combinator counter
combinator counter
the combinator counter adds positive inputs even if the condition is "signal < 0". although I have use for this and this option needs to remain, its confusing for creating network conditions. "unequal" seems to be more fitting in this condition and I do think this needs to be implemented as well.
Re: combinator counter
I am afraid the bug description is too chaotic.
Please describe what entity is connected to what exactly and what the inputs and outputs are specificly, best use screenshots to do so.
Please use the Action > Result > Expectation method to structure your post.
Please describe what entity is connected to what exactly and what the inputs and outputs are specificly, best use screenshots to do so.
Please use the Action > Result > Expectation method to structure your post.
Re: combinator counter
I basicly set up a counter.
the filter inserter gets the filter command from the constant combinator to the right minus the item count in the chest (steel icon).
the filter inserter scans the incoming items, gives the signal via red cable to the arithmetic combinator. the signal gets multiplied with -1 by the arithmetic combinator and is sent to the input of the compairing combinator via red cable. so i get a negative impulse.
the blue fast inserter gives positive item counts to the compairing combinator output and input via green cable.
the condition for action in the compairing combinator is ""Pro signal" < 0" (german version). The signal from the filter inserter (from the left) meets this requirement, because the signal gets turned from positiv to negative "-1 < 0". the signal from the fast inserter (from the right) does not meet the requirement "item < 0" as it is "1 > 0". yet the signal gets used as well (which isnt bad for what I intended to do, but its wrong on the condition). at the end of the test the belt is empty and the counter is 0, which is what I wanted, but is wrong on the logic set up (and it confuses me). the counter should have stayed at -50 (I used 50 test items each of steel, iron and cooper)
(the way it works right now is like "item unequal 0" and not "item < 0) So either way this is intended and the condition sign is wrong or there is something strange going on in the background, that also positive signals get used.
I hope the picture gives you everything u need.
the little black boxes in the picture were done with paint to filter out any unnecessary objects for the problem.
the filter inserter gets the filter command from the constant combinator to the right minus the item count in the chest (steel icon).
the filter inserter scans the incoming items, gives the signal via red cable to the arithmetic combinator. the signal gets multiplied with -1 by the arithmetic combinator and is sent to the input of the compairing combinator via red cable. so i get a negative impulse.
the blue fast inserter gives positive item counts to the compairing combinator output and input via green cable.
the condition for action in the compairing combinator is ""Pro signal" < 0" (german version). The signal from the filter inserter (from the left) meets this requirement, because the signal gets turned from positiv to negative "-1 < 0". the signal from the fast inserter (from the right) does not meet the requirement "item < 0" as it is "1 > 0". yet the signal gets used as well (which isnt bad for what I intended to do, but its wrong on the condition). at the end of the test the belt is empty and the counter is 0, which is what I wanted, but is wrong on the logic set up (and it confuses me). the counter should have stayed at -50 (I used 50 test items each of steel, iron and cooper)
(the way it works right now is like "item unequal 0" and not "item < 0) So either way this is intended and the condition sign is wrong or there is something strange going on in the background, that also positive signals get used.
I hope the picture gives you everything u need.
the little black boxes in the picture were done with paint to filter out any unnecessary objects for the problem.
- Attachments
-
- Screenshot für Forum Bugreport.png (2.44 MiB) Viewed 5365 times
- TruePikachu
- Filter Inserter
- Posts: 978
- Joined: Sat Apr 09, 2016 8:39 pm
- Contact:
Re: combinator counter
EDIT: The "+" combinator is adding zero, each/each, right?
All combinators add together their RED and GREEN inputs before doing whatever operation is desired; this currently appears to be Not a Bug.
Re: combinator counter
the arithmetic combinator on the right to the compairing combinator just adds 0 in order to split the network appart. the signal then gets sent to the chest and constant combinator. the constant combinaror, minus the counter signal and minus the chest signal, sends the demand signal to the filter inserter, which sets the filter for those demanded items. the items on the belt and in the chest are counted together and subtrackted from the constant combinator. if the demand is met, the count will go to 0 and the icon in the filter will disappear so that the filter inserter stops grabbing those items. like that I intend to make on point delivery without drones.TruePikachu wrote:It might be useful to attach the save file, especially since there are wires from the network in question leading out of what's visible. Also it is somewhat hard to tell what's going on, at least for me.
EDIT: The "+" combinator is adding zero, each/each, right?
All combinators add together their RED and GREEN inputs before doing whatever operation is desired; this currently appears to be Not a Bug.
the filter inserter is supposed to give a negative item impulse and the fast inserter is supposed to give a positive item impuls. the belt in between the inserters is supposed to act as a temporary transporting chest, where every item input and output is recorded (the goal is to divert items of the exact number to the requested chests. in order to do so i have to know the number of items in every place at all times (if I check the item count at the power pole it doesnt matter for me if the item count is neative or positive as long as I know what the count is right)
the thing that confuses me is the condition for the compairing inserter "item < 0". "item < 0" is supposed to work with negative impulses, because negative < 0. yet, the positive impulses get used as well, which is the condition "item > 0", which isnt the shown condition of the combinator.
since the inserters only sent impulses, the "adding together before the combinator" shouldnt happen and if it does 1 - 1 = 0 doesnt affect the count, which is good, because it means 1 item left the belt and 1 item reached the belt.
The condition by which the combinator works right now would rather be "unequal 0" but not "item < 0". This is not unimportant, since these conditions need to do what they show. It can generate mistakes, that can only be found by doubting the game code, which isnt what id like to do.
- TruePikachu
- Filter Inserter
- Posts: 978
- Joined: Sat Apr 09, 2016 8:39 pm
- Contact:
Re: combinator counter
What happens if you remove the GREEN wire between the < and +, and replace it with a RED wire? As it stands in the screenshot, pulses from the fast inserter will reach the + combinator.
Re: combinator counter
I can't understand what the issue is. Your explanations is confusing. It appears that maybe you are misunderstanding how the decider combinator in the middle should work? You created a green loop from output to input. The loop will only hold negative signals because that's what the condition says. Adding a positive signal of 1 will increase the looped signal by 1 and the looped signal will be kept in the loop if it's still less than 0
Can you upload the save please? Will make it much easier for me to reproduce and understand the issue.
Can you upload the save please? Will make it much easier for me to reproduce and understand the issue.
Re: combinator counter
...or remove all parts that are not necessary to show the issue.
Re: combinator counter
Twinsen wrote: It appears that maybe you are misunderstanding how the decider combinator in the middle should work? You created a green loop from output to input. The loop will only hold negative signals because that's what the condition says. Adding a positive signal of 1 will increase the looped signal by 1 and the looped signal will be kept in the loop if it's still less than 0
Can you upload the save please? Will make it much easier for me to reproduce and understand the issue.
well I guess you got the issue. my mistake was that I thought the decider would permit only new negative impulses. I left out the addition of the impulse signal and the loop signal before the decider makes the decision. sorry for the fuss. the blank adding of 2 values just by combining 2 wires on a pole confuses me at times. sorry, made a fuss about unnecessary stuff.