Remove or Miniaturize Circuit Networks

[ssilk]

A) Wube has a community manager (Klonan).
B) "No, because..." is quite different from "No".
C) 600 hours/60 day = 10 hours per day. With weekends and so on. O.K. that explains all.

[OdinYggd]

I'd agree with minaturization. Even in real life, a simple if X > Y do this process decision was usually performed in a space smaller than a shoebox, although high power and high pressure applications would need more room than that to house the actual switching/valving hardware.

While early digital computers were room-filling monstrosities that had less computing power than a modern calculator, before even those were around we had industrial automation based on electromechanical solutions and crude analog circuits that would have operated on levels comparable to the factorio combinators. Things like generator field regulators from the 1930s, that operated by a mix of electrical and mechanical means to perform a surprisingly sophisticated task in an operating panel footprint

Quite a few times now I've offered suggestions for adding functionality to the combinators, or revisiting how a particular function of them was implemeted. I think the hardest thing about them is getting started, once you understand the basics they can be logically mapped to real world equivalents and building with them becomes easy. At the same time, the lack of circuit density is problematic in a lot of situations, and expanding them to provide PLC-like functionality would be a big help for a lot of people. I've seen a lot of graphics programming interfaces for PLCs designed to be super intuitive, such a thing should be possible. But to do anything of this nature we would probably need more circuit wire channels, at least give us 8 channels to work with instead of only 2.

Removing the circuit stuff? No thank you! I use that constantly to streamline my operations, usually just by connecting a box to an inserter to regulate the quantity of a material in the system. And if you are playing without logistics bots, that is the only way to perform such material flow regulation.

[siggboy]

the lack of circuit density is problematic in a lot of situations

I think it just gets plain ugly after a certain grade of circuit complexity. Any circuit that uses more than a handful of combinator looks awful in the game world. It just doesn't fit in any longer. The wiring gets really messy too, and it's extremely difficult to rearrange things.

Being able to put it into a small "box" and then have a nice interface to play around with the basic elements would be a lot more fun.

[Harkonnen604]

Wanna factorio OPC server?

[Aatch]

How about an official response to this idea/suggestion:

No.

Seriously? If you're not going to elaborate, why even bother with the response in the first place? Also, there are two suggestions here, is that "No" for both of them or just one? Did you look through the rest of thread or is this just a response to the initial post (which I do disagree with)?

This response is dismissive and borders on condescending. I understand how annoying poor suggestions can be, but the second post in this thread lays out what I consider to be a pretty reasonable idea. If you disagree with that, then, please, elaborate.

[Prymal]

Klonan already gave the "no, because" explanation back in March; check page two of the thread. Perhaps the response from Rseding91 was to help clarify the "no" part for those who are still arguing.

[siggboy]

Perhaps the response from Rseding91 was to help clarify the "no" part for those who are still arguing.

Supposedly, the only real reason for his response was that he was venting. I don't think it was to clarify anything (it's difficult to actually "clarify" something in one word).

[Rseding91]

Perhaps the response from Rseding91 was to help clarify the "no" part for those who are still arguing.

Supposedly, the only real reason for his response was that he was venting. I don't think it was to clarify anything (it's difficult to actually "clarify" something in one word).

No venting. Just trying to get everyone to stop posting since we won't do it anyway.

[siggboy]

No venting. Just trying to get everyone to stop posting since we won't do it anyway.

And now we're still posting and talking about you telling us to stop posting. Oh the irony

[Klonan]

Can we please stay on topic,

Try to convince us why we should remove or miniaturize the circuit network,
We do listen to all ideas (even ones we don't agree with), and maybe you will inspire us

[OdinYggd]

Removal would be awful, its useful in so many ways and needs to be expanded upon with more wire colors, more signal channels to use on a wire (color coded signals would be huge for increasing functional density), and minaturization of the components. Quite simply it makes no sense that an inserter contains a functional decider plus its inserter functionality in a single tile while a decider combinator takes up 2 tiles.

Miniaturizing the components to a level where instead of just math or logic operators you have instead 1 tile Microcontroller units and Programmable Logic Controller units would really bring the circuit density much closer to being in line with the scale of factorio, as well as greatly simplifying the often tangled and numbingly complicated circuitry needed to perform complex automation tasks.

