[Help Thread] Multi-Blind Discussion

[Dylan Swarts]

Hi Blinders! (and other cubers )
Look here!

Result	Time[memo]	Points	Avg memo/cube	Avg exec/cube	Total avg/cube	Date	Errors	Comment
23/30	1:01:37[41:20]	16	1:22.66	40.56	2:03.22	31 March 2020	C1: did NK instead of KN, exec error. C3:inversed TX edges. C5: inversed UM corners. C9: incorrectly exec'd SV edges lol. C19: recalled SN instead of MK, realised when I got to a cube with SN, also inversed SU edges. C20: inversed QR, thinking I inversed all QT and QR in the before cubes, but only this one was actually wrong haha. C22: inversed SP corners.	Lots of inverses, can be fixed with more 3bld! 27 SUBHOURED, 2 solved over the hour, 1 dnf

Starting from left: Result is 2nd best with 30, not good but I am working on it.
Time: Fastest by almost 2 minutes! Both memo and exec are faster.
Avg memo & avg exec: OVERALL PB on both avg memo, exec, and total/cube!! I find it hilarious though, that this (and all other 30 cube attempts I've done) are better than my usual with low/mid 20 cube attempts. Imagine going back there, maybe I could get sub 2/ cube. Anyway.
Errors: As you can see, very stupid errors, almost only inverses. Definitely almost fixed, I'm at least doing almost 100% 3 style.

Conclusion: Going to start doing 33 and 30 cube attempts now, doing 30 until comfortably sub 1 (meaning avg per cube will have to be sub 2) and then I'll pick an amount after 33 to work on simultaneously.

From the like 5 posts of mine upwards ^^, I was doubtful of 42. Think we MIGHT be able to reach it again.
Also the videos of these attempts are being uploaded, around 4 min vids (I talk at the end sometimes, that's why), so if you would like to check them out, feel free to do so!
Thanks for reading, happy cubing!

 

[dudefaceguy]

My first multiblind attempt was a success! 2/2 cubes. I can't really describe how good that feels.

I memorized and executed edges first and used Graham Siggins's "weak swap" technique for parity. I had odd parity on both cubes and it worked well for me. Does anyone else here use this?

 

[Dylan Swarts]

I watched that 3 part series yesterday! I already do the parity thing, have been for a long time. Need to rewatch lots of it, the more advanced things are still not in my head, but pretty cool things, could save me some memo!

Grats on the multi, that is great! Took me a couple of tries to get a success

 

[dudefaceguy]

I watched that 3 part series yesterday! I already do the parity thing, have been for a long time. Need to rewatch lots of it, the more advanced things are still not in my head, but pretty cool things, could save me some memo!

Grats on the multi, that is great! Took me a couple of tries to get a success

Is edges-corners the standard memorization and execution order for multi blind? I'm just wondering if it's also viable to do corners first, since you then know whether the edges have odd parity. But then you would have to leave one corner unsolved, which could lead to errors.

Those are definitely great videos with a lot of advanced concepts. I'm going to re-watch them a few times now that I've had a chance to play around with weak swap. I'm still using flip commutators, which are slow and difficult to work into my memo.

 

[Habsen]

Is edges-corners the standard memorization and execution order for multi blind? I'm just wondering if it's also viable to do corners first, since you then know whether the edges have odd parity. But then you would have to leave one corner unsolved, which could lead to errors.

Those are definitely great videos with a lot of advanced concepts. I'm going to re-watch them a few times now that I've had a chance to play around with weak swap. I'm still using flip commutators, which are slow and difficult to work into my memo.

First of all, congratulations on your MBLD success. It took me a couple of attempts to achieve this.
Regarding the MBLD order, you should memorize and execute in the same order. I did it the other way for a while and I can tell you that this back and forth jumping in your memorized images is annoying and error prone. The beauty about the UF/UFR buffer combination is that you don't have to know whether you have parity or not. When you finish your edge memo and you realize that you have an odd number of targets, just append UR (B in Speffz) to your last letter pair. I am also using the weak swap described by Graham and I am really glad that I found these videos.

 

[dudefaceguy]

First of all, congratulations on your MBLD success. It took me a couple of attempts to achieve this.
Regarding the MBLD order, you should memorize and execute in the same order. I did it the other way for a while and I can tell you that this back and forth jumping in your memorized images is annoying and error prone. The beauty about the UF/UFR buffer combination is that you don't have to know whether you have parity or not. When you finish your edge memo and you realize that you have an odd number of targets, just append UR (B in Speffz) to your last letter pair. I am also using the weak swap described by Graham and I am really glad that I found these videos.

Great, thank you! Multi blind is so much fun.

 

[Keroma12]

I memorized and executed edges first and used Graham Siggins's "weak swap" technique for parity. I had odd parity on both cubes and it worked well for me. Does anyone else here use this?

I've been using weak swap for a long time. It's so automatic now, occasionally I accidentally do it in 3BLD.

Graham and I actually co-created the method. He showed me a special case, then I realized it could be done more generally, and together we ironed out the details. It has a few other theoretical applications, such as avoiding corner twist algs in multi, but they tend not to be worth it as they require more thought and only give you different algs instead of reducing the number of algs. But if you're not as interested in speed, as I think you mentioned, they might be worth checking out for fun.

 

[Dylan Swarts]

I've been using weak swap for a long time. It's so automatic now, occasionally I accidentally do it in 3BLD.

