[3BLD] Old Pochmann Dilemma

[Timona]

I rewatched JPerm's Old Pochmann video from 5 years ago and I'm a little confused.

The way I remember OP for corners, the swapping alg is (R U' R' U') R U R' F' (R U R' U') R' F R. According to his video, the buffer piece is UBL and you swap with E (according to Speffz) with V of DFR.
But the way I have always done it or I remember doing it, you swap A (buffer) with P (target).

For example, F (R U' R' U') R U R' F' (R U R' U') R' F R F'.
According to JPerm's tutorial, the setup, swap and setup above swaps E and M.
But from my own perspective, the way I learnt it, it swaps A and D.

If I do it this way, does it change any other part of the solve, like does it affect parity or does everything remain the same?

 

[IsThatA4x4]

I rewatched JPerm's Old Pochmann video from 5 years ago and I'm a little confused.

The way I remember OP for corners, the swapping alg is (R U' R' U') R U R' F' (R U R' U') R' F R. According to his video, the buffer piece is UBL and you swap with E (according to Speffz) with V of DFR.
But the way I have always done it or I remember doing it, you swap A (buffer) with P (target).

For example, F (R U' R' U') R U R' F' (R U R' U') R' F R F'.
According to JPerm's tutorial, the setup, swap and setup above swaps E and M.
But from my own perspective, the way I learnt it, it swaps A and D.

If I do it this way, does it change any other part of the solve, like does it affect parity or does everything remain the same?

I'm pretty sure it works just fine, but I'm by no means a BLD expert.

 

[Timona]

I'm pretty sure it works just fine, but I'm by no means a BLD expert.

I mean, it should because as far as I know, it does the exact same thing for the exact same pieces, just different stickers.

 

[AbsoRuud]

Everyone on the planet who uses Old Pochman swaps A with P (or L, in case you use the KRZ letter scheme). IMO Jperm uses different letters for the sole purpose of being different.

 

[Garf]

Everyone on the planet who uses Old Pochman swaps A with P (or L, in case you use the KRZ letter scheme). IMO Jperm uses different letters for the sole purpose of being different.

Not everyone... I use E with V. It is easier for me to visualize pieces swapping to that location.

 

[ender9994]

Everyone on the planet who uses Old Pochman swaps A with P (or L, in case you use the KRZ letter scheme). IMO Jperm uses different letters for the sole purpose of being different.

Back when I used old pochman, I had no letter for the buffer, and my targets were either A, D, or G depending on which of the 3 other top corners I was targeting (aka target piece would either be UFL, UFR, or UBR).

I still use a custom letter scheme. It makes no difference really, except when trying to write out the sequence for someone else

 

[mookiemu]

Why are you wasting time and effort learning pochman? Old Pochman is a dinosaur and will be limiting. And because there are so many moves just to solve one piece, it is ripe for mistakes and makes it nearly imposible to be sub 2minutes. 3-style is only slightly harder to learn, but more efficient and it teaches you how the cube works. Sub 20 second solve times are attainable with 3-bld.
I started out learning Pochman and I did. When I came upon Noah's tutorials, I realized what a waste of time learning Pochman was. 3-style is fast and once you get the concepts of commutators and conjugates, you will be making up algorithms on the fly.
Noah Cubes has some great tuts. Check out Noah's tutorials. they are a little outdated because most high-level BLD solvers today use hybrid and freestyle methods and maybe even 5-style. But what you learn here will carry you further if you want to go. Old Pochman will not allow you to go further.
This one here is the first in a series of 10 that will have you up and running in no time.

