programmable buildings (assembler, furnaces, plants)

[MeduSalem]

I love these ideas!! call it "Improvement Everywhere" and make another thread on it ^^
I would love to elaborate a bit

I don't think that the devs would really consider any of the many things I would have ideas on how to "generalize" or "improve" them because if they really did they would have done it that way a long time ago (or they skipped the ideas for other reasons they never mentioned and they are not really obligated to explain why they did some things they did), because some of these things I suggested at various points or have been discussed multiple times over, like for example the belt-thing.

Also if they really considered all of my ideas... it might as well not be called Factorio anymore because I think kovarex and his team have their own vision on how the game should look like which is totally right so, even if it is not to everyone's liking or leaves room for improvements.

That said I would code my own game with a similar, but yet different theme or write a mod but - and that's my biggest weakness - I don't really have the endurance do such a monstrous project despite my bazillion of ideas... I'm more the think-tank and problem-solver type... I'm easy with developing vast concepts, but I don't really want to be the guy who has to implement all that... I'm not a hatchet man, I'm not mentally stable enough to endure the ugly, dirty work that is necessary at times. Some may find fun in the programming part, but I'm more a designer, not a programmer, though I have to a certain degree of knowledge about programming.

residue fluids inside assemblers nowadays get deleted. Is it possible to pump fluids out of the assembler when it is not active? Ill have to experiment..

... Nah deleting the residue fluids is wasteful and an obvious nasty solution that begs for redemption or a coup de grâce. Might as well not enable recipe switching for that kind of recipes then if it boils down to an ugly solution as "delete resources".

Also I don't really know if one can pump fluids out of an assembler/chemplant when its not active, but I believe not... I think once it's inside the fluid is inside.

[Yoyobuae]

to conclude:
there is only one signal into the assembler for recipe change, say green wire, and one out of the assembler for the residue handling, say red wire
no feedback loops, no signal chaos

IMO, it's not a good idea to use green and red wire like that. It's inconsistent with all other circuit network connections. And red/green wire signals adding up is useful feature in other circuit setups (so not OK to change how red/green wire behaves globally).

Other option would be to have two connection ports for the signals (like with combinators). But that might a bit confusing at first (which is input and which is output?).

[MeduSalem]

Other option would be to have two connection ports for the signals (like with combinators). But that might a bit confusing at first (which is input and which is output?).

Lable it "CN IN" and "CN OUT" obviously...

I truly wish that the combinators had such a lable because often it's not clear with them too.

[Gertibrumm]

to conclude:
there is only one signal into the assembler for recipe change, say green wire, and one out of the assembler for the residue handling, say red wire
no feedback loops, no signal chaos

IMO, it's not a good idea to use green and red wire like that. It's inconsistent with all other circuit network connections. And red/green wire signals adding up is useful feature in other circuit setups (so not OK to change how red/green wire behaves globally).

Other option would be to have two connection ports for the signals (like with combinators). But that might a bit confusing at first (which is input and which is output?).

Thats not my point, sorry I meant different, I doenst matter on which of the wires the signals come and go.

In a practical example the incoming product/set-recipe signal would be on one line to the assembler and product-inserter, the residue-signal would be irrelevant on this line (use diodes to one-way the signal).
The assembler residue item signal would go to the filter inserter that exserts this residue

[Yoyobuae]

residue fluids inside assemblers nowadays get deleted. Is it possible to pump fluids out of the assembler when it is not active? Ill have to experiment..

... Nah deleting the residue fluids is wasteful and an obvious nasty solution that begs for redemption or a coup de grâce. Might as well not enable recipe switching for that kind of recipes then if it boils down to an ugly solution as "delete resources".

See, I would prefer if assembler/chem plant would simply jam in that case. Prevent jam is possible, a whole range of recipes being disabled is not (except by modding).

I, as a player, don't need the devs to keep me away from potentially problematic things. I just need to have options, I can decide for myself if an option is worth it or not. And then we (the players) can have endless discussions about which option is best.

Thats not my point, sorry I meant different, I doenst matter on which of the wires the signals come and go.

