dailysudoku.com

[DonM]

Asellus, I get your point, but it strikes me that a) when it comes to patterns, certain patterns for a given construct (ie. SDC) will appeal to different brains differently and b) trying to describe the backstory of these patterns (no matter what one's interpretation) makes them seem more complicated than they really are.

When it comes to these particular SDC, I quickly see that there are 5 digits, 23589, in 3 cells at r2c789 ie. 'the core'. There is a bivalue cell, (58)r1c9, in the same box (B3), both digits occurring in 23589. Then there is a a bivalue cell, (39)r2c5, both digits also occurring in 23589, but not in (58)r1c9. That leaves the digit 2 in r2c789.

Thus all the other 5s and 8s in B3 (except those in the core) can be removed. Likewise all the other 3s and 9s in R2 (except those in the core) can be removed. Finally, all the 2s can be removed from B3 and R2 (except those in the core). Finding 5 digits in 3 cells or 4 digits in 2 cells and a suitable bivalue cell in the same box and same line is pretty simple and these are, by far, the commonest forms of SDC one will come across.

But it's much more difficult to describe when one tries to cover all the possible bases (which is why terms like aals and aaals have to be used to describe 'the core'). I pity the new solver trying to understand these general descriptions (mine in the tutorial and everyone else's) and, in addition, figure out what's going on from rather obscure text-based grid examples. Which is why there are a number of colored graphics examples in my tutorial (mentioned above) which are IMO far more informative.