 

[AnthonyRochester]

Swapping A with P achieves exactly the same thing as swapping E with V, you're just looking at a different side of the piece

 

[abunickabhi]

Why are you wasting time and effort learning pochman? Old Pochman is a dinosaur and will be limiting. And because there are so many moves just to solve one piece, it is ripe for mistakes and makes it nearly imposible to be sub 2minutes. 3-style is only slightly harder to learn, but more efficient and it teaches you how the cube works. Sub 20 second solve times are attainable with 3-bld.
I started out learning Pochman and I did. When I came upon Noah's tutorials, I realized what a waste of time learning Pochman was. 3-style is fast and once you get the concepts of commutators and conjugates, you will be making up algorithms on the fly.
Noah Cubes has some great tuts. Check out Noah's tutorials. they are a little outdated because most high-level BLD solvers today use hybrid and freestyle methods and maybe even 5-style. But what you learn here will carry you further if you want to go. Old Pochman will not allow you to go further.
This one here is the first in a series of 10 that will have you up and running in no time.

I still think it is better to go step by step, and learn Pochmann or Orozco first before jumping to 3-style.

 

[mencarikebenaran]

I still think it is better to go step by step, and learn Pochmann or Orozco first before jumping to 3-style.

yeah i think old pochmann still need to learn.
because sometimes u will meet some cases that can be solved using old pochmann method.

 

[mookiemu]

I still think it is better to go step by step, and learn Pochmann or Orozco first before jumping to 3-style.

I don't know, I learned old pochman first and it turned out to be a waste of time. 3-style and R2M2(R1M1-R'1M'1) variations are just as easy to learn, and they are interchangeable with each other allowing you to pick what is more efficient for a cycle. And this way, way, way more efficient than Pochman. Once I learned 3style, I was like, what the hell did I waste my time on Pochman for? It's nearly impossible to get sub 2 minutes on the solve portion of BLD with Pochman. And having learned Pochman and 3-style I found that 3style is just as easy/hard to learn as Pochman. There are just too many moves in Pochman for each piece solved and too much room for error. Not only that, learning 3-style and RM solves teaches you how the cube works and gives you an understanding of the cube that pochman does not. You can literally create algs on the fly, and sub 10 for the solving portion is very doable. That's impossible with Pochman. The only thing pochman is good for is as one way to handle parity.

 

[Thom S.]

I don't know, I learned old pochman first and it turned out to be a waste of time. 3-style and R2M2(R1M1-R'1M'1) variations are just as easy to learn, and they are interchangeable with each other allowing you to pick what is more efficient for a cycle. And this way, way, way more efficient than Pochman. Once I learned 3style, I was like, what the hell did I waste my time on Pochman for? It's nearly impossible to get sub 2 minutes on the solve portion of BLD with Pochman. And having learned Pochman and 3-style I found that 3style is just as easy/hard to learn as Pochman. There are just too many moves in Pochman for each piece solved and too much room for error. Not only that, learning 3-style and RM solves teaches you how the cube works and gives you an understanding of the cube that pochman does not. You can literally create algs on the fly, and sub 10 for the solving portion is very doable. That's impossible with Pochman. The only thing pochman is good for is as one way to handle parity.

Mostly true, but you have to acknoledge that OP still has the least high learning curve out of all Methods
M2 (R2 too) has you worrying about if your on an odd or even target.
Kind of the same of Orozco
Turbo needs you to think about 2 Targets at a time.
And 3-Style / BH, despite becoming more and more intuitive over time, still needs you to be comfortable with the whole concept of commutators.

I think to remember old forum posts of Stefan telling about how before OP came to the scene, people solved mostly with commutators too, but not in the current BH way and it was very difficult to explain and learn relative to how difficult OP is to learn.

 

[mookiemu]

Mostly true, but you have to acknoledge that OP still has the least high learning curve out of all Methods
M2 (R2 too) has you worrying about if your on an odd or even target.
Kind of the same of Orozco
Turbo needs you to think about 2 Targets at a time.
And 3-Style / BH, despite becoming more and more intuitive over time, still needs you to be comfortable with the whole concept of commutators.

I think to remember old forum posts of Stefan telling about how before OP came to the scene, people solved mostly with commutators too, but not in the current BH way and it was very difficult to explain and learn relative to how difficult OP is to learn.

