyed out anymore, see figure 1.2A. By simply left-clicking you could choose the output direction of the robot, in the figure set to right output. The long handed robot will not have the ability to work angled, instead you can set if in and outputs are taken from one or two tiles away, see figure 2.2A. The example takes from one tile away putting it two tiles away. Otherwise it still will use standard rules (inner/outer belt, north-west rule).

3) With the Robotics smart lane picking upgrade we get the ability to define from where to take and where to place solids, taking out the standard rules of inner/outer lanes and north-west rule. As you can see in figures 1.2B and 2.2B the in/output fields get quartered with the ability to place checkers (north-west, north-east, south-west, south-east).
*Checkers: About the checkers there are different possibilities what setting a checker would define in detail. I will come to this later.

4) With both, Robotics flexibility upgrade and Robotics smart lane picking both gets combined as you can see in figures 1.3 and 2.3.

5) I don’t know how far this suggestion should go now. I think till yet it is still intuitive. But if we want go on, the next researchable technology would be ‘Robotics multi inserting’. As you can see in figure 1.4 you could set two or more output directions, so you could supply two assemblers for example. Or as for figure 2.4 for the long handed robot we could take from both tiles behind and/or give to both tiles in front.
Problems we may get with multi inserting is how priorities would be handled. Would it be left-right-left-right for example, and what happens when one lane is full, etc.

6) After there will be “Smart robotics” to research. This will bring us the options of the current smart inserter. But instead of being its own robot the options would be available to the existing ones, yellow, blue and red.

7) Last but not least I want to suggest to include a fast long handed robot. This could replace the green one as this got obsolete with my suggestion. So we would still have four robots, one short, one long and both in slow and fast.

And to the checkers:
In my first thoughts the system with the four checkers was rather easy and intuitive. But as I thought a bit longer about it is more complicated than I thought. It has to be discussed how much abilities we should get there, how much checkers are allowed to set, and priorities could be handled. But I don’t want to include that yet, as it would make the suggestion maybe twice as long while I don’t know if the devs are actually interested in such an ornate suggestion.

[Boogieman14]

I like. I've occasionally wondered why inserters could only do that 180 thing and no 90 degree insertions. Also, lane selection would be nice.

[Khyron]

@OP
I'll be honest and say I didn't read your whole post. But I have read a number of posts like this. The general idea is: a device (inserter, logistics drone, chest, whatever) has some basic, hard-coded functionality which results in certain behavioural traits, such as inserters preferring to take items from the near side of a belt. The idea/suggestion is to enhance the device by introducing either a) player-configurable settings, such as option to take from the far side of the belt or b) alternate copies of the device, such as the addition of a 90° inserter.

On the surface, these ideas generally seem worthwhile because most players have encountered a situation where the default behaviour of a certain device causes a problem for the factory designer. However, most of these ideas & suggestions don't consider two important points...

• Adding new stuff creates a complexity barrier for new players. This becomes important when you consider just about every device in the game has already had a suggestion for one, two or three alternate modes or settings. If we implemented all these ideas a new player would end up with 100s of devices many of which do what appear to be similar things. Suddenly there's an excessive burden of choice for new players. A lot of posters usually respond to this problem by hiding their device/settings somewhere in the research tree or the menu/options. I think that mostly transfers the problem rather than resolving it.
• Conversely, the introduction of new modes and settings makes the game easier for experienced players. They already have a working knowledge of all the devices and settings, so they can just pick and choose the extra ones to suit their situation. That diminishes the puzzle aspect of the game.

So in most cases these kind of ideas make the game harder for new players and easier for experienced players. I would say that's the opposite of where the game should be heading in the long run. I'd rather have the game appear simple and accessible to new players but the more you play, the more design factors you become aware of which challenges your decision making.

Of course, the first and most obvious reason many of these requests go unanswered is that each would require development time, which means delaying some other feature like multiplayer. Add to that the introduction of new devices and settings creates opportunities for bugs within the device but also bugs as that device interacts with other devices or systems.

