So today I got the idea on how to make a long delay without using a huge repeater chain. It takes two clocks that will only set off an AND gate when they match. What ever numbers you choose on each side will give you the delay of the circuit.
So you start with 4 tick repeaters and lets say 2 on one clock and 3 on the other. 4x2x3=24 So after 24 ticks the clock will be aligned and trigger. 4x5x6=120 So the equation is (Repeater Delay Setting) X (Clock1) X (Clock2) = (Delay) Now this doesn't work for all cases. If you use a 3 and 6 the clock will align only after 2 cycles of the 3 clock so you cannot use numbers that has one as a factor of the other.
These 2 delays are nearly identical but one takes 30 repeaters and the other only takes 11. This picture the circuit doesn't save much space but it grows considerably. With clocks of 9 and 11 you would need 99 repeaters to equal the same delay as apposed to 20.
Now I don't have everything ironed out yet. I don't know exactly how much extra delay the clock system adds into itself. The key is to get as close as you want to your target number and then add delay to reach the target. This system also currently acts as a clock so you'll need an rs_nor latch to hold power, and potentially disable the clocks. It's also a one way delay so this is better in a lot of cases. In a repeater chain you might not want to wait for the entire chain to clear before being able to operate the circuit.
Good idea. The total delay for this sort of circuit should be the least common multiple of the individual delays, so anyone using this, make sure your delays are relatively prime.
This allows you to get delays over a minute long without needing hundreds of repeaters. There are contraptions that need really long delays, such as ones dealing with flowing water or lava... that's why.
Rollback Post to RevisionRollBack
Playing Minecraft since [Friday, March 19, 2010, 9:20:21 PM] (First indev world save)
I have seen that before, by CX Gamer (I think thats his name).
The only thing I could find was one with the mine cart and cob web by GreyZap. Sure it's a long delay but it's not the same at all. My circuit let's you dial in a lot more specific delays in actual measurable tics and not "how long it takes a mine cart to fall." Sure that's a nice way to get long delays in a small time but the guy admits it's not accurate.
Rollback Post to RevisionRollBack
Playing Minecraft since [Friday, March 19, 2010, 9:20:21 PM] (First indev world save)
I put an entry up for this exact method on the wiki yesterday (coincidence).
Here's a video:
Rollback Post to RevisionRollBack
Apparently I'm a complete and utter jerk and come to this forum just like to make fun of people, be confrontational, and make your personal life miserable. If you think this is the case, JUST REPORT ME. Otherwise you're just going to get reported when you reply to my posts and point it out, because odds are, I was trying to be nice.
I put an entry up for this exact method on the wiki yesterday (coincidence).
Here's a video:
Huh, I didn't even realize anybody else had done these before... I just fired up Minecraft and started brainstorming.
A few Notes:
The version in the video:
Will not work in Minecraft 1.0 without modification (-)
Doubles as a really long monostable circuit. (+)
Can easily be modified to stop additional signals (+)
My version:
Will shut it's self off if another input is received. This could be very useful if you ever want to stop the delay. (+)
Is a lot easier to calculate and set the total delay. The clock is repeaters only and the total circuit only adds 4 ticks of delay compared to a comparable repeater chain. (+)
It's not easy to adapt mine to stop additional signals, they just stop the counter (-,+)
One uses AND gates with an rs_nor latch and the other (mine) uses t-flipflops. They both do the same job and seem to be just as compact as each other. So it really depends on your exact use for it which is better.
Will not work in Minecraft 1.0 without modification (-)
Doubles as a really long monostable circuit. (+)
Can easily be modified to stop additional signals (+)
Haven't looked at it in 1.0, you're probably right.
Modified to stop signals: uh. It does that already.
My version:
Will shut it's self off if another input is received. This could be very useful if you ever want to stop the delay. (+)
Is a lot easier to calculate and set the total delay. The clock is repeaters only and the total circuit only adds 4 ticks of delay compared to a comparable repeater chain. (+)
It's not easy to adapt mine to stop additional signals, they just stop the counter (-,+)
I fail to see how your repeater clocks differ from his. They're both doing the same thing: racing each other around two different length tracks. Calculating his delay is exactly the same: total the amount of delay on both tracks, multiply (excepting when they have common factors).
Apparently I'm a complete and utter jerk and come to this forum just like to make fun of people, be confrontational, and make your personal life miserable. If you think this is the case, JUST REPORT ME. Otherwise you're just going to get reported when you reply to my posts and point it out, because odds are, I was trying to be nice.