In a practical example the incoming product/set-recipe signal would be on one line to the assembler and product-inserter, the residue-signal would be irrelevant on this line (use diodes to one-way the signal).
The assembler residue item signal would go to the filter inserter that exserts this residue

If set-recipe signal can be isolated from residue item output signal from the same assembler then it's all good. The two wires would only be used to isolate the residue exserter from the set-recipe signal.

Still, need to isolate the set-recipe signal from the residue output, since that signal needs to be connected to the normal output inserter. That's easy to do with a buffer combinator, though.

Other option would be to have two connection ports for the signals (like with combinators). But that might a bit confusing at first (which is input and which is output?).

Lable it "CN IN" and "CN OUT" obviously...

I truly wish that the combinators had such a lable because often it's not clear with them too.

The detailed info (ALT key) is clear enough, IMO.

[MeduSalem]

See, I would prefer if assembler/chem plant would simply jam in that case. Prevent jam is possible, a whole range of recipes being disabled is not (except by modding).

I, as a player, don't need the devs to keep me away from potentially problematic things. I just need to have options, I can decide for myself if an option is worth it or not. And then we (the players) can have endless discussions about which option is best.

Okay well I can live with the case of "jamming" too... Then it's my task as a player to consider not using these recipes if they may cause jamming.

[mrvn]

residue fluids inside assemblers nowadays get deleted. Is it possible to pump fluids out of the assembler when it is not active? Ill have to experiment..

... Nah deleting the residue fluids is wasteful and an obvious nasty solution that begs for redemption or a coup de grâce. Might as well not enable recipe switching for that kind of recipes then if it boils down to an ugly solution as "delete resources".

Also I don't really know if one can pump fluids out of an assembler/chemplant when its not active, but I believe not... I think once it's inside the fluid is inside.

If the pipe is still connected to the assembler then the fluid could simply flow back into the pipe till it is full (which might already happen). Anything left after that would still be lost unless changing recipes takes several ticks so the fluid network can carry away excess fluids. It's similar to removing a fluid tank. If connected to another fluid tank the fluids are transferred. Any overflow is lost.

Maybe overflow could create a puddle on the ground that slowly seeps away.

[mrvn]

On the subject of having multiple connections:

An assembler is 3x3 tiles big. One signal could be located in the center tile and the other on the outside. That way it would be clear which is which no matter how the assembler is rotated.

[MeduSalem]

Maybe overflow could create a puddle on the ground that slowly seeps away.

Yeah, dump the chemical waste and stuff into the wilderness, the biters need a replacement for all the water sources we are filling with concrete anyways.

Somehow I had to think about "The Animals of Farthing Wood" now. Sad memories.

An assembler is 3x3 tiles big. One signal could be located in the center tile and the other on the outside. That way it would be clear which is which no matter how the assembler is rotated.

I could live with that. Otherwise we would have to use the ALT overlay to make it clear.

[Yoyobuae]

An assembler is 3x3 tiles big. One signal could be located in the center tile and the other on the outside. That way it would be clear which is which no matter how the assembler is rotated.

I could live with that. Otherwise we would have to use the ALT overlay to make it clear.

Didn't we already conclude it wasn't necessary to have two connection points? As long as the output doesn't feedback to the input a single connection point is sufficient.

IMO, adding multiple connection points is an ugly solution unless the entity really has clearly defined input/output role (ie. combinators).

[MeduSalem]

Didn't we already conclude it wasn't necessary to have two connection points? As long as the output doesn't feedback to the input a single connection point is sufficient.

IMO, adding multiple connection points is an ugly solution unless the entity really has clearly defined input/output role (ie. combinators).

*scolls up, reads again*

You mean with Gertibrumm's idea of using either the red OR the green wire to write the Set-Recipe-Signal and the other wire to read the residue output?

I understood it so that you both disagreed on that since you were on the stance that using both wire colors to do something different isn't a good idea since everywhere else in the game red/green gets added up instead.

[Yoyobuae]

You mean with Gertibrumm's idea of using either the red OR the green wire to write the Set-Recipe-Signal and the other wire to read the residue output?

I understood it so that you both disagreed on that since you were on the stance that using both wire colors to do something different isn't a good idea since everywhere else in the game red/green gets added up instead.