I acknowledge that OP is easier to understand because it solves piece by piece and there are fewer (but much longer algorithms). But, even with OP, you are still solving in terms of cycles, ie two pieces because even in OP you have to know whether the cycles are odd or even. And you still have to deal with cycles in the memo. And you have to deal with cycle breaks and parity. So thinking of 2 targets at a time is no big deal.
Once you get the idea that you are solving in cycles, (two pieces at a time), R2M2 (and R and M), turbo, BH, and commutator/conjugate solves become much easier to understand because you are solving the word (that you've used for your memo). And if you do your setup before each cycle, it's ridiculously easy and efficient.
For instance, for every face you solve, using old pochman, you have to perfom something like F (R U' R' U' R U R' F') (R U R' U' R' F R) F'), plus setups which usually add at 1 -3 moves to set up and 1-3 moves to put back. That means for every half cycle you have to do an average of 20 moves! That's 40 moves per cycle! That's 40 moves for every two solved pieces!!!
Now consider this, I want to cycle UBL>UBR>DFL. In Old Pochman, that's at least 40 moves for a tiny payday! That same cycle using a 3-style alg is L'U'L R2 U'L'U R2! 8 moves!
If you think in terms of cycles then all of the different techniques are interchangeable as long as you use one technique per cycle. And the idea would be to use whatever technique is most efficient for that cycle. It could be Turbo, R2, BH, commutators, even OP and if you add setups before each cycle, it's possible to solve each cycle in 10 or less moves! That's 10 moves for every two pieces!
It's worth going that little extra mile to start with Commutators like in Noah's 3-style video I posted above.

 

[mookiemu]

I wanted to add that the other advantages of going an extra mile and learning some sort of 3-style first, is that the idea of cycles becomes ingrained right from the beginning. You need to understand cycles for your memo (no one I know does their memo one letter at a time). Learning and understanding cycles allows you to do things like floating buffers, cancellations, floating parity (which mitigates the parity disadvantage by subtracting a whole cycle from your solve if you get parity).
If you do (setup)+(cycle)+ (undo setup). This gives you the power to pick and choose which technique you use for each cycle. You could for instance do (setup)+(R2R2)+(undo setup), and next cycle if it's more convenient, you could do (setup)+(commutator)+(undo Setup), and the next one (setup)+(turbo)(undo setup), all in one solve.
You can also do possible 5-style solutions. For instance If I have DF buffer for my edges and I come across two cycles I need to solve RU>UL and UB>FU (ID and AE) in my naming schedule, it would be nearly 80 moves to solve those four piece using pockman. It would be roughly 18 moves using M2, about 17 moves using just R and M moves, about 20 moves with Turbo. If I recognize those two memo words together ID (as in photo ID) and AEEEE! (As in the Fonz), I know that I can just do UM'U'M and I've solved 4 pieces! If I see RU-BL and then UB-FU (Inn and AEEE in my memo names) then would I do a (setup)+(5style cycle)+(undoSetup) which in this case is L(UM'U'M)L'. 6 moves and I've solved 4 pieces! There are many possible variations of UM'U'M and lots of setup opportunities for when you recognize two words that are solvable this way. Of course this is advanced, but understanding cycles can take you there!

 

[EngiNerdBrian]

I still think it is better to go step by step, and learn Pochmann or Orozco first before jumping to 3-style.

I agree with this. While there are surely more efficient methods OP is a great introduction with a lower barrier to entry into BLD. BLD is unquestionably difficult for many cubers to learn and OP has the least upfront information that needs presented and mastered.