Modified to stop signals: uh. It does that already.
He builds the circuit, demonstrates it, shows how to extend it, and then finally adds the over flow safety. It's a completely optional addition, and as to why I said "Can easily be modified to stop additional signals (+)"
I fail to see how your repeater clocks differ from his. They're both doing the same thing: racing each other around two different length tracks. Calculating his delay is exactly the same: total the amount of delay on both tracks, multiply (excepting when they have common factors).
It's simple math, or at least mine is. My total circuit adds 4 ticks of delay, the same as a full delay repeater. The system has no other delay in it so it truly is clock1 x clock2. So the equation is (clock1) x (clock2) + 1. So if you have a 9 repeater clock and an 11 repeater clock you get the same total delay EXACTLY equal to a chain of 100 repeaters.
On your posted version there is so much that throws off the equation. The AND gates add two ticks of delay. No telling what exactly that does to the final time. This could theoretically be compensated for by setting a repeater on both clocks to 2 instead of 4 (but who else besides me would know to do that). The other problem is that the clock start and output are separated by repeaters, and these are off set as to not trigger the and gate on the first pass. All this will throw off the final delay time. I have to build this and run a lot of tests to see exactly what these extra inconsistencies do... unless you know exactly how to calculate the actual delay... but i doubt it.
Well I built the thing, modified it to work with 1.0 and checked it out. The biggest difference I see now is mine was built to figure out delays in a number equivalent to that of a repeater chain.
The version posted by Draco only handles it's equation purely on ticks alone. It just starts two clocks, one being on tick slower and it will race them until the faster clock over takes the slower clock and sets off the AND gate. I built it with 8 repeaters.
(clock1) x (clock2) + (i can only assume a 4 tick delay like mine)
(8x4+2) x (8x4+2-1) + 4 = (number of tick delay). Now there is also the the 4 tick delay in starting clock2, I'm not sure how that effects the timing, as well as the output being in the center of the clock instead of at the exact starting point.
Edit: yeah you can disregard any of that equation garbage. I set the circuit up with the fewest number of repeaters as I could, crunched the numbers and figured out what the delay would be using the equation. Then I set up a chain and the race was on. The end result very very sad. Not only did the circuit not finish the race when it was supposed to, but closer to 5 times longer then predicted. When it did finally finish the race the pulse from the dueling clocks was to fast to actually set off the rs_nor latch and just kept pulsing my output. I'm sure the clocks would eventually align and provide a long enough pulse set the rs_nor latch but the test was already a long off failure. Also if you watch the tutorial video the one running in the back round that is "about 2 minutes" runs well past the 5 minute mark. If you cannot accurately predict the delay it is useless as a circuit component and only good if you "want something to happen eventually"
Edit again:
I went back and re-tried the circuit and it sort of worked this time, however I have come the the conclusion that this ONLY, I repeat ONLY works as a monostable circuit in the demonstrated configuration. If you try to output off the AND gate like mine it will pulse the output before it finally pulses long enough to reset the RS_Nor latch. With a pulse limiter connected to the displayed output then this can be used as a long delay, but I'm still not sure how to calculate the delay for any kind of accurate use.
So my version = long accurate delay circuit.
Minecraft Addicts version = long monostable circuit.
The only thing I could find was one with the mine cart and cob web by GreyZap. Sure it's a long delay but it's not the same at all. My circuit let's you dial in a lot more specific delays in actual measurable tics and not "how long it takes a mine cart to fall." Sure that's a nice way to get long delays in a small time but the guy admits it's not accurate.
AHEM, I siad this thread (the repeater thingy) and not the minecart though cobweb. I have only read your post, but I think I remember a video from MinecraftAddict
Also, setting 3 and 6? Please tell me how you get 6 on a repeater setting, but I think you mean 2. Besides, 1,2,3,4 are all factors of 12.
AHEM, I siad this thread (the repeater thingy) and not the minecart though cobweb. I have only read your post, but I think I remember a video from MinecraftAddict
Yeah the MinecraftAddict version is the one that draco posted. I've been playing with it for most of the evening and it is proving to be VERY unreliable. I could only get it to work when I switched back to 1.8.1 and then it only worked in the first test, several tests after and it would run for several minutes and never shut off. I even let the thing run for an hour and it was still ticking away. Any attempt to modify the delay failed. I even tried the smallest number of repeaters and it just fails. If you watch the video MinecraftAddict mentions "it runs for about two minutes" when he starts it up at 30 seconds and you can see it clearly running still at the 5 minute mark.
