• Welcome to the Speedsolving.com, home of the web's largest puzzle community!
    You are currently viewing our forum as a guest which gives you limited access to join discussions and access our other features.

    Registration is fast, simple and absolutely free so please, join our community of 40,000+ people from around the world today!

    If you are already a member, simply login to hide this message and begin participating in the community!

Random Cubing Discussion

ryanj92

Member
Joined
Dec 26, 2011
Messages
1,575
Location
Sheffield, UK
WCA
2012JONE03
YouTube
Visit Channel
has there ever been much thought put into devising a 'recognition metric' to compare different methods/steps?

a simple example might be 'this step scores 1 point for every sticker you need to look at in order to determine the case'

so OLL would be perhaps 6 (8 u layer stickers, but you only -need- to see 6 of them), and also PLL (as PLL can be recognised from 2 sides). and then OLLCP would be 9-10 (6 for OLL, 3-4 for CP), and EPLL 2.

needs a bit more thought, but i'd be interested to hear if anyone else has tried this before :)
 

YouCubing

Member
Joined
Mar 31, 2015
Messages
2,420
Location
Roswell GA
WCA
2015JOIN01
Whether one hour off was a +2 or a DNF would've still cost him the record, so I don't really know what you're on about.

Hmm, good point.
But I still don't like the Clock regs :p

imo you should really be complaining about the clock companies

This is the truest post ever.

I think as a community we should learn from this
(video)

The post that got Justin banned.
Well really his post got him banned but you know what I mean.
 

ottozing

Platinum Member
Joined
Aug 13, 2011
Messages
3,289
Location
Canberra, Australia
WCA
2012MCNE01
YouTube
Visit Channel
has there ever been much thought put into devising a 'recognition metric' to compare different methods/steps?

a simple example might be 'this step scores 1 point for every sticker you need to look at in order to determine the case'

so OLL would be perhaps 6 (8 u layer stickers, but you only -need- to see 6 of them), and also PLL (as PLL can be recognised from 2 sides). and then OLLCP would be 9-10 (6 for OLL, 3-4 for CP), and EPLL 2.

needs a bit more thought, but i'd be interested to hear if anyone else has tried this before :)

This sounds really cool, and I'd like to see this explored for sure. Another thing that would have to be taken into account is the number of places you have to look for such pieces. Stuff like second block of Roux, F2L pairs, basically anything where what you solve after recognition leaves unsolved pieces (Obviously OLL is different because you're involving all of the pieces and only involving one sticker on each piece). Not sure how that would work out though, and I'm not really cunning enough to think of anything more codified beyond that :p
 

AlphaSheep

Member
Joined
Nov 11, 2014
Messages
1,083
Location
Gauteng, South Africa
WCA
2014GRAY03
a simple example might be 'this step scores 1 point for every sticker you need to look at in order to determine the case'
An obvious extension would be +1 point for every additional face involved. Because a hypothetical recognition system that involves looking at one sticker on each of the 6 faces is far inferior to OLL recognition where all 6 stickers are on the same face.

Also needs a means of taking into account blocks, because recognising a 2x2x1 block is easier than recognising 3 different stickers (excluding a centre?)

Needs more thought, but I like the idea.
 

ryanj92

Member
Joined
Dec 26, 2011
Messages
1,575
Location
Sheffield, UK
WCA
2012JONE03
YouTube
Visit Channel
This sounds really cool, and I'd like to see this explored for sure. Another thing that would have to be taken into account is the number of places you have to look for such pieces. Stuff like second block of Roux, F2L pairs, basically anything where what you solve after recognition leaves unsolved pieces (Obviously OLL is different because you're involving all of the pieces and only involving one sticker on each piece). Not sure how that would work out though, and I'm not really cunning enough to think of anything more codified beyond that :p

An obvious extension would be +1 point for every additional face involved. Because a hypothetical recognition system that involves looking at one sticker on each of the 6 faces is far inferior to OLL recognition where all 6 stickers are on the same face.

Also needs a means of taking into account blocks, because recognising a 2x2x1 block is easier than recognising 3 different stickers (excluding a centre?)

Needs more thought, but I like the idea.

very good points! my first thought is to suggest 1 point for each 'layer' involved - so last layer stuff would score 1, <R,U> stuff 2, etc.
either added on to the total or displayed like x+y or x/y...
i'm surprised nobody has thought about this much before - i might start a thread about this tomorrow when i've given it some more time :)
 

mark49152

Premium Member
Joined
Oct 29, 2012
Messages
4,719
Location
UK
WCA
2015RIVE05
YouTube
Visit Channel
https://gyazo.com/410123700d45cf0f66a04a30b6f1be82

when did this change? are the 6 and 7 icons painfully small on other peoples monitors too? i can barely count the layers, thinking a square with a 6/7 in it would be more clear...

EDIT: oh i see if you hover over it tells you the event i guess that's okay
What an awful, pointless change. Yes the icons look OK, but the information is not as well laid out or clear, and it is horribly slow and buggy on mobile. A classic case of making cosmetic changes for the sake of it at the expense of practical utility. I do hope they don't ruin the rest of the WCA site by making similar changes. IMHO the current WCA site design is one of the best I have seen - simple, clear, compact and uncluttered.
 

whauk

Member
Joined
Sep 28, 2008
Messages
464
Location
Germany
WCA
2008KARL02
YouTube
Visit Channel
has there ever been much thought put into devising a 'recognition metric' to compare different methods/steps?

a simple example might be 'this step scores 1 point for every sticker you need to look at in order to determine the case'

so OLL would be perhaps 6 (8 u layer stickers, but you only -need- to see 6 of them), and also PLL (as PLL can be recognised from 2 sides). and then OLLCP would be 9-10 (6 for OLL, 3-4 for CP), and EPLL 2.

needs a bit more thought, but i'd be interested to hear if anyone else has tried this before :)

Recognizing blocks is easier than recognizing difficult patterns of adjacent/opposite colors. So this metric should probably include the type of comparison that you have to apply to the stickers. Not sure if this was your intention or how to formalize it though.
Example: E-perm recognition (Do you have to do a U before the alg or not?) is way harder than PLL-skip recognition.
 

guysensei1

Member
Joined
Nov 24, 2013
Messages
5,143
Location
singapore
WCA
2014WENW01
I do not own a rubiks clock, I am signed up for clock in a comp that's less than a month away, there's no time to get a physical one, so I intend on borrowing one on the day itself to use. But I know how to solve one (I did it once on a simulator)
I did this for pyraminx too last year

Anyone else do this? Lol
How easy is clock to sub-30? Because that's the cutoff.
 

qqwref

Member
Joined
Dec 18, 2007
Messages
7,834
Location
a <script> tag near you
WCA
2006GOTT01
YouTube
Visit Channel
guysensei: Clock is very easy to sub-30, especially if you are familiar with the method and recognition. The hardest part for you will be knowing how much you have to move your fingers to turn the corners a given amount. If you warm up a bit with the clock you're going to use you should be able to do it with only minor corrections. Each solve has only 14 turns so you have plenty of time.


A 3x3x3 alg I found in a word document I made on 3/31/2000:
LB’T’RBTR’BL’D2RFD2B’F’DBDL’D’LD2R’DRDFD’FT2B’T2B’F2DBDRD’R’L’DLDBD’B’F’D’FLDL’F’DFDR’D’RD’R’D2RD’
Or in modern notation:
LB'U'RBUR'BL'D2RFD2B'F'DBDL'D'LD2R'DRDFD'FU2B'U2B'F2DBDRD'R'L'DLDBD'B'F'D'FLDL'F'DFDR'D'RD'R'D2RD'
 
Top