• Welcome to the Speedsolving.com, home of the web's largest puzzle community!
    You are currently viewing our forum as a guest which gives you limited access to join discussions and access our other features.

    Registration is fast, simple and absolutely free so please, join our community of 40,000+ people from around the world today!

    If you are already a member, simply login to hide this message and begin participating in the community!

deepSubDiver's Square-1 Scrambler

Joined
Apr 22, 2006
Messages
54
Location
Oranjestad, Aruba
WCA
2003BLON01
YouTube
Visit Channel
[
The discussion originally began in in this thread by Michiel. I'm moving it here to separate discussion for the sake of both programs.
Michiel didn't participate in the discussion, but since this is a modified copy of the old thread, this post needs to remain (due to the forum software).

-Lucas
]
 
Last edited by a moderator:

Mike Hughey

Administrator
Staff member
Joined
Jun 7, 2007
Messages
11,303
Location
Indianapolis
WCA
2007HUGH01
SS Competition Results
YouTube
Visit Channel

I haven't been following the WCA Forum for a long time, so I don't know - has this particular scrambler been discussed by the WCA? I think it has been mentioned before there that they would like to have something like this, and here it is! In any event, I really appreciate it; I've never fully trusted the 40 move scrambles anyway (seemed like they favor non-parity slightly), and it will be nice to be able to have scrambles that are a little shorter for all of my BLD attempts.

Has anyone verified (by reviewing the code, or statistics, or both) that this is a valid random state scrambler?

Edit: I just counted - this thing generates scrambles that are about 30% shorter than the 40 move scrambles. Nice! Not only would it be fairer; it would also help speed up competitions.

Edit 2: I sure hope this proves to be a valid scrambler - it brought me good luck. I just got my first ever sub-4 square-1 BLD solve with the first scramble I ever tried from this scrambler: 3:43.16 (1:43 memorization)! And it was a pretty rare case that I hadn't ever solved before: case 2 (star on one side, edges on the other side 2-6, probability 16/3678).
 
Joined
Apr 22, 2006
Messages
54
Location
Oranjestad, Aruba
WCA
2003BLON01
YouTube
Visit Channel
Are you serious? You think generating the algs is a bottleneck? Note how even my simple javascript scrambler produces 500 scrambles pretty much instantly.

I think this one does (haven't tried it myself):
http://m.cubing.dk/

No, I am not serious. I knew it would tick you off. No way I am going to program C (maybe never again).
Thanks, I'll try it, it looks good.
 

deepSubDiver

Member
Joined
Jun 17, 2009
Messages
161
Location
Germany
[...] Has anyone verified (by reviewing the code, or statistics, or both) that this is a valid random state scrambler?
I can verify this myself, it uses random permutations of each piece and finds a solution for that state. I think Lucas Garron once did a statistical research on hon random sq1 scrambles can be and especially about their distribution.
 

Mike Hughey

Administrator
Staff member
Joined
Jun 7, 2007
Messages
11,303
Location
Indianapolis
WCA
2007HUGH01
SS Competition Results
YouTube
Visit Channel
I can verify this myself, it uses random permutations of each piece and finds a solution for that state. I think Lucas Garron once did a statistical research on hon random sq1 scrambles can be and especially about their distribution.

Did you see the latest posts by Lucas here? Did you do option 1 here (the correct option - from your description, it sounds like you might have, but it's hard to be sure), or did you do option 2 or 3?

It's interesting that Lucas mentions that the official scrambler does star shapes more commonly than it should, and then I got 3 star shapes out of my first 10 scrambles using your scrambler. But of course, that might just be luck.
 

deepSubDiver

Member
Joined
Jun 17, 2009
Messages
161
Location
Germany
I continuously add random pieces from a pool, appending them to my current state. I think this corresponds to option 3.
 

blade740

Mack Daddy
Joined
May 29, 2006
Messages
851
WCA
2007NELS01
YouTube
Visit Channel
I haven't been following the WCA Forum for a long time, so I don't know - has this particular scrambler been discussed by the WCA? I think it has been mentioned before there that they would like to have something like this, and here it is!

The biggest problem is that, being PHP, this scrambler is even less portable than cube explorer. You wouldn't be able to just download the script and run it without installing a web server and PHP. Basically, the only way to get scrambles would be to have an active internet connection.

I agree, though. This scrambler would be much better than what we have now.
 

Mike Hughey

Administrator
Staff member
Joined
Jun 7, 2007
Messages
11,303
Location
Indianapolis
WCA
2007HUGH01
SS Competition Results
YouTube
Visit Channel
The biggest problem is that, being PHP, this scrambler is even less portable than cube explorer. You wouldn't be able to just download the script and run it without installing a web server and PHP. Basically, the only way to get scrambles would be to have an active internet connection.

I agree, though. This scrambler would be much better than what we have now.

I guess I was wondering how hard it would be to move the code to something more portable. If deepSubDiver would be interested in it, that is.
 

deepSubDiver

Member
Joined
Jun 17, 2009
Messages
161
Location
Germany
The biggest problem is that, being PHP, this scrambler is even less portable than cube explorer. You wouldn't be able to just download the script and run it without installing a web server and PHP. Basically, the only way to get scrambles would be to have an active internet connection.

I agree, though. This scrambler would be much better than what we have now.
The scrambler is a .NET application, not PHP. All I do in PHP is sending requests to the service and parsing the results.

I will have a look into porting it into js in the forthcoming weeks. Keep tuned!

/edit: Oh, and thanks for seperating this discussion from the old thread, Lucas.
 
Top