a little tough, but do-able

You know, I think they should really think about re-configuring how they do the rankings here.

It's not a very interesting system.

what would be more interesting to you?

I don't know... it's just that the way it is now, there really isn't anywhere for me to move unless I pay... but there's no challenge in getting a high ranking if you pay because you have access to all of the puzzles - you just have to play them all.

For me to reach the 90% section, there would have to be a total of 4,300 total available points and I'd have to finish every single game from now on.

it's not the most competitive rankings, it's true

I guess the way the rankings are set up now makes me feel LESS inclined to be "pro" because I don't see a challenge there.

but it isn't a very competitive site

I guess that's true. I'm just a competitive person. :)

Oh well. I guess I shouldn't complain unless I have a good solution. Off to bed!

gnight!

done

ekspert?? very easy

start

this certainly doesn't look expert!

done

go

done

hi all

a bit tricky

Looks fun!

I was hoping for an "easy" today.

I hate experts on weekdays, I rarely have time to finish them before I have to go.

Done! Had to use greenies but not too bad.

Difficulty score 60. Required a naked quad and some guessing.

Guessing is never required.

Isn't that the highest it's been?

No, we had an 81 once, if I remember right.

No. See http://www.pauls-pc-repair.com/sudoku
follow the table of scores link.

9. Going green.

August 2 is a 94, no wonder I can't solve it. hah

There are 2 puzzles that scored 94, which is the highest score awarded so far.

8 puzzles have been given scores greater than 60.

1 other puzzle received a score of 60 (Feb. 7).

Ah, you're right. I've solved October 1, though. Perhaps there's hope for August 2 after all, but I've tried (and failed) about four times on it.

How about since February, which is when I started?

March 24th was a 71.

I must have done it by guess.

So since Feb. 1st there have only been 3 puzzles with scores greater than or equal to 60.

Maybe I didn't see your score on those.

I'm going to have to put the rest of this one off until the afternoon.

I see a hidden pair. But its compliment is not naked.

Oh yeah, I guess it is naked.

Guess I'll take a break.

go

upper right Square is the key

13 filled. Stuck. Naked pairs all over the place. Looking for a wing or something.......

done. with greens and guessing

I have less trouble with most experts than I do mediums! I'll never understand my brain. lol

I think the reason is that experts involve different solution techniques than mediums.

I know I tend to over think the mediums.

I'm glad I'm not the only one. I guess I've just not grasped the techniques of the mediums. Figures...lol

Done. Had to make 1 guess. Rats.

I still need to read up on techniques. Would be nice to know what I am doing.

Try this site. It's one of the best I've found. http://www.angusj.com/sudoku/hints.php

Will check it out. Thanks.

somebody wanna help me out? a hint? please?

Once I figured out the upper left it all fell into place. HTH, I don't want to give away too much. ;o)

thank you

no problem!

there we go.

Guessed. Go it. I always feel ashamed when I guess.

wow...got it...finally!

hi folks

wow - really fell into place!

hi everyone

go

there are alot of number here for this to be an expert puzzle

Done, my first Expert , Yes yes yes, One small guess and was on my way. In record time, less time than any easy :) See you guys on Babble. Bye

Having to go green, don't feel too bad about that though.

going green just now too and don't feel bad at all ;)

I hate when I botch a technique, find that I messed up, but then it winds up unlocking the puzzle anyways because it eliminated a number from the square

I mean, I'm done now and it was good and all, but, I screwed up a technique and seeing the error showed the puzzle -- it doesn't feel like a good solve.

lol, at least you manage to solve it. I am still staring at the screen, not knowing how to go on without guessing. And I don't want to guess.

go

errrm, I'm a bit puzzled by this being expert, as it's the fastest I've ever done

errrm, I'm a bit puzzled by this being expert, as it's the fastest I've ever done

and no guessing?

well. I don't think it was a guess

I could start to hate you, LOL. Ok, seriously, so it can be solved without guessing. I will NOT give up then. Congrats!

there's one point at which it stops for a while, but I suppose if you then choose a square with two choices and follow them through to eliminate one

... then I must have been lucky in that choice of square

ah.... ok, lol. Trial and error.

how long did it take you andy?

7 or 8 minutes

very good!!

indeed

feels like a fluke

lol bet it does

you did great, wish I could too... still banging my head against a brick wall

finished in 20 mins :)

yay gemma! Guess, trial and error, or certainty all over?

guess lol

i guessed halfway through

at least I admit it lol

nothing wrong with guessing, lol... I just want to solve it, get the techniques down well enough... but for this one, I am just not good enough

aww you'll do it!! how long have you been working on it?