Graham and I actually co-created the method. He showed me a special case, then I realized it could be done more generally, and together we ironed out the details. It has a few other theoretical applications, such as avoiding corner twist algs in multi, but they tend not to be worth it as they require more thought and only give you different algs instead of reducing the number of algs. But if you're not as interested in speed, as I think you mentioned, they might be worth checking out for fun.

Yeah, I struggle grasping the concept but I believe just taking it a bit slower and rewatching last 2 vids should get me all good for it.

 

[Habsen]

Yeah, I struggle grasping the concept but I believe just taking it a bit slower and rewatching last 2 vids should get me all good for it.

Think of it as treating the UF and UR pieces the same (assuming these are the pieces that are swapped when you have parity). The one that comes first during tracing will be speculatively solved to UR (or RU if it's flipped) and the other one becomes your buffer piece which you use to break into new cycles. If at the end of tracing you realize that your speculation was wrong, i.e. you have an odd number of targets, you solve your buffer piece to UR and thereby do or undo the UF/UR swap.

 

[Dylan Swarts]

Basically that yes, thought I missed something else important, cause I pretty much know that . Thanks for the summary !!

 

[Dylan Swarts]

Sorry I didn't even crop out useless things but okay. Thought I'd share this. Look bottom right, and then top left, at those two vids.
Coincidence hahah. Of course Graham is at a comp, and that was long ago, in no way am I trying to say Rowe is Graham level (because that's untrue at this point in time)
Just thought I'd share \o/

 

[ProStar]

View attachment 11607
Sorry I didn't even crop out useless things but okay. Thought I'd share this. Look bottom right, and then top left, at those two vids.
Coincidence hahah. Of course Graham is at a comp, and that was long ago, in no way am I trying to say Rowe is Graham level (because that's untrue at this point in time)
Just thought I'd share \o/

ohmygosh I'm so glad you shared I need those micro farms!!

 

[Dylan Swarts]

Fun little epic fail from me. Okay well only timewise is it really a fail.

23/33

1:20:01[55:25]

13

1:40.75

44.69

2:25.44

4 April 2020

C2: inversed WJ corners smart boi. C3: memo'd half of edges completely wrong. C6: recalled 2 pairs out of order. C10: no idea, 3c. C16: wrong corner parity letter. C17:couldn't recall corners. oops C19: honestly idc anymore lol. C21: some dumb memoing. C26: missed a move in QA edges. C idk. Missed a cube lol this attempt is rubbish.

I are feel big stupid. 4 solved subhour woop.

Comparing this time to 30 cubes in 1:01 makes you see my problem. Definitely just a bad day. Bad focus. I don't think it has much to do with it being so many cubes, but it definitely might have had a small impact. Sub 1:07 soon I hope.

 

[brododragon]

ohmygosh I'm so glad you shared I need those micro farms!!

Nah areogel is where it's at.

 

[dudefaceguy]

I've been using weak swap for a long time. It's so automatic now, occasionally I accidentally do it in 3BLD.

Graham and I actually co-created the method. He showed me a special case, then I realized it could be done more generally, and together we ironed out the details. It has a few other theoretical applications, such as avoiding corner twist algs in multi, but they tend not to be worth it as they require more thought and only give you different algs instead of reducing the number of algs. But if you're not as interested in speed, as I think you mentioned, they might be worth checking out for fun.

Thanks for your contribution to multiblind! I'm using weak swap for 3BLD as well as MBLD, since my main goal is accuracy. I want to use the same set of techniques for both events right now. And since I'm so slow (7 minutes for 3BLD) memorizing corners first doesn't really yield benefits for me.

 

[Dylan Swarts]

I still need to work on implementing weak swap. Trying to do that and push for high numbers isn't working lol. I'm going to do a bunch of small attempts using it.

 

[fun at the joy]

I just did my first attempt with 10 cubes: 6/10 in 52:27.04 [39:58.15]

DNFs:
Cube2: off by 4E
some stupid slice execution mistake
Cube3: off by 5C
This cube had 4 twisted corners and I memoed the U/D stickers of the twists but then I didn't exec PVMCFDSX and only did PMFS rip
Cube4: off by 4E4C
It looks like I did an extra D/ missed a D'
Cube9: off by 4E4C
another move

The only thing I am happy about is that this was faster than my 8 cube attempt around a month ago and I only did a 5 Cube and a 6 Cube attempt in between.

 

[Dylan Swarts]

Well done @fun at the joy
While I'm here I'd like to share that I had successfully used Weak Swap in my multiblind yesterday! No dnfs were caused by it so I'm happy. And I think I saved 2 whole letterpairs in memo on my 30 cube attempt. Great technique, really enjoy using it!

 

[Rafaello]

Just done my first attempt with 3 cubes, 3x3 in 24:11, could be a lot faster (memo about 19:00). When I change edges to M2, I'll try with more cubes

 

[fun at the joy]

I wanted to do 10 cubes for C@H2 but I accidentally applied the same scramble to the last 2 cubes so I just started exec after I memoed 9.

7/9 in 51:14.93 [40:36.92]
DNFs:
Cube1: 6E6C - some stupid exec error because memo/recall was correct. I was pretty sure that this wasn't a success during exec.
Cube5: 4E4C - another move, looks like a missed F/ extra F

pretty bad time too, Idk what happened there.