It does seem that we are making fairly similar errors in our DNFs. When you execute the cycles in the wrong direction, do you find that you execute the easier of the two directions? I tend to do this when I cycle in the wrong direction because the incorrect direction is actually easier to execute than the true direction.
I'm not sure I really think there is an easier direction most of the time. I suppose there sort of is for centers, but only in terms of the way I think, not in terms of execution. And I don't think I've really noticed this. But I suppose I can look for it. For the ones this week, the OU corners always confuse me - I think the wrong direction. So I'd be just as likely to get UO backwards. I really need to stop and think on those 9-movers sometimes - my muscle memory still doesn't know which way they go. The others were just weird - I don't normally make so many wrong direction mistakes; I don't know what happened this week.
Mike, I have noticed that your philosophy of always doing post mortems if helping me. I find that I still make the same errors, but I tend to catch them either as they are happening, or just afterward. For example, I am still making the same mistakes as you describe here, but I realize it either mid execution, or before moving onto the next cycle. This makes it possible to correct it before moving on. I wonder if approaching 100% accuracy does not mean not making mistakes at all, but rather having the ability to catch those mistakes before the end of the solve and correct them.
What are your thoughts on this, as you have been doing post mortems much longer than I have?
Actually, it was since you've been showing your post-mortems so much lately that I decided to go into detail this week. (And it helped that I had so many mistakes this week to do it with .) But in general, I do at least try to mostly figure out what happened on each solve I do at home, and I think it probably has helped my accuracy. I certainly do catch mistakes fairly often after I make them, and often I can fix them. However, sometimes I don't - on my 2/3 multi this week, I realized I had turned the cube in the wrong way to fix the flipped edges, but I couldn't remember which wrong way, so it didn't help. It seems like, especially in competition, it's still better not to make the mistakes at all than it is to correct them, since you can easily get confused when correcting. But I'm happy to get benefits either way. If nothing else, I suppose now you see what I told you before: taking a really long time to memorize a cube is not necessarily going to get you near 100% accuracy, since so many mistakes are made in execution.
I'm not sure I really think there is an easier direction most of the time. I suppose there sort of is for centers, but only in terms of the way I think, not in terms of execution. And I don't think I've really noticed this. But I suppose I can look for it. For the ones this week, the OU corners always confuse me - I think the wrong direction. So I'd be just as likely to get UO backwards. I really need to stop and think on those 9-movers sometimes - my muscle memory still doesn't know which way they go.
Is it an error of viewing, or of executing? I've tried to follow Chester's advice to stop viewpoint shifting, and it has helped me for some of the cycles. However, some of them I still have a hard enough time with that I still viewpoint shift them. They give me enought trouble that viewpoint shifting is actually faster than trying to remember which "sub-class" of AnI and not AnI this case is.
If nothing else, I suppose now you see what I told you before: taking a really long time to memorize a cube is not necessarily going to get you near 100% accuracy, since so many mistakes are made in execution.
That's weird that you mention that, as I have been seeing a positive effect on my accuracy because I've slowed down my memorization phase. I wonder if that shows that I was making MANY more memorization errors than execution errors previous to my post mortems though. My 5x5x5 right now is very close to a 2:1 split for memorization and execution when I get solves around 14:30-15:30. For faster solves the memorization just "sticks" and I guess I get more to a 1.5:1 or 1.25:1 (forgive the unsimplified ratios, I feel this way makes more sense for comparison).
All of my 4x4x4's this week were done at a "memorize slower than what feels comfortable, and solve at slightly faster than what feels comfortable" pace. I feel that, for me, this is a perfect balance of speed and accuracy that I could use in competition. This would be the pacing I would use in a competition, but I may go back to trying to take more risks at home for a fast single.
For 5x5x5 my goal is to shoot for a 2:1 split for memo:solving, but if I happen to memorize faster due to things sticking easier then that is just gravy. So far this seems to be helping, and actually this was the pacing I was attempting to use even on my 11:29 solve. It just happened that all of my memorization was sticking quite easily for me.
I know you recommend not to use this pacing in competition, but I think the compromise of how I do it for 4x4x4 would help account for the effects of being nervous and possibly making more errors in competition: memorizing slightly slower than what I feel I can comfortably do, and solving slightly faster than what I feel I can comfortably do. I'm not attempting to go "break neck" speed, I'm moreso attempting to go just a little too quickly - if that makes any sense.
Clock: 2:39.84 [0:34], 2:11.83 [0:28], 2:07.05 [0:26], 2:15.61 [0:29], 2:46.97[0:22] = 2:22.43
Comment: Big memory recall pause on the last one – it was really an easy one – might have been a PB without that pause.
What order do you memorise? Do you have any tips for memo, because for me, on a typical 2:10 solve, my memo is normally about 50 seconds. Do you think you are fast at memo just because of lots of practise on other stuff? Do you use any form of story or do you just memorise the numbers?
