Odder's URL based puzzle scrambler

Hello cubers!

So today I made an URL-based scrambler. This makes it possible to define in the address bar what type of scrambles you want, how many of them you want and how you want to group them.

I'll throw a linky here: http://scrambler.odderen.dk

So how does it work? Well if you just click that link, you'll get no scrambles at all. You need to define the scrambles you want.

The syntax are following:
?puzzle1=N&puzzle2=N&...&group=N

the "puzzle" list is:

• 222 (random state, suboptimal)
• 333 (random state)
• 444 (random moves)
• 555 (random moves)
• 666 (random moves)
• 777 (random moves)
• pyra (random state, suboptimal)
• mega (random moves)
• clock (WCA)
• sq1 (random state)

N can be set to any number, if you leave it blank, it will count as 0.
group is set with 5 as default.


Let's say you want:
3x3x3: 28 scrambles
pyra: 14 scrambles
2x2x2: 56 scrambles
Grouped in 7 scrambles
Then you have you use this link:
scrambler.odderen.dk?333=28&pyra=14&222=56&group=7

Let's say you want:
sq1: 5 scrambles
2x2x2: 500 scrambles
Grouped in 50 scrambles
Then you have you use this link:
scrambler.odderen.dk?sq1=5&222=500&group=50





To be honest I have no idea who will be using it, but I thought it would be fun to make this

(Disclaimer: it's using Mark2 to generate scrambles)

Awesome. I just screwed around with it a little, and it works like a charm. The organization is great as well

On the one hand, there's no reason that anyone actually needs this.

On the other hand, it works great. Nice work.

JoYE IS AWESOME

Specs112 said:

On the one hand, there's no reason that anyone actually needs this.

Click to expand...

you never know. But yeah, I don't see what it can be used for.

May I suggest placing a limit on the number of scrambles? My first thought was, "OOHHHH.... 100,000 SQ-1 scrambles?!"

Carson said:

May I suggest placing a limit on the number of scrambles? My first thought was, "OOHHHH.... 100,000 SQ-1 scrambles?!"

Click to expand...

The first thing I did after reading your post was try that.

The second thing I did after reading your post was restart my browser.

Specs112 said:

The first thing I did after reading your post was try that.

The second thing I did after reading your post was restart my browser.

Click to expand...

What browser? A good browser should make it easy to stop.

In any case, this really needs web workers. Easy, transparent web worker usage is on the to-do list for the Mark 2 scramblers.

Awesome, but when I saw the thread title I thought you meant <U,R,L> gen

ben1996123 said:

Awesome, but when I saw the thread title I thought you meant <U,R,L> gen

Click to expand...

That's normally shown as <R,U,L>, not <U,R,L>

Lucas Garron:

I've now done a little change in the randomSource based on what Chen Shuang told me to do

Works Great!

Odder said:

the "puzzle" list is:

• 222 (random state, suboptimal)
• 333 (random state)
• pyra (random state, suboptimal)
• clock (WCA)
• sq1 (random state)

Click to expand...

2x2x2 might be MRSS, unless you tweaked it.
3x3x3 is MRSS
pyra is MRSS, I think, unless you tweaked it.
clock is obviously MRSS, and it's optimal enough (14 dial turns).
sq1 is MRSS "suboptimal".

I don't know why people care so much about suboptimal vs. optimal, if the scrambles are short enough. Random-state matters way more, including MRSS vs. naive random-state (Square-1), and using a good random seed.

Odder said:

Lucas Garron:

I've now done a little change in the randomSource based on what Chen Shuang told me to do

Click to expand...

Coo'. You should use the same MersenneTwister object for all scramblers, though, else they all get the same seed. (Else if you ever have two random scramblers that generate random scrambles similarly, they might generate them the same way, because they might have been seeded at the same time.)

Lucas Garron said:

I don't know why people care so much about suboptimal vs. optimal, if the scrambles are short enough.

Click to expand...

... and if they're long enough. At least for home training, optimal is bad because you don't want to know in advance that there's a three moves solution. Intentionally suboptimal is better there. I don't think that's actually done here, though, cause I just got a three moves scramble (F U2 R).

Lucas is right about scramble types as I stole all the scramblers from Mark2

Stefan, you got a 3 move scramble with this?

Yes, like I said (it was for 2x2x2, in case that was unclear).

(I actually skimmed 1000 2x2x2 scrambles before asking without finding any "short" scrambles) Okay, I see what you mean, "13. F' U F R U' " also kinda short. But as I said, it's Mark2 scrambles, so I cannot describe how the scrambler works to be honest.

Odder, could you plug a skewb scrambler into this as well? I don't know how well it would work, but just a functional one would work great.

theZcuber said:

Odder, could you plug a skewb scrambler into this as well? I don't know how well it would work, but just a functional one would work great.

Click to expand...

Odder's also already ahead of you on that. He doesn't have one done yet, but he certainly will soon - he's probably going to use Chen Shuang's optimal random-state scrambler for this.

Dunno if I should post or not but thanks to Justin Jaffray's skewb skillz it now supports skewb