Weekly competition 2007-30 (2,3,3OH,4,5,BF,SQ1,MG,FMC,MM,PM)

Mike Hughey said:
I really like Arnaud's method for edge matching. You might consider trying it. I have video of him, ExoCorsair, and Lars Vandendbergh doing 5x5x5 solves at the US Open, and the thing you can't help noticing is how much slower Arnaud turns the cube relative to the others. And yet his edge-matching is almost as fast as theirs! I'm a bit suspicious that maybe Arnaud's method of edge-matching is actually superior to the BigCubes.com method, but since his method isn't currently used by any of the really fast-hands people, we can't see just how efficient it is.
Click to expand...

can i find those vids anywhere? If not, could you mail me them or something? Do you have a lot other USO vids?
 
mitchell has some on his site www.opticubes.com, but that's why I've been trying to get relearn Arnaud's method. He showed me at the US Open, but I forgot the intricacies of the method. Granted, Arnaud does not really move any cube extremely fast, and he still is there. I just finished doing a bunch of 5x5 solves, and I was able to get 30-35 second centers and sub 30 3x3s, but my edges killed everyone of my times (65-70). Getting Arnaud's method into a fast-handed person could prove to be powerful.

Nvm... mitchell didn't have any 5x5 vids. O well.
 
Erik said:
can i find those vids anywhere? If not, could you mail me them or something? Do you have a lot other USO vids?
Click to expand...

I still just have them on the computer in original DV format; I haven't tried converting them or uploading them yet. I'll have to see if I can get around to it - sorry, I'm rather lazy about it. I mostly taped the blindfold finals, and other than that I just have a few scattered videos, like the 3 5x5x5 solves I recorded.

By the way, ExoCorsair, I've tried using both methods and counting moves, and I never expect it while it's happening, but it almost always works out to almost exactly the same number of moves. Arnaud's method actually seems worse at first, but the last 4 tredges in the bigcubes method wind up being enough moves to compensate so it works out the same. (Of course, the last 4 tredges are memorized algorithsm which go quickly, so from an execution perspective, you'd think it'd be quicker.) So the benefit one way or the other either is due to my not executing one of them optimally (which is certainly possible - I might even be doing both of them wrong!), or due to recognition or execution being quicker per edge one way or the other. But I'm not good enough to make any determination on that, unfortunately. Which is why it would be interesting for someone who's better at 5x5x5 to try to figure it out.

ExoCorsair, maybe you were suggesting you were going to do just that. If not, would you consider it? I'd like to see how it works out in your (much more capable) hands. Or maybe you should do move counts for the bigcubes method and Arnaud should do them for his method, and then we can compare averages (to make sure good versions of the method are used in both cases).
 
Arnauds method does indeed have the same amount or sometimes a bit more moves, but this small disatvantage is blown away by the nice lookahead of the method, if I perfect my style and improve my lookahead skills a bit there should be exactly 0 delays in the pair-up phase :)
 
Mike Hughey said:
ExoCorsair, maybe you were suggesting you were going to do just that. If not, would you consider it? I'd like to see how it works out in your (much more capable) hands. Or maybe you should do move counts for the bigcubes method and Arnaud should do them for his method, and then we can compare averages (to make sure good versions of the method are used in both cases).
Click to expand...

I meant that I was going to experiment and count moves. :cool:

Either way, results should be interesting (as yours is already). Although your results pretty much match what I initially suspected...
 
I finally had a little while to do the comp.:
3x3 Speedsolve:
16.86 (14.08) 20.94. 16.84 (POP)
Avg: 18.21
The 20.94 was a minor pop but I was able to fix quick. The last POP was huge, so I just said screw solving. All were full non-lucky solves, without warmups. I am satisfied.

BTW Arnaud, I have still be confirming the FM posts each week, just so you know ;)
 
About my 5x5x5 edge pairing method: I made a comparison with Ron and it turned out that he uses about 75 moves using the bigcubes method and I use about 80 moves. The move count of my method is greatly depending on the 50% chance of parity. If you end up with 3 semi-pairs you don't have parity and save 12 moves.

As Erik explained the good part about my method is the look-ahead. You don't start with a center and then try to find to wings (2 out of 24 pieces), but you start with a wing and try to find the center (1 out of 12). Then while you are setting up that center you know what the the next wing is so you can already look for the next center (1 out of 12) and so on. Finding 1 out of 12 pieces is easier than 2 out of 24 pieces and you can find the pieces while you move the cube.

Doing 80 moves in about 60 seconds should be possible for everyone that practises for a while.

Finally for those that don't know, Erik is using my method and has fast hands. His times for this week:
5x5: (1:39.91), (2:01.06), 1:50.59, 1:53.25, 1:44.69 => 1:49.51
 
Wow - I didn't know Erik was using your method either, Arnaud. I guess that proves it's at least as good as the bigcubes method. And there's still room to believe it might possibly be better.