What order do you memorise? Do you have any tips for memo, because for me, on a typical 2:10 solve, my memo is normally about 50 seconds. Do you think you are fast at memo just because of lots of practise on other stuff? Do you use any form of story or do you just memorise the numbers?
I'm sure my practice on other BLD stuff helps, yeah. A long time ago, I came up with a correlation of numbers to letters for other reasons (I can't even remember why now), and so I just used them for clock: 1=O, 2=W, 3=H, 4=R, 5=V, 6=X, 7=S, 8=E, 9=N, 10=T, 11=L, 0(12)=A. I do 5 letter pairs for all the edges and centers as follows: front u edge, front r edge, front d edge, front l edge, front center, (flip puzzle), back center, back u edge, back r edge, back d edge, back r edge. I put those 5 letter pairs (using my letter pair images) in a story, so it's quite secure. Then I flip the puzzle back to the front and do the last four just verbally remembering the letters: ul corner, ur corner, dr corner, dl corner. Since it's verbal, it probably only takes 2 or 3 seconds to do the corners, if it's a good day. Overall, it's significantly less to memorize than a 3x3x3, so it's pretty easy to get it all quickly.
@Chris: The particular 9-mover corners that give me trouble do so because I try to do them from muscle memory. If I actually stop and think about where the pieces move, I can see them fine and get them correct, but I try to go faster than that, which gets me in trouble. By the way, this is the thing that kills me most with pyraminx BLD - I try to do them from muscle memory, and I miss way too often. Recently I've started to force myself to actually "see" the pyraminx cycle before I start, and as a result my accuracy went way up, without really slowing me down at all. I suspect I should just do the same on these 3x3x3 algs that are giving me trouble.
And I think that 2:1 memo to execution would kill me, because I just can't execute that fast. I got a sub-7 4x4x4 BLD solve in competition this weekend, and I think that even if I spent half an hour memorizing first, I couldn't execute a 4x4x4 BLD in sub-2:30! I'm just not that fast at execution. I'd be completely unable to compete with you with a 2:1 ratio, I think.
@Chris: The particular 9-mover corners that give me trouble do so because I try to do them from muscle memory. If I actually stop and think about where the pieces move, I can see them fine and get them correct, but I try to go faster than that, which gets me in trouble. By the way, this is the thing that kills me most with pyraminx BLD - I try to do them from muscle memory, and I miss way too often. Recently I've started to force myself to actually "see" the pyraminx cycle before I start, and as a result my accuracy went way up, without really slowing me down at all. I suspect I should just do the same on these 3x3x3 algs that are giving me trouble.
And I think that 2:1 memo to execution would kill me, because I just can't execute that fast. I got a sub-7 4x4x4 BLD solve in competition this weekend, and I think that even if I spent half an hour memorizing first, I couldn't execute a 4x4x4 BLD in sub-2:30! I'm just not that fast at execution. I'd be completely unable to compete with you with a 2:1 ratio, I think.
Well, to be fair that is the ratio I shoot for, and only for 5x5x5. For 4x4x4 I don't know my actual breakdown. All I know is the feeling of: go slightly slower than what feels comfortable through memorization, then go slightly faster than what feels comfortable through the solving phase. I'll try to time a couple breakdowns for my 4x4x4 and give you what it is. I imagine with my attempted pacing that it's much closer to 1:1, but with more time weighted slightly more on the memorization end.
My philosophy so far has been that 4x4x4 and 5x5x5 are completely different beasts, and are not to be handled the same way in terms of discovering an ideal pacing at all.
My philosophy so far has been that 4x4x4 and 5x5x5 are completely different beasts, and are not to be handled the same way in terms of discovering an ideal pacing at all.
Interesting. It's worked well for you so far, considering your great results on 4x4x4. I must admit I've always assumed the opposite - that they are exactly the same and should be treated identically. Maybe you're 100% right and I'm just fighting against reality. It wouldn't be the first time.
Interesting. It's worked well for you so far, considering your great results on 4x4x4. I must admit I've always assumed the opposite - that they are exactly the same and should be treated identically. Maybe you're 100% right and I'm just fighting against reality. It wouldn't be the first time.
To be honest, I'm just trying to try different things and take what works. I was excited, yes, about getting a much better result this week, but that could also just be a good week. I've learned from collaborating with Daniel on the BH method that sharing ideas is the best way to get faster. That's mostly what I am doing here, even if it comes across more gruffly than I intend.
I in no way am trying to imply that I believe what I am doing to be 100% right, and everyone else to be 100% wrong. My intention is only to state what I have been doing, and what it's effects have been so far by choosing to do that. I hope I haven't gotten on your nerves or offended you by any of my posts. From discussing ideas with Daniel for so long we have both learned to be fairly blunt with each other about our approaches. It is just understood from us sharing ideas for so long that we are trying to help each other to improve, not trying to shoot each other's ideas down. If I did not come across as having this intention, then just know that it is how I was intending to say what I said in the posts I've made in this thread.