 

[qaz]

I don't know, I learned old pochman first and it turned out to be a waste of time. 3-style and R2M2(R1M1-R'1M'1) variations are just as easy to learn, and they are interchangeable with each other allowing you to pick what is more efficient for a cycle. And this way, way, way more efficient than Pochman. Once I learned 3style, I was like, what the hell did I waste my time on Pochman for? It's nearly impossible to get sub 2 minutes on the solve portion of BLD with Pochman. And having learned Pochman and 3-style I found that 3style is just as easy/hard to learn as Pochman. There are just too many moves in Pochman for each piece solved and too much room for error. Not only that, learning 3-style and RM solves teaches you how the cube works and gives you an understanding of the cube that pochman does not. You can literally create algs on the fly, and sub 10 for the solving portion is very doable. That's impossible with Pochman. The only thing pochman is good for is as one way to handle parity.

This is simply not true. I averaged ~50s for memo+execution using M2/OP when I did BLD. 3-style is obviously much faster, but you are exaggerating the gap between them.

 

[mookiemu]

This is simply not true. I averaged ~50s for memo+execution using M2/OP when I did BLD. 3-style is obviously much faster, but you are exaggerating the gap between them.

You said you average 50seconds for memo and execution. Let's unpack that. Assuming you get good scramble with no parity and no flipped edges, the average BLD solve is 4 cycles for corners and 6 cycles for edges. OP corners average 20 turns per face or 40 turns per cycle (two faces). That's 160 turns just to solve the corners. Then if one uses OP for edges that's average 240 turns to solve the edges. That's 400 turns just to solve a cube.

OK, you are using M2 and saving some turns. Straight up M2 averages 16 moves per cycle except 4 bad cases which add another 8 turns to every cycle. But that is mitigated by the really good UB case, so we'll only count 3 bad cases. So 16x6 plus 3x8 turns for bad cases. That's 96+24=116. 116 turns just to solve the edges using M2. 116+240=356 moves just to solve the cube using OP/M2. You said you do memo and execution in 50 seconds. OK, the world class, cream of the crop memo averages about 10 secs, Tommy Cherry averages 8 seconds for the memo. Let's give you the benefit of the doubt and say your memo is an incredible 15 seconds. That leaves you with a solve time of 35 seconds. 356/35 is roughly 10TPS! World class average TPS in regular cube 3x3 is about 8TPS. So, if you are being truthful, you have a TPS of about 10 TPS. Right? If your memo is 15 seconds then that's 11TPS, and if it's 20 seconds, your TPS is about 12TPS! That's cream of the crop amazing. OK, I have no reason to doubt that you are telling the truth and you have incredible memo and TPS. How many people are gonna come close to that?

If your memo and TPS is that good, imagine how much better you would be if you learned R2? If you took the time to learn M2 (Which personally I think is easier than OP. I taught my 9 year old son M2), you could have learned R2 as well. R2 and M2 are pretty much the exact same concept. It's as easy to learn as M2 because it's the same exact idea! Each R2 cycle averages about 16 turns as well, except the 4 L-slice bad cases which add 8 moves apiece, minus the one UL R2 case. That's 16x4+(8x3)=88 turns just to solve corners. 88 turns +120 turns = 208 turns. Already a huge drop in turns. 208/40 is a much more reasonable 5TPS to get to a 40 second solve with a 10 second memo. At 10TPS you'd be a sub 30 BLD cuber. If you know M2, you already know R2 because it's the same thing! Why aren't you learning R2?

OK, you said I'm exagerating the difference between 3style and OP/M2. Let's unpack that too. If you were to learn 3 style, you could solve every cycle in 8-12 turns a cycle. Let's average that out to about 10 turns per cycle because there aren't that many 12 turn BLD cases. (6 cycles x 10 moves)+(4 cycles x 10 moves). That's 100 turns to solve the cube (could be a lot less if you learn floating buffers, floating parity, cancellations, and a few 5 style algs). 100/10TPS gives you and execution time of 10 seconds and with a memo of 10 seconds you would average sub-20. You don't think there is a big deal of difference between Sub 50 and sub 20? 30 seconds is an astronomical gap in the world of cubing.

