Big cubes time breakdown

[Pedro]

I did some big cube solves today, timing the separate steps so I could know how bad/good I'm on each one...

the 5x5x5 went like this:

1:13 | (1:21) |- 1:12 -|- 1:08 -| (1:03) = 1:11 |
2:10 | (2:27) | (2:04) |- 2:11 -| 2:10 = 2:10 |-> 4:01
0:42 | (0:44) |- 0:38 -| (0:37) | 0:40 = 0:40 |
----- | ----- | ----- | ----- | -----
4:05 | (4:30) | 3:54 | 3:56 | (3:53) = 3:58

the 4x4x4:

0:20 -|- 0:25 -|- 0:22 -|- 0:21 -|- 0:20 -|- 0:23 -| (0:14) |- 0:17 -|- 0:15 -|- 0:18 -|- 0:22 -| (0:26) = 20.3
0:37 -|- 0:36 -|- 0:28 -|- 0:44 -| (0:45) |- 0:44 -|- 0:36 -|- 0:30 -| (0:24) |- 0:36 -|- 0:44 -|- 0:30 = 35.5
0:34 -|- 0:36p | 0:27p*|(0:44)op*|0:42op| 0:30 -|- 0:33 -| 0:34p -| (0:25) | 0:29p -|- 0:30 -| 0:38op = 33.3
------ | ------ | ------ | ------ | ------ | ------ | ------ | ------ | ------ | ------ | ------ | ------
1:31 -|- 1:37 -|- 1:17 -| (1:49) |- 1:47 -|- 1:37 -|- 1:23 -|- 1:21 -| (1:04) |- 1:23 -|- 1:36 -|- 1:34 = 1:30.6

where "o" means oll parity, "p" means pll parity and * means that step was skipped

(I hope I layed it out in a good way )

so, as you can see, I'm far from good on the 5x5...need to cut like 20 seconds on the centers and at least 1 minute on the edges... but that would make me still a high-2 minutes solver...

I'm not that bad on the 4x4...getting centers and edges a few seconds better would make me average like 1:10...

so, what about you guys? how does the normal guys' times break? and the good ones? Frank says on bigcubes that master centers on the 4x4x4 should be under 20, and edges under 30, but that would give you almost 50 seconds...to be sub-60 you'd need to do the 3x3x3 part (probably with parity(ies) ) in less than 10s...

 

[pjk]

I am no good on the 5x5, and decent on the 4x4, here are my breakdowns:
4x4:
Centers: 16
Edges: 31
3x3 Stage: 30
---------
77 Seconds = 1:17 Avg

5x5:
Centers: 55
Edges: 100
3x3 Stage: 30
-----------
185 Seconds = 3:05 Avg

I need to practice both more, and I will drop my times drastically.

Pedro, your 3x3 solving stages are really slow compared to your real 3x3 avg. You should be able to shave off a good 10+ seconds on each.

 

[AvGalen]

4x4x4 = 25 + 30 + 45 = 1:40
5x5x5 = 50 + 60 + 50 = 2:40

For step 2 I don't understand what happens to most people. Consider this:
On 4x4x4 you have to pair 12 edge-wings with 12 edge-wings that can be in any of 22 places
On 5x5x5 you have to pair 24 edge-wings with 12 edge-centers that can be in any of 12 places.
Shouldn't 5x5x5 edge-pairing be relatively faster than 4x4x4 edge-pairing? As you can see, mine is exactly twice as slow with room for improvement if I could move faster.

 

[Pedro]

Pedro, your 3x3 solving stages are really slow compared to your real 3x3 avg. You should be able to shave off a good 10+ seconds on each.

Yeah...but I guess that has 2 main reasons:
on the 4x4x4, I'm counting parity fix(es) within the 3x3 part...so that makes it a little slower...and I can't move like the 3x3, despite my 4x4x4 turning good...

for the 5x5x5 I can only blame my cube...it's not half as loose/smooth as the ones I see guys doing in videos...I think I could get faster if my cube gets better...

4x4x4 = 25 + 30 + 45 = 1:40
5x5x5 = 50 + 60 + 50 = 2:40

For step 2 I don't understand what happens to most people. Consider this:
On 4x4x4 you have to pair 12 edge-wings with 12 edge-wings that can be in any of 22 places
On 5x5x5 you have to pair 24 edge-wings with 12 edge-centers that can be in any of 12 places.
Shouldn't 5x5x5 edge-pairing be relatively faster than 4x4x4 edge-pairing? As you can see, mine is exactly twice as slow with room for improvement if I could move faster.

isn't that 24 places?

and for the 5x5x5 you have to pair 12 edge-centers that can be in any of 12 places with 24 edge-wings that can be in any of 24 places...so I guess that's reason...a lot more stuff to search

 

[ExoCorsair]

I am very slow on the 4x4x4...

