elrog
Veteran
Using corners first on the 3x3 is less efficient than other methods because solving edges would have bad recognition time when you have none of them solved and can't see the whole cube all at once. It is better to solve some edges and some corners both before solving all of the edges and then all of the corners because either one becomes increasingly harder the more of the other you have solved. The problem with this for you would be that your 2x2 CLL algs would not work on a 3x3 without messing other peices up. I think your best bet would be the Columns first method because the algorithms would be the most similar to those that you use on the 2x2. In the Columns first method, you solve 4 corner edge pairs and place them like in fridrich, but without the cross. You then solve the top 4 corners with CLL. Having the middle layer solved will vastly improve recognition time for the last 8 edges rather than doing the last 12 of them.
There is a method called EG for the 2x2. You may have haerd of it or not, but you basically solve the top 4 corners while permuting the bottom layer corners. Essentially you solve the bottom layer with 2 corners needing swapped or all solved, then solve the top 4 corners using double the amount of algorithms for CLL. Remember that half of these are the CLL algs. There is a columns first method that solves 4 pairs, and permutes them while solving the top 4 corners. It said this is accomplished by using the 2x2 algs and treating the pairs as a single corner piece, using double turns to keep them together. You do not by any means have to learn the extra algs for EG, but you could do this same double move technique. For a better explanation of this, follow this link: http://www.speedsolving.com/wiki/index.php/PEG
As for solving the last 8 edges, you should be able to intuitively solve 3 of the bottom edges and the centers. You may then proceed to orient the rest of the edges with setup moves for L5E as used in Roux. Then you could solve the last bottom edge and do EPLL. EPLL had very few algs. Here is a link for those orientation algs of which there are 3 of: http://www.speedsolving.com/wiki/index.php/L5E
Here are EPLL algs of which there are 4 of with one being an inverse: http://www.speedsolving.com/wiki/index.php/EPLL
To solve with very few algorithms, you could always learn Heise. I know you probably won't want to because theres no good tutorials out there on it because it can be hard to explain and understand. It does use 0 algorithms though. It is more meant for fewest moves solving than speedsolving. I just use it for fun and so I understand what I`m doing. I don't really expect you to learn this, but I think its cool, so here it is: http://www.ryanheise.com/cube/
Another option would be the waterman method, It blockbuilds one layer, then solves the top 4 corners, then does the last 8 edges. It woul have the most similar algs to your 2x2 ones 2nd to columns first. I understand that this method uses a quite a few algorithms though. I haven't ever cared to much to learn more about it than that, but if you want to, I'd check here first to get a basic idea of the method: http://www.speedsolving.com/wiki/index.php/Waterman
Follow the last link in the link for more information. You may also do the first layer by solving the corners then edges, or edges then corners.
There is a method called EG for the 2x2. You may have haerd of it or not, but you basically solve the top 4 corners while permuting the bottom layer corners. Essentially you solve the bottom layer with 2 corners needing swapped or all solved, then solve the top 4 corners using double the amount of algorithms for CLL. Remember that half of these are the CLL algs. There is a columns first method that solves 4 pairs, and permutes them while solving the top 4 corners. It said this is accomplished by using the 2x2 algs and treating the pairs as a single corner piece, using double turns to keep them together. You do not by any means have to learn the extra algs for EG, but you could do this same double move technique. For a better explanation of this, follow this link: http://www.speedsolving.com/wiki/index.php/PEG
As for solving the last 8 edges, you should be able to intuitively solve 3 of the bottom edges and the centers. You may then proceed to orient the rest of the edges with setup moves for L5E as used in Roux. Then you could solve the last bottom edge and do EPLL. EPLL had very few algs. Here is a link for those orientation algs of which there are 3 of: http://www.speedsolving.com/wiki/index.php/L5E
Here are EPLL algs of which there are 4 of with one being an inverse: http://www.speedsolving.com/wiki/index.php/EPLL
To solve with very few algorithms, you could always learn Heise. I know you probably won't want to because theres no good tutorials out there on it because it can be hard to explain and understand. It does use 0 algorithms though. It is more meant for fewest moves solving than speedsolving. I just use it for fun and so I understand what I`m doing. I don't really expect you to learn this, but I think its cool, so here it is: http://www.ryanheise.com/cube/
Another option would be the waterman method, It blockbuilds one layer, then solves the top 4 corners, then does the last 8 edges. It woul have the most similar algs to your 2x2 ones 2nd to columns first. I understand that this method uses a quite a few algorithms though. I haven't ever cared to much to learn more about it than that, but if you want to, I'd check here first to get a basic idea of the method: http://www.speedsolving.com/wiki/index.php/Waterman
Follow the last link in the link for more information. You may also do the first layer by solving the corners then edges, or edges then corners.
Last edited:
