Minecraft Forums

I tried implementing your Flip-Flop in my world and found that I really need to memorize it. Truth be told, I still don't fully "understand" the design; which parts do which things. I dissected it in the simulator and so I understand the principle it's operating under, but I can't quite look at any given part of the gate and know what it does and where it should connect to when I'm constructing it. This has made debugging it...difficult. I've assembled 2, and the first one worked ok, but the second one...has failed me and I'm not quite sure why. However it's an understandable difficulty given that I was trying to construct it in a mirrored form.

I may resort to making a sort of "hybrid" flip-flop that's based on a more intuitive (for me) design, but incorporating some of the miniaturization techniques that you've used.

By the way, does anybody know how many ticks a minecraft day is? I need to get an idea of how many counters I'm going to need for my clock.

Ticks deppends on your fps

I am building a simple prison system. Each door have a switch that opens closes. All doors are connected to a master switch that overrides the individual doors, and closes all the doors. My basic layout seems like this.



I know I sort of have to use AND gates, but I can not really get it... Sigh

Now the brown pieces are single doors with stone between them. Just for curiosity would it be possible to make the doors work in this fashion?
Stepping on a pressure plate opens the door, and keeps it open, until you hit the plate again. Or pushes the button on the other side. If the door is closed. Pushing the button will open it, and keep it open until you either hit the button again, or walk over the plate.

I know this has something to do with flip flops and states. But I cant work my head around it. Atleast not in the space i have.

Any help will be greatly appreciated.

hi um i need positioning help on a long row of doors z+1 level of operating/house level. as in the set of doors will turn into a wall when switch is on.


[]
[]

something roughly like that for the door level, it'll be longer but i just need a startup example.

Quote from Hans Lemurson »

I tried implementing your Flip-Flop in my world and found that I really need to memorize it. Truth be told, I still don't fully "understand" the design; which parts do which things. I dissected it in the simulator and so I understand the principle it's operating under, but I can't quite look at any given part of the gate and know what it does and where it should connect to when I'm constructing it.

I made this color coded picture comparing it to the expanded T-flipflop on the wiki page. It might help you visualize how it works more easily. They both operate the same way. (except the T input for the one on the right is actually the C input)

Can anyone explain how "charges" flow through wires when they come to junctions, such as a 4 way junction?

I don't understand why the flow goes "that way" instead of "this way"

Like in VincentLaw's post above me.
On the bottom right picture, why doesn't the light blue/cyan line, from 3, go to Q instead of going to 2?
Or if it's traveling from 2 to 3, why doesn't it just turn to Q?

Why doesn't the purple do that?

It seems like having a 4 way junction like that should just send a signal in all 4 directions which can just screw up the circuit's output

Quote from VincentLaw »

Quote from Hans Lemurson »

I tried implementing your Flip-Flop in my world and found that I really need to memorize it. Truth be told, I still don't fully "understand" the design; which parts do which things. I dissected it in the simulator and so I understand the principle it's operating under, but I can't quite look at any given part of the gate and know what it does and where it should connect to when I'm constructing it.

I made this color coded picture comparing it to the expanded T-flipflop on the wiki page. It might help you visualize how it works more easily. They both operate the same way. (except the T input for the one on the right is actually the C input)

I've had trouble using the more compact t flip flops. If the orientation changes the toggle will not work. I had one structure facing south and one facing west. The toggle in the north/south works fine but the same toggle in the east/west one does not. However the less compact version has no issues in either structure. It took me forever and several re-builds to determine the cause of the problem. This is due to the processing order that has been discussed by Baezon and one other guy (can't remember who off hand)

Quote from Ydoow »

Can anyone explain how "charges" flow through wires when they come to junctions, such as a 4 way junction?

I don't understand why the flow goes "that way" instead of "this way"

Like in VincentLaw's post above me.
On the bottom right picture, why doesn't the light blue/cyan line, from 3, go to Q instead of going to 2?
Or if it's traveling from 2 to 3, why doesn't it just turn to Q?

Why doesn't the purple do that?