Centers: 20
Edges: 30
3x3x3: 50+

I'm a a little better at 5x5x5.
Centers: 35-40
Edges: 50-55
3x3x3: 35-40

I think that on the 4x4x4, solving the centers + edges in sub-40 is a good goal: centers in sub-20, edges in 20. That would probably give enough room for sub-60 solves.

 

[AvGalen]

isn't that 24 places?

No, the first edge-wing can be anywhere of that 24 places. The edge-wing that you want to match the first edge-wing with cannot be at the same place and it cannot be connected to it because then it would be already matched/solved. That leaves 22 places.

and for the 5x5x5 you have to pair 12 edge-centers that can be in any of 12 places with 24 edge-wings that can be in any of 24 places...so I guess that's reason...a lot more stuff to search

That is the big-cubes method. I was talking about my own method. You pick an edge-wing at a fixed position (1/1) and you match it with its center (1/12 or actually 1/11 places because of the semi-pair start)

 

[Jack]

I think I am better at 4x4.

4x4
Centers: 16
Edges: 30-35
3x3: 25-30 (depends on parities mostly)

I have had 3 solves before that were turned into 3x3 in about 37 seconds, one of them had parity and the other two were pops...

5x5
Centers: 45
Edges: 1:35
3x3: 35

 

[Pedro]

That is the big-cubes method. I was talking about my own method. You pick an edge-wing at a fixed position (1/1) and you match it with its center (1/12 or actually 1/11 places because of the semi-pair start)

hmm...I only took a look at your method, didn't read all...but by what you're saying, you're not solving the whole edge triplet just wing+center...isn't there another step after that?

 

[AvGalen]

No, no other step.
Bigcubes method: Match 1 center with 2 wings. Repeat 12 times.
My method: Match 1 wing with its center on the forward slice, match another wing with its center on the return slice. Repeat 12 times.

Bigcubes method saves a few moves, my method makes finding pieces easier, has better look-ahead and you only need 1 parity-fix.

 

[Pedro]

oh...

well, I still think edge "pairing" in 5x5 should really take longer...because on the 4x4 you have to do just one "matching" for each edge group, while you have to do 2 "matching" in the 5x5...

I think doing it your way is kinda like doing 2 at once in the 4x4, but you do just 1 group instead...

 

[AvGalen]

My method is exactly like 2-at-a-time on the 4x4x4. You match an edge-wing with another piece with a d-move, then replace the matched pieces, then match another edge-wing with another piece with a d'-move.

On a 4x4x4 you have to do 2-at-a-time 6 times.
On a 5x5x5 you have to do 2-at-a-time 12 times.
On a 6x6x6 you have to do 2-at-a-time 18 times.
On a 7x7x7 you have to do 2-at-a-time 24 times.

As you can see it is a lineair scale and on 5x5x5 it should take twice as long as on the 4x4x4.

P.S. My method is not really exactly like 2-at-a-time because I start with a semi-pair to avoid weird cases.

 

[pjk]

Pedro, my times listed included parities as well, hence why it is 30 seconds instead of 20. My 5x5 is really tight too, but that allows me to lower the move count which compensates somewhat. With a better cube 25 seconds wouldn't be hard.

To take off 10 seconds on 3x3 stage, just work on working in your cubes. Good luck.

 

[DanHarris]

Arnaud, I am "most people" in your book, at the moment I seem to be much slower (relatively) on the 5x5x5 edges than on the 4x4x4 edges. There are two main reasons for this I think: firstly there is just so much more to "see" on the 5x5x5, I take our point about only having to look in a certain number of places for the piece you are looking for, but when everything is still scrambled it can be very hard to focus on the right spots (if you see what I mean). Also, I think with the 5x5x5 there are more tricky cases, such as when pieces are stuck in your working "M-ring", and could even be the wrong way round, just working these out might cause delays. I am still trying to get the hang of your semi pair method but it's not that fast for me still

As for time breakdowns, my absolute best for the 4x4x4 is around 12, 18, 20, my fastest solves have been when edges+centres are done in 30-32 seconds. On average, it is more like 15, 23, 25.

For the 5x5x5, average is about 40, 70, 25, So I still need to work on my edges of course.

Another point to make, I find one of the hardest parts about solving the big cubes is making the cross directly after pairing up the edges. Does anyone have some useful advice for this "no-inspection" cross?

Dan

 

[AvGalen]

firstly there is just so much more to "see" on the 5x5x5, I take our point about only having to look in a certain number of places for the piece you are looking for, but when everything is still scrambled it can be very hard to focus on the right spots

With my method, "the right spots" are the 12 center edges. With the Big Cubes 5x5x5 Method and also on 4x4x4 "the right spots" are the 24 wing edges.