During that same redesign, it should be plenty possible to make the whole system a great deal more intuitive to work with. The present system is not very helpful in that regard, and I think a lot of players avoid using it or do not use it to its full potential because it is difficult to understand and time consuming to configure.

This thread has indeed been going on for a few months though, by now some of the earlier replies probably aren't as relevant as they were at the time because the way people play the game might have changed. Perhaps it would be more to the point to lock this thread and replace it with a new one to revisit this topic, carrying over ideas presented in this thread while getting rid of previous bad feelings and opinions attached.

[Engimage]

I am totally against removing the circuit network. It extends gameplay and makes a lot of room for creativity.

However miniaturizing is a good idea as anything more complex than a handful combinators becomes a real mosstrosity.

However moving circuits into a "black box" makes much sense. It still gives you a space for creativity but does not require map space for that. And ofcourse a limit of 2 output wires is a real pain. My idea would be creating a universal signal cable (which would be multicolored to differentiate it from common cables) and on network connector you should just pick a number of wire to connect to.

I have seen a mod for such thing at some point where you could pack some parts of your factory into a black box but can't find it now. But the idea is the same.

I do understand that initial idea of Factorio as a game is to give you a single area for creating and mixing everything. But I have to agree that combinators make your creations look unpleasant at best.

[siggboy]

Most of the good points have been made already, even just now by Pacifyer and Odin or earlier in the thread.

Remove the circuit network? Of course not, are you kidding me.

Miniaturize? Yes, please.

Reasons:

• Combinators do not fit the theme of the game. They're simply too big. It's completely immersion breaking. Odin made the point in more words and very eloquently so I won't repeat it here.
• From a practical game-play perspective, as soon as you make something which is actually complex, i.e. does execute an algorithm, the "user interface" of the combinators becomes utterly inadequate. It's not even possible to move combinators around. In order to clean up or refactor your design you have to rebuild and rewire, it's very error prone and time consuming. Yuck.
• In many cases, some really compelling uses for combinators are not practical because they use so much space. For example, in a Smart Furnace setup you usually want to have good beacon coverage around the furnaces. There simply is no space to fit in multi-combinator contraptions, but there are some clever setups that could use them.
• User Interface. That's the big one. If you made a "combinator box" entity, which then actually contains the circuit, you'd have a lot more freedom with designing an adequate user interface for creating circuits. It would also be a bit more compelling to add more circuit "entities" to the game, because those wouldn't be entities that the player had to craft, they'd only exist as virtual "circuit elements" inside the circuit box.

The basic combinators could still be a part of the game, for executing very simple tasks. They should all be only 1 tile big, though, like the constant combinator.

[Harkonnen604]

I suggest a new forum thread is opened just to discuss blackboxing of circuits. Just because name of this one is evil, as the thread itself has become evil too

As for my personal opinion for blackboxing - I'm for it. But it should still put some challenge on placing combinators (black-boxes) and not be able to hide A LOT inside. So I am for blackboxing of latches, timers, counters and other fundamental contraptions, but not the whole core-i7 inside

Actually this can work down to have combinators occupy just 1 tile, and require no power (get those volts from the bus).

[OdinYggd]

I suggest a new forum thread is opened just to discuss blackboxing of circuits. Just because name of this one is evil, as the thread itself has become evil too

As for my personal opinion for blackboxing - I'm for it. But it should still put some challenge on placing combinators (black-boxes) and not be able to hide A LOT inside. So I am for blackboxing of latches, timers, counters and other fundamental contraptions, but not the whole core-i7 inside

Actually this can work down to have combinators occupy just 1 tile, and require no power (get those volts from the bus).

What I had in mind for microprocessor capabilities would be on the level of the Arduino. Simple sequential logic programs that change outputs in response to inputs, with token amounts of ram and storage just so that it can actually model real world programming behaviors.

For PLCs, these would use the standard ladder logic notation. There would need to be limits on the number of lines and the total number of virtual contacts implemented, but I haven't yet figured out where these limits should be in terms of size and complexity because after a certain point a microprocessor does it better anyway.

[siggboy]

But it should still put some challenge on placing combinators (black-boxes) and not be able to hide A LOT inside.

Why should there be limits on how much you can put inside the "black box"? If you have an application for a very large circuit, and you're able to implement it, you should be able to do so; as you are now.