It seems like having a 4 way junction like that should just send a signal in all 4 directions which can just screw up the circuit's output

You are partly right, the color separation is just for visualization purposes. In the picture on the right, the wire that shares the purple and the cyan has the same signal. Wires don't have a flow direction, they just switch between two states, either on or off. If any piece of wire is on, all of the connected wire within 15 blocks of the torch that turned it on is also on. So at a junction, the signal goes every way. Logic gates are the only way to enforce a direction. you can think of every redstone torch in a circuit as a Not gate, which has an input that is only an input, and an output that is only an output. This forces signals to "flow" in the direction of the outputs, instead of going backwards.

Quote from Sven_The_Slayer »

I've had trouble using the more compact t flip flops. If the orientation changes the toggle will not work. I had one structure facing south and one facing west. The toggle in the north/south works fine but the same toggle in the east/west one does not. However the less compact version has no issues in either structure. It took me forever and several re-builds to determine the cause of the problem. This is due to the processing order that has been discussed by Baezon and one other guy (can't remember who off hand)

When you make things more compact glitches due to the design are more likely to be present. There is a specific combination of two not gates reversing direction that makes this problem most severe by losing a tick (I posted details on page 41 of this thread). The original design of this T-flipflop had the orientation issue you describe due to this, but the latest design was built this in mind and has been tested at both N/S and E/W orientations. It functions normally either way and can be found on page 42 or quoted on page 44. I do not recommend building any of the T-flipflop versions that you find on page 41.

Smaller circuits inherently require more thorough in-game testing.

Thanks Vincent. That makes a lot more sense.

I always imagined electrons flowing through the wires and turning wherever there was a junction. I thought that these circuits worked off of the flow of electrons turning at specific points in a junction.

So in a sense, I imagined the "charge" 0 or 1 moving either left or right (or at least that's how it seemed, to me, when I looked at diagrams such as yours).
It's a lot more logical, easier, and more correct to imagine all of the wire being in an equal state of "charge" with gates changing the state of the wires thereafter.

That's gonna help me a lot in designing new stuff.

Also, I read that signals won't be carried over 100 blocks (unless in online mode and players are near the wire) is this true?

So I can't have buttons that change minecart track direction from really really far away?

This is true. The minecraft wiki says its about 300 blocks, but I've never tested the actual distance to find out. Redstone will only function in chunks that are currently loaded, and since they load 16 blocks long at a time, there will not be a single precise number from where you are standing at any given time. I'm sure people have gone into more detail on signal distance previously in this thread.

A question that I hope someone can help with:

I'm trying to design a combination lock where a series of buttons must be pressed a certain number of times each to gain entry. I'm using a series of T-flops as my binary counter. (Specifically, the 6x5x2 one designed by VincentLaw).

What I want to do is add a reset button that will clear the counter's total. I believe what I need is a way to toggle each T-flop so that its output is off, but I don't know how to do that without already knowing the output of each individual T-flop. And even if I did know, I still can't see how to reset the system with only one button. Can anyone offer any advice on how this could be done, if at all?

Thanks in advance.

Quote from Shiny_Toad »

Is this what you mean? This is one reset button which resets all the counters.



And yes, my second counter is inverted for easier access. Or something.

Yes, that's it exactly. I'd tried hooking up a button to the output directly, and it didn't seem to do anything. The setup you have with the inverters works perfectly, though. Thank you very much.

Could someone give a quick explanation of what happens each time you press the input button?

I more of mean in terms of what will be shown by output. If you want to give me a rundown of exactly what will happen at each point, go for it. But I think I can trace it all if I know what the output will be after each press.

I haven't really looked into T-Flops yet, but I think I get the general idea of them now.

Oh wow. That's fancy

So I connected a clock generator to a T-Flip Flop.
I'm considering using this as a light house beacon that constantly flashes (that's assuming I manage a way to cram this all in a small space somehow).

T-Flip Flops are still frying my brain though.

The best I can figure is that it uses two loops to preserve a charge. Then the input causes it the stored charged to change for a second, before reverting back to the old state because of the loops

Also, someone make design for a T-Flip Flop that can fit in less than half the space of the one I showed you there :biggrin.gif:

(my lighthouse dream seems hopeless now lol)

You don't actually need a flip-flop in order to make a lighthouse; just a clock. Flip-flops act as "period doublers", turning a 5-clock input into a 10-clock output.

