Reduction Centers

[Dene]

Hey guys.
I've been thinking about my centers in reduction for a while and I am displeased. My times are around thus:
4x4x4: 10-20 seconds (this is a guess)
5x5x5: 30-40 seconds
6x6x6: 1:30-2:00 minutes
7x7x7: 2:45-3:45 minutes

These times aren't exactly flash.

Here is my thinking:
When I am doing centers, my chain of thought is purely intuitive (as opposed to (e.g.) OLL and PLL where I blindly follow an algorithm). As in, I don't see a few pieces and think "now I do this algorithm". Instead, I am always improvising my formulation (even though there are obvious patterns, I still don't think of my actions as algorithms).
An example: The 7x7x7 (seeing as this is the one I have the least experience with, it is the best example here). I generally start the center by doing the middle 3x3x3 block, then extend that to a 3x3x4 block, then 3x3x5, then 4x4x5, then finish it (a fairly standard technique, I guess).
But then I am extremely prone to being creative if I see a nice block. For example, I might start the 3x3x3 by having a 1x1x4 bar, instead of just a 1x1x3 bar. I might then think "I may as well just make this a 1x1x5 bar, then skip the extention later on". Surely this sort of variability is a good thing, but if it detains from following a standard technique it might not pay off (where the standard technique is certain to be of a reasonable speed, being creative might end up costing time).
What I would like to know is how do the "faster" guys out there view things when they do centers?
Do you allow for creativity?
Do you blindly follow algorithms, or do it all intuitively?

I feel I am losing speed in not having the "pre-designated" algorithms to follow in doing the centers. Thus any opinions (preferably of the "faster" guys) are greatly appreciated.

I know this is quite garbled, sorry. I am not very good at clarity. If you don't understand, please point it out.

 

[Pedro]

well, you're much better than me at centers, but what I do is:

4x4 - really can't decide between Pochmann and "normal" way. I'm leaning more towards the standard one (center 1, opposite, others)

5x5 - I do 1x2 or 1x3. Usually like: 1x2 with middle center, 2x2, 3x2, 3x3
or 1x3 with middle, 2x3, 3x3

6x6 and 7x7 - for this I use "bars". That is, 1x4 in 6x6 and 1x5 in 7x7
I don't know if it's the best way, but I like it...I usually start from middle to outside. So it's either:
6x6 - inner, inner, outter, outter ... or ... inner, outter, inner, outer
7x7 - middle, inner, inner, outter, outter ... or ... middle, inner, outter, inner outter

I think having a standard way makes you focus better. Of course you should be a little flexible for easy cases, but having a pre-set path helps you not think too much

 

[Mike Hughey]

6x6 and 7x7 - for this I use "bars". That is, 1x4 in 6x6 and 1x5 in 7x7
I don't know if it's the best way, but I like it...I usually start from middle to outside. So it's either:
6x6 - inner, inner, outter, outter ... or ... inner, outter, inner, outer
7x7 - middle, inner, inner, outter, outter ... or ... middle, inner, outter, inner outter

I think having a standard way makes you focus better. Of course you should be a little flexible for easy cases, but having a pre-set path helps you not think too much

I do exactly the same for 6x6x6 and 7x7x7, Pedro. In the same order, and I even vary between the two options just like you. I always go middle, inner, outer, inner, outer for the last 2 centers, since that leaves no more than 2 pieces that need commutators at the end. And I agree - for me a pre-set path keeps me from indecision, which is the biggest killer for me.

I'm terrible at centers on 7x7x7 - typically around 5 minutes. Then 3 minutes for edges, and about a minute for 3x3x3 phase. Over half my solve is centers.

 

[Swordsman Kirby]

Hey guys.
I've been thinking about my centers in reduction for a while and I am displeased. My times are around thus:
4x4x4: 10-20 seconds (this is a guess)
5x5x5: 30-40 seconds
6x6x6: 1:30-2:00 minutes
7x7x7: 2:45-3:45 minutes

These times aren't exactly flash.

Here is my thinking:
When I am doing centers, my chain of thought is purely intuitive (as opposed to (e.g.) OLL and PLL where I blindly follow an algorithm). As in, I don't see a few pieces and think "now I do this algorithm". Instead, I am always improvising my formulation (even though there are obvious patterns, I still don't think of my actions as algorithms).
An example: The 7x7x7 (seeing as this is the one I have the least experience with, it is the best example here). I generally start the center by doing the middle 3x3x3 block, then extend that to a 3x3x4 block, then 3x3x5, then 4x4x5, then finish it (a fairly standard technique, I guess).
But then I am extremely prone to being creative if I see a nice block. For example, I might start the 3x3x3 by having a 1x1x4 bar, instead of just a 1x1x3 bar. I might then think "I may as well just make this a 1x1x5 bar, then skip the extention later on". Surely this sort of variability is a good thing, but if it detains from following a standard technique it might not pay off (where the standard technique is certain to be of a reasonable speed, being creative might end up costing time).
What I would like to know is how do the "faster" guys out there view things when they do centers?
Do you allow for creativity?
Do you blindly follow algorithms, or do it all intuitively?

I feel I am losing speed in not having the "pre-designated" algorithms to follow in doing the centers. Thus any opinions (preferably of the "faster" guys) are greatly appreciated.

I know this is quite garbled, sorry. I am not very good at clarity. If you don't understand, please point it out.

For me, it's all intuitive, yet my center phase isn't that fast compared to other people in my time window.

 

[Kenneth]

I normaly solve the first two opposite at once and that from placing stripes that I build in the U-face, the bigger the cube the better it works I think.

For 4x4x4 I sometimes use "Guimond" = place mixed opposite centres and then sort them out.

For 5x5x5 I got a special for the first two: solve the side stripes for both sides first, then build both the middle ones, one in the U-face and one in D, place both in a single slice... sometimes you got a single pice stuck between the bars at one of the sides you are solving, then do a Niklas and place the bars in one go.

 

[Dene]

I think having a standard way makes you focus better. Of course you should be a little flexible for easy cases, but having a pre-set path helps you not think too much

Yes this is what my thinking is, but you never know. Sticking strictly to one routine might be better, or maybe it is better to do it purely intuitive. I guess it's hard to know >.<

 

[whauk]

a 3x3x3 block in a center??
that doesnt work! I guess you wanted to say 3x3!?