 

[Abram Grimsley]

You said you average 50seconds for memo and execution. Let's unpack that. Assuming you get good scramble with no parity and no flipped edges, the average BLD solve is 4 cycles for corners and 6 cycles for edges. OP corners average 20 turns per face or 40 turns per cycle (two faces). That's 160 turns just to solve the corners. Then if one uses OP for edges that's average 240 turns to solve the edges. That's 400 turns just to solve a cube.

OK, you are using M2 and saving some turns. Straight up M2 averages 16 moves per cycle except 4 bad cases which add another 8 turns to every cycle. But that is mitigated by the really good UB case, so we'll only count 3 bad cases. So 16x6 plus 3x8 turns for bad cases. That's 96+24=116. 116 turns just to solve the edges using M2. 116+240=356 moves just to solve the cube using OP/M2. You said you do memo and execution in 50 seconds. OK, the world class, cream of the crop memo averages about 10 secs, Tommy Cherry averages 8 seconds for the memo. Let's give you the benefit of the doubt and say your memo is an incredible 15 seconds. That leaves you with a solve time of 35 seconds. 356/35 is roughly 10TPS! World class average TPS in regular cube 3x3 is about 8TPS. So, if you are being truthful, you have a TPS of about 10 TPS. Right? If your memo is 15 seconds then that's 11TPS, and if it's 20 seconds, your TPS is about 12TPS! That's cream of the crop amazing. OK, I have no reason to doubt that you are telling the truth and you have incredible memo and TPS. How many people are gonna come close to that?

If your memo and TPS is that good, imagine how much better you would be if you learned R2? If you took the time to learn M2 (Which personally I think is easier than OP. I taught my 9 year old son M2), you could have learned R2 as well. R2 and M2 are pretty much the exact same concept. It's as easy to learn as M2 because it's the same exact idea! Each R2 cycle averages about 16 turns as well, except the 4 L-slice bad cases which add 8 moves apiece, minus the one UL R2 case. That's 16x4+(8x3)=88 turns just to solve corners. 88 turns +120 turns = 208 turns. Already a huge drop in turns. 208/40 is a much more reasonable 5TPS to get to a 40 second solve with a 10 second memo. At 10TPS you'd be a sub 30 BLD cuber. If you know M2, you already know R2 because it's the same thing! Why aren't you learning R2?

OK, you said I'm exagerating the difference between 3style and OP/M2. Let's unpack that too. If you were to learn 3 style, you could solve every cycle in 8-12 turns a cycle. Let's average that out to about 10 turns per cycle because there aren't that many 12 turn BLD cases. (6 cycles x 10 moves)+(4 cycles x 10 moves). That's 100 turns to solve the cube (could be a lot less if you learn floating buffers, floating parity, cancellations, and a few 5 style algs). 100/10TPS gives you and execution time of 10 seconds and with a memo of 10 seconds you would average sub-20. You don't think there is a big deal of difference between Sub 50 and sub 20? 30 seconds is an astronomical gap in the world of cubing.

Dude, chill. @Timona was just asking a question. No need to start an argument.

 

[Cuberstache]

@mookiemu I don't think anyone is trying to say that 3-style isn't the best method for BLD because it obviously is. What people are trying to say is that it's not a good method to learn for someone who literally doesn't know how to solve a cube blindfolded. They simply want to get a success; speed is completely irrelevant. You mentioned movecount differences, but the point of OP is that the moves are extremely easy. You literally just do the same alg over and over again and only have to worry about figuring out and remembering 2ish setup moves per piece. 3-style is so much harder because you have to consider how to make an interchange, how to insert a piece into the interchange correctly, and which direction to do the alg in. For someone new to BLD, 3-style would not even be faster than OP.

 

[mookiemu]

Dude, chill. @Timona was just asking a question. No need to start an argument.

I'm not arguing. I thought we were having an discussion and I was honestly enjoying it and taking in the points being made. My bad.