The only advanced technique needed here are a couple of XY chains.

almost an hour, but chatting at the same time, so that is not good

no its not. you need full concentration on this one!! plus I had help off my boyfriend :)

see, I still don't know what XY chain means

what is an xy chain? I have my own techniques & don't know what they are called.

ah drwho, that is a technique I just don't get. Read ugh explanations and I don't get it. I wish I could see someone DO it.

done

goofed up once & made an 'educated, haha' guess

yay helen :)

Better yet, learn how to do alternate inference chains and most of the other chain techniques come along for the ride.

well done helen :))

27 minutes of my life gone . . .

ok, solved it with one guess, guessed right first time, but still it does not feel like solving. Drwho, alternate inference chains? Never heard of those!

sprite, i'm with you on having to see a demonstration, the words just start sounding like blah, blah, technoblah

I never bothered about techniques, just reasoned it out for myself from the bottom up. But I find that I am lacking for some puzzles now.

yep, educated guess sounds much better

The explanation of alternate inference chains is here:
http://www.sudoku.com/forums.html?utm_s
ource=301/viewtopic.php

thanks!

I just don't know what the education is

see ya later, bye

Here is the idea of AIC: begin with the assumption this can't be true. The if this can't be true it implies that something else must be true.

ok, that is the kind of reasoning that I like! That makes sense to me.

drwho, is it in the topic advanced techniques?

If you chain together as many pairs of such inferences as you want, you always have a chain where the last assumption is true or your first assumption was false.

I found it, thanks :)

So you have a chain with a proposition on both ends, at least one of which must be true. This can then lead to eliminating possibilities from common buddies of the ends of the chain.

I so wish I could sit next to someone doing this. I get lost in the words.

I have the other problem, I can understand the words, I just can't spot the instances of the techniques on a sudoku grid!

perhaps I should try and find a dutch explanation. Or I must tackle a real good sudoku solver and have him sit next to me through some tough sudoku ;)

Oops, the explanation of AIC left out a step.

One more try...

Assume A is false.

A is false implies B must be true.

B is true implies C must be false.

Now you can repeat the process from C to D to E by the same 3 steps that got you from A to C.

You can repeat those 3 steps as many times as you need (0 or more repetitions) to get to a chain where the 2 ends both see a cell or cells that you can eliminate something from.

ok, I copy/pasted this. I can see the logic, which is a start. Thank you very much, drwho!

Just take the 3 step chain. Assuming A is false implies C must be true. Now it turns out that you can reverse the logic and show that if C is false A must be true.

is this the same as single value chaining?

So either the proposition A or C is true, maybe both.

that is the kind of reasoning I like. Not just using a technique, but knowing what you are doing and why.

Skizzy - it is a generalization of it.

It is also a generalization of XY chains and most, if not all the other known chain techniques.

i'm still stuck on last week's expert -- i seem to be misapplying a rule that i thought i understood

I really like your explanation, drwho. Will work on that and try this ugly expert again :)

To really understand the technique you need to know the difference between strong and weak inference.

was just reading about that... so far it is not clear to me

The alternate inference chain alternates strong and weak inferences in a chain.

The first and last inference links in the chain must be strong.

Basically, you can think of a strong inference as meaning that if A is false it implies B must be true.

A weak inference may be thought of as meaning that if A is true it implies that B must be false.

you just lost me there, I am sorry. Why is the one stronger than the other?

If in a strong inference, A is false implies B is true, you cannot say if B is true A must be false. Rather if B is false implies A is true.

ah, ok! Now I am starting to get it. A strong reference tells you more than a weak one.

If in a weak inference, A is true implies B is false, you cannot say A is false implies B is true, rather B is true implies A is false.

Nor can you say in the previous example of weak inference that if B is false A must be true.

In strong inference, both propositions could be true.

In weak inference both propostions could be false.

right! Thanks for your patience. Not there yet, but it is starting to dawn on me.

One confusing point: most strong inferences that you will encounter in Sudoku are also weak inferences!

arrrggghhhhh! LOL! ok, back to square one for me... I copy/pasted it all and will take it line for line. Thanks again :)

Okay, if exactly one of 2 propositions must be true and the other false you have a situation that satisfies the requirements for both weak and strong inference.

Weak inference is used when you have a set of propositions where no more than 1 proposition can be true.

Strong inference is used when you have a set of propositions where at least 1 proposition must be true.

key words there are _can_ and _must_ ?

Yes. In weak inference nothing has to be true. In strong inference something must be true.

thanks, that helps.

Or even more strongly, in weak inference everything could be false, in strong inference everything could be true.

and in strong inference everything _could_ be true, but at least something _must_ be true?

Yes.

thank you, for the explanation and your patience. Will go work on this. :)

You're welcome.

Try this. If an empty cell has only 1 and 2 as candidates; assume it cannot be 1.

Strong inference says it must be 2.

Now suppose another empty cell in the same row has 2 and 3 as the only candidates.

The proposition that the first cell is 2 creates a weak inference that the second cell is not 2. And hence from that a strong inference that it is 3.

Now suppose that in the same column as the 2nd empty cell is an empty cell with only 1 and 3 as candidates.

The proposition that the 2nd cell is a 3 creates a weak inference that the 3rd cell is not a 3 which creates a strong inference that it is a 1. This is an XY wing or a short XY chain.

You can start with the 3rd cell and assume that it is not a 1, which will lead to the inference that the 1st cell must be a 1.

Now you know that one of the first and last cells in the chain is a 1. You can eliminate 1 as a candidate from all common buddies of the 2 ends of the chain.

go

woot expert... go

11 solved... going green

go

done

done
good one

done. no greens. one unused checkpoint and a wild guess that worked out.