So put these three considerations in one hand and what you think the game would gain from your proposal in the other and see how it weighs up. It's not for me to say because, as I said, I didn't read your entire post.

[ssilk]

@Khyron: <bookmark>

@FatMcK: I think this suggestion is in that kind remarkable, that it is quite well thought through.

The problem I see with it: I need some concrete use case, which can't be made the current setup, where I would say "yes, this is so useful, that the game needs it". I doubt that there is one, which is also fun and easy ... I think it would be cool for a mod.

[silman]

DyTech Mod already introduces a little of what you are describing. It has craftable insters that only pick up on the near/far side and only place on the near/far sice (pemutations of each also). What it doesn't have is the 90 Degree placement and cant lane-pick when the belts are in-line with an inserter interrupting them. so in order to take advantage of this feature you need the belts to be not aligned. Also the fact that one must craft an entirely different inserter just to get this behavior is quite redundant to me.

I agree that it is rather lame that the inserters in game are pretty dumb, and I have never liked the North-West rule that factorio currently has (and minecraft has had for years), because it introduces proprietary-ness into the game. It makes it so you have to exploit the games inner-workings to get a factory working just right instead of having full control.

+1 i hope the devs see this thread and consider implementing something like this in the future.

[FatMcK]

@Khyron: <bookmark>

@FatMcK: I think this suggestion is in that kind remarkable, that it is quite well thought through.

Thanks for your answer.

The problem I see with it: I need some concrete use case, ..

Well, as I read this I spontaneously thought there are dozens, if not hundreds of cases..

.. which can't be made the current setup, ..

Here I started to think about concrete cases. I really thought long about, but well, currently anything is doable with the current setup. Anything that came up in my mind could be 'workarounded' with what we have now.

.. where I would say "yes, this is so useful, that the game needs it". I doubt that there is one, which is also fun and easy ... I think it would be cool for a mod.

So if it is usefull or not I think depends on what everyone individually expects of the game and how someone wants to play it.

I want to refer to Minecraft here. I played it long time and I really tried to do everything possible what redstone could offer. And well, with redstone stuff in Minecraft you could literally do anything, but just not in a practical way. If you have build a medium sized house and want to include some very simple redstone things like auto opening/closing doors, a door bell and a light switch you actual need to construct something out of redstone thats even bigger then the house itself. Lets don't even get started with a automated four-way train station. No matter what you plan to construct, don't even think about it being compact.

In Factorio we are luckily far away from that. But still, sometimes we have to do very tricky things to get things working, to get the solids on the belts how we need them. There are situations were you may think 'wow, found a nice solution for this', but actually I have more moments where I think 'k, it's working, but it looks terrible'. Situations where you need that steel 5 tiles away for you new assembling line, but we don't have the abilities to get it without a bulwark of splitters and undergound belts.

[CrushedIce]

I like your idea. I think everyone of us had a situation where we had to build some "workarounds" to get the items on the right side of the belt or something like that.
Of course this works, but it needs a lot of extra space and doesn't look that pretty (in my opinion)
I also think that it woudn't be too confusing for new players, because you could make it completely optional. By default, without touching the configuration gui, all the inserters could work just like they do now.
So i would like to see the developers think about such an idea maybe after the multiplayer update

[FatMcK]

I also think that it woudn't be too confusing for new players, because you could make it completely optional. By default, without touching the configuration gui, all the inserters could work just like they do now.

I think that I put thoughts in an userfriendly usability here. You don't have to 'make' it optional, it is optional. By default they work like the current ones.

Actualy the current smart inserter was the first thing I realy wondered about when I first used it in my first game. I placed it to see what it can, and it simply refused to do anything before I found out about its gui.

[silman]

I think it's dumb that inserters can only place on one side of a belt if its inserting from the side of the belt instead of the bottom.