Wireless transmission and Step-limit exploits

redxaxder · 2014-02-17 21:09:50

iceman wrote:

For instance, I'm pretty sure it's actually possible to transmit 1 bit of information without wires, if you use the 32 step limit.

Oh my god. We have to build this.

Before I found out about the cycle limit, I was considering the possibility of a trap that crashes the game by taking a gigantic number of cycles to simulate (don't think badly of me ). A mere 32 steps is easily attainable! A period 7 oscillator and a period 5 oscillator running at the same time will hit the limit.

How do electronics behave when the step limit is reached?

Last edited by redxaxder (2014-02-17 21:10:25)

iceman · 2014-02-17 21:29:34

If it hits 32 steps without hitting a loop, it treats the *whole* 32 steps as a loop (i.e., everything resets to the lowest state ever in that turn, instead of just during the loop).

To make a "wireless" switch, you'd need to have a button that pushes a circuit over the 32 step limit. Then, the receiver would be something that is unpowered for 1 step, but powered every step after that. Before the button is pressed, the receiver would be powered during a loop (say, steps 20-23). When the button is pressed, the whole turn counts as a loop, the receiver reverts to its lowest power, and voila! Wireless communication

I think something like this will work. The cascade on the right takes 29 steps (if my count is right), then starts looping. If the button is pressed, it adds 5 steps, putting it over the limit. So, if you press the button... the door opens. The transmitter can be reduced quite a bit, by, like you said, using a couple of oscillators instead of a cascade... this is pretty amazing!!!

iceman · 2014-02-17 21:43:29

http://www.twitch.tv/theiceman76/c/3743523

This is amazing

redxaxder · 2014-02-17 21:46:00

I think this should also be sufficient:

http://castledraft.com/editor/m49oKq

Since 2, 3, and 5 are prime, those three oscillators have a collective period of 30. If something that takes 3 steps to resolve happens at the beginning of the turn it should be enough to go over the limit. Of course, that button only exceeds the limit on the turn that it's pressed .

Holy crap it actually does it. Wireless transfer. Here's the test house I used:

http://castledraft.com/editor/ONscDF

Last edited by redxaxder (2014-02-17 21:53:01)

vahn · 2014-02-17 21:50:19

You crazy electronics nerds! Very cool!!

iceman · 2014-02-17 21:50:57

Yeah, I just wanted to do something quick that I knew would work instead of figuring out smaller circuits (I'm too tired to do that right now =P). In case you were writing that when I posted, look up - I have a video of it working!!! Really awesome feeling... the first wireless switch in TCD history!

iceman · 2014-02-17 21:59:56

It seems like you've tested your version as well... isn't it awesome?!? Great call on the prime-period circuits, btw... I thought you would have to have the whole transmitter circuit in one place, but you can have a 3 and 5 period circuit hidden somewhere, and have a button that ONLY is connected to the period 2 circuit trigger the whole thing!

Last edited by iceman (2014-02-17 22:00:41)

redxaxder · 2014-02-17 22:10:39

I'm having trouble getting it to work with 2,3,5. I've only gotten 7,5 to work.

Scratch that. I've only gotten 8,5 to work. Maybe I'm miscounting something. Ah. I found it; I'm miscounting a lot.

Last edited by redxaxder (2014-02-17 22:16:28)

Blip · 2014-02-17 22:22:11

Well, the game may have just changed forever. Now the weakness of wired walls has been overcome; switches inside a concrete box can affect stuff outside of it. And, with divided prime period circuits, also in their own concrete boxes, that would be very, very hard to crack.

Electronics have come very far since I built the latch-based press counter way back in v6, or the read/write lock in v8. And this is really crazy. But at least it's just one bit of storage...