Also, I think with the 5x5x5 there are more tricky cases, such as when pieces are stuck in your working "M-ring", and could even be the wrong way round

Those tricky cases also happen when you do 6-at-a-time on 4x4x4. If you do 2-at-a-time on 5x5x5 and only use d (not d2) to match you avoid all bad cases. All special cases that remain aren't bad because they either give you good look-ahead (flipped edge) or 3-at-a-time (parity before end, 3-cycle)

I believe that edges on 4x4x4 can be done in 20 seconds by an expert, so I also believe they can be done on a 5x5x5 in 40 seconds by an expert. I am talking averages here! That also means that a 1:30 averages in total is possible (30+40+20)

Does anyone have some useful advice for this "no-inspection" cross?

I am also having trouble with this, but I guess you should just do inspection during the parity-fix (or last 3-cycle) or train "no-inspection" cross on the 3x3x3

 

[Jack]

When I do the no inspection cross, I just usually look for a white edge piece during the last few moves, then put it in place as soon as I am done the edges, then while I put that in place I can usually see another one, and soon it is just like a normal 3x3 solve. I find that most of the time it is quite smooth from edges to 3x3.

 

[Pedro]

I think Jack is right...I just try to look for a white edge while doing my last alg on the 4x4x4 or 5x5x5...

Pedro, my times listed included parities as well, hence why it is 30 seconds instead of 20. My 5x5 is really tight too, but that allows me to lower the move count which compensates somewhat. With a better cube 25 seconds wouldn't be hard.

To take off 10 seconds on 3x3 stage, just work on working in your cubes. Good luck.

can you do finger tricks on your 5x5? I can barely turn an outer layer with my indexes dunno why it's not good yet, having it for some time and shaving the wing edges...Ron said he could prepare a good 5x5 in like 2 weeks...and mine is about 2 or 3 months

 

[Mike Hughey]

So you guys can have a good laugh, here's the breakdown from someone who's really terrible at both (me):

4x4x4: 30 55 65 = 2:30
5x5x5: 75 105 55 = 3:55

I really think I might wind up better at 5x5x5 than at 4x4x4 someday, at my current rates of improvement. As you can see, I'm better at edge-matching on the 5x5x5 than on the 4x4x4, relatively speaking. I know it doesn't make much sense, but it's true.

It would obviously help a lot if I got better on 3x3x3. Hey, I'm pretty happy with the 3x3x3 part of 5x5x5 at 55 seconds, considering I'm 35 seconds at best with 3x3x3.

 

[Pedro]

Hey folks

I did another avg today (this time 12 solves) for the 5x5, and here's what I got

1:06 | 1:10 |(1:18)| 1:07 | 1:08 | 1:04 | 1:14 |(0:55)| 1:02 | 1:02 | 1:08 | 1:10 => 1:07
1:57 | 1:45 | 1:50 | 1:38 | 1:46 | 1:50 |(2:00)| 1:40 | 1:50 |(1:32)| 1:33 | 1:33 => 1:44
0:44 | 0:35 | 0:42 | 0:39 | 0:32 | 0:40 |(0:30)| 0:42 | 0:33 | 0:35 | 0:32 |(0:51) => 0:37
----------------------------------------------------------------------------------
3:47 | 3:30 |(3:50)| 3:24 | 3:26 | 3:34 | 3:44 | 3:17 | 3:25 |(3:09)| 3:13 | 3:34 => 3:29

all I can say is...WOW! What's that?! I improved like 25 seconds on the edges part!

I tried 5 solves with Arnaud's technique, and got 2:14, 2:14, 2:43, 2:06 and 2:05...then I went back to my normal way, and got this

sorry Arnaud, but I think I'll stick with my current way...

I'm really impressed...don't know what happened...

 

[Mike Hughey]

I tried 5 solves with Arnaud's technique, and got 2:14, 2:14, 2:43, 2:06 and 2:05...then I went back to my normal way, and got this

I have had the opposite happen on several occasions - I try 5 solves or so with the bigcubes.com method, and then go back to Arnaud's method, and suddenly get much better times. It's funny how using a different method can speed you up on your regular method. I may have to try it again and see if I can get another improvement.

It looks like the big difference between you and me at the moment is the 3x3x3 solve time.

 

[ExoCorsair]

can you do finger tricks on your 5x5? I can barely turn an outer layer with my indexes dunno why it's not good yet, having it for some time and shaving the wing edges...Ron said he could prepare a good 5x5 in like 2 weeks...and mine is about 2 or 3 months

Get a Mefferts.

My competition cube for 5x5x5 will most likely remain as the Eastsheen (size, color scheme), but Mefferts 5x5x5 cubes are SO much better than Rubik's.