I later mentioned that the same wire could carry both signals if assembler didn't feed back it's own output signals to the set-recipe logic (which won't ever make sense, unlike inserters and such).

[MeduSalem]

I later mentioned that the same wire could carry both signals if assembler didn't feed back it's own output signals to the set-recipe logic (which won't ever make sense, unlike inserters and such).

Ah yeah, I seem to have missed out on that part, my bad... But I would find it somewhat nasty to have to filter the set recipe signal out of the residue signals so that I don't accidently take the finished items out of the assembler with the inserter that handles the residuals... it's like I can already imagine people would be moaning and crying about "why can't they be seperate, why do I have to put that additional combinator there". I might be one of them.


But I think you get why I'm personally thinking Red and Green networks shouldn't be summed up by default except if you explicitely state an item to do that. In the case of above the seperation would have come in quite handy as you would only have to say "read recipe signal on Red wire and write residual items to green wire" or vice versa.

That said an option to "restore" the old pre-0.13 behaviour of treating the networks as seperate would be worth its own suggestion. Where you can hook up Red and Green wires to the same combinator and it does whatever arithmetic/comparison seperately for each wire color and then writes the individual outputs back to Red and Green respectively without any information crossing over from one network color to the other and vice versa... or you can even specify that both outputs are than added up before output (like a complete reversal of the current behaviour of adding the wires before the input). Or even do a crossover like do arithmethic on green wire but output to Red while in parallel doing arithmetic on red wire but output to green. Would surely open up several more possibilities or make existing contraptions easier.

[Yoyobuae]

Ah yeah, I seem to have missed out on that part, my bad... But I would find it somewhat nasty to have to filter the set recipe signal out of the residue signals so that I don't accidently take the finished items out of the assembler with the inserter that handles the residuals... it's like I can already imagine people would be moaning and crying about "why can't they be seperate, why do I have to put that additional combinator there". I might be one of them.

Just use green wire for set-recipe input and red wire for reading residue signals. The set-recipe input won't cross over into red wire (that's what wire colors do). That happens without any new special rules for wire colors.

The only place where the two signals would really mix is in assembler itself. But the assembler should be able to tell appart it's own output signals.

It's just like enabling/disabling an inserter with some virtual signal while at the same time reading hand contents. If two wire colors are used the input virtual signal wont cross over into inserter output signals.

[MeduSalem]

Just use green wire for set-recipe input and red wire for reading residue signals. The set-recipe input won't cross over into red wire (that's what wire colors do). That happens without any new special rules for wire colors.

The only place where the two signals would really mix is in assembler itself. But the assembler should be able to tell appart it's own output signals.

It's just like enabling/disabling an inserter with some virtual signal while at the same time reading hand contents. If two wire colors are used the input virtual signal wont cross over into inserter output signals.

AHHH. *facepalm

So basically the combinator would be only on the Set Recipe Signal input... so that the Residual Item signals don't travel back up where the Set Recipe signal is coming from (because it might be otherwise a problem for other assemblers who get the same Recipe signal). No need for filtering then of course... yeah, now that makes sense.

[Gertibrumm]

It is funny that I dont really play factorio the way intended but instead am waiting for this feature since proposed for 0.14 …
I am still preparing a better CPU and better layouts for my sequencer factory

[golfmiketango]

I think a more interesting and amusing solution to this would be to implement the ability of an inserter to reverse direction based on criteria from the circuit network -- in other words, to rotate itself 180 degrees on demand. This would also allow certain things to be done with trains that are not currently possible.

[ssilk]

It is funny that I dont really play factorio the way intended but instead am waiting for this feature since proposed for 0.14 …
I am still preparing a better CPU and better layouts for my sequencer factory

I would not bet on that feature.

[Gertibrumm]

It is funny that I dont really play factorio the way intended but instead am waiting for this feature since proposed for 0.14 …
I am still preparing a better CPU and better layouts for my sequencer factory

I would not bet on that feature.

At least there is a mod for it

[voxel]

Hey

I posted a new thread here with some of my ideas: viewtopic.php?f=6&t=38783
I didn't want to derail this thread too much (though if a MOD feels it should be merged with this thread feel free to do so).