Also, setting 3 and 6? Please tell me how you get 6 on a repeater setting, but I think you mean 2. Besides, 1,2,3,4 are all factors of 12.
3 and 6 is not the delay of the repeaters but the number of repeaters in each clock. All the repeaters are set to 4 and the number of repeaters in each clock determines the delay [(clock1) X (clock2) + 1]. 4 & 5 would give you a delay equivalent to 21 repeaters (the 4 torches in the system adds 4 tics) Following the same math you might think that 3 and 6 gives you a delay of 19 repeaters but you only get 7 because when the 3 clock will cycle twice and line up with the 6 clock when it finishes it's first cycle.
My final compact version is 7x5x4 (not counting if the clock sticks out)
MinecraftAddicts version is 10x5x5 (moot point considering I've still yet to get it to work reliably)
Rollback Post to RevisionRollBack
Playing Minecraft since [Friday, March 19, 2010, 9:20:21 PM] (First indev world save)
So you start with 4 tick repeaters and lets say 2 on one clock and 3 on the other. 4x2x3=24 So after 24 ticks the clock will be aligned and trigger. 4x5x6=120 So the equation is (Repeater Delay Setting) X (Clock1) X (Clock2) = (Delay) Now this doesn't work for all cases. If you use a 3 and 6 the clock will align only after 2 cycles of the 3 clock so you cannot use numbers that has one as a factor of the other.
These 2 delays are nearly identical but one takes 30 repeaters and the other only takes 11. This picture the circuit doesn't save much space but it grows considerably. With clocks of 9 and 11 you would need 99 repeaters to equal the same delay as apposed to 20.
Now I don't have everything ironed out yet. I don't know exactly how much extra delay the clock system adds into itself. The key is to get as close as you want to your target number and then add delay to reach the target. This system also currently acts as a clock so you'll need an rs_nor latch to hold power, and potentially disable the clocks. It's also a one way delay so this is better in a lot of cases. In a repeater chain you might not want to wait for the entire chain to clear before being able to operate the circuit.
Update Video:
Compacted version with self shutoff:
On and Off Power delay version:
Playing Minecraft since [Friday, March 19, 2010, 9:20:21 PM] (First indev world save)
Playing Minecraft since [Friday, March 19, 2010, 9:20:21 PM] (First indev world save)
An added t-flip flop to the output and two alternating pulse limiters and this will set a delay on and off power
Playing Minecraft since [Friday, March 19, 2010, 9:20:21 PM] (First indev world save)
This allows you to get delays over a minute long without needing hundreds of repeaters. There are contraptions that need really long delays, such as ones dealing with flowing water or lava... that's why.
Playing Minecraft since [Friday, March 19, 2010, 9:20:21 PM] (First indev world save)
The only thing I could find was one with the mine cart and cob web by GreyZap. Sure it's a long delay but it's not the same at all. My circuit let's you dial in a lot more specific delays in actual measurable tics and not "how long it takes a mine cart to fall." Sure that's a nice way to get long delays in a small time but the guy admits it's not accurate.
Playing Minecraft since [Friday, March 19, 2010, 9:20:21 PM] (First indev world save)
Here's a video:
Huh, I didn't even realize anybody else had done these before... I just fired up Minecraft and started brainstorming.
A few Notes:
The version in the video:
Will not work in Minecraft 1.0 without modification (-)
Doubles as a really long monostable circuit. (+)
Can easily be modified to stop additional signals (+)
My version:
Will shut it's self off if another input is received. This could be very useful if you ever want to stop the delay. (+)
Is a lot easier to calculate and set the total delay. The clock is repeaters only and the total circuit only adds 4 ticks of delay compared to a comparable repeater chain. (+)
It's not easy to adapt mine to stop additional signals, they just stop the counter (-,+)
One uses AND gates with an rs_nor latch and the other (mine) uses t-flipflops. They both do the same job and seem to be just as compact as each other. So it really depends on your exact use for it which is better.
Playing Minecraft since [Friday, March 19, 2010, 9:20:21 PM] (First indev world save)
Haven't looked at it in 1.0, you're probably right.
Modified to stop signals: uh. It does that already.
I fail to see how your repeater clocks differ from his. They're both doing the same thing: racing each other around two different length tracks. Calculating his delay is exactly the same: total the amount of delay on both tracks, multiply (excepting when they have common factors).