The limits should only be posed by the user interface. At some point it would become unmanageable, but in principle the game should allow circuits as big as you want.

I also don't see where the "challenge" in placing the combinators should lie, that you're talking about.

[Harkonnen604]

Well, then there can be several tiers of blackboxes with biggest occupying like like 4x4 tiles outside and 16x16 tiles inside (off-the-wall numbers).

[siggboy]

Well, then there can be several tiers of blackboxes with biggest occupying like like 4x4 tiles outside and 16x16 tiles inside (off-the-wall numbers).

I still don't understand why you'd want to restrict it in this way. The limits of combinator designs are not the footprint of the combinators, it's the ability of the player to create complicated circuits.

Right now the pleasure in making a design smaller is not that you can then more easily deploy it -- it's only because optimizing low-level implementations is a feat.

In the cases where combinators are not practical due to size considerations the limits that you suggest wouldn't make a difference. They'll only, again, needlessly put constraints on the players who want to make really complex designs.

The point of the miniaturization would be to remove those constraints, not introduce new ones to take their place.

[ssilk]

Removing is no option.
Which reduces this to miniaturizing. I think needed space is not an issue, cause we really have enough space to create big circuits.

I think the circuits lack some points, which brings obviously players to think about this suggestions (which doesn't solve the problems btw in my eyes).
The things are:
- Needed time for wiring
- Needed time for configuration
- Size/Clarity in overview
- Debugging
- Predefined circuits.

I explain the points.

Needed time for wiring
- you need to have the right wires. Don't have them: Craft them. Or order them. In every case the wires took up inventory space and time to craft/configure request.
- The wiring itself is much better with 0.13, but still not perfect enough. I cannot say exactly, what to improve, but I think for example to highlighting of connection points. Or intelligent autowiring.

Needed time for configuration
- The configuration feels still a bit clumsy. A normal configuration needs 6 clicks minimum. 10 is quite normal.
- I need to run and run and run if the setup is spread over a bigger area (cause it makes sense). BTW this is also a big point against the miniaturization, because it doesn't solve this problem.

Size/Clarity of overview
- One point is that it is still difficult to see what is connected with what. Much better with 0.13. But I think I don't like the game, when fully zoomed in, but when working on circuits that is useful/needed.
- I think I would like to concentrate to the devices connected to a circuit only. The things around confuse me.

Debugging
- You search for a bug in your circuits and it is sometimes quite nasty to find it.

Predefined circuits
- some set of blueprints that displays always repeating tasks.


Some ideas as consequence of above:
1. No wires/cables needed. Wires/cables can be zipped out of the devices with some kind of tool/mode. Works also for electric.
2. Easy cut and rearrange wires. The behave more like real wires. For example: I drag on one end, until the connection is loosed and I can plug it then into another device.
3. Much longer (endless?) reach for combinator/wire-setup.
4. Some kind of shadow that blends out unconnected devices (more or less dark, depending of level of connection). See the circuits and the connected devices only. See point 1.
5. The view changes when zooming into combinators: You can see then the internal configuration. This depends on some kind of view-mode perhaps. See viewtopic.php?f=80&t=18153 Information-Layers (for map and/or normal view)
6. More than one way to configure the combinators/inserters etc. For example some kind of command-line. Much more fine-tuning of that surface.
7. Intelligent wiring. Hm. This belongs to point 4. I have a set of devices, that belongs to this circuit. Now I add a device and this needs of course to be connected to that circuit.
8. I can make graphical boxes around circuits and name/describe that. A useful feature for much more cases than only this.
9. Predefined Blueprints is based on 8: When printing them, the description is also printed.
10. A debugger: Copy all the devices in a circuit and depending of that into an own view. The network is automatically uncluttered in that view, the devices are rearranged in a nice grid. And then I can look at each device/wire, down to a step-by-step mode, what happens.

Just ideas. Nothing seriously needed or could be achieved in a different way (see this suggestion).

[British_Petroleum]

Removing is no option.
Which reduces this to miniaturizing. I think needed space is not an issue, cause we really have enough space to create big circuits.

Yes there is plenty of space to build giant combinator setups in your factory, however then if you want to build things either side of this setup, you then have a really long boring work, whereas if you can miniaturize your combinator setup, it is now a short walk