My advice would be to just construct a long-period clock at the base of the lighthouse and then wire it up to a bunch of lights at the top of the tower. If you wanted to make a REALLY long-period beacon for your lighthouse then you might need to make use of a flip-flop, but consider this:

Redstone cost of:
-One optimum flip-flop to double the period of your clock: 26
-Adding 2-more to the period of your existing clock: 4 (basic clocks must have odd periods, so increases come 2 at a time)

You don't break-even in terms of materials cost of using a flip-flop until you want to make a clock whose period is greater than 25. (The price of a flip-flop is equal to that of a 13-clock)

In other news, I've been playing around with an alternate architecture for T Flip-Flops based off of the D-Latch design. Loop the output back around to where the input is, and then make the "clock" input the new regular input and voila, it's a Flip-Flop. It's not very space-efficient, but it's interesting to make one with what I believe to be a different architecture. I'll post pics eventually.

Condensed version of a full adder. Currently the size is 5x7x3 but it's a work in progress. The Cin and Cout ports are aligned for easy stacking.

.

Any feedback is appreciated.

Quote from Hans Lemurson »

You don't actually need a flip-flop in order to make a lighthouse; just a clock. Flip-flops act as "period doublers", turning a 5-clock input into a 10-clock output.

My advice would be to just construct a long-period clock at the base of the lighthouse and then wire it up to a bunch of lights at the top of the tower. If you wanted to make a REALLY long-period beacon for your lighthouse then you might need to make use of a flip-flop, but consider this:

Redstone cost of:
-One optimum flip-flop to double the period of your clock: 26
-Adding 2-more to the period of your existing clock: 4 (basic clocks must have odd periods, so increases come 2 at a time)

You don't break-even in terms of materials cost of using a flip-flop until you want to make a clock whose period is greater than 25. (The price of a flip-flop is equal to that of a 13-clock)

In other news, I've been playing around with an alternate architecture for T Flip-Flops based off of the D-Latch design. Loop the output back around to where the input is, and then make the "clock" input the new regular input and voila, it's a Flip-Flop. It's not very space-efficient, but it's interesting to make one with what I believe to be a different architecture. I'll post pics eventually.

Yeah I basically wanted a longer period for the torch.
Eventually I just went with a clock though.

It was a real shame because I was literally only 1 torch away from making my flip flop fit, but it was just one block too close to the right and there was no way to go up or down :l

I hate to burst your bubble, but I'm pretty sure that's a 5-clock. The torch you have set-up as your "output" is actually part of the loop.

The reason the design for a 4-clock is so complicated is because simple loop-clocks only work for an odd number of torches. If you have an even number of torches, then all you get is a bloated Latch (since the patters can hold one of two self-stable states). In fact, I first discovered the Latch on my own when playing around with clock designs.

So anyways, I don't want to make you feel bad after all the enthusiasm you have for this, but I'm pretty sure the wiki will need to be reverted. It's a 5-clock.

EDIT:
I just read ALL of your comments, and realize that this might have something to it. I think you could have been a bit more explicit with the "quirk in the game whereby if two torches are placed above one another on alternate blocks facing in a North/South direction they will both change instantaneously with no delay." bit in your post.

I will have to check this out for myself. I'm thinking of making 3 clocks: an EW and NS version of these as well as a conventional 5-clock.

First I will see if the EW clock and default 5-clock sync up with each other, and then make the NS clock to see if it is unsynchronized with the other two. I fortunately just got back from a redstone-hunting expedition and so am in no shortage of the stuff.

Quote from DEADBEEF »

Quote from Hans Lemurson »

I will have to check this out for myself. I'm thinking of making 3 clocks: an EW and NS version of these as well as a conventional 5-clock.

First I will see if the EW clock and default 5-clock sync up with each other, and then make the NS clock to see if it is unsynchronized with the other two. I fortunately just got back from a redstone-hunting expedition and so am in no shortage of the stuff.

I'd already done this and made a short video, I just hadn't bothered uploading it as it didn't really show much that you couldn't get from a photo.
Anyway, I just uploaded it....

*snip*
The rightmost lamp is connected to the 4-clock design orientated in a N-S direction.
The middle lamp is connected to a standard 5-clock.
The leftmost lamp is connected to the 4-clock design orientated in a E-W direction (making it a 5-clock).

As you can see from the latter half of the video, the left & middle lamps are in phase with one another whereas the right is out of phase and flashing slightly faster.

That's a 5-Clock, as the guy two spaces up said. The torch you have as you ouput is the only reason your loops works, and is therefore part of the loop. Count all the torches, and you get 5