He builds the circuit, demonstrates it, shows how to extend it, and then finally adds the over flow safety. It's a completely optional addition, and as to why I said "Can easily be modified to stop additional signals (+)"
It's simple math, or at least mine is. My total circuit adds 4 ticks of delay, the same as a full delay repeater. The system has no other delay in it so it truly is clock1 x clock2. So the equation is (clock1) x (clock2) + 1. So if you have a 9 repeater clock and an 11 repeater clock you get the same total delay EXACTLY equal to a chain of 100 repeaters.
On your posted version there is so much that throws off the equation. The AND gates add two ticks of delay. No telling what exactly that does to the final time. This could theoretically be compensated for by setting a repeater on both clocks to 2 instead of 4 (but who else besides me would know to do that). The other problem is that the clock start and output are separated by repeaters, and these are off set as to not trigger the and gate on the first pass. All this will throw off the final delay time. I have to build this and run a lot of tests to see exactly what these extra inconsistencies do... unless you know exactly how to calculate the actual delay... but i doubt it.
Playing Minecraft since [Friday, March 19, 2010, 9:20:21 PM] (First indev world save)
The version posted by Draco only handles it's equation purely on ticks alone. It just starts two clocks, one being on tick slower and it will race them until the faster clock over takes the slower clock and sets off the AND gate. I built it with 8 repeaters.
(clock1) x (clock2) + (i can only assume a 4 tick delay like mine)
(8x4+2) x (8x4+2-1) + 4 = (number of tick delay). Now there is also the the 4 tick delay in starting clock2, I'm not sure how that effects the timing, as well as the output being in the center of the clock instead of at the exact starting point.
Edit: yeah you can disregard any of that equation garbage. I set the circuit up with the fewest number of repeaters as I could, crunched the numbers and figured out what the delay would be using the equation. Then I set up a chain and the race was on. The end result very very sad. Not only did the circuit not finish the race when it was supposed to, but closer to 5 times longer then predicted. When it did finally finish the race the pulse from the dueling clocks was to fast to actually set off the rs_nor latch and just kept pulsing my output. I'm sure the clocks would eventually align and provide a long enough pulse set the rs_nor latch but the test was already a long off failure. Also if you watch the tutorial video the one running in the back round that is "about 2 minutes" runs well past the 5 minute mark. If you cannot accurately predict the delay it is useless as a circuit component and only good if you "want something to happen eventually"
Edit again:
I went back and re-tried the circuit and it sort of worked this time, however I have come the the conclusion that this ONLY, I repeat ONLY works as a monostable circuit in the demonstrated configuration. If you try to output off the AND gate like mine it will pulse the output before it finally pulses long enough to reset the RS_Nor latch. With a pulse limiter connected to the displayed output then this can be used as a long delay, but I'm still not sure how to calculate the delay for any kind of accurate use.
So my version = long accurate delay circuit.
Minecraft Addicts version = long monostable circuit.
Playing Minecraft since [Friday, March 19, 2010, 9:20:21 PM] (First indev world save)
Here:
Playing Minecraft since [Friday, March 19, 2010, 9:20:21 PM] (First indev world save)
AHEM, I siad this thread (the repeater thingy) and not the minecart though cobweb. I have only read your post, but I think I remember a video from MinecraftAddict
Also, setting 3 and 6? Please tell me how you get 6 on a repeater setting, but I think you mean 2. Besides, 1,2,3,4 are all factors of 12.
Yeah the MinecraftAddict version is the one that draco posted. I've been playing with it for most of the evening and it is proving to be VERY unreliable. I could only get it to work when I switched back to 1.8.1 and then it only worked in the first test, several tests after and it would run for several minutes and never shut off. I even let the thing run for an hour and it was still ticking away. Any attempt to modify the delay failed. I even tried the smallest number of repeaters and it just fails. If you watch the video MinecraftAddict mentions "it runs for about two minutes" when he starts it up at 30 seconds and you can see it clearly running still at the 5 minute mark.
3 and 6 is not the delay of the repeaters but the number of repeaters in each clock. All the repeaters are set to 4 and the number of repeaters in each clock determines the delay [(clock1) X (clock2) + 1]. 4 & 5 would give you a delay equivalent to 21 repeaters (the 4 torches in the system adds 4 tics) Following the same math you might think that 3 and 6 gives you a delay of 19 repeaters but you only get 7 because when the 3 clock will cycle twice and line up with the 6 clock when it finishes it's first cycle.
My final compact version is 7x5x4 (not counting if the clock sticks out)
MinecraftAddicts version is 10x5x5 (moot point considering I've still yet to get it to work reliably)
Playing Minecraft since [Friday, March 19, 2010, 9:20:21 PM] (First indev world save)