Chris and Mike you two seem to be good at Blindfold solving can either of you tell me where to find a good 3op tutorial. I know about Macky's one but I can't follow how to do the corners a video tutorial would probably be better
Not at all - I think you may have misinterpreted my last comment. (Maybe I just expressed it badly.) I really meant that I suspect I'm wrong in thinking that 4x4x4 and 5x5x5 should be treated the same. I've often wondered why everyone else's 4x4x4 times are so much better relative to their 5x5x5 times than mine are, and I'm suspecting that's because I've taken this approach of treating them the same. Perhaps if I treated 4x4x4 differently, I could get faster results on it.
And yet, I must admit that deep down, I still kind of believe they should be treated identically. I still have this gut feeling that a 5x5x5 BLD solve should take significantly less than twice the time of a 4x4x4 BLD solve, if you're really really good at both. (My guess is about 5/3 the time.) Despite the fact that all the WCA results data (and even most people's home results) go completely against that.
I think I also made my comment because I was a little surprised: on most things associated with big cubes BLD, I have eventually wound up doing things the same way you do, sometimes by accident. It's unusual that we have such a different view on this, since we agree on so many other things relative to big cubes BLD.
Anyway, please don't believe I'm offended - I'm really very much enjoying the discussion. I think I've learned a lot from you in this thread.
First time I've competed in every event 2x2x2: 4.86, (5.04), (3.05), 3.68, 4.96 = 4.50 3x3x3: 14.59, (16.83), 13.48, (13.07), 14.01 = 14.03 4x4x4: 1:00.02, (1:05.91), (56.40), 59.31, 1:03.05 = 1:00.79 5x5x5: 2:03.67, 2:09.60, (2:14.20), 1:58.30, (1:56.83) = 2:03.86 6x6x6: 5:01.10, 4:54.96, 4:56.49, (4:52.22), (5:11.28) = 4:57.52 7x7x7: 7:33.71, 7:29.42, (7:03.47), 7:11.19, (1:33:19.00) = 7:24.27
Last one was my bld one http://www.speedsolving.com/forum/s...ld-Accomplishment-Thread!&p=483235#post483235 2x2x2BLD: 27.30, 6.01, DNF = 6.01 Haha 3x3x3BLD: 1:25.22, DNF, 1:36.68 = 1:25.22 4x4x4BLD: 5:41.84, DNF, DNF = 5:41.84 My accuracy is bad this week 5x5x5BLD: 14:06.45, DNF, 12:36.12 = 12:36.12 MultiBLD: 9/10 40:31 Didn't feell like going for a new PB OH: 27.40, (29.58), (27.06), 28.56, 29.18 = 28.38 Feet: (1:25.30), 1:23.14, 1:18.06, (1:15.99), 1:24.11 = 1:21.77 MTS: 45.10, (42.27), 47.24, (49.85), 47.43 = 46.59 2x2x2+3x3x3+4x4x4: 1:26.01 2x2x2+3x3x3+4x4x4+5x5x5: 3:31.18 Magic: (1.69), 1.52, 1.51, 1.49, (1.44) = 1.51 Master Magic: 5.49, (4.39), 4.69, 4.96, (6.18) = 5.05 Clock: 7.91, (8.33), 8.25, (7.49), 7.58 = 7.91 Megaminx: (2:36.10), 2:35.72, 2:28.58, 2:29.04, (2:26.49) = 2:31.11 Pyraminx: 6.20, 7.49, (8.10), (5.86), 6.83 = 6.84 Square-1: 31.18, 28.59, (24.06), (32.47), 31.09 = 30.29 FMC: 36 moves U' F R2 D L2 R2 B' R B' R D R B' D' R D B R2 D' R' D2 B' D' B U R' U' R2 U D R' U' R D' R2 B
Normal scramble with PM B
2x2x2: U' F R2 D L2 (5)
2x2x3: R2 B' R B' R D (11)
Blocks: R B' D' R D B R (18)
EO and F2L: R D' R' D2 B' D' B (24)
Finish: U R' U' R2 U D R' U' R D' R2 (35)
z' R U L R' L F' D2 R' U2 R B' U' B F' U' F2 U F' R U' R' U B U2 B' R' F R F' R B U B' U' R' B' U2 B U2 B' R B U B' U' B' R' B2
OH results were absolutely terrible (Left my OH cube at Princeton Open). 4x4 went very well. FMC was found in 20 seconds. Magic times are from competition.
I really need to stop and think on those 9-movers sometimes - my muscle memory still doesn't know which way they go. The others were just weird - I don't normally make so many wrong direction mistakes; I don't know what happened this week.
I suppose that is a good point!
