#I fear that this might just move the
1 messages · Page 1 of 1 (latest)
The way my brain approaches this problem is to only generate a random signal when you want to overwrite, because otherwise you would need a way to prioritize signals
If you generate a new signal every tick, but only allow it to pass through a decider while the reset signal is true
it takes the random things and when the green checkmark comes in it passes whatever it's on right now for a single tick
the issue is that the memory doesn't get reset, but only overridden with an item that's higher up in the item list
dunno, maybe it's prioritizing item IDs or something
I just need a way to properly reset the memory
can you show me how you are generating the random signal?
could be biased
it's in the video
the selector combinator
it gets a list of items
and then picks one at random
that's out of my hands
if it's biased it's the selector being biased
Yea my b thats probably not the problem
the problem is the memory
it's taking an anything signal, and if an item comes in with a higher ID it overrides it because it's prioritizing that in picking the the anything signal
would be my guess
I don't know how it's selecting the anything
but that doesn't matter, I just need to clear the output and let it accept a new signal from the decider before it
and that's what I'm struggling with
okay so your memory cell is the bottom combinator on the left side there
yes
what if you added a condition that didn't save what is being saved when the reset is given
right now all i see is * > 0 as a condition
if you added the reset then it would erase the memory right?
yes
green* > 0 and reset = 0
Don't know if im still not understanding your problem but if you just erased the memory when adding a new signal it sounds like that might fix it
in like an hour ill be able to test this on my machine
just ping me then I'll ask one more time in the channel just to see if someone picks it up, just in case, but yeah I appreciate the help
@zenith aurora Is this it?
sorry for weird glare hdr monitors dont record well
@zenith aurora
that's bright 
one sec lemme check
it does looks like it, I don't see what's different in your system tho
your memory cell is saving t he value, but mine isn't
weird
maybe try toggling all network filters to on?
mine works without filtering out red or green anywhere
OH
I got it!
I had another decider inbettween the assembler and the memory to filter out the green signal and that must've delayed the signal by a frame and caused it to do it wrong
I fucking LOVE how finicky circuits are
okay one last issue tho
how do we get the system started 
so here's the issue
the memory output sets the recipe of an assembler, but that assembler never starts without this input, so it never sends the finished checkmark
seed it with a constant combinator and toggle the const off after its running?
depends on how often you turn it on and off I g
u could automate turning it on with a clock
that's what I thought, but it seems a bit finniky if the system shuts down it won't start up automatically
yeah kinda
1 sec
so this is the system when there's nothing do do and it gets a new order in
oh wait
how do I make the clock?
but it can be interrupted by the assembler finishing something
put factorio on non hdr monitor lol
basically the constant combinator adds 1 to a memory cell every tick
memory cell erases data when your value is at given number 1800 tics = 30 seconds
the recipe finishing also resets the clock
Oh shit I think I have it
and the bottom decider will actually send the reset symbol to the system that randomly selects and stores your recipe
I made it a bit different
I have global clock that counts to 120
and it sends a heart signal
this heartbeat kickstarts the recipe if the system is idle
and since it's a global signal all machines get it
and because it's not the checkmark it doesn't interfere with the resetting of the recipe
it works but doesn't solve the issue it was supposed to solve 
oh well thank you so much anyway