Now that I know Erik uses it, I'm pretty certain I'm sticking with this method long-term. (Although I'll probably still practice the bigcubes method sometimes.)

And Arnaud, thanks for the move counts. That lets me check to see if I'm doing it reasonably close to right.
 
by the way Arnaud I do have some small differences/improvements now. Like in the beginning when you don't have a half pair paired up from the beginning or I don't see one right away I just do 1 times 2 half-pairs from any spot, it's a bit better than only pairing up a single one I think... Also I have a small improvement for the overall pairing but I'll explain it to you when I see you again.
 
Erik has made the comments in the past competitions that there were scrambles with his old method and his new method ... he saw the results of the new methods (Arnaud's) catch up to his real fast with little effort. I guess once he fixed himself on Arnaud's pairing, he stopped saying "new method"/"old method" because now it's his only method.
 
Like in the beginning when you don't have a half pair paired up from the beginning or I don't see one right away I just do 1 times 2 half-pairs from any spot, it's a bit better than only pairing up a single one I think.
Click to expand...
This is my thought process
0: If I see a semi-pair during the final moves of the centers-fase I move it (to the D-layer) to preserve it. This happens about 50% of the times
1a: After the centers-fase, If I see a semi-pair in the U or F layer I use that one, otherwise I use a preserved one (step 0) or quickscan for one on the LDRB-layers. This happens about 60% of the time.
1b: ) If I don't see a semi-pair I look for a wing and center and set them up
2a: If I see the return-semi-pair during the setup I will do 2-edges at a time
2b: Otherwise I do 1 edge at a time (d RUR' d' F') while scanning for the next matchup

Also I have a small improvement for the overall pairing but I'll explain it to you when I see you again.
Click to expand...
I am very curious about this. The only improvement I actually use is to switch to a 4-edges-at-a-time or even 6 edges-at-a-time matchup/return system if I see that there are very many semi-pairs and hardly any "double unmatched" pairs. That hardly ever happens and sometimes backfires.

I will try to write down an example edge-pairing solution tonight.
 
Yes,
I'd have to say I was not that good at the now 'old method' but not too bad since I did some sub-2's with it. Arnaud explained it at German Open and later on msn too a bit. He told me his times with it and I was amazed so I tried it and the first results were not bad at all and it improved later on quite much. The most difficult thing is to keep yourself from trying to go too fast and really only turn faster when you see the next edge already.
 
Kiyoshi Takahashi

33:
33.09 (40.19) 38.70 (31.30) 32.81 => 34.87
wow :) my first competition average under 35 seconds ! (last time, it was my first under 40 seconds average). On the other hand I really don't believe that I will manage to get under 25 seconds one day :(...

33 - OH:
1:31.73 1:25.67 1:23.12 (DNF) (1:22.48) => 1:26.84
My first competition ever in OH so I am really happy :) . DNF was a bad "memory leak" mistake (two in a row) so I gave up... 1:22 and 1:23 were lucky cases...
 
I thought Erik's solution for fewest moves was the coolest thing ever. But for me, it didn't quite work at the end. Instead of a B, I had to do B2 L B' to get it solved. I hope it's just a typo on his part entering the solution, or I'm doing it wrong (which is most likely)...

Anyway, it's really an amazing solution! What a cool way to solve it.
 
Jefferson James

2x2x2
Average= 16.05
(12.64), 15.58, (17.83), 15.66, 16.91
With the little time I do have to practice I will not be working on this cube.

3x3x3
Average= 37.70
(41.25), 37.33, 38.48, 37.27, (32.72)
I was rushing and not looking ahead having bad times then the last solve I went slow not even trying and it came out best. I just get impatient sometimes.

4x4x4
Average= 3:58.93
(4:18.33), 3:59.14, 3:49.92, (3:31.14), 4:16.14
Few seconds better lol.
 
You must log in or register to reply here.

Similar threads

AvGalen
Weekly competition 2007-46 (2,3,3OH,4,5,BF,SQ1,MG,MMG,FMC,MM,PM,Relay)
2 3
Replies
56
Views
8,277
AvGalen
AvGalen
AvGalen
Weekly competition 2007-45 (2,3,3OH,4,5,BF,SQ1,MG,MMG,FMC,MM,PM,Relay)
2
Replies
37
Views
5,540
fourdown00
F
AvGalen
Weekly competition 2007-44 (2,3,3OH,4,5,BF,SQ1,MG,MMG,FMC,MM,PM,Relay)
2 3
Replies
51
Views
7,276
malcolm
M
AvGalen
Weekly competition 2007-43 (2,3,3OH,4,5,BF,SQ1,MG,MMG,FMC,MM,PM,Relay)
2 3
Replies
57
Views
7,804
mrCage
mrCage
AvGalen
Weekly competition 2007-42 (2,3,3OH,4,5,BF,SQ1,MG,MMG,FMC,MM,PM,Relay)
2 3
Replies
54
Views
7,730
cmhardw
cmhardw
Back
Top