colcolpicle's puzzles

[Rene]

For Third Rubiks Rubicon (dedixak): secigut

It is not the fastest of solutions  , but it does solve all combinations. I am most annoyed that the correct combination (3201) is almost the slowest to solve.
The moves I used are:

• E in the right, followed by E in the left: cycles all crates clockwise
• E in the right, followed by D in the left: keeps the bottom-right crate fixed, cycles all others clockwise
• D in the left, followed by FF in the left: swaps the top two crates

[Rene]

For cacypym: diticop

Edit: I like the cleanliness of Rene's solution. I'll have to study it some before I understand how that table works though.

Thanks 
Short explanation: the table is quite straightforward; it has one barrel for each value of a crate. An 'F' barrel means "don't use the crate", '0' means "add the crate", '1' means "subtract the crate". The "E0" vibrator at the right converts the barrels into timing, which then directs the crate to the right side of the combined add/subtractor. The add/subtractor actually comes from an earlier solution I did a long time ago.

I like the simple trick for the '0' crate in your solution. I should have done it that way as well. That would have saved me all the trouble of moving it into the add/subtractor.

[jf]

For Third Rubiks Rubicon (dedixak): zusesos.

I use a simple dictionary lookup, and use only two simple moves: either drop a heavy barrel (i.e. E or F) on the left or drop it on the right, and let it run to completion before dropping the next. The dictionary is indexed by the first three output barrels, and amazingly fits in 144 barrels (there are two more extras just to push the last entry through the reader). The largest sequence is six barrels, and all others are shorter averaging less than four. I put in an entry for 3201 just for style, but it's a no-op.

The main difficulty I had was a lack of working space, though in the end after spending most of my design effort on the dictionary lookup device, I ended up with loads of free real estate.

[colcolpicle]

statistics befydid:dubinyn

Extra credit if you can get it to work 100% of the time(I didn't, but I'm pretty sure it's theoretically possible  )

[Rene]

statistics befydid:dubinyn

Extra credit if you can get it to work 100% of the time(I didn't, but I'm pretty sure it's theoretically possible  )

For Statistics (befydid): kimihax

If the accuracy is not enough, add another stage to the counters - but be prepared to wait a long time for the answer. If you don't want to wait this long, then remove a stage from the counters - but be prepared that accuracy will drop.


I'm pretty sure it is not possible to make it work 100% of the time theoretically. There is always a small chance that a '1' never comes up, for example, and then you cannot distinguish between a 1 and a 2 crate. The only thing you can theoretically check with 100% certainty, is that the crate is not equal to zero.
That said, I am pretty sure that you will never see my solution fail 
 
Colcolpicle, if you want to make your solution more accurate you should:
Start the left-most counter at 8 instead of 0. Right now, you have about 50% chance to get it wrong, because the average number of barrels that are counted for a certain value of the crate falls exactly on the moment the right counter steps. For example, if the crate is '1', the number of barrels is on average 1*16 = 16. So if the number of barrels is below average, the right counter will still be at 0 when it is read. Starting with 8 will put the average in the middle of the range.

[colcolpicle]

I stand by what I said:

but I'm pretty sure it's theoretically possible   

But the solution wouldn't really have an end and couldn't really be proven to be solved from that point on. I was thinking that you could make a thing that could change the number at the end so that if the machine comes up with a more likely answer, it could be changed. Although now that I think about it, that machine might need an infinite amount of memory.........

Anyway, new puzzle, another rubiks, but a bit easier.
Easy rubiks genexas

[colcolpicle]

New puzzle! 

Simple comparison: panutab : dynenyx

[Werbad]

For Simple comparison (panutab): mypecyx

[Rene]

For Simple Comparison (panutab): zoderom

[colcolpicle]

I seem to have over-thought that one. Haven't had time to solve this one yet, but:

Simple Comparison 2 fixopeg : vevovar


Edit: solve posted (you might almost consider my solution cheating)

[Werbad]

For Simple Comparison 2 (fixopeg): mokufoh
Your solution fails for a left F by the way.