I'm thinking of a way to use this for multiple bits now. You could have each bit be transmitted wirelessly on a different step, essentially performing a slow multiplex that then, in the reciever, stores the transmitted bits in latches. That could be then used for anything you want. I'll try this tomorrow, gotta sleep now.

joshwithguitar · 2014-02-17 22:24:26

My brother and I have been running experiments as well after reading iceman's first proposed idea of wireless communication. Basically it seems the whole thing is totally broken - you can have it so that things directly connected to power are unpowered. Sadly, it looks like Jason will have to fix this one.

iceman · 2014-02-17 22:27:16

Blip wrote:

Well, the game may have just changed forever. Now the weakness of wired walls has been overcome; switches inside a concrete box can affect stuff outside of it. And, with divided prime period circuits, also in their own concrete boxes, that would be very, very hard to crack.
Electronics have come very far since I built the latch-based press counter way back in v6, or the read/write lock in v8. And this is really crazy. But at least it's just one bit of storage...
I'm thinking of a way to use this for multiple bits now. You could have each bit be transmitted wirelessly on a different step, essentially performing a slow multiplex that then, in the reciever, stores the transmitted bits in latches. That could be then used for anything you want. I'll try this tomorrow, gotta sleep now.

You know how I said I felt like a scientist before? Well, now I feel like I just discovered the atomic bomb...

redxaxder wrote:

I'm having trouble getting it to work with 2,3,5. I've only gotten 7,5 to work.
Scratch that. I've only gotten 8,5 to work. Maybe I'm miscounting something. Ah. I found it; I'm miscounting a lot.

I think the circuits in your original design aren't 3,5,7... they're 6,10,14. I think a wave of power takes 3 steps do go down, then a wave of no power takes 3 steps to go down; 6 in total.

EDIT:

joshwithguitar wrote:

My brother and I have been running experiments as well after reading iceman's first proposed idea of wireless communication. Basically it seems the whole thing is totally broken - you can have it so that things directly connected to power are unpowered. Sadly, it looks like Jason will have to fix this one.

Awwwww, but we were having so much fun! =P He can probably hamper it by putting it back to 256 steps - that way, you'll need more circuits to get it to work. However, I don't think he's going to be able to limit it completely, without changing the electronic system. You can increase an 11 period circuit to 13 easily, and it'll increase the amount of steps by 60 (60 if you have 2, 3, and 5 backing it up) or more, easily.

Last edited by iceman (2014-02-17 22:53:07)

Hippasus · 2014-02-17 22:33:22

Broken!

This is definitely broken, just saying.

redxaxder · 2014-02-17 22:35:19

Are you sure the step limit is 32? Because I don't think this one should work, but it tested fine.

http://castledraft.com/editor/xFNr6s

Pandamonium · 2014-02-17 22:39:44

Seriously, I love watching this thread develop!

awesome research guys!

redxaxder · 2014-02-17 22:49:49

redxaxder wrote:

Are you sure the step limit is 32? Because I don't think this one should work, but it tested fine.
http://castledraft.com/editor/xFNr6s

Ah! I know why it works! The left oscillator is period 5. The right one is period 3. The pulse generator already contains something of period 2, but it takes it a few turns for the output to reach a stable state.

Since the total period of the oscillators is 30, the stabilization time of the pulses is enough to bring it over the limit.

Hippasus: what did you do to make the power go out like that?

Last edited by redxaxder (2014-02-17 22:58:28)

Hippasus · 2014-02-17 22:53:27

Just reproduced your results redxaxder. It has got me confused as well. I can't imagine this has a limit of more than 13 cycles. Are we missing something?

Edit: Ok, that makes sense. Got it now.

Last edited by Hippasus (2014-02-17 22:55:07)

Hippasus · 2014-02-17 23:00:24

It might be a little bit too game ruining to post on the forum, suffice to say that you're 90% of the way there. I can PM you, but with great power comes great responsibility...

redxaxder · 2014-02-17 23:02:15

PM me. I'm headed down this road anyway [=

*edit. Oh. I found it. That so cray.

Last edited by redxaxder (2014-02-17 23:09:34)

Hippasus · 2014-02-17 23:05:15

Send your email to mine and I'll tell all.

Pandamonium · 2014-02-17 23:15:01

Hippasus wrote:

Send your email to mine and I'll tell all.

I don't get any of the electrical engineering aspects of this all, but if your effectively saying once you surpass the 32 loops, everything turns "off" then effectively you can "lock out" a section of the home if it's not solved within 30 steps of a button being pressed? or are these the sub-steps and not physical ones?

Last edited by Pandamonium (2014-02-17 23:15:24)

redxaxder · 2014-02-17 23:21:51

Wat.

At the very least, that bug is going to be fixed.

And if Jason wants to prevent wireless communication, I guess it may be enough to check each connected component for loops separately. Merely increasing the cycle limit would be futile, though. A period of 2730 is attainable by just using 2,3,5,7, and 13.

Last edited by redxaxder (2014-02-17 23:31:36)

Hippasus · 2014-02-17 23:32:00

Redxaxer, it seems you figured it out before I sent my email!

Pandamonium wrote:

I don't get any of the electrical engineering aspects of this all, but if your effectively saying once you surpass the 32 loops, everything turns "off" then effectively you can "lock out" a section of the home if it's not solved within 30 steps of a button being pressed? or are these the sub-steps and not physical ones?

Sub-steps not steps, so this can be from the very start. The consequences are totally game breaking, but awesome.

Hippasus · 2014-02-17 23:34:55

redxaxder wrote:

And if Jason wants to prevent wireless communication, I guess it may be enough to check each connected component for loops separately.

Yep, it would be pretty much the only way. With a paradox connected to a long binary counter you can make cycle lengths into the billions.

Last edited by Hippasus (2014-02-17 23:35:42)

Hippasus · 2014-02-17 23:51:41

As long as there is an arbitrary limit to the number of cycles used to calculate the 'stability' of the state of the electronics globally the criterion for stability when it hits this limit will also be arbitrary. Since the stabilising criterion is arbitrary at 32 cycles there will always be a way of detecting when it hits the limit. Because of this wireless communication is basicly an inevitable by-product of a cycle limited, global scope system. Since you can't make cycles unlimited without opening up the game to other colourful exploits, the only way of stopping wireless communication would be to make each electronic devices state calculated independently.

joshwithguitar · 2014-02-17 23:52:26

redxaxder wrote:

Before I found out about the cycle limit, I was considering the possibility of a trap that crashes the game by taking a gigantic number of cycles to simulate (don't think badly of me ).

Yeah, Hippasus and I discovered you could crash the house last year and Jason added the limit when we informed him.

The Castle Doctrine Forums

#1 2014-02-17 21:09:50

Wireless transmission and Step-limit exploits

#2 2014-02-17 21:29:34

Re: Wireless transmission and Step-limit exploits

#3 2014-02-17 21:43:29

Re: Wireless transmission and Step-limit exploits

#4 2014-02-17 21:46:00

Re: Wireless transmission and Step-limit exploits

#5 2014-02-17 21:50:19

Re: Wireless transmission and Step-limit exploits

#6 2014-02-17 21:50:57

Re: Wireless transmission and Step-limit exploits

#7 2014-02-17 21:59:56

Re: Wireless transmission and Step-limit exploits

#8 2014-02-17 22:10:39

Re: Wireless transmission and Step-limit exploits

#9 2014-02-17 22:22:11

Re: Wireless transmission and Step-limit exploits

#10 2014-02-17 22:24:26

Re: Wireless transmission and Step-limit exploits

#11 2014-02-17 22:27:16

Re: Wireless transmission and Step-limit exploits

#12 2014-02-17 22:33:22

Re: Wireless transmission and Step-limit exploits

#13 2014-02-17 22:35:19

Re: Wireless transmission and Step-limit exploits

#14 2014-02-17 22:39:44

Re: Wireless transmission and Step-limit exploits

#15 2014-02-17 22:49:49

Re: Wireless transmission and Step-limit exploits

#16 2014-02-17 22:53:27

Re: Wireless transmission and Step-limit exploits

#17 2014-02-17 23:00:24

Re: Wireless transmission and Step-limit exploits

#18 2014-02-17 23:02:15

Re: Wireless transmission and Step-limit exploits

#19 2014-02-17 23:05:15

Re: Wireless transmission and Step-limit exploits

#20 2014-02-17 23:15:01

Re: Wireless transmission and Step-limit exploits

#21 2014-02-17 23:21:51

Re: Wireless transmission and Step-limit exploits

#22 2014-02-17 23:32:00

Re: Wireless transmission and Step-limit exploits

#23 2014-02-17 23:34:55

Re: Wireless transmission and Step-limit exploits

#24 2014-02-17 23:51:41

Re: Wireless transmission and Step-limit exploits

#25 2014-02-17 23:52:26

Re: Wireless transmission and Step